GU Hao ,DU Zhan-yu ,Eduard MURANl ,Enrico D’ALESSANDRO ,CHEN Cai ,WANG Xiao-yan ,MAO Jiu-de,Klaus WlMMERS,SONG Cheng-yi

1 College of Animal Science &Technology,Yangzhou University,Yangzhou 225009,P.R.China

2 Leibniz Institute for Farm Animal Biology (FBN),Dummerstorf 18196,Germany

3 Department of Veterinary Science,Unit of Animal Production,University of Messina,Messina 98122,Italy

4 Bond Life Science Center,University of Missouri,Columbia,MO,65211,USA

Abstract Retrotransposons,a type of DNA fragment that can mobilize itself on genome,can generate genetic variations and develop for molecular markers based on the insertion polymorphism.Zinc finger proteins (ZNFs) are among the most abundant proteins in eukaryotic animals,and their functions are extraordinarily diverse and particularly important in gene regulation.In the current study,bioinformatic prediction was performed to screen for retrotransposon insertion polymorphisms (RIPs) in six ZNF genes (ZNF2,ZNF3,ZNF7,ZNF8,ZNF10 and ZNF12).Six RIPs in these ZNFs,including one short interspersed nuclear element (SINE) RIP in intron 1 and one long interspersed nuclear element 1(L1) RIP in intron 3 of ZNF2,one SINE RIP in 5′ flanking region and one SINE RIP in intron 2 of ZNF3,one SINE RIP in 3′ UTR of ZNF7 and one L1 RIP in intron 2 of ZNF12,were discovered and their presence was confirmed by PCR.The impact of the SINE RIP in the first intron of ZNF2,which is close to the core promoter of ZNF2,on the gene activity was investigated by dual-luciferase assay in three cell lines.Our results showed that the SINE insertion in the intron 1 of ZNF2 repressed the core promoter activity extremely significantly (P＜0.01) in cervical cancer cells and porcine primary embryonic fibroblasts (HeLa and PEF),thus SINE may act as a repressor.This SINE RIP also significantly (P＜0.05)affected the corrected back fat thickness in Yorkshire pigs.The corrected back fat thickness of individuals with SINE insertion in the first intron of ZNF2 was significantly (P＜0.05) higher than that of individuals without SINE insertion.In summary,our data suggested that RIPs play important roles in the genetic variations of these ZNF genes and SINE RIP in the intron 1 of ZNF2 may provide a useful molecular marker for the screening of fat deposition in the pig breeding.

Keywords: retrotransposon,ZNF2,polymorphism,repressor,fat deposition

1.lntroduction

Transposable elements (TEs) are a type of DNA element that can mobile on the genome (Doolittle and Sapienza 1980;Orgel and Crick 1980),which was firstly discovered in maize in the 1940s (McClintock 1950).Their mobility or transposition can generate genetic variations and alter the phenotypes of organisms (Fedoroffet al.1983).With the advance of whole genome sequencing,it is common to accept that TEs extensively distribute in prokaryotes and eukaryotes and play important roles in the evolution of organisms (Jacqueset al.2013).TEs are generally classified into two groups based on their mobilization intermediates (Finnegan 1989).Class I elements mobilize by a RNA intermediate,also designated as retrotransposons,while Class II elements mobilize by a DNA intermediate associated with a transposase,also known as the DNA transposons.Generally,two retrotransposon groups are distinguished by the absence or presence of long terminal repeats (LTRs).The LTR elements (including endogenous retroviruses,ERVs) range in size from a few hundred base pairs to over 10 kb and are structured similarly to retroviruses.The non-LTR superfamily mainly includes long interspersed nuclear elements (LINE),SINE and SINE-VNTR-Alu (SVA) elements (Lexaet al.2014).Retrotransposons are biologically important components of mammalian genomes and have been a major force in shaping the structure and function of mammalian genomes(Richardsonet al.2015).LINEs account for about 20% of mammalian genomes,while LTRs and SINEs expansions vary significantly in different mammals (LTRs account for about 0.1-10% of mammalian genomes;SINEs account for about 8-20% of mammalian genomes) (Chenet al.2019).The mobility of LINE-1s (also known as L1s) and non-autonomous SINEs such as Alu and SVA continues to generate both intra-individual and inter-individual genetic variation in the human population (Smit 1993;Haveckeret al.2004;Peastonet al.2004;Becket al.2011).

Retrotransposons provide an excellent opportunity to develop the molecular marker system due to their long,defined,conserved sequences and new insertional polymorphisms generated by active members (Kalendaret al.1999).Multiple retrotransposon-based molecular markers,such as retrotransposon-based insertion polymorphisms (RBIPs) (Flavellet al.1998),interretrotransposon amplified polymorphism (IRAP) and retrotransposon-microsatellite amplified polymorphism(REMAP) have been developed and used to determine pedigrees and phylogenies in plants.In addition,SINE(Alu) insertion polymorphisms have been developed and successfully applied for the population genetic analysis in humans (Terreroset al.2009;Cherniet al.2011;Abyzovet al.2013).RIP markers have also been developed for several domesticated species including chicken (Leeet al.2017),sheep (Chessaet al.2009),deer (Ellederet al.2012;Hronet al.2020),pig (Liet al.2015;Liuet al.2018;Zhenget al.2020),horse (Rooneyet al.2018) and dog(Kallaet al.2020).

Zinc fingers are small protein domains that fold around one or more zinc ions (Matthews and Sunde 2002).Functionally,these motifs carry out a wide variety of tasks within cells by providing stable structural scaffolds and driving critical binding interactions,especially among proteins,DNA and RNA.The motifs are particularly important in the gene regulation,where many proteins employ them to bind DNA in a sequence-specific manner,to activate or inhibit specific genes.Zinc fingers are extremely versatile,as can be seen by their abundance in nature: over 700 proteins contain Cys2-His2 zinc fingers in the human genome alone and they are thought to be the largest protein superfamily in metazoans (Laityet al.2001;Iserniaet al.2020).

It is commonly accepted that retrotransposon insertions can mediate structural variations (SVs) in genome,and they may exert impact on gene activity.It was hypothesized in the present study that there would be structural variations derived from retrotransposon insertions inZNFprotein family genes,which might affectZNF’s promoter activity as well as Yorkshire’s age at 100 kg body weight and carcass traits.To verify this,in the current study,the contribution of RIPs to the large structural variations of sixZNFgenes (ZNF2,ZNF3,ZNF7,ZNF8,ZNF10andZNF12) was evaluated,and six RIPs in sixZNFgenes were confirmed by PCR.The study focused on the function of a SINE RIP in the first intron ofZNF2gene,its breed distribution and its effect on age at 100 kg body weight and carcass traits of Yorkshire pigs.

2.Materials and methods

2.1.Multi-alignment of ZNF gene sequences and structural variation prediction

The genic sequences (including 5 kb of the 5′ flanking regions and 3 kb of the 3′ flanking regions) ofZNF2,ZNF3,ZNF7,ZNF8,ZNF10andZNF12from the assembled 16 pig breed genomes,including one reference genome (Duroc) and 15 assembled nonreference genomes (Cross-bred of Yorkshire_Landrace_Duroc,Landrace,Yorkshire,Pietrain,Berkshire,Hampshire,Rongchang,Meishan,Bamei,Jinhua,Bama,Wuzhishan,Tibetan,Goettingen and Ellegaard Gottingen minipig),were downloaded from NCBI Whole Genome Shotgun (WGS) Database,or re-assembled based on the contigs.Then the structural variations were manually checked based on the multi-sequence alignment of these genes by ClustalX Program (in order to facilitate multi-sequence alignment,N in the spliced sequence is removed).Large structural variations (＞50 bp)presented in one or more genome were used for analysis.The custom repeat library (Chenet al.2019)was used to annotate the retrotransposon (SINE,LINE and ERV) insertions in theseZNFgenic sequences by using RepeatMasker Program (version 4.0.9),and only the large structural variations (＞50 bp) overlapping with retrotransposons at least 50 bp were designated as RIPs.

2.2.Sample preparation,primer design and PCR verification of RlPs

The predicted RIPs based on the sequence alignment of theseZNFgenes across 16 pig breed genomes were submitted for PCR detection by using 11 domestic pig breeds (Landrace,Yorkshire,Duroc,Sujiang,Erhualian,Meishan,Fengjing,Tibetan,Bama,Wuzhishan and Diannan small-ear) and wild boar DNA samples.The source information of the above pig tissue samples was shown in Appendix A.In order to reduce the workload of PCR and improve the efficiency of polymorphism detection,pool DNA is used as the template for PCR amplification.Two DNA pools were prepared for each breed,whereby each pool included three individual DNA sample (50 ng μL-1) mixtures.

The forward and reverse primers were designed in upstream and downstream sequences flanking the retrotransposon insertions as shown in Fig.1-A,and the expected electrophoresis fragment length polymorphism is presented in Fig.1-B.Primers were designed by Oligo7 Software and synthesized by Nanjing Qingke Biotechnology Co.,Ltd.(Nanjing,China).All PCR primers were listed in Appendix B.

Fig.1 Schematic diagram of primer design and predicted electrophoresis results.A,PCR primers designed for detection of polymorphic insertion.B,expected electrophoretic result for different genotypes.+/+,homozygous with retrotransposon insertion;-/-,homozygous without retrotransposon insertion;+/-,heterozygote.

The pooled DNA samples were used as templates for PCR amplification in a 20-μL PCR reaction system.The specific PCR reaction system components are as follows: 1 μL of DNA (50 ng μL-1),each 1 μL of upstream and downstream primers (10 μmol μL-1),10 μL of 2×TaqMix,and 7 μL deionized water.PCR reaction procedure is as follows: 95°C pre-denaturation for 3 min,95°C denaturation for 30 s,60°C annealing for 30 s,72°C extension for 30-100 s,35 cycles from denaturation to extension,72°C extension for 10 min,and storage at 4°C.

The PCR products were separated by electrophoresis on 1.2% agarose gel in 1× TAE buffer.The gels were stained with ethidium bromide for 30 min,washed twice with distilled H2O and photographed under UV light by Gel Image System (Tanon 3500).A DL2000 DNA ladder was used as a molecular weight marker.

2.3.Comparative genomics analysis and promoter prediction

TheZNF2genic sequences including the flanking regions of cattle,sheep,horse,dog,human,pig and mouse,and the coordinates were downloaded from the NCBI database,and then submitted to mVISTA (http://genome.lbl.gov/vista/index.shtml) for comparative genomics analysis.The promoter region ofZNF2gene on Duroc was predicted by using the EPD (https://epd.epfl.ch//index.php),BDGP (https://fruitfly.org/seq_tools/promoter.html) and Promoter 2.0 Prediction Server (http://www.cbs.dtu.dk/services/Promoter/).The ChIP-seq data (Zhaoet al.2021) were also downloaded and used to combine with the promoter prediction.

2.4.Vector construction and dual-luciferase reporter assay

Three conserved genomic fragments (Frag.1/713 bp,Frag.2/1665 bp and Frag.3-SINE-/989 bp) and Frag.3-SINE+/1303 bp closest to the predicted promoter ofZNF2were amplified by PCR from the Duroc DNA samples and further verified by sequencing.Then they were cloned (primers were listed in Appendix C) into the pGL3-enhancer vector (catalog,E1771,Promega) to verify whether these fragments had promoter activity.In addition,Frag.2/1665 bp,Frag.2-Frag.3-SINE-/2630 bp and Frag.2-Frag.3-SINE+/2944 bp were cloned (primers were listed in Appendix C) into the pGL3-basic vector(catalog,E1751,Promega) to explore the function(enhancer or repressor) of SINE/314 bp in Frag.3.All the recombinant plasmids were transfected into HeLa,PK15 (porcine kidney cells) and PEF cells.The specific operations were as follows: a total of 3×105cells each for PK15,PEF and HeLa cell lines were plated in 6-well plates and transfected with the plasmids (recombinant plasmids with pGL3-enhancer and pGL3-basic as framework,plasmid amount was 500 ng) by using FuGENE?HD transfection reagent (catalog,E2311,Promega).After 48 h,cells in each well of the six-well plates were collected for luciferase activity evaluation by using the Dual-Luciferase?Reporter Assay System Kit (catalog,E1910,Promega) according to the manufacturer’s protocol.Three wells for each vector were measured for each experiment(repeated observation number=3).Finally,the dualluciferase (renilla,firefly) signal was collected through the multi-mode microplate detection system (EnSpire,Perkin Elmer,USA),then relatively luciferase activity (firefly luciferase/renilla luciferase) was calculated,and the signal values of pGL3-enhancer and pGL3-basic were used as negative controls.

2.5.Population genetic analysis

Three commercial breeds (Yorkshire,Landrace and Duroc),one hybrid (Sujiang) and five Chinese native breeds (including Mi pig,Erhualian,Meishan and Fengjing in Jiangsu Province and Laiwu black pig in Shandong Province) and one Italian native breed(Sicilian black pig in Sicilia) were used to detect breed distribution ofZNF2-SINE-RIP (24 samples per pig breed except Yorkshire).The source information of the samples was shown in Appendix A.Hardy-Weinberg equilibrium were tested by Chi-square test as follows:The formula is the sum of all genotypes at a given locus,O(observed) represents the observed number of each genotype,E(expected) represents the expected number of each genotype under the assumption that Hardy-Weinberg equilibrium law holds.Polymorphic information content (PIC) is an indicator of the degree of microsatellite DNA variation.The formula of PIC is as follows:In the formula,PiandPjwere theith andjth allele frequencies in the population,respectively;nis the number of alleles.

2.6.Trait determination

The traits including age at 85-130 kg body weight,back fat thickness and loin eye muscle area were recorded.The back fat thickness and loin eye muscle area were measured at the 3rd-4th last ribs,5.0 cm off the midline by using LX8000 B-Mode Ultrasonic Equipment (Beijing Kangchengda Technology Co.,Ltd.,Beijing,China).Both of back fat thickness and age were adjusted to 100 kg using a breed-specific correction factor based on a formula according to the National Swine Genetic Evaluation Center (NSGEC) guidelines and reference(Binet al.2016).The correction formula is as follows:Age at 100 kg body weight=Measured age-[(Measured body weight-100)/CF],where CF is the correction factor,CFboar=(Measured body weight/Measured age)×1.826040,CFsow=(Measured body weight/Measured age)×1.714615.Corrected back fat thickness=Measured back fat thickness×CF,where CF=A/[A+B×(Measured body weight-100)],Aboar=12.402,Bboar=0.106530;Asow=13.706,Bsow=0.119624.

2.7.Statistical analysis

The experimental data were processed by SPSS18.0 Software Package (SPSS,Chicago,USA).The data from cell experiments were analyzed by one-way ANOVA,then the least significant difference (LSD) method was used to test differences based on means,and then Graphpad Software was used to draw a histogram.The experimental unit was each well in cell experiments.The data from animal experiments were analyzed by independent sampleT-test (Duet al.2020).The experimental unit was each pig in animal experiments.TheP＜0.05 was considered to be statistically significant andP＜0.01 was considered to be extremely significant.

3.Results

3.1.Six RlPs generated by retrotransposon insertions in six pig ZNF genes

Select the C2H2 zinc finger protein gene on pigs for research.Starting with the genomic alignment of sequences of sixZNFgenes and thier flanking regions obtained from 16 breeds,by analysis with the RepeatMasker,47 large SVs were identified,of which 24 SVs were highly overlapped with SINEs and LINEs.The RepeatMasker results of these 24 predicted RIPs were shown in Appendix D.Data on these SVs were summarized in Table 1.Details on genomic coordinates ofZNFgenes on Duroc genome,IDs of contigs for assembling theZNFgenes including the flanking regions and the length of the assembledZNFgenes including flanking sequences were shown in Appendices E,F and G,respectively.

Then,all large SVs generated by RIPs were investigated by PCR with genomic DNA samples from 11 domesticated pig breeds and wild boars.Finally,six RIPs including two RIPs inZNF2,two RIPs inZNF3,one RIP inZNF7and one RIP inZNF12were obtained(Fig.2-A;Table 2).All these RIPs were further confirmed by cloning and sequencing.Annotation by RepeatMasker and sequencing results revealed that RIPs inZNF2were generated by one 314-bp SINE insertion in intron 1 and one 127-bp L1 insertion in intron 3.ForZNF3,both RIPs were generated by SINE insertions: one was a 418-bp SINE insertion in the 5′ flanking region and the other was a 364-bp SINE insertion in the intron 2.The RIP inZNF7was generated by a 318-bp SINE insertion in 3′ UTR,and the RIP inZNF12was an insertion of L1 (1 148 bp) in intron 2 (Fig.2-B;Table 2).

Fig.2 The agarose electrophoresis result for the verified retrotransposon insertion polymorphisms (RIPs) and their positions in ZNF genes.A,six predicted RIPs detected with pool DNA samples by PCR.M,DNA marker DL2000.B,the position of six confirmed RIPs in ZNF genes.The arrow represents the confirmed RIPs,and the direction of the arrow indicates the retrotransposon insertion direction relative to the gene.Length refers to the total length of the gene including its 5′ and 3′ flanks.

Table 1 Predicted structural variations and retrotransposon insertion polymorphisms (RIPs) in ZNF genes and their flanking regions

3.2.SlNE insertion may act as a repressor to regulate activity of ZNF2 gene

The 314-bp SINE insertion in the first intron ofZNF2is close to the first exon (based on the longest transcript of pigZNF2gene),with an 1 484-bp distance to the transcription start site (TSS).But it is a bit far away from the second exon,with an 8 719-bp distance to the translation initiation codon (ATG) (Fig.3-A).The prediction results ofZNF2’s promoters were shown in Appendix H.The insertion was close to the putative core promoter region ofZNF2,which was predicted by promoter 2.0 (score＞1),BDGP (score＞0.8) and EPD and highly conserved across multiple mammal species including pig,cattle,dog,horse,human,and sheep(Fig.3-B),indicating that this insertion might influence the core promoter activity ofZNF2gene.

Fig.3 Structure of ZNF2 and its sequence conservation analysis across seven species (pig ZNF2 gene as reference).A,distribution of retrotransposon insertion polymorphisms (RIPs) in ZNF2 gene and its flanking regions (5-kb 5′ flank and 3-kb 3′ flank).Yellow is the flanking region;light blue is untranslated region (UTR);dark blue is coding sequence (CDS);gray is intron;blue arrows represent RIPs (SINEA1 RIP is located in intron 1 of ZNF2;L1D8 RIP is located in intron 3 of ZNF2);and stars indicate the promoter signals of ZNF2 gene on Duroc predicted by promoter 2.0,EPD and BDGP online programs.B,the results of the sequence conservation analysis.Frag.1,Frag.2,Frag.3-SINE- and Frag.3-SINE+ were cloned genomic fragments for dual-luciferase reporter assay.Red is conserved genome sequences in multiple species;light blue is UTR;and dark blue is CDS.

To further evaluate the impact of SINE insertion on the promoter activity ofZNF2,four genomic fragments (Frag.1,Frag.2,Frag.3-SINE-and Frag.3-SINE+) in the conserved region ofZNF2and their combination were amplified by PCR and cloned into the pGL3-basic and pGL3-enhancer vectors,respectively (Figs.3 and 4-A and B),then transfected into cells (HeLa,PK15 and PEF) and submitted for luciferase assay evaluation (Fig.4-C and D).The pGL3-enhancer vector contains SV40 enhancer in the downstream of expression box,while the pGL3-basic backbone is absence of this functional element (Fig.4-A and B).

The luciferase assay based on the pGL3-enhancer backbone vectors revealed that Frag.2 had promoter activity in all cell lines (HeLa,PK15 and PEF),while Frag.1 displayed weak promoter activity in HeLa and PEF cells,while the promoter activity both of wild type Frag.3 (Frag.3-SINEA1-) and mutant Frag.3 (Frag.3-SINEA1+) was not detectable in HeLa,PK15 and PEF cells (Fig.4-C).These data suggested that Frag.2 is the core promoter region ofZNF2,which was well supported by ChIP-seq data (Appendices I and J) (Zhaoet al.2021).

To further evaluate the effect of SINE1 insertion on the promoter activity ofZNF2,the Frag.2,the region covered the Frag.2 and Frag.3 (Frag.2-Frag.3-SINE-),and the mutant containing the insertion allele (Frag.2-Frag.3-SINE+) were cloned and inserted into the pGL3-basic backbone vectors and submitted for luciferase assay evaluation.We found that the SINE insertion repressed the promoter activity extremely significantly(P＜0.01) in HeLa and PEF cell lines,and Frag.3-SINEalso repressed the core promoter (Frag.2) activity ofZNF2(Fig.4-D),indicating that SINE insertion may act as a repressor ofZNF2’s promoter (the difference between Frag.2-Frag.3-SINE-and Frag.2-Frag.3-SINE+is only 314 bp (SINE),which has a negligible effect on the transfection efficiency).

Fig.4 Verification of the ZNF2 promoter activity and evaluation of the function of the SINE insertion by dual-luciferase reporter assay.A,vectors to verify the activity of ZNF2’s promoters.Luc+ represents firefly luciferase coding gene;En represents SV40 enhancer;RLuc+ represents renilla luciferase coding gene;TK-pro represents HSV-TK promoter;and pRL-TK is the internal reference plasmid.B,vectors for verifying the function of SINE insertion.pGL3-Frag.2-Luc+ was used as the positive control.C,the activity of ZNF2’s promoters verified by dual-luciferase reporter assay.pGL3-enhancer was used as the negative control.D,the impact of SINE insertion on the activity of ZNF2’s promoters.pGL3-basic was used as the negative control.Three wells for each vector were measured for each experiment.Bars represent mean±SD (n=3).Different letters indicate significant differences (P＜0.01).

3.3.RlP distribution in different pig breeds and effect on age at 100 kg body weight and carcass traits

The distribution of the polymorphism generated by the 314-bp SINE insertion in the first intron ofZNF2(ZNF2-SINE-RIP) was evaluated in 10 pig breeds,including three commercial breeds (Yorkshire,Landrace and Duroc),one hybrid (Sujiang),five Chinese native breeds (Erhualian,Meishan,Fengjing and Mi pig in Jiangsu Province;Laiwu black pig in Shandong Province) and one Italian native breed (Sicilian black pig in Sicilia).The number of pigs used for each breed,breed origins,genotype frequencies and allele frequencies were listed in Table 3.The results revealed that theZNF2-SINE-RIP displayed polymorphism in five breeds,including three commercial breeds(Yorkshire,Landrace and Duroc),one hybrid (Sujiang)and one Italian local pig breed (Sicilian black pig),with low or medium frequencies of SINE+allele in most breeds except in Duroc,where this breed displayed a very high frequency of SINE+allele (＞89%;Table 3),indicating that it may originate from this breed (Duroc).In contrast,the polymorphism was not observed for five Chinese native breeds (Erhualian,Meishan,Mi pigs,Laiwu black pig and Fengjing).Hardy-Weinberg test revealed that this SINE RIP in all detected populations including Chinese native breeds (Erhualian,Meishan,Fengjing,Mi and Laiwu black pig),the imported pig breeds (Yorkshire,Landrace and Duroc),the hybrid (Sujiang) and Italian local pig breed(Sicilian black pig) were in line with Hardy-Weinberg equilibrium (P＞0.05;Table 3).PIC was moderately polymorphic (0.25＜PIC＜0.5) in the hybrid pig (Sujiang)and commercial pig (Landrace),while PIC was low polymorphic (PIC＜0.25) in other breeds of pigs (Table 3).

Table 2 The detailed information for six retrotransposon insertion polymorphisms (RIPs) in ZNF genes

The effects ofZNF2-SINE-RIP on age at 100 kg body weight,corrected back fat thickness and loin eye muscle area were analyzed in Yorkshire pigs.We found thatZNF2-SINE-RIP significantly (P＜0.05) affected the corrected back fat thickness (Table 4).The corrected back fat thickness of heterozygous individuals with SINEA1+/-genotype was significantly (P＜0.05) higher than that of homozygous individuals (SINEA1-/-).The loin eye muscle area of pigs with SINEA1+/-genotype was lower than that of pigs with SINEA1-/-genotype,but the difference was not significant (P＞0.05),which indicated thatZNF2-SINE-RIP might influence the deposition of fat and lean in pigs.

Table 3 Analysis of genetic diversity of ZNF2-SINE-RIP in different breeds

Table 4 Effect of ZNF2-SINE-RIP on age at 100 kg body weight and carcass traits

4.Discussion

The insertion of retrotransposons might affect the promoter activity and expression of genes (Chenet al.2019).Thus,it would be very meaningful to study the effect of RIP onZNFprotein family genes.The results from the present study partially supported our hypothesis that there would be structural variations derived from retrotransposon insertions inZNFprotein family genes,which might affectZNF’s promoter activity as well as Yorkshire’s age at 100 kg body weight and carcass traits.We found six RIPs inZNFgenes.Among them,ZNF2-SINE-RIP could inhibit the promoter activity and increase the corrected back fat thickness of pigs.These results supported the hypothesis of the source of structural variations and the influence of RIP on promoter activity and corrected back fat thickness,provided a theoretical basis for studying the effect of RIP on gene expression and also suggested thatZNF2might be involved in fat regulation.However,partial results of animal experiments didn’t support the hypothesis of the influence of RIP on the age at 100 kg body weight and loin eye muscle area because there wasn’t significant difference.It was suggested thatZNF2might not be involved in regulating the growth rate of the body weight and loin eye muscle of pigs.

Retrotransposition not only led to structural variations in mammalian genomes,but also presented a new opportunity for the evolution of a new gene or the modificatiovn of an existing function at the site of insertion(Ewinget al.2013).It has been found that the RIPs had extensive impact on the gene activity and phenotypes in plants and animals.In the flower of Phalaenopsis Orchids (Hsuet al.2019),a retrotransposon inserted in the promoter of genePeMYB11,which caused Harlequin/Black and appeared as the new color in Phalaenopsis Orchids.Meanwhile,it has been found that promoter derived from SINE insertion could drive biallelic expression (Smithet al.2017).In the animal genome,the SINE insertion in follicle stimulating hormone beta (FSHβ)and protein disulfide isomerase associated 4 (PDIA4) genes affected the litter size of pigs (Magotraet al.2015;Liuet al.2018).These studies indicated that retrotransposons had a profound effect on genes,which urged us to study the significance of retrotransposons insertion intoZNFgenes.

The zinc finger is found widespread in nature,including 3% of the genes of the human genome.Several hundreds ofZNFgenes had been identified in mammals(Klug 2010),and the exact function of mostZNFgenes was not known.However,zinc-finger proteins were known to interact with DNA and RNA and functioned as transcription regulators (Theunissenet al.1992;Iserniaet al.2020).ZNF2is a protein coding gene,which belongs to C2H2-type zinc-finger protein family.ZNF2also had transcriptional repression activity mediated with the Krüppel-associated box-A (KRAB-A) domain,which was highly conserved in mostZNFproteins in mammals (Witzgallet al.1994;Eccoet al.2017).So it was important to studyZNFgenes.Thus,we investigated the structural variation ofZNFgenes caused by the retrotransposon insertion.Based on the multiple alignments ofZNFgenic sequences and PCR analysis,24 RIPs were predicted,and six RIPs were confirmed by PCR inZNFgenes.These RIPs might exert an effect on the transcriptional activities of genes.

On the basis of the above results,we focused on theZNF2-SINE-RIP.We constructed vectors and transfected into HeLa,PK15 and PEF cellsin vitroto identify the core promoter region ofZNF2and the effect of SINE insertion on the activity of the core promoter.The results showed that Frag.2 was the core promoter and the insertion of SINE in its downstream inhibited the core promoter activity.At the same time,studies had found that SINE B1 elements could influence the activity of downstream gene promoters,with acquisition of DNA methylation and loss of activating histone marks,thus resulting in a repressed state (Estécioet al.2012).Thus,the SINE inserted near the core promoter might obtain DNA methylation,thereby inhibiting the activity of the promoter.

In order to explore the distribution and origin ofZNF2-SINE-RIP in different pig breeds,we conducted PCR testing on the samples.The population genetic analysis revealed theZNF2-SINE-RIP in all breeds were in Hardy-Weinberg equilibrium.The high frequency of SINE insertion allele (SINE+) was detected in over 89% in the lean pig breed Duroc,and low frequency of SINE+allele in Yorkshire(3.59%),Landrace (33.33%) and Sicilian black pigs(14.58%),and medium frequency of SINE+allele in Sujiang(50%) were observed.However,the SINE+allele was totally absent in Chinese native pig breeds.These data indicated that the SINE insertion allele (SINE+) might originate from the lean type pig (Duroc).Sujiang was a hybrid with 62.5%Duroc bloods and 37.5% local pig bloods,thus the medium frequency of SINE+allele was observed.

On the other hand,PCR detection was carried out in the genomes of Yorkshire pigs and combined with the data of age at 100 kg body weight and carcass traits to identify its influence.We found thatZNF2-SINE-RIP significantly affected the corrected back fat thickness.It had been suggested thatZNF423,another zinc-finger protein,might play role in the intramuscular fat deposition inlongissimus dorsimuscle of Mashen pigs,which was a Chinese native breed (Zhaoet al.2017).Taken together,these data suggested that someZNFgenes,includingZNF2andZNF423,as important transcription regulators,might be involved in the regulatory pathways of fat and muscle metabolism,and they could be candidate genes for fat deposition.TheZNF2-SINE-RIP,as genetic marker,might also have application potential for pig molecular breeding.

5.Conclusion

In summary,six RIPs,located in the different genic regions of fourZNFgenes (ZNF2,ZNF3,ZNF7,andZNF12),were identified in the present study.One 314-bp SINE insertion in the intron 1 ofZNF2mainly distributed in three introduced commercial pig breeds (Duroc,Yorkshire and Landrace),one Europe native pig breed (Sicilian black pigs) and one hybrid (Sujiang) containing Duroc blood.TheZNF2-SINERIP significantly (P＜0.05) affected the corrected back fat thickness in Yorkshire pigs and could repress the promoter activity ofZNF2extremely significantly (P＜0.01).Our data suggested that the 314-bp SINE insertion might act as a repressor to be involved in the regulation ofZNF2gene activity and fat deposition in pigs.

Acknowledgements

This research was supported by the National Natural Science Foundation of China (32002146 and 31872977),the China Postdoctoral Science Foundation(2020M671630),the Jiangsu Postdoctoral Science Foundation of China (2021K221B) to Chen Cai,the Jiangsu Agriculture Science and Technology Innovation Fund,China [CX (19) 2016],and the Priority Academic Program Development of Jiangsu Higher Education Institutions and the High-end Talent Support Program of Yangzhou University,China to Song Chengyi.

Declaration of competing interest

The authors declare that they have no conflict of interest.

Ethical approval

All treatments and protocols involving animals in this study were strictly done in accordance with the guidelines of the Animal Experiment Ethics Committee of Yangzhou University,China (approval number: YZUDWSY2018-12).

Appendicesassociated with this paper are available on http://www.ChinaAgriSci.com/V2/En/appendix.htm