Portal de Periódicos da CAPES

Long Terminal Repeats Are Used as Alternative Promoters for the Endothelin B Receptor and Apolipoprotein C-I Genes in Humans

2001; Elsevier BV; Volume: 276; Issue: 3 Linguagem: Inglês

10.1074/jbc.m006557200

1083-351X

Patrik Medstrand, Josette‐Renée Landry, Dixie L. Mager,

Peroxisome Proliferator-Activated Receptors

To examine the potential regulatory involvement of retroelements in the human genome, we screened the transcribed sequences of GenBankTM and expressed sequence tag data bases with long terminal repeat (LTR) elements derived from different human endogenous retroviruses. These screenings detected human transcripts containing LTRs belonging to the human endogenous retrovirus-E family fused to the apolipoprotein CI (apoC-I) and the endothelin B receptor (EBR) genes. However, both genes are known to have non-LTR (native) promoters. Initial reverse transcription-polymerase chain reaction experiments confirmed and authenticated the presence of transcripts from both the native and LTR promoters. Using a 5′-rapid amplification of cDNA ends protocol, we showed that the alternative transcripts of apoC-I and EBR are initiated and promoted by the LTRs. The LTR-apoC-I fusion and native apoC-I transcripts are present in many of the tissues tested. As expected, we found apoC-I preferentially expressed in liver, where about 15% of the transcripts are derived from the LTR promoter. Transient transfections suggest that the expression is not dependent on the LTR itself, but the presence of the LTR increases activity of the apoC-I promoter from both humans and baboons. The native EBR-driven transcripts were also detected in many tissues, whereas the LTR-driven transcripts appear limited to placenta. In contrast to the LTR of apoC-I, the EBR LTR promotes a significant proportion of the total EBR transcripts, and transient transfection results indicate that the LTR acts as a strong promoter and enhancer in a placental cell line. This investigation reports two examples where LTR sequences contribute to increased transcription of human genes and illustrates the impact of mobile elements on gene and genome evolution. To examine the potential regulatory involvement of retroelements in the human genome, we screened the transcribed sequences of GenBankTM and expressed sequence tag data bases with long terminal repeat (LTR) elements derived from different human endogenous retroviruses. These screenings detected human transcripts containing LTRs belonging to the human endogenous retrovirus-E family fused to the apolipoprotein CI (apoC-I) and the endothelin B receptor (EBR) genes. However, both genes are known to have non-LTR (native) promoters. Initial reverse transcription-polymerase chain reaction experiments confirmed and authenticated the presence of transcripts from both the native and LTR promoters. Using a 5′-rapid amplification of cDNA ends protocol, we showed that the alternative transcripts of apoC-I and EBR are initiated and promoted by the LTRs. The LTR-apoC-I fusion and native apoC-I transcripts are present in many of the tissues tested. As expected, we found apoC-I preferentially expressed in liver, where about 15% of the transcripts are derived from the LTR promoter. Transient transfections suggest that the expression is not dependent on the LTR itself, but the presence of the LTR increases activity of the apoC-I promoter from both humans and baboons. The native EBR-driven transcripts were also detected in many tissues, whereas the LTR-driven transcripts appear limited to placenta. In contrast to the LTR of apoC-I, the EBR LTR promotes a significant proportion of the total EBR transcripts, and transient transfection results indicate that the LTR acts as a strong promoter and enhancer in a placental cell line. This investigation reports two examples where LTR sequences contribute to increased transcription of human genes and illustrates the impact of mobile elements on gene and genome evolution. long terminal repeat endogenous retrovirus human ERV apolipoprotein C-I endothelin B receptor polymerase chain reaction reverse transcription-PCR rapid amplification of cDNA ends kilobase pair(s) base pair(s) hepatic control region splice donor endothelin A very high proportion of mammalian DNA consists of retroelements that have arisen via RNA reverse transcription and reintegration into the genome (1Smit A.F.A. Curr. Opin. Genet. Dev. 1999; 9: 657-663Crossref PubMed Scopus (728) Google Scholar). Retroelements are found in most, if not all, species, where they have amplified to high copy numbers during evolution (2Boeke J.D. Stoye J.P. Coffin J. Hughes S.H. Varmus H.E. Retroviruses. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1997: 343-435Google Scholar). The sheer number of such mobile elements suggests that they affect the host genome, and several observations indicate that retroelements impact on the species in a number of ways by acting as insertional mutagens or contributing regulatory functions to genes (3Kidwell M.G. Lisch D. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 7704-7711Crossref PubMed Scopus (438) Google Scholar). While transposable elements can be harmful to their host, the vast majority of transposable elements present in humans are derived from ancient transpositional events which are fixed in Old World primates. Potential long term effects of the majority of these elements must be either neutral or beneficial; otherwise they would be eliminated by selection. Human DNA contains essentially two classes of retrosequences, (i) the non-long terminal repeat (non-LTR)1 retroposons represented by LINE and Alu sequences, and (ii) the LTR retroelements in which the endogenous retroviruses (ERVs), solitary LTRs derived from ERVs, and other LTR-like sequences fall (4Smit A.F.A. Curr. Opin. Genet. Dev. 1996; 6: 743-748Crossref PubMed Scopus (488) Google Scholar). Human ERVs (HERVs) are classified into different families based on sequence similarity and monophyletic clustering (5Larsson E. Kato N. Cohen M. Curr. Top. Microbiol. Immunol. 1989; 148: 115-132Crossref PubMed Scopus (128) Google Scholar, 6Tristem M. J. Virol. 2000; 74: 3715-3730Crossref PubMed Scopus (266) Google Scholar). The thousands of ERVs and solitary LTRs that are present in human DNA are the result of infections and transposition events during primate evolution. Solitary LTRs are common features in the human genome, and they probably arose from a recombination event between the 5′ and 3′ LTR of a full-length provirus. Despite their evolutionary age, many ERVs are still transcriptionally active in human cells, where different ERV families show quite different sites and levels of transcription (7Wilkinson D. Mager D.L. Leong J.-A.C. Levy J. The Retroviridae. Plenum Press, New York1994: 465-535Crossref Google Scholar). The LTR and ERV elements are especially interesting in this regard, since they naturally possess enhancer, promoter, and polyadenylation functions within their LTRs, which probably accounts for differences in transcription of the various HERV families. Besides promoting transcription of retroviral genes, several studies have demonstrated that ERVs and LTRs can assume gene regulatory functions (8Britten R.J. Gene (Amst.). 1997; 205: 177-182Crossref PubMed Scopus (159) Google Scholar, 9Sverdlov E.D. FEBS Lett. 1998; 428: 1-6Crossref PubMed Scopus (109) Google Scholar, 10Brosius J. Gene (Amst.). 1999; 238: 115-134Crossref PubMed Scopus (304) Google Scholar). For example, the paratoid-specific expression of amylase in humans is dependent and under control of an HERV-E element (11Ting C.N. Rosenberg M.P. Snow C.M. Samuelson L.C. Meisler M.H. Genes Dev. 1992; 6: 1457-1465Crossref PubMed Scopus (232) Google Scholar). HERV-E also appears to be involved in the expression of human pleiotrophin in placenta (12Schulte A.M. Lai S. Kurtz A. Czubayko F. Riegel A.T. Wellstein A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 14759-14764Crossref PubMed Scopus (177) Google Scholar). It has also been demonstrated that a HERV-K LTR encodes the last 67 amino acids of one form of the leptin receptor OBR (13Kapitonov V.V. Jurka J. J. Mol. Evol. 1999; 48: 248-251Crossref PubMed Scopus (67) Google Scholar). These findings indicate that an LTR insertion adjacent to or within a gene could have a variety of effects without destroying gene function. Such new insertions may alter tissue specific gene expression or enhance the general transcription levels of the gene, which could be selectively advantageous. We are using LTR sequences to study the involvement of retroelements in gene regulation. Specifically, we have searched the expressed sequence tag and transcribed subset of GenBankTM for chimeric retroviral gene sequences. Here, we report two human genes that are affected by ERV LTRs, the apolipoprotein C-I (apoC-I) gene and the endothelin B receptor (EBR) gene. We show that these two genes use a HERV-E LTR as an alternative promoter, demonstrate the presence of the chimeric transcripts in human tissues, and test the significance of the LTRs at the genomic loci of apoC-I and EBR. Reverse transcription was done with Superscript II (Life Technologies, Inc.) using the same reaction conditions as described previously (14Medstrand P. Lindeskog M. Blomberg J. J. Gen. Virol. 1992; 73: 2463-2466Crossref PubMed Scopus (101) Google Scholar). PCR was carried out using 0.1–0.5 volumes of each cDNA (0.1–0.5 μg of the initial RNA) per reaction. RNA samples were either obtained fromCLONTECH or prepared from different sections of placenta, as described previously (15Wilkinson D.A. Freeman J.D. Goodchild N.L. Kelleher C.A. Mager D.L. J. Virol. 1990; 64: 2157-2167Crossref PubMed Google Scholar). The following primers were used to detect the different transcript forms shown in Fig. 2 (see Table I for primer sequences): LTR-apoC-I fusion transcript, primers APO-LTR1/APO-Ex1; native apoC-I transcript, primers APO-N/APO-Ex1; LTR-EBR fusion transcript, primers EBR-L1/EBR-Ex1; native EBR transcript, primers EBR-N/EBR-Ex1. Amplification was done by using 0.5 volumes of cDNA (see above) with the following cycling profile: one initial incubation of 95 °C for 1 min followed by 35 cycles (for the apoC-I amplifications) or 30 cycles (for the EBR amplifications) of 95 °C for 30 s, 63 °C for 30 s, and 72 °C for 30 s, and one final elongation at 72 °C for 5 min. In the semiquantitative RT-PCR of different EBR transcript forms (Fig. 5), the following primer combinations were used: LTR-EBR fusion transcript, primers EBR-L1/EBR-Ex2; native EBR transcript, primers EBR-N/EBR-Ex2; total EBR transcript, primers EBR-Ex3/EBR-Ex2. In these experiments, 0.15 volumes cDNA was used in the same PCR profile as described above but with lower cycling (25–28 cycles) to avoid saturation effects during the amplification. The intensity of the amplification products was measured after 25 cycles from ethidium bromide-stained gels using the 1D Image Analysis software (Eastman Kodak Co.).Table IPrimers used for RNA analysisPrimerSequencePosition 1-aPositions relative to the first nucleotide of the translational initiation; positions upstream (−) in the genomic DNA or downstream (+) in the cDNA.APO-ex15′-AGCCGCATCAAACAGAGTGAACTT-3′+157 /180APO-ex25′-TCCTCCTGCTACATTCTGAGTGG-3′−477 /−455APO-ex35′-ACGTGCCTTGGATAAGCTGAAG-3′+95 /+114APO-LTR15′-GTCTGAGGAATTTTGTCTGCGGCT-3′−500 /−477APO-N5′-CCAAGCCCTCCAGCAAGGATTC-3′−182 /−161EBR-ex15′-AGTCTATGTGCTCTGAGTATTGAC-3′+571 /+594EBR-ex25′-GACTGGCCATTTGGAGCTGAGAT-3′+496 /+518EBR-ex35′-CTTCTGGAGCAGGTAGCAGCATG-3′−20 /+3EBR-ex45′-GACGCCACCCACTAAGACCTTATG-3′+138 /+161EBR-ex55′-GACGCCTTCTGGAGCAGGTAGCA-3′−25 /−3EBR-L15′-CATGGAGGATCAACACAGTGGCT-3′−21,500 1-bApproximate position (see "Experimental Procedures.").EBR-N5′-TTACTTTTGAGCGTGGATACTGGC-3′−166 /−1431-a Positions relative to the first nucleotide of the translational initiation; positions upstream (−) in the genomic DNA or downstream (+) in the cDNA.1-b Approximate position (see "Experimental Procedures."). Open table in a new tab Figure 5RT-PCR of EBR transcripts in placenta.The relative abundance of the LTR and native transcript forms was estimated by low cycle RT-PCR. The LTR-driven form was detected by using primers derived from exon1/exon3, the native form with primers from exon 2A/exon3, and the total EBR was amplified using primers derived from exon 2B/exon3 (see Fig. 3). RT-PCR was done on cDNA of different section of placenta: villi (V), decidua (D), chorion (C), amnion (A). Sizes of the DNA markers are shown to the left. Expected amplification product sizes were obtained using the different primer combinations. The relative abundance of the different transcript forms is shown to the right of the gel. Total EBR levels of the different samples are indicated with a bar. Filled and open rectangles show the values of the native and LTR-driven forms, respectively. All values were adjusted to the total seen in decidua.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Placental and brain Marathon-ready cDNA libraries were purchased from CLONTECH, and a 5-μl cDNA library was used in 5′-RACE analysis as described in the protocol supplied with the library. The first PCR amplification was performed using EBR exon-specific primer (Table I) EBR-ex4 and the AP1 primer (provided by the supplier) and with the apoC-I primer APO-ex1 and primer AP1. The nested PCR was performed by EBR primer EBR-ex5 and the AP2 primer (provided by the supplier) and with the apoC-I primer APO-ex2 and AP2. The following temperature profile was used for all amplifications: one initial denaturing at 95 °C for 1 min followed by 35 cycles at 95 °C for 30 s and annealing and extension at 68 °C for 4 min. The 5′-RACE products were cloned using the pGEM-T vector system I (Promega). Clones were selected for sequencing after hybridization using retrovirus-specific oligonucleotides APO-LTR1 and EBR-L1 (Table I). The oligonucleotide primer APO-ex3, complementary to exon 3 of apoC-I, was radiolabeled with γ-32P, and 1.2 × 106 dpm of the labeled primer was incubated with 5 μg of total RNA in a 10-μl solution containing 50 mm KCl, 20 mm Tris-HCl, pH 8.4, 2.5 mm MgCl2, 0.1 mg/ml bovine serum albumin at 61 °C for 20 min. The samples were then transferred to ice, and 300 units of Superscript II and 15 units of RNase inhibitor were added to the reaction and adjusted to a volume of 20 μl with a final concentration of 50 mm KCl, 20 mm Tris-HCl, pH 8.4, 2.5 mm MgCl2, 0.1 mg/ml bovine serum albumin, and 0.5 mm dNTPs. The primer extension reaction was performed by incubating samples at 25 °C for 10 min, 42 °C for 50 min, and 95 °C for 10 min. The reaction products were separated on a 6% polyacrylamide gel containing 7 m urea and visualized by exposing to x-ray film at −70 °C. The intensity of the extension products were measured using the ImageQuant software after incubation on PhosphorImager plates (Molecular Dynamics, Inc., Sunnyvale, CA). Locus specific PCR was performed essentially as described previously (16Medstrand P. Mager D.L. J. Virol. 1998; 72: 9782-9787Crossref PubMed Google Scholar). Genomic DNA prepared from marmoset (New World Monkey), baboon (Old World Monkey), gibbon, orangutan, gorilla, chimpanzee, and human cell lines (17Goodchild N.L. Wilkinson D.A. Mager D.L. Virology. 1993; 196: 778-788Crossref PubMed Scopus (72) Google Scholar) was used in PCR. Primers APO-P1 (5′-GGTTTTTACAGTGTCATCCAGCT-3′)/APO-P2 (5′- GATTCAGGTTGGTGCTGAGTG-3′) were used to detect the presence or absence of the solitary LTR in the apoC-I locus of different primates. The LTR upstream of the EBR locus was amplified by primer EBR-F1 (5′-AACATCCTCTGTCTCTCTCC-3′; sequence flanking the LTR integration) and primer EBR-LTR1 (5′-GATCGACCCCTGACCTAACC-3′; sequence from the LTR). The apoC-I and EBR primers were specified from GenBankTMaccession numbers AB012576 and AL139002, respectively. The 5′-flanking regions of apoC-I and EBR were amplified from human genomic DNA and subcloned upstream of the luciferase gene in the promoterless luciferase reporter plasmid pGL3B (Promega). To facilitate directional cloning into pGL3B, primers were designed with terminal sequences specific for restriction enzyme recognition. The restriction enzyme adaptor is indicated after each primer (see below), where the following suffixes are used: K,KpnI; B, BglII; Ba, BamHI; X,XbaI; Xh, XhoI. The numbers in parenthesis after each primer indicate the start and end positions of the primer upstream in the flanking DNA sequence, relative to the first nucleotide of the initiation codon of the two genes. The primer sequences are available upon request. Because the exact distances of the EBR LTR and the hepatic control region (HCR) to EBR and apoC-I are uncertain, the primer sequences used to amplify the EBR LTR and the HCR are shown below, where primer sequence in lowercase type indicates the restriction enzyme recognition sequence. The following EBR constructs were made. For pEBR-NP, the 5′-flanking region of the native EBR transcription initiation site was isolated using primers EBR-NP1K (−1259/−1239) and EBR-NP2B (−208/−187). Digested and purified amplification products were inserted intoKpnI/BglII-digested pGL3B. For pEBR-LTR, the complete LTR was amplified with primers EBR-LTR1K (5′-ggggtaccTAAGGGAGGATACCACC-3′)/EBR-LTR2B (5′-GCAGCTTCTCCTGCTACAagatcttc-3′) and inserted intoKpnI/BglII-digested pGL3B. pEBR-NP+LTR-S and pEBR-NP+LTR-A were made by introducing the LTR at a distance of the native promoter region of construct pEBR-NP. The full LTR was amplified with LTR-specific primers, EBR-LTR1Ba (5′-cgggatccTAAGGGAGGATACCACC-3′)/EBR-LTR2Ba (5′-GCAGCTTCTCCTGCTACAggatcccg-3′). Purified andBamHI-digested amplification products were introduced into the BamHI site of construct pEBR-NP, which is located 2 kb from the KpnI/BglII site on pGL3B. Constructs introduced either in sense (LTR-S) or antisense (LTR-A) with respect to the native EBR promoter region were isolated. The location of the primers was based on the initiation codon at position 1260 of GenBankTM accession number D13162. The LTR primers were derived from GenBankTM accession number AL139002. The following apoC-I constructs were made. For pAPO-P, the 5′-flanking region of the apoC-I transcription initiation site was isolated using primers APO-P1K (−1271/−1249)/APO-P2B (−165/−145) and inserted intoKpnI/BglII-digested pGL3B. This construct contains both the native and LTR promoter regions. For pAPO-LTR, the complete LTR of the apoC-I locus was amplified with primers APO-LTR1K (−920/−901)/APO-LTR2B (−484/−465) and introduced in theKpnI/BglII site of pGL3B. For pAPO-P-noLTR, the LTR was removed from the apoC-I locus by amplifying the non-LTR parts of pAPO-P with primers APO-P1K (−1271/−1249)/APO-P3X (−941/−924) and APO-P4X (−455/−439)/APO-P2B (−165/−145). The two amplification products were digested with XbaI, ligated together, and introduced after KpnI/BglII digestion into pGL3B. This construct has the same structure as the pAPO-P except that the LTR is absent. pAPO-P-noLTR+LTR-S and pAPO-P-noLTR+LTR-A were made by introducing the LTR at a distance of the native apoC-I promoter region lacking the LTR (construct pAPO-P-noLTR). The full LTR was amplified with LTR-specific primers, APO-LTR1Ba (−920/−901)/APO-LTR2Ba (−484/−465). Purified andBamHI-digested amplification products were introduced into the BamHI site of construct pAPO-P-noLTR. Constructs introduced either in sense (LTR-S) or antisense (LTR-A) with respect to the apoC-I promoter region were isolated. The location of the primers is with respect to the apoC-I initiation codon at position 27457 of GenBankTM accession number AB012576. The following baboon apoC-I constructs were made. For pBAPO-P, the baboon apoC-I locus was amplified from baboon genomic DNA with primers APO-P1K (−782/−760)/APO-P2B (−160/−140) and inserted intoKpnI/BglII-digested pGL3B. For pBAPO-P+LTR, construct pBAPO-P was amplified with primers APO-P1K (−782/−760)/APO-P5Ba (−463/−444) and APO-P6Ba (−435/−414)/APO-P2B (−160/−140). The two amplification products were digested with BamHI and ligated together. This religated fragment was inserted into pGEM-T (construct pBAPO-GEM). The LTR that was amplified with primers APO-LTR1Ba/APO-LTR2Ba (see above) was introduced into the BamHI site of pBAPO-GEM. After selection of LTR-positive clones, the KpnI/BglII cassette (containing the LTR in the baboon apoC-I at the same orthologous site as in humans) was subcloned into pGL3B. Positions of the primers are from GenBankTM accession number L13176 and with respect to the initiation codon of the baboon apoC-I at position 1017. The HCR was isolated from human DNA using PCR and primers HCR1Xh (5′-ccgctcgagTTAGAGAACAGAGCTGCAGGCT-3′) and HCR2Xh (5′-ATGCCCCGACCCCGAAGCctcgagcgg-3′). The primer sequences were derived from positions 36815–36836, and positions 37201–37218 of GenBankTM accession number AF050154, respectively. The PCR product was digested with XhoI, and the purified fragment was introduced into the pGL3B SalI site of all apoC-I constructs, which is 3′ to the luciferase gene. Plasmid DNA was purified by using the Qiagen plasmid midi kit (Qiagen) prior to transfections. HepG2 (human hepatoblastoma cells; ATCC HB-8065) cells were cultured in Dulbeccos's minimal essential medium supplemented with 10% fetal calf serum. Cells were seeded 24 h prior to transfections in six-well plates at a density of 3 × 105 cells/well. Transient transfections of HepG2 were done by cotransfecting 1.5 μg of plasmid DNA and 50 ng of pRL-TK vector (Promega) using calcium phosphate (Cellphect; Amersham Pharmacia Biotech) as described in the protocol supplied with the reagent. JEG-3 cells (human choriocarcinoma; ATCC HTB-36) were maintained in RPMI supplemented with 5% fetal calf serum. JEG-3 cells were seeded in six-well plates at a density of 2 × 105 cells/well and cotransfected 24 h later with 1.0 μg of plasmid DNA and 200 ng of pRL-TK using 7 μl of LipofectAMINE (Life Technologies, Inc.), as described in the protocol from the supplier. After 24 h, the cells were lysed, and the luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega) and normalized to the internal control. Transfections were performed in triplicates and repeated at least twice. Double-stranded plasmid DNAs were sequenced using an ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction kit and an ABI 373 sequencer (PerkinElmer Life Sciences). Using the strategy outlined in Fig.1, we searched GenBankTM and the human expressed sequence tag data bases using the LTR and the leader region of published HERV elements (7Wilkinson D. Mager D.L. Leong J.-A.C. Levy J. The Retroviridae. Plenum Press, New York1994: 465-535Crossref Google Scholar). Transcripts with only the U3-R part of the LTR and no other HERV sequence probably represent mRNA polyadenylated by the LTR, whereas transcripts with R-U5 or R-U5-leader probably represent promotion by an LTR. During data base screenings, we encountered two transcripts where HERV-E (18Repaske R. Steele P.E. O'Neill R.R. Rabson A.B. Martin M.A. J. Virol. 1985; 54: 764-772Crossref PubMed Google Scholar) sequences were fused to the EBR gene (accession number D90402) and the apoC-I gene (accession number W79313). The structure of these transcripts suggests that the EBR utilizes the splice donor (SD) in the leader region of the HERV element, which is located downstream of the 5′ LTR (Fig. 1) of an integrated provirus. The same SD is used in subgenomic splicing of HERV-E envelope transcripts, suggesting that this represents the original SD of HERV-E (19Rabson A.B. Hamagishi Y. Steele P.E. Tykocinski M. Martin M.A. J. Virol. 1985; 56: 176-182Crossref PubMed Google Scholar). The apoC-I fusion transcript represents another possible LTR-driven transcript type, which is derived from a solitary LTR and reads into the flanking non-HERV region. To authenticate the presence of fusion transcripts, we synthesized primers corresponding to the retroviral and the gene-specific regions of the identified transcripts. By using this primer combination in RT-PCR, it is possible to detect the presence and the relative abundance of the fusion transcripts in human tissues. Both of the genes were previously reported as having a different promoter region, separated from the retroviral LTR (20Lauer S.J. Walker D. Elshourbagy N.A. Reardon C.A. Levy-Wilson B. Taylor J.M. J. Biol. Chem. 1988; 263: 7277-7286Abstract Full Text PDF PubMed Google Scholar, 21Nakamuta M. Takayanagi R. Sakai Y. Sakamoto S. Hagiwara H. Mizuno T. Saito Y. Hirose S. Yamamoto M. Nawata H. Biochem. Cell Biol. Commun. 1991; 177: 34-39Google Scholar, 22Ogawa Y. Nakao K. Arai H. Nakagawa O. Hosoda K. Suga S. Nakanishi S. Imura H. Biochem. Cell Biol. Commun. 1991; 178: 248-255Google Scholar). We will refer to these two regions as the "native" apoC-I and EBR promoter. To detect any biases of the LTR and native transcripts, we also used a primer unique to transcripts of the native promoters of the two genes. Results of the RT-PCR on a panel of RNAs derived from different human tissues are shown in Fig. 2. The LTR-promoted EBR transcript is restricted to placenta, where its levels appear comparable with that of the widely expressed native transcript (Fig.2 B). In the case of apoC-I, transcripts from the native promoter in liver were high as expected (20Lauer S.J. Walker D. Elshourbagy N.A. Reardon C.A. Levy-Wilson B. Taylor J.M. J. Biol. Chem. 1988; 263: 7277-7286Abstract Full Text PDF PubMed Google Scholar) but are also detectable by PCR in many of the other RNAs tested (Fig. 2 A). Transcripts from the solitary LTR were detected in two distinct forms (see Fig.3 A), both of which were also detected in many tissues. The result of this experiment clearly demonstrates the presence of fusion transcripts between LTRs of HERV-E and the genes for EBR and apoC-I. The LTRs at the EBR and apoC-I loci vary in their tissue specificity, with the EBR LTR being much more restricted in activity. Sequencing of the PCR products verified the nature of the fusion transcripts, where the two fusion transcript forms of apoC-I are the result of differential splicing in the 5′ UTR (Fig.3). To confirm that the apoC-I and EBR fusion transcripts initiate within the LTRs and do not represent transcripts from a promoter located upstream of the LTRs, we isolated the 5′-ends of both LTR fusion gene transcripts. Using a 5′-RACE protocol, we established that both the apoC-I and EBR fusion transcript initiate within their LTRs (see below and Fig. 3). Sequencing of several 5′-RACE clones showed that the apoC-I and EBR initiation site is located downstream of a previously reported TATA box of HERV-E (18Repaske R. Steele P.E. O'Neill R.R. Rabson A.B. Martin M.A. J. Virol. 1985; 54: 764-772Crossref PubMed Google Scholar, 19Rabson A.B. Hamagishi Y. Steele P.E. Tykocinski M. Martin M.A. J. Virol. 1985; 56: 176-182Crossref PubMed Google Scholar). This is the TATA also used by other HERV-E proviruses because a full-length transcribed HERV-E element (GenBankTM accession number M74509) starts 2 bp downstream of the initiation site of the apoC-I LTR. In the case of the EBR fusion transcript, the sequence representing the longest 5′-UTR also began within the LTR, but at a position 3′ (90 bp) to the apoC-I initiation site. Both the apoC-I and EBR genomic loci were partially characterized at the time of our initial studies. The only retroviral remnant of the original proviral insertion at the apoC-I locus is a solitary LTR, which is located 300 bp upstream of the native apoC-I promoter. The two initiation sites are separated by 390 bp, where the initiation sites of the native and LTR promoters are located 180 and 575 bp upstream of the apoC-I initiation codon, respectively (Fig. 3). The EBR LTR was not present in the reported 2-kb sequence upstream of the EBR native promoter (GenBankTM accession D13162), which is located ∼250 bp upstream of the EBR initiation codon (21Nakamuta M. Takayanagi R. Sakai Y. Sakamoto S. Hagiwara H. Mizuno T. Saito Y. Hirose S. Yamamoto M. Nawata H. Biochem. Cell Biol. Commun. 1991; 177: 34-39Google Scholar, 22Ogawa Y. Nakao K. Arai H. Nakagawa O. Hosoda K. Suga S. Nakanishi S. Imura H. Biochem. Cell Biol. Commun. 1991; 178: 248-255Google Scholar). A genomic clone containing both the HERV-E proviral element and the EBR genomic locus was recently deposited in GenBankTM (accession numberAL139002). The sequence of this clone is in a preliminary state of annotation and contains unordered pieces of DNA. The parts of the LTR and leader region that were identified in EBR 5′-RACE (see above) are identical to the proviral element of AL139002. It has been previously reported that several other alternative transcripts (named EDNRBΔ) are created by initiation 560 and 940 bp upstream of the ATG codon and alternative splicing of the 5′-UTR (23Tsutsumi M. Liang G. Jones P.A. Gene (Amst.). 1999; 228: 43-49Crossref PubMed Scopus (23) Google Scholar). The 5′ LTR leader of the HERV-E element is located over 20 kb from the EBR gene and is joined by splicing to the same splice acceptor in the 5′-UTR as are the spliced EDNRBΔ transcripts. The genomic organization and the structures of the different transcripts at these two loci are shown in Fig. 3. Due to the retroviral sequence, the fusion transcripts have partially different 5′-UTRs compared with the native forms, but all maintain the same apoC-I and EBR coding regions. Using primers flanking the integration sites of the LTRs in PCR of different primate DNAs, we earlier assigned the time of integration of various HERV-K elements during primate evolution (16Medstrand P. Mager D.L. J. Virol. 1998; 72: 9782-9787Crossref PubMed Google Scholar). Using the same approach, we were able to determine when the two HERV-E LTRs integrated in the primate lineage. The apoC-I LTR was detected in all hominoids, whereas Old and New World monkeys did not have this LTR integrated in the apoC-I locus,