RsaI repetitive DNA in Buffalo Bubalus bubalis representing retrotransposons, conserved in bovids, are part of the functional genes

Background Repetitive sequences are the major components of the eukaryotic genomes. Association of these repeats with transcribing sequences and their regulation in buffalo Bubalus bubalis has remained largely unresolved. Results We cloned and sequenced RsaI repeat fragments pDp1, pDp2, pDp3, pDp4 of 1331, 651, 603 and 339 base pairs, respectively from the buffalo, Bubalus bubalis. Upon characterization, these fragments were found to represent retrotransposons and part of some functional genes. The resultant clones showed cross hybridization only with buffalo, cattle, goat and sheep genomic DNA. Real Time PCR, detected ~2 × 104 copies of pDp1, ~ 3000 copies of pDp2 and pDp3 and ~ 1000 of pDp4 in buffalo, cattle, goat and sheep genomes, respectively. RsaI repeats are transcriptionally active in somatic tissues and spermatozoa. Accordingly, pDp1 showed maximum expression in lung, pDp2 and pDp3 both in Kidney, and pDp4 in ovary. Fluorescence in situ hybridization showed repeats to be distributed all across the chromosomes. Conclusions The data suggest that RsaI repeats have been incorporated into the exonic regions of various transcribing genes, possibly contributing towards the architecture and evolution of the buffalo and related genomes. Prospects of our present work in the context of comparative and functional genomics are highlighted.


Background
Different families of repetitive DNA contribute towards architectural organization of the mammalian genomes [1]. They represent both, tandemly arranged and interspersed sequences [2]. Based on their size and mode of propagation, Interspersed elements can be divided into two separate classes, the long terminal repeat (LTR) and non-LTR. The non LTR LINEs (long interspersed repeat elements) and SINEs (Short interspersed repeat elements) are widely distributed occupying a substantial fraction of the eukaryotic genomes. These elements replicate and proliferate themselves through a "copy and paste" mechanism called retrotransposition [3,4]. In this process, transcription of their genomic copies is followed by an RNA intermediate resulting cDNAs reintegration at a new location in the genome [5]. Approximately, 100 LINE and SINE families have been reported to date in various eukaryotic genomes [6]. In mammals, LINE, L1 repeats are dominant retrotransposons type both in the common ancestor and in extant species [7]. In addition to L1, an element belonging to the retrotransposable element family of autonomous retrotransposons (RTE-1) has been reported in mammals [8]. Few mammals have active non LTR LINE other than L1 that contribute significantly to repeat composition.
Species specific retrotransposons have been widely used as a tool for Phylogenetic analysis and population studies [9,10]. Many retrotransposons are inactive, found in the non-coding regions of the genome and are subjected only to the neutral evolution. Thus, rare new insertions have led to some form of advantageous or noteworthy phenotypic variations [11][12][13]. These and other such discoveries have resulted in a shift from earlier thought of them being "parasite" to functional elements cultivated in the genome for their beneficial attributes.
Retrotransposons have gained novel functions, providing alternative splice sites and/or polyadenylation signals or modifying gene expression [14][15][16]. These elements account for 46.5% of bovine genome [17]. The bovine genome is very different in its repeat composition compared to other mammalian genomes. It has unusual composition of LINE RTE type Bov B and its associated SINE elements which together account for 25% of the bovine genome [17]. The impact of the interspersed repeats on the genomes of human [18][19][20], dog [21,22], cow [17,23], mouse [24] and opossum [25,26] has been studied. However fate of these interspersed elements and their association with mRNA transcriptomes in buffalo remains still unclear. Here, we report RsaI family repetitive DNA in the genome of water buffalo "Bubalus bubalis" and their copy number status. We also studied their expression in somatic tissues and spermatozoa. The repeat fraction pDp1, pDp2 and pDp3 were used for fluorescence in situ hybridization (FISH) with buffalo metaphase chromosomes. In addition, we isolated and sequenced ACOT11 (Acyl-coenzyme A thioesterase 11) gene harboring part of pDp1 repeat.

Results
RsaI enzyme digestion uncovers four repeat fractions Digestion of buffalo genomic DNA with RsaI enzyme, besides minor ones, showed four prominent bands ranging from 1331 base pairs, pDp1; 651, pDp2; 603, pDp3; to 339; pDp4 ( Figure 1). Approximately, 15-20 recombinant clones subjected to restriction digestion and slot blot hybridization screening yielded ten positive clones for each repeat. Five-six clones from each fragment were then sequenced. The accession numbers of the recombinant clones are given in the Additional File 1. All the four major repeat elements were AT rich but sequence-wise, were different from one another (Additional File 2). No inter-clonal variations were detected in these sequences. However, random repeats were present in the nucleotide sequences (  Differential expression of RsaI fragments in somatic tissue and spermatozoa of buffalo RT-PCR analysis using internal primers of pDp1, pDp2, pDp3 and pDp4 (see Table 3) and cDNA from somatic tissues and spermatozoa of buffalo showed amplification of band across tissues and sperm confirming their   transcriptional potentials (Additional File 8). Quantitative Real Time PCR analysis of these sequences showed differential expression of RsaI related transcripts across somatic tissues and spermatozoa ( Figure 4). pDp1 showed highest expression in lung, pDp2 and pDp3 in Kidney and pDp4 in ovary. Summary of the relative expression (in folds) derived from 2 -ΔΔCt values obtained for various transcripts based on Real Time PCR are given in Table 4.

Multiple copies of RsaI fragments in bovids
The pair of internal primers deduced from pDp1-4 used for expression study was also employed for copy number analysis of these fragments in cattle buffalo, goat and sheep using Real Time PCR. Standard curve slope value was between 3.2-3.5. Single melting peak on dissociate curve confirmed primer specificity. pDp1 showed 2 × 10 4 copies, pDp2 and pDp3,~3000 copies each and pDp4 showed~1000, in these species (Table 5 and Additional File 9). pDp1, pDp2 and pDp3 sequences are dispersed throughout the buffalo genome In silico analysis of pDp1, pDp2, pDp3 and pDp4 sequences using reference cattle genome revealed its multiple locations on the cow chromosomes ( Figure 5). Probing of pDp1 to buffalo genomic DNA digested with RsaI enzyme detected a strong hybridization signal (Additional File 10) giving rise to a single isomorphic band in all the samples. FISH mapping of pDp1, 2, 3 spectrum red labeled cloned probes showed ubiquitous discernible signals over buffalo metaphase chromosomes (Figures 6 and 7). All the chromosomes showed dispersed pattern with all the three probes used independently. In several chromosomes, signals in the centromeric regions were absent or reduced giving rise to inconsistent pattern. FISH with pDp1 sequence showed more localized signals on the metaphase chromosomes as compared to that detected by pDp2 and pDp3. FISH with spectrum red labeled clone pDp4 probe showed background signals even after washing the slides under high stringent conditions (60°C in 0.1× SSC). This might be due to the short length (339 bp) of the probe pDp4 used coupled with its dispersed genomic organization.
Full length cDNA sequence of ACOT11 gene in Buffalo Blast search using reference mRNA sequence revealed RsaI repeats to be part of Acyl-coenzyme A thioesterase 11(ACOT11), Vacuolar Protein Sorting 24 (VPS24) and Solute carrier organic anion transporter family member 1A2 (SLCO1A2) genes (Additional File 1) Full length Buffalo ACOT11 cDNA was generated using end point PCR and gene specific (B. taurus ACOT11) primers (Table 3). Assembled cDNA sequence of 2592 base pair fragment lacking poly A tail representing six exons (Additional Files 11 and 12) were deposited in the Gen-Bank (HQ848649 and HQ848650).

Discussion
We have studied four (pDp1, pDp2, pDp3 and pDp4) RsaI fragments from the buffalo genome which are AT rich, though buffalo genome on the whole is GC rich (40.69%, NC_006295). Database searches with the repeat-maskers revealed the presence of several LTR, LINE and SINE element in the four sequences (Table  2). Apparently, SINEs occupy the (G+C)-rich regions while LINEs are mainly located on the (A+T)-rich regions. Reports suggest that very large number of highly truncated insertions of L1 have occurred in the bovine genome [8]. Full length copies of the human L1 contain two open reading frames, ORF1 and 2. ORF1 encodes a DNA binding protein and ORF2 includes endonuclease and reverse transcriptase domains [27]. Presence of partial reverse transcriptase and endonuclease domains in pDp1, pDp2 and pDp4 reported herein led to the hypothesis that RsaI repeats might be related to a novel retrotransposable element. Interspersed repeats get inserted into a new genomic location through the process of retrotransposition [28,29]. This is reflected by our FISH results of pDp1, 2 and 3 showing signals on all over the chromosomes with varying intensity. In silico analysis of pDp3 on the reference cattle genome showed fewer distribution suggesting its poor characterization in the cattle genome. This is supported by the fact that using real time PCR, we detected similar copy number of pDp3 in the bovids. Repetitive sequences in centromeric regions are dynamic components, ever prone to mutation, recombination, deletion, and translocation leading eventually to their alterations [30]. Absence of RsaI sequences in the centromere of several chromosomes may be undergoing  Cb-calibrator such events. All the four sequences showed no homology with each other as revealed by clustalW alignment (Additional File 2). Startlingly, these four repeat elements were not detected in any of the non-bovid species. This suggests that irrespective of their origin and biological significance, their evolutions have been confined to limited number of species. Most likely, in nonbovid species, they were not favored evolutionarily and therefore purged slowly and gradually in due course of time. We presume that amplifications of repeat elements could have originated from its discrete blocks. Reintegration of extra chromosomal copies of these repeat elements [31] could have allowed its further dissemination in the genome. However, it is not clear whether all the repeats have similar significance in the buffalo genome. It is likely that these repeats are collectively involved in the evolution and sustenance of bovid chromosomes. Reports suggest that new retrotransposons are conserved within the same group of species [32]. This is corroborated by our Slot blot and PCR results. Real time PCR results showed approximately similar copy numbers for pDp1, 2, 3 and 4 in cattle, goat and sheep genomes as mentioned earlier. The copy number assessment of these repeats in different known and nondescript breeds of buffalo may enable to establish a correlation, if any, towards the delineation of different breeds.
Retrotransposons copies are reportedly involved in the regulation of transcription [33][34][35][36]. Presence of RsaI repeats in exonic database of various transcribing genes suggests that these sequences function as parts of mRNA. Since repeat sequences were also present in the introns, it is possible that they are transcribed as pre-mRNA and contribute to the processing of mRNA and splicing. Thus, differential expression of transcripts in somatic tissue and spermatozoa might be under the influence of post transcriptional regulation required for various cellular processes. Presence of transcribing retroelements within the buffalo spermatozoa reported here seems to be first such observation. Studies have shown existence of an RT-dependent mechanism operating in the spermatozoa, responsible for the genesis of new biologically active retrogenes [37]. These retrogenes may be delivered to the embryos during fertilization and propagated subsequently in the tissues of adult individuals [37].
RsaI elements were found to be part of the three functional genes (ACOT11, VPS24 and SLCO1A2) mostly present in 3' UTR. Our work corroborates recent reports that most part of retrotransposons inserts themselves in first and last exons and in untranslated regions (UTRs) [38]. In human, this type of insertion has been shown to create new non-conserved polyadenylation signals [39], influencing the level of gene expression [40]. However, how these insertions affect expression of buffalo transcriptomes is still a matter of speculation.

Conclusions
Buffalo has several known and non-descript breeds of which a few are considered to be superior with respect to productivity and economic return. Whether, RsaI repeat in different breeds of buffalo would show similar Average ± standard deviation organization and expression pattern is not known. However, if informative in breed delineation, these would prove to be useful biomarkers.

Species sample and DNA extraction
Approximately, 10 ml blood samples from both the sexes of buffalo, goat and sheep were collected from local slaughterhouse, Delhi following the guidelines of institute's Ethical and Biosafety Committee. Cattle blood sample was procured from the owner of the animal. Blood samples of human, fish, bird, rat, jungle cat, bonnet monkey, were available from other projects in the lab. Tiger, Indian rhinoceros and leopard samples were obtained with due permission from the competent authorities of the state and union government of India following strictly the guidelines of the Institute's Ethical and Biosafety Committee. Genomic DNA was extracted according to standard phenol-chloroform procedure [41].

Restriction digestion of buffalo genomic DNA, cloning and sequencing
Approximately, 5 μg of buffalo genomic DNA from both the sexes was digested with RsaI restriction enzyme following supplier's (NEB) specification. Fragments were separated on 1% agarose gel in 1× TBE. Distinct bands within the smear were sliced from the gel, purified and cloned into dephosphorylated pBluescript II SK+ vector (Stratagene, USA), using standard protocol [42]. Approximately, ten clones, representing each fragment, were screened with restriction enzyme Xho1/Nde1 for the presence of insert. Slot-blot hybridization was conducted using RsaI fragments as probe labeled by random priming (RediprimeTM II kit, Amersham Pharmacia biotech, USA). Finally, positive clones were selected for sequencing.

pDp1 (B) pDp2 (D) pDp4 (C) pDp3
Color key scores 50-80 80-200 >= 200 (A) Figure 5 In silico mapping of RsaI sequences, pDp1 (A), pDp2 (B) pDp3 (C) and pDp4 (D) using existing cattle genomic map data (cow build 4 genome database). Note presence of pDp1, pDp2 and pDp4 sequences at multiple locations on cattle chromosomes. Color bar having percent homology of RsaI sequences with cattle genome is shown on right side of the figure. In silico analysis Multiple sequence alignment and Phylogenetic tree construction were carried out using ClustalW program.

Buffalo genomic DNA analysis
For cross hybridization studies, approximately 500 ng heat denatured genomic DNA from 14 species each mentioned earlier were slot-blotted onto the nylon membrane (Amersham) along with cloned plasmid as positive control and 2× SSC as negative control following standard protocol [41]. For Southern hybridization, approximately, 4-5 μg of buffalo, cattle, goat and sheep genomic DNA were subjected to restriction digestion using RsaI enzyme following supplier's (NEB) specifications. The digested DNA was resolved on 1.5% agarose gel and transferred onto the nylon membrane (Amersham) following standard protocol [42]. Membranes were rinsed in 2× SSC, dried and UV cross-linked. Blots were hybridized at 60°C overnight with α-32P-dCTP labeled recombinant plasmid (25 ng) using random priming method (rediprimeTM II kit, Amersham Pharmacia biotech, USA). Washing of the membranes was done using standard protocols and signals were recorded by exposure of the blot to X-ray film [41].

PCR
The extent of sequence conservation across bovid genomes was further determined by PCR analysis. Primers were designed from ORF using Primer 3 software (Table  3). 50 ng of genomic DNA from buffalo, cattle, goat and sheep were PCR amplified using reaction conditions 95°C for 5 minutes followed by 35 cycles each consisting 95°C for 1 minute, 60°C for 1 minute and 72°C for 1 minute, and final extension at 72°C for 10 minutes for all the four fragments. Amplified fragments were resolved on 1.5% agarose gel in 1% TAE buffer (Table 3).

RNA isolation and synthesis of cDNA
Total RNA was extracted from testis, kidney, liver, spleen, lung, heart, ovary, brain and sperm using TRIzol (Molecular Research Center, Inc., Cincinnati, OH) following manufacturer's instructions [43,44]. Tissues were procured from the local slaughterhouse, Delhi. Fresh ejaculated semen samples from buffalo bulls were obtained from the animal farm, Lucknow, U.P., India, strictly following the guidelines of the Institutes Ethical and Biosafety Committee. To check the contamination of mRNA from the cells other than spermatozoa, RNA extractions from the sperms were tested by RT-PCR for both the CDH1 (E-cadherin) GenBank Accession no. AJ400864 and CD45 (tyrosine phosphatase) GenBank Accession no. NM_001002763 [45]. Similarly, presence of DNA was ruled out by PCR using β-actin primers, GenBank accession no. DQ661647 (Table 3). Following this, approximately 10 μg of RNA from different tissues and spermatozoa was reverse transcribed into cDNA using commercially available high capacity cDNA RT kit (Applied Biosystems, USA). The success of cDNA synthesis was confirmed by 35 cycles of PCR amplification using buffalo derived β-actin primers.

RT-PCR and Relative expression analysis
Expression analysis of pDp1, pDp2, pDp3 and pDp4 transcripts using 50 ng cDNA from different somatic tissues and spermatozoa of buffalo was done with sequence specific internal primers designed by Primer 3 software (Table 3). PCR conditions were same as that mentioned earlier. β-actin was used as positive control. Relative expression analysis for the four transcripts was done using SYBR Green and Real time PCR Sequence Detection System-7500 (ABI, USA). The primers specific to the sequence were designed using Primer Express Software V2.0 (ABI) ( Table 3). Primers for the housekeeping gene GAPDH (Glyceraldehyde 3-phosphate dehydrogenase) GenBank Accession no. XR_083674.1 was used to normalize the values for each sample. The specificity of each primer pair and efficiency of the amplification were tested by assaying serial dilutions of the cDNA. Each reaction was performed in triplicates and the mean value was used for the analyses [46]. The cyclic conditions comprise 50°C for 2 min and 95°C for 10 min, followed by 40 cycles each of 95°C for 10 s and 60°C for 1 min. Each experiment was repeated three times to ensure consistency of the results. The expression level of the desired sequence in different tissues and spermatozoa was calculated in the form of 2 -ΔΔCt value http://www3. appliedbiosystems.com/cms/groups/mcb_support/documents/generaldocuments/cms_042176.pdf [46].

Copy number calculation
Real-time qPCR assays were performed in a 25 μl reaction volume containing 12.5 μl 2× SYBR Green ® PCR Master Mix (Applied Biosystems, Foster City, CA, USA), genomic DNA (0.1, 1.0 and 5 ng), forward and reverse primers at final concentration of 100 μM. Copy number estimation for four RsaI fragments in buffalo, cattle, goat and sheep genomes were done using 10 fold dilutions series of recombinant plasmids in the range 30,00,00000 to 3 copies (assuming haploid genome of farm animals =3.3 pg, wt per base pair = 1.096 × 10 -21 gm). Reactions were performed in 96-well MicroAmp Optical Reaction Plates (Applied Biosystems) in triplicates using the Real time PCR Sequence Detection System-7000 (ABI, USA) and SYBR Green dye. Primers and assay conditions were similar to those used for Relative expression studies (Table 3). Reaction specificity was confirmed with melting curves analysis. The standard curve was prepared using 10 folds dilution series of the recombinant plasmids and buffalo, cattle, goat and sheep genomic DNA [45,46].

Chromosome preparation and fluorescence in situ hybridization (FISH)
Approximately, 200 μl of the whole blood from buffalo was cultured for chromosome preparation following standard protocols [46]. FISH was conducted with spectrum red labeled pDp1, pDp2 and pDp3 and pDp4 cloned probes (Abbott Molecular) on the metaphase chromosomes using Nick Translation Kit, Abbott Molecular, (Illinois, USA). Hybridization was carried out in 20 μl volume containing 50% formamide, 10% Dextran sulphate, Cot 1 DNA and 2× SSC, pH 7 for 16 hours at 37°C in a moist chamber. Post hybridization washes were done in 2× SSC at 37°C (low stringent condition) and then at 60°C in 0.1× SSC (under high stringent condition). Slides were counterstained with DAPI, screened under Olympus Fluorescence Microscope (BX51) and images were captured with Olympus U-CMAD-2 CCD camera. Chromosome mapping was done following the International System for Chromosome Nomenclature (ISCND 2000) for Bovids [47].
Generation of full length buffalo ACOT11 mRNA using endpoint PCR Blast search with pDp1 sequences showed 91% homology with B. taurus ACOT11 gene from nucleotide position 730-1331 encompassing 602 bp. Full length buffalo ACOT11 mRNA was generated using primers designed from B. taurus ACOT 11 (Accession No. NM_001103275). Details of Primer sequences and product size are given in Table 3. PCR amplified products were cloned into and pGEMT-easy vector and sequenced. Finally, buffalo ACOT11 gene sequences were assembled and full length sequence was deposited in the GenBank.