Virology 377 (2008) 88–99

Contents lists available at ScienceDirect

Virology

j ourna l homepage: www.e lsev ie r.com/ locate /yv i ro

brought to you by COREView metadata, citation and similar papers at core.ac.uk

provided by Elsevier - Publisher Connector

The RNA transport element RTE is essential for IAP LTR-retrotransposon mobility

Andrei S. Zolotukhin a, Ralf Schneider b, Hiroaki Uranishi c, Jenifer Bear a, Irina Tretyakova a,Daniel Michalowski a, Sergey Smulevitch a, Sean O'Keeffe b, George N. Pavlakis c, Barbara K. Felber a,⁎a Human Retrovirus Pathogenesis Section, Vaccine Branch, Center for Cancer Research, National Cancer Institute-Frederick, Frederick, MD 21702-1201, USAb Institute of Experimental Genetics, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Germanyc Human Retrovirus Section, Vaccine Branch, Center for Cancer Research, National Cancer Institute-Frederick, Frederick, MD 21702-1201, USA

⁎ Corresponding author. Fax: +1 301 846 7146.E-mail address: felber@ncifcrf.gov (B.K. Felber).

0042-6822/$ – see front matter. Published by Elsevier Idoi:10.1016/j.virol.2008.04.002

A B S T R A C T

A R T I C L E I N F O

Article history:

Wepreviously identified an Received 6 February 2008Returned to author for revision17 March 2008Accepted 3 April 2008Available online 15 May 2008

Keywords:RetrotranspositionRetroelementmRNA transportNuclear exportCTERetrovirusHIVSRVNXF1RBM15

RNA transport element (RTE) present at a high copy number in themouse genome.Here,we show that a related element, RTE-D, is part of amobile LTR-retrotransposon, which belongs to a familyof intracisternal A-particle related elements (IAP). We demonstrate that RTE-D is essential for the mobility ofthe retrotransposon and it can be substituted by other known RNA export signals. RTE-deficient IAP transcriptsare retained in the nucleus, while the RTE-containing transcripts accumulate in the cytoplasm allowing Gagprotein expression. RTE-D acts as a posttranscriptional control element in a heterologous reportermRNA and isactivated by the cellular RNA binding protein 15 (RBM15), as reported for the previously described RTE. Weidentified a complex family of RTE-containing IAPs in mouse and mapped the active RTE-D-containing IAPs tothe Mmr10 group of LTR-retrotransposons. These data reveal that, despite a complex evolutionary history,retroelements and retroviruses share the dependency on posttranscriptional regulation.

Published by Elsevier Inc.

Introduction

The retroviral full-lengthmRNAhas to be exported to the cytoplasmin its unspliced form, since this transcript encodes the Gag-Pol poly-protein and also serves as genomic RNA. Such a transport mechanismhas been first described for HIV, where the viral nuclear Rev proteinbinds to the cis-acting Rev-responsive element (RRE) and promotes theexport and expression of these mRNAs. Rev and the Rev-like proteinsfrom the other complex retroviruses bind to the nuclear receptor CRM1that enables interactionwith components of the nuclear pore complexresulting in export of their RNAs prior to splicing (Cochrane, 2004;Cochrane, McNally, and Mouland, 2006; Cullen, 2003; Felber, Zolotu-khin, and Pavlakis, 2007; Kutay and Guttinger, 2005). In contrast toHIV-1 and other complex retroviruses, the export of unspliced mRNAsof simple retroviruses {e.g. the simian type D retroviruses (SRV/D) suchas Mason–Pfizer Monkey virus (MPMV), simian retrovirus 1 (SRV-1)and 2 (SRV-2) (Bray et al., 1994; Tabernero et al., 1996)} is mediatedsolely by cellular factors. The export of CTE-containing RNAs is me-diated by NXF1, which binds directly to CTE (Grüter et al., 1998). NXF1is a conserved cellular mRNA export receptor (Grüter et al., 1998;

nc.

Herold, Klymenko, and Izaurralde, 2001; Segref et al., 1997; Tan et al.,2000; Wilkie et al., 2001), which directly interacts with the nuclearpore complex andmediates export of the CTE-containingmRNAs to thecytoplasm. Thus, retroviruses have adapted to interactwith the nuclearexport receptors either directly (via NXF1) or indirectly (via CRM1-binding viral proteins).

We have previously identified a potent RNA transport element(RTE) in the mouse genome, by using a Rev–RRE deficient HIV clone asmolecular trap (Nappi et al., 2001). RTE is able to replace the lentiviralRev–RRE regulatory system generating infectious HIV and SIV (Nappiet al., 2001; Smulevitch et al., 2006). RTE belongs to a family of ele-ments that are present abundantly in the mouse genome and are likelyassociated with intracisternal A-particle retroelements (IAP) (Nappiet al., 2001). The RTE is functionally analogous to CTE, yet structurallyunrelated, and both elements utilize the cellular mRNA export ma-chinery (Nappi et al., 2001; Smulevitch et al., 2005).We recently showedthat the RNA binding motif 15 protein (RBM15) binds to RTE directlyand tethers the RTE-containing mRNAs to the NXF1 export machinery(Lindtner et al., 2006).

The discovery of retrotransposition-competent IAP LTR-retroele-ments (Dewannieux et al., 2004) led us to investigate their post-transcriptional regulation. Here, we report that such IAPs contain aconserved element, termed RTE-D, which is closely related to theoriginally identified RTE (Nappi et al., 2001). We show that RTE-D is

89A.S. Zolotukhin et al. / Virology 377 (2008) 88–99

fully functional and is essential for the life cycle of the IAP LTR-retroelements. Thus, retroelements, like simple and complex retro-viruses, depend on posttranscriptional regulation for the nucleocy-toplasmic transport of their full-length RNA, which is essential tocounteract the negative-acting determinants located within the IAPretroelement transcript.

Results

Mobile LTR-retrotransposons contain the RNA transport element RTE

The identification of autonomouslymobile IAP LTR-retrotransposonsRP23-231P12, RP23-324C9, RP23-440N1 and RP23-92L23 (Dewannieuxet al., 2004) prompted us to investigate whether they contain RNAexport elements. Sequence alignment of the originally identified RTE(termed RTE-A) (Smulevitch et al., 2005) and these four active IAPsequences revealed the presence of RTE-like elements located betweenthe pol open reading frame (ORF) and the polypurine tract (ppt) up-stream of the 3′ LTR in all these retroelements. Fig. 1A shows the struc-ture of the LTR-retrotransposon RP23-92L23 as representative example,with the RTE-like element (RTE-92L23) located between nt 6446–6687.This RTE-like element shares a relatively low sequence homology of69.8% with RTE-A (Fig. 1B). We modeled RTE-92L23 (Fig. 1A, bottompanel) onto the experimentally identified secondary structure of RTE-Areported previously (Smulevitch et al., 2005). Despite the sequencediversity (nucleotide changes shown in red in Fig. 1B), the predictedstructure of RTE-92L23 shares similarities with that of RTE-A. Wenoticed a high occurrence of compensatory changes, leading to thepreservation of the overall RTE structure (indicated by asterisks; Fig.1A).Based on the variation in structural features (stem–loops SL I through SLIV), the RTEs were previously classified into 4 groups A, B, C, and D(Smulevitch et al., 2005) (see also Fig. 6A, top panel). In accord with theabove classification, we placed the RTE-92L23 into the RTE group D,and hence, RTE-92L23 is referred to as RTE-D in this report. The dif-ferences between the RTEs of group A (RTE-A) and groupDmostly affectthe hairpin loop IV and the bottom portion spanning SL I and SL II,and these structures were shown to be not essential for RTE-A function(Smulevitch et al., 2005).

Comparison of RTE-92L23 to those identified within the othermobile IAP LTR-retrotransposons RP23-324C9, RP23-440N1 and RP23-92L23 (Dewannieux et al., 2004) showed a high degree of conservationin RP23-324C9 (100%), RP23-231P12 (99.2%) and RP23-440N1 (97.5%)(Fig.1C).We also found that the IAP LTR-retrotransposonMIA14 (Mietzet al., 1987) contains an RTE identical to that of RTE-92L23, while RTE-3061 that harbors the prototype RTE-D (Smulevitch et al., 2005) shows99.6% identity. We further identified RTEs in IAP LTR-retrotransposonsisolated from C3H/He inbred mice with myeloid leukemia (Ishiharaet al., 2004). The RTE in the intact clone Q14 (IAP type I) is 97.5%identical to RTE-92L23, while the RTEs in cloneswith large deletions ingag/pol (IAP types IΔ1 to IΔ4 (Ishihara et al., 2004)) show 94–99%identity.We noted that the nucleotide changes are located in predictedsingle-stranded regions and mostly in the bottom part of the element.Fig. 1D shows the phylogenetic relationship of the different group DRTEs. Taken together, group D RTEs are present in IAP LTR-retrotrans-posons and are conserved in both intact and deleted IAPs, suggesting abiological role in the IAP LTR-retrotransposons life cycle.

RTE is essential for retrotransposition

We next chose one of the autonomously mobile IAP LTR-retro-transposons, the molecular clone 92L23, to study the role of RTE in theIAP life cycle, using a previously described mobility assay in HeLa cells(Dewannieux et al., 2004). Briefly, 92L23neoTNF contains a neo gene en-coded in theminus strand of LTR-IAP provirus,which is interrupted byanintron in the plus orientation. HeLa cells were transfected with the wildtype molecular clone or the mutant clone 92L23neoTNFΔRTE, where the

RTE homologous region was replaced with a polylinker (Fig. 2A),respectively, and the cells were subjected to G418 selection. G418resistance can be only established upon integration of reverse transcrip-tionproductoriginating fromsplicedplus-strandRNA, and thenumberofG418 resistant colonies provides a count of retrotransposition events.Wefound that thewild type IAP clonewasmobile as expected (Dewannieuxet al., 2004),whereas theΔRTEmutant producedonly fewG418-resistantcolonies (Fig. 2A). Fig. 2B shows a quantitative analysis of a representativeretrotransposition assay. To count the colonies, serial dilutions oftransfected cells were seeded and subjected to G418 selection. TheΔRTE-D mutant generated b1% of colonies compared to the wild type92L23neoTNF. Similar data were obtained in 5 independent experiments,and demonstrate that RTE is essential for retrotransposition.

To test whether the function of the RTE-D can be complemented byhomologous and heterologous RNA export signals, we inserted suchsignals into 92L23neoTNFΔRTE and measured the mobility of the re-sultant hybrids (Fig. 2B). We found that insertion of the originallyidentified RTE-A (Nappi et al., 2001; Smulevitch et al., 2005) inΔRTE-D/RTE-A(+) restored the mobility. We next asked whether a heterologousRNAexport signal, the constitutive transport element CTE of SRV-1 (Brayet al.,1994; Zolotukhin et al.,1994) could also support retrotransposition.Fig. 2B shows that the presence of CTE in cis in ΔRTE-D/CTE(+) alsorestored themobility of theΔRTEmutant. Thus, these data demonstratethat the presence of an active RNA export element (RTE, CTE) can sub-stitute for RTE-D and support the finding that posttranscriptionalregulation is essential for retrotransposition.

RTE-D promotes the cytoplasmic accumulation and expression of LTR-IAPprimary transcript

We next investigated the mechanisms leading to RTE-dependenceof the LTR-retrotransposon transcript. By analogy to retroviruses, wespeculated that IAP LTR-retrotransposons require posttranscriptionalregulation in order to increase the cytoplasmic levels and expression oftheir primary transcripts. We first examined the effects of RTE-D on thegag/prt/pol expression using the full-length LTR-IAP molecular clone92L23. For this purpose, the neo cassette was removed from boththe original IAP 92L23neoTNF construct (Dewannieuxet al., 2004) and itsΔRTE mutant, restoring the authentic LTR-IAP retrotransposon struc-ture.WeusedRT-PCR andNorthern blot analysis to analyze themRNA inthe nuclear and cytoplasmic extracts of cells transfected with the wildtype or ΔRTE molecular clone (Fig. 3). Using a primer pair that detectsthe gag-encoding primary transcript (Fig. 3A, s1 and a1), we found thatboth clones produced similar levels of nuclear gag mRNAs. In contrast,the lack of RTE led to a dramatic reduction of the gag mRNA accu-mulation in the cytoplasm. Northern blot analysis (Fig. 3B) confirmedthat the RTE deletion led to a strong reduction of cytoplasmic gag/prt/polmRNA level. In further support, we found that the RTE-D deletionalso led to about 5-fold lower levels of Gag protein production, whencompared to the wild type clone (Fig. 3C). Thus, the impaired cy-toplasmic accumulation of the primary transcript and the low Gagproduction are key to the loss of the mobility of the ΔRTE IAP.

Interestingly, both the wild type and ΔRTE full-length transcriptswere found mostly in the nucleus and the RTE deletion did not sig-nificantly affect their nuclear levels, suggesting the presence of a strongnuclear retention determinant that was partly counteracted by RTE. Infurther support, it was previously reported that expression of the gaggene from the related IAP M1A14 required the presence of the CTE(Wodrich et al., 2001). Similarly, we found that subgenomic transcriptscontaining the individual IAP 92L23 gag and pol genes were expressedonly poorly, while the presence of the cis-acting RNA export elementsCTE or RTE activated the protein expression (data not shown). Thesefindings confirm the notion that IAP gag/pol primary transcript, like thatof HIV gag/pol, is subject to analogous posttranscriptional restriction.

We further found that a portion of both the wild type and ΔRTEprimary transcript were subject to splicing, but the RTE deletion did not

90 A.S. Zolotukhin et al. / Virology 377 (2008) 88–99

affect the splicing efficiency (compare lanes 6 and 7, Fig. 3B). The splicedmRNAs were not retained in the nucleus, suggesting that splicing re-moved the responsible retention determinant. RT-PCR and sequencingof the spliced products revealed a splice donor and splice acceptorlocated at nt 546 and nt 5632, respectively, generating a non-codingsmall transcript lacking the gag/prt/pol region. In conclusion, RTE-D isrequired to promote cytoplasmic accumulation and expression of theLTR-IAPprimary transcript, and thus, thisfindingprovides themolecularbasis for the absolute requirement of RTE for retrotransposition.

RTE-D promotes heterologous reporter gene expression

We next tested the function of RTE-D in heterologous reportertranscripts. We inserted RTE-D into different RNA export reporterplasmids that are otherwise expression-defective, since they lack aposttranscriptional control element. In the HIV reporter, pNLgag(Felber et al., 1989), the gag gene is flanked by the major HIV splicedonor and a cryptic splice acceptor site, and protein can be onlyproduced from the unspliced mRNA. In this or similar HIV-gag basedreporter systems, the level of protein expression upon insertion of acis-acting RNA export element (e.g. RRE plus Rev, CTE or RTE) providesa quantitative measure of the posttranscriptional regulatory potencyof the cis">cis-acting element (Felber et al., 1989; Hammarskjold,

Fig. 1. LTR-retrotransposon IAP RP23-92L23 contains an RTE-D element. (A) Genomic strupolymerase (pol) genes, RTE element (RTE-D) and the polypurine tract (ppt) are indicatedstructural features (stem–loop (SL) I through SL IV) are indicated (bottom panel left). RTE-9(bottom panel right). Letters in red indicate divergent nucleotides in RTE-92L23 comparedelements. (B) Nucleotide sequence alignment of RTE-A and RTE-92L23. Divergent nucleotidindicated. (C) Multiple nucleotide sequence alignment shows conservation of RTE elemenindicate the hairpin loop sequences. Black shading indicates the nucleotides that are divergmobile LTR-IAPs (Dewannieux et al., 2004). (D) Neighbor-joining, unrooted phylogram showivalues are indicated. RTE-3061 corresponds to the prototype RTE-D sequence reported in S

2001; Smulevitch et al., 2006; Smulevitch et al., 2005; Tabernero et al.,1997; Tabernero et al., 1996; Wodrich et al., 2001; Wodrich,Schambach, and Krausslich, 2000).

We found that the presence of the RTE-containing fragment of92L23 (nt 5860–6921) led to a ~10-fold activation of Gag expressioncompared to the ‘empty’ pNLgag reporter (Fig. 4). RTE-D is morepotent than the related RTE-A in this assay (Fig. 4). Similar data wereobtained upon insertion of the RTE-D fragment into cat reportermRNA (DM138) (Huang et al., 1991), where the cat gene is insertedinto a modified HIV env fragment and flanked by the env splice sites,and CAT protein can be only expressed from the unspliced mRNA,analogous to the pNLgag (data not shown). We also found the RTE-D isactive in the splice site deficient gag reporter mRNA pNLCgag (Felberet al., 1989) (data not shown). Thus, RTE-D, similarly to the originallyidentified RTE-A (Nappi et al., 2001; Smulevitch et al., 2005) or the CTE(Bray et al., 1994; Tabernero et al., 1996), acts in heterologous settingsas a transferable posttranscriptional control element.

RBM15 activates expression of the RTE-D containing reporter mRNA

Since we previously identified RBM15 as a trans-acting factor thatfacilitates RTE-A activity, we asked whether RBM15 also acts on therelated RTE-D. Fig. 4 shows that cotransfection of RBM15 activated the

cture of LTR-IAP RP23-92L23. The long terminal repeat (LTR), gag, protease (prt) and. The structure of the RTE-A RNA from (Smulevitch et al., 2005) is shown and its key2L23 folds into a structure that is typical of the group D RTE (Smulevitch et al., 2005)to RTE-A, and asterisks depict compensatory changes preserving the shared structurees in RTE-92L23 are shown in red. Regions spanning the hairpin loops I through IV arets belonging to the subgroup D from intact IAP LTR-retroelements. Grey shaded areasent from the consensus. RTE-92L23 is underlined. Asterisks denote the RTEs from theng nucleotide sequence conservation of RTE-D elements in LTR-IAPs. Percent divergencemulevitch et al. (2005).

Fig. 1 (continued ).

91A.S. Zolotukhin et al. / Virology 377 (2008) 88–99

Fig. 2. Retrotransposition depends on posttranscriptional regulation. (A) RTE is essential for retrotransposition. IAP LTR-retrotransposons 92L23 is shown on top. For the mobilityassays, the wild type 92L23 reporter construct (92L23neoTNF) (Dewannieux et al., 2004) that contained the G418 selection cassettes inserted at nt 5673 used. We generated the ΔRTEmutant (92L23neoTNFΔRTE) as indicated. To the right, the resultant G418 selection from plates seeded with 106 cells are shown, and the number of clones provides a mobility count. Atypical experiment is presented, and similar results were obtained in several independent experiments. (B) RTE-A and CTE substitute for RTE-D inmobility assay. Insertion of RTE-A orCTESRV-1 into 92L23neoTNFΔRTE-D generated ΔRTE/RTE-A(+) and ΔRTE/CTE(+), respectively, and the resulting plasmids were tested for their retrotransposition ability as described inpanel A. Serial dilutions of the transfected cells were seeded into plates, subjected to selection. The G418-resistant colonies were counted in triplicate plates and the averageretrotransposition frequencies per 105 cells including standard deviations are presented. Also shown are the average values in percent to those obtained with thewild type construct.A typical experiment is shown, and similar data were obtained in three independent experiments.

92 A.S. Zolotukhin et al. / Virology 377 (2008) 88–99

expression of pNLgag transcripts that contained RTE-A as expected(Lindtner et al., 2006). Similarly, we found that gag expression of theRTE-D containing reporter is also increased by RBM15. We noted thatthe RTE-D fragment spanning nt 5860–6921 showed ~3 fold activationby RBM15, while the fragment spanning nt 6097–7324, that has lowerbasal level activity, showed ~10-fold activation. We concluded thatRTE-D and RTE-A share the trans-acting activator RBM15 as well as theunderlying mechanism of function as predicted from their conservedprimary and secondary structure.

Classification of active RTE-containing LTR-retroelements

We next studied the relationship of RTE-containing LTR-IAPs to theknown transposon categories. LTR-retrotransposons of M. musculuswere previously categorized into 21 distinct families that form threeclasses (McCarthy and McDonald, 2004), based on amino acid ho-mologies between reverse transcriptase proteins (Fig. 5). One suchfamily, Mmr10_IAP, belongs to a clade comprising of seven additionalfamilies that are closely related (Fig. 5, asterisk), and members of any

Fig. 3. RTE-D is essential for the IAP gag/pol mRNA export and expression. (A) Nuclear and cytoplasmic mRNA was extracted from 293 cells transfected with 4 µg of the IAPplasmids plus 1 µg of the GFP plasmid FRED25 and analyzed by RT-PCR using primers located in IAP gag (primers s1 and a1 shown in the cartoon below) and with primers in GFPto amplify the gag and GFP mRNAs, respectively. As control, the PCR reactions were run with samples to which the reverse transcriptase (RT) was not added, as shown to the left.Both the regular (25 cycles) and the extended (32 cycles) RT-PCR analyses were performed as indicated. Nuc and cyto indicate nuclear and cytoplasmic enriched subcellularfractions. (B) Northern blot analysis of cells transfected with 1 µg of the IAP plasmids plus 0.2 µg of the GFP plasmid FRED25. The blots were hybridized with a probe spanning the3′ end of the transcript (see panel A). Short and long exposures of the portion of the blot containing the cytoplasmic fractions are shown. The location of the unspliced and thespliced transcripts are indicated. The blot was rehybridized with a probe for GFP. (C) Human 293 cells were transfected with 10 µg of the plasmids expressing the authentic 92L23(wt) or its ΔRTE mutant, or mock transfected as indicated. The GFP expression plasmid pFRED25 (0.3 µg) was cotransfected as internal control. Two days later, the cells wereharvested and the IAP-Gag proteins were analyzed on immunoblot using IAP-Gag antiserum (αMIA-gag), and the same blots were subsequently reprobed with GFP antibodies asindicated. Arrowheads show the positions of processed IAP-Gag proteins. On the graph to the right, the quantificated signals fromwestern blots are plotted on the X-axis, as rawpixel intensities (arbitrary units). Bold numbers show the gag expression values that were normalized to those of GFP.

93A.S. Zolotukhin et al. / Virology 377 (2008) 88–99

of the eight families have been described as IAP by various authors(McCarthy and McDonald, 2004).

We compared the sequencesof themobile clonesRP23-92L23, RP23-231P12, RP23-324C9 and RP23-440N1 (Dewannieux et al., 2004) tothose of representative eight IAP families and found that they belongexclusively to Mmr10_IAP (Fig. 5). Comparison of the pol proteins ofthese retrotransposons to those of the prototypeMmr10_IAP (McCarthyand McDonald, 2004) showed a compelling ~99% homology (Fig. 5),whereas those of other Mmr families were far more distant (data notshown). In further support of this assignment, we found that theprototypeMmr10_IAP (McCarthy andMcDonald, 2004) also contains anRTE that shares 95–96% sequence identity with the RTEs of the knownmobile LTR-IAPs (Fig. 1C). Together, these data place the fourmobile IAPLTR-retrotransposons RP23-92L23, RP23-231P12, RP23-324C9 andRP23-440N1 unambiguously into the Mmr10_IAP family (Fig. 5). Se-quences representative of otherMmr families (McCarthy andMcDonald,2004) did not show significant homology to RTE, indicating that RTE-Duniquely belongs to Mmr10_IAP LTR-retrotransposons.

Additional RTE-containing LTR-retrotransposons in the mouse genome

The assignment of RTE-D containing LTR-retrotransposons to theMmr10_IAP family (Fig. 5) was based on the phylogeny drawn fromreverse transcriptase amino acid homology, and hence, it was limitedto the species that possessed the RTcoding sequence. However, a largenumber of LTR-retrotransposons and their derived sequences containlarge deletions affecting the RT open reading frame, and thus wereexcluded from these analyses. Therefore, we undertook a comprehen-sive screen of the mouse genome for loci that had properties expectedof recently integrated RTE-containing LTR-retroelements, but did notnecessarily contain the recognizable RT coding sequences.

We previously reported the presence of more than 100 elementsidentical to the originally described RTE-A andmore than 3000 closelyrelated elements (RTE belonging to groups B, C, and D, see Fig. 6A, toppanel) with at least 70% sequence identity, distributed on all mousechromosomes (Smulevitch et al., 2005). We compiled the representa-tive samplings of RTEs of each group, which reflected the sequence

Fig. 4. RBM15 activates RTE-A and RTE-D. pNLgag reporter (0.2 µg) containing RTE-A orRTE-D were transfected into HLtat cells in the absence or presence of 0.25 µg RBM15expression plasmid. Two days later, the cell extracts were analyzed for Gag and GFPproduction and the data are displayed with standard deviation (SD). The fold inductionin the presence of RBM15 is indicated.

94 A.S. Zolotukhin et al. / Virology 377 (2008) 88–99

diversitywithin such groups (Fig. 6A, initial RTE homology sequences),and these samplings were then used to map the matching loci in themouse genome (Fig. 6A, RTEmatches). In order to identify the putativeRTE-containing LTR-retrotransposons, our queries required that (i) themapped RTE loci (Fig. 6A) must be flanked by two LTRs of the sametype and in the same orientation, and (ii) these LTRs should bewithina distance of up to 12 kb. Such queries identified about 250 RTE-containing putative LTR-retrotransposons (Fig. 6A, retrotransposoncandidates). Strikingly, we found that RTEs of the different groupswere linked in an obviously nonrandommanner to distinct LTR types(Fig. 6A, LTR type, named according to their prototypic sequencesin Repbase). Thus, the group A RTE associated exclusively withRLTR10, whereas 90–95% of the group C and group D RTE werepreferentially associated with IAPLTR2-MM and IAPLTR1-MM,respectively. Interestingly, 92% of the identified RTE-D-containingloci were highly homologous to an LTR-IAP internal sequence of thetype IAPLTR1a_MM-int located between their LTRs. We found that

Fig. 5. Phylogenic analysis of the IAP LTR-retrotransposons containing RTE-D. DendrogramMcDonald, 2004) is shown (left). The assignment of RTE-containing IAP LTR-retrotransposonthe four mobile IAP LTR-retrotransposons to the prototype Mmr10_IAP clone. Asterisk indic

the closest match of the IAPLTR1a_MM-int prototype (81% identity,1.8% insertions and 0.7% deletions) was to Mmr10_IAP (Fig. 5),establishing that the identified candidate loci indeed represent theMmr10_IAP retrotransposons, and further verifying the validity ofour approach. Only one of the RTE-C-containing loci showed internalmatches to retroelement-like sequences (data not shown), likely dueto a high degree of truncation of these IAPs (see also Fig. 6B). Wenoticed that a large proportion of RTE-A-containing candidates wereof complex descent, combining the retroelement-like integrase, LINE,RTE-A and CTE-like sequences in one cassette (corresponding toRLTR10-int in Repbase) that were flanked by RLTR10 LTRs (see alsoNappi et al., 2001).We did not find any association of RTE-Bwith LTR-retrotransposons, likely because of its overallmuch lower abundance.In summary, the RTE-containing LTR-retrotransposons fall into threegroups, characterized by the presence of distinct RTEs associatedwith distinct LTRs, suggesting a complex evolutionary history.

To obtain an independent validation of our findings, we back-mapped the identified Repbase prototypes of RTE-containing candi-date retrotransposons onto mouse genome. This analysis resulted inthe recovery of the same number of hits (Fig. 6A, back-mapping) asobtained with our initial queries (Fig. 6A, retrotransposon candi-dates), strongly suggesting that our candidate loci are authenticretrotransposons. By studying their length distributions, we foundthat the RTE-A-containing retrotransposons peaked at ~3.0 kb,closely matching the length of the Repbase prototype LTR-retro-transposon (RLTR10-int plus two RLTR10 type LTRs), as expected(Fig. 6B). The RTE-C-containing loci also peaked at ~3.0 kb, andsince they did not display significant internal matches to the knownsequences (data not shown), we speculate that they representedhighly truncated retrotransposon-derived elements. In contrast, theRTE-D-containing sequences showed two peaks at ~5.2 and ~7.0 kb,closely matching the length of the deleted IAP LTR-retrotransposontype IΔ1 to IΔ4 and the full-length Mmr10_IAP retrotransposons,respectively (Fig. 6B).

To trace independently the full-length Mmr10_IAP, we back-mapped the autonomously mobile family member 92L23 sequence(7084 nt) onto themouse genome (Fig. 6C) and found 196 high-scoringsequences. Their length distribution exhibited a compound peak at~6.5 to 7 kb, in agreement with the data shown in Fig. 6B, indicating ahigh incidence of full-length or nearly full-length Mmr10_IAPs.

To address the questionwhether the identified RTE-containing locihad amplified via retrotransposition or gene duplication, we studiedtheir chromosomal distribution. As shown in Fig. 6D, we found thatthese candidates, and hence the associated RTEs, are found in allchromosomes, except the Y chromosome. Thus, retrotransposition is

based on reverse transcriptase amino acid sequence (modified from McCarthy ands to the Mmr10_IAP family is illustrated in a table (right), presenting the homologies ofates the IAP clade.

95A.S. Zolotukhin et al. / Virology 377 (2008) 88–99

themost likelymechanism that led to the spread of the RTE-containingLTR-retrotransposons.

Together with our phylogenetic data, these results revealed a novelclass of murine retrotransposons that are characterized by the presenceof RTE-like posttranscriptional regulatory element.

Discussion

Mobile elements have a major impact on genome structure, evo-lution and expression, and their mobility contributes to the geneticdiversity. LTR-retrotransposons and their related sequences account forabout 10% of the mouse genome (Maksakova et al., 2006; Waterstonet al., 2002), and also constitute a significant portion of the humangenome (for recent reviews see Bannert and Kurth, 2004; Bannertand Kurth, 2006). To gain further insight into the biological role of retro-elements, it is important to understand the mechanisms by which theyregulate their own expression in order to support efficient propagation.

RTE is essential for the mobility of IAP LTR-retrotransposons

Here, we report that mouse IAP LTR-retrotransposons encode cis-acting RNA export signals that are essential for their mobility. As amodel, we studied the autonomously mobile IAP LTR-retrotransposon92L23 (Dewannieux et al., 2004) and found that it contains a post-transcriptional regulatory element related to RTE-A (Nappi et al., 2001)and belongs to a previously described structural RTE subgroup D(Smulevitch et al., 2005). RTE-D is absolutely required for the LTR-IAPmobility, but can be replaced by CTE, a direct NXF1 binding ligand, orRTE-A, the ligand for RBM15 (Lindtner et al., 2006). Both CTE and RTEact via the NXF1 pathway (Lindtner et al., 2006) and serve as con-stitutive RNA export signals, allowing nuclear export prior to andindependently of splicing. RBM15 protein is thought to provide a directlink between RTE-A and the nuclear export receptor NXF1 (Lindtneret al., 2006).In agreement with this, we found that RBM15 promotesRTE-D mediated reporter gene expression, as we have previouslyshown for RTE-A (Lindtner et al., 2006).

RTE-D and IAPE are distinct elements

Another posttranscriptional regulatory element, the “IAP expressionelement” (IAPE), was previously identifiedwithin the 3′ portion of the polgene of the murine MIA14 LTR-IAP (Wodrich et al., 2001) and was shownto promote expression of HIV gag in a heterologous reporter assay.Sequence comparison revealed amatching regionwithin 92L23 (nt 4566–6267)with 98% identity compared to the IAPE ofMIA14. Importantly, IAPEand RTE represent two distinct regions within the LTR-retrotransposon.While both elements display posttranscriptional regulatory function, theycould contribute independently to retrotransposon expression. It isnoteworthy, that the presence of the IAPE within the RTE-deleted92L23neoTNFΔRTE was not sufficient to mediate retrotransposition (seeFig. 2). Thus, our data suggest that RTE-D is the major posttranscriptionalregulatory element in the IAP LTR-retrotransposons.

Retroviruses and retroelements depend on posttranscriptional regulation

Taken together, (i) the absolute requirement of RTE for LTR-IAPmobility, (ii) the reduced cytoplasmic levels of the RTE-deleted IAPtranscript, (iii) the ability of CTE and RTE to increase IAP LTR-retrotransposon expression, and (iv) the inability of the IAP gag or polgene to be expressed in the absence of a cis">cis-acting posttranscrip-tional control element, demonstrate that posttranscriptional regulation isessential for the LTR-IAP lifecycle. Thisfinding further groups the LTR-IAPretrotransposons together with retroviruses such as HIV-1 (Feinberg etal., 1986; Felber et al., 1989; Hadzopoulou-Cladaras et al., 1989; Sodroskiet al., 1986), the HTLV retroviruses (Hidaka et al., 1988; Inoue, Seiki, andYoshida, 1986; Inoue, Yoshida, and Seiki, 1987; Seiki et al., 1988), the

human endogenous retrovirusHTDV/HERV-K (Bogerd et al., 2000; Loweret al.,1995;Magin et al., 2000;Magin, Lower, and Lower,1999; Yanget al.,2000; Yang et al.,1999), themousemammary tumor virus (MMTV) (Bar-Sinai et al., 2005; Indik et al., 2005; Mertz et al., 2005) as well as SRV/MPMV(Ernst et al.,1997; Tabernero et al.,1997) forwhichdependencyonposttranscriptional regulation for their life cycle has been demonstrated.In all cases, it has been demonstrated that export and expression of theunspliced transcript requires a positive-acting regulatory system. In HIV,this step is executed via the viral Rev–RRE system (Fig. 7). In othercomplex retroviruses, this step is executed by theviral factors such as Rexin HTLV, Rec in HTDV/HERV-K and Rem in MMTV. In contrast, in SRV-1/MPMV/SRV-2 and in the IAP LTR-retroelements, this step is controlled bycellular factors interactingwithviral cis">cis-actingRNAelements (NXF1-CTE and RBM15-RTE, respectively) (Fig. 7).

Our understanding of the reason for the dependency on posttran-scriptional regulation is based on the findings that many retroviraltranscripts contain RNA signals that negatively affect the trafficking ofthe transcript in the absence of the positive-acting regulatory systems.The presence of such signals is responsible for the posttranscriptionalfate of the transcript in the absence of positive regulation,which can bemanifested at different steps as poor export, oversplicing, low stability,and poor translation (Cochrane, McNally, and Mouland, 2006; Felber,Zolotukhin, and Pavlakis, 2007; Stoltzfus and Madsen, 2006). It ispossible that the nuclear retention determinants of the IAP LTR areanalogous to the previously described INS/CRS elements of HIV. Insupport of this idea, we found that these determinants are transferableand confer expression defects when present in heterologous reportertranscripts, acting in the same fashion as HIV INS elements (data notshown). Thus, the presence of negative-acting RNA signals is func-tionally conserved inwidely divergent retroviruses fromHIV and SRV-1/MPMV to IAP LTR-retrotransposons,which further corroborates theirbiological importance. Therefore RNA elements like RRE, CTE, and RTEprovide an essential function to the retroviral transcript. Interestingly,we also found that inactive retroelements that lack intact gag/polgenes preserved the presence of active RTE or CTE. We previouslyidentified the CTEORG, a CTE-related element located within the ORG-IAP retroelement (Tabernero et al., 1997) (Fig. 7) that shares with SRV-1/MPMV CTE the secondary structure and NXF-1 binding sites, and isan active RNA export element (Tabernero et al., 1997; von Gegerfeltet al., 2006). In this report, we also found that the highly truncated IAPLTR-retroelements types IΔ1–IΔ4 (Ishihara et al., 2004) preservedRTE-D. Thus, these truncated elements contain the minimal set of cis">cis-acting signals that support retrotransposition, enabling their mobilityvia protein trans-complementation (see references inMaksakova et al.,2006). Although the region responsible for nuclear retention of thefull-length LTR-IAP RNA (nt 546–5632 of 92L23) is largely deleted inIΔ1–IΔ4, even the shortest transposon, IΔ4, still retains two stretchesof homology (nt 564–981 and 1620–1956 of 92L23) and these regionstogether comprise 32% of IΔ4′s transcript length. It is therefore plau-sible that such sequences in IΔ4 are sufficient to confer nuclear re-tention, explaining the preservation of RTE.

In conclusion, similarly to infectious retroviruses, the LTR-IAPretrotransposons can utilize distinct, structurally unrelated but func-tionally analogous RNA export elements (RTE and CTE). These findingsillustrate the likely convergent evolution of retroelements towardsacquisition of functionally analogous RNA regulatory signals, furthersupporting the view that the evolution of retrotransposons may haveinvolved acquisition and exchange of pre-existing genetic modules(Coffin, Hughes, and Varmus, 1997).

Materials and methods

Plasmids

The following full-length LTR-IAP clones have been described:RP23-231P12, RP23-324C9, RP23-440N1 and RP23-92L23

96 A.S. Zolotukhin et al. / Virology 377 (2008) 88–99

Fig. 7. Retrovirus and retroelement expression depend on posttranscriptional regulation. Structures of HIV-1, SRV-1/MPMV, ORG-IAP retroelement, IAP LTR-retroelement and itsdeleted mutant types IIΔ4 are shown. There is functional conservation of negative-acting RNA elements (filled boxes) embedded within gag/pol of all retroviruses and retroelements.Only complex retroviruses like HIV-1 contain such determinants within env. Because the ORG-IAP retroelement is severely mutated it was not further analyzed, and the indicatedputative negative-acting sequences (indicated in grey) are shown by analogy to the other retroviruses and retroelements having active RNA export signals. These retroviruses/retroelements depend on the presence of a positive-acting mechanism and contain RNA export elements located upstream of the ppt and 3′ LTR like RRE, CTE, CTEORG, RTE-A, RTE-D.The RNA export factor can be viral (Rev for RRE) or cellular (NXF1 of CTE and CTEORG; RBM15-NXF1 for RTE).

97A.S. Zolotukhin et al. / Virology 377 (2008) 88–99

(Dewannieux et al., 2004),MIA14 (GenBank accession no.M17551), theIAP clone Q14 (GenBank accession no. AB09818) (Ishihara et al., 2004),and the IAP mutants type IΔ1 to IΔ4 (Ishihara et al., 2004). For RP23-92L23 constructs, the numbering starts from the first nucleotide of its5′ LTR (nt 161601; GenBank accession no. AC012382). The RTE-92L23 islocated between nt 6446–6687. ΔRTE-92L23 has nt 6269–6689replaced by a polylinker. The originally identified RTE-A spans nt391–616 (GenBank accession no. AF250999). TheRTE-DprototypeRTE-3061 has been identified between nt 34124–34365 in GenBankaccession no. AC003061 (Smulevitch et al., 2005) and is located withinan intact IAP (nt 27673–34761). RTE-MIA14 is located between nt6460–6701 (GenBank accession no. M17551) isolated from clonepMIA60 (Mietz et al., 1987) (a gift from K. Lueders). We resequencedthis region of MIA14 and noted the following changes: T6569 deletion;G6577C; C6580T; T6582C; insertion of G between 6602 and 6603;G6618 deletion; C6643 deletion; G6672 deletion. The reporterplasmids pNLgag and pNLCgag were described previously (Felberet al., 1989). Briefly, pNLgag consists of the HIV-1 5′ LTR and the 5′untranslated region containing the major splice donor of HIV-1followed by the gag gene (nt 1–2621 of NL4-3 starting with the firstnt ofU3 as +1) and theHIV-1 sequence fromnt 8886 (XhoI) to endof the3′ LTR (nt 9709) including a cryptic splice acceptor. This plasmidproduces an unspliced gag-encoding and spliced non-coding mRNA(Felber et al., 1989). pNLCgag lacks the splice donor and produces only

Fig. 6. RTE-containing LTR-retrotransposons identified in the mouse genome. (A) The secSmulevitch et al., 2005) are presented on top. The table below indicates for each RTE group thsequences); the number of RTE homology loci mapped in the mouse genome (RTE matches);candidates) which are further split into subtypes (Subtype) according to their associated LTRthe Repbase prototype sequences onto mouse genome. (B) The histogram shows the lretrotransposons. Relative frequencies are plotted on Y-axis. (C) The histogram shows the lenas in panel B. (D) Chromosomal distribution of the LTR-retrotransposons from panel A. The

unspliced gag-encodingmRNA (Felber et al., 1989). The GFP expressionplasmids pFRED25 and pFRED143 (Stauber et al., 1998) were describedpreviously. The IAP-92L23 gag and pol coding sequences were PCR-amplified usingprimers that added a 3XFLAG tag at the C-terminus andwere inserted into pCMVkan (Rosati et al., 2005) between the CMVpromoter and the bovine growth hormone (BGH) polyadenylationsignal. The inserted pol gene contains an optimized translationinitiation signal. The CMV promoter (SpeI–SacII) was replaced by theHIV-1 LTR to generate LTR-IAP-92L23 gag and pol expression plasmids.The IAPE-containing regions of MIA14 (nt 4684–6195) (Wodrich et al.,2001) and the corresponding region from RP23-92L23 (nt 4566–6267)as well as the RTEs of 92L23, MIA14, 3061 were inserted into pNLgagbetween the HIV-1 gag gene and the cryptic splice acceptor, or into asimilar location within pNLCgag plasmid. The RBM15 expressionplasmid was described (Lindtner et al., 2006).

Transfection and expression analysis

For mobility assays, HeLa cells were grown in 60 mm plates andtransfected with 4 µg of the IAP molecular clones as described(Dewannieux et al., 2004). The cells were seeded in 100 mm platesat several dilutions ranging from 5×105 to 104 followed by G418selection. After 10 days the cells were fixed and the colonies countedusing 5×105 plate for the inactive IAP and the 104 plate for the active IAP.

ondary structures of the previously identified RTEs (group A, B, C, D; modified frome number of representative sequences used for genomic mapping (Initial RTE homologyand the number of the RTE-containing putative LTR-retrotransposons (Retrotransposonsequences (LTR type). Back-mapping, the number of hits recovered by back-mapping ofength distribution (plotted on X-axis, in nt) of the candidate RTE-containing LTR-gth distribution of high-scoring matches to IAP 92L23 in the mouse genome, presentednumber of matches is plotted on the Y-axis.

98 A.S. Zolotukhin et al. / Virology 377 (2008) 88–99

The HeLa derived cell line HLtat that constitutively produces HIVtat, or human 293 cells, were transiently transfected with HIV gagreporter plasmids using Superfect (Qiagen) or calcium coprecipitationtechnique, respectively, as described previously (Lindtner et al., 2006).For 293 cells a tat expression plasmid pBstat was cotransfected.Two days later, the cells were harvested and analyzed for protein orRNA expression. HIV gag expression was monitored using a p24gagantigen capture assay (Zeptometrix). GFP was measured as described(Lindtner et al., 2006). IAP expression was monitored on Westernimmunoblot using a rabbit anti-MIA14 gag antiserum 3607.X (1:1000dilution) (Mietz et al., 1987) (a gift from K. Lueders), followed by HRP-conjugated anti-rabbit antiserum (dilution of 1:10,000) and detectedby Enhanced Chemi-Luminescence (ECLplus; Amersham, GE Health-care). FLAG-tagged gag and pol were detected using anti-FLAG anti-body (dilution of 1:10,000) followed by HRP-conjugated anti-mouseantiserum (dilution of 1:20,000) and detected by ECLplus. Quantifica-tion of Western blots was performed by using NIH image software.GFP fluorescence was measured as in Bear et al. (1999). RNA wasisolated from the nuclear and cytoplasmic fractions (Zolotukhin et al.,2002) and subjected to semi-quantitative RT-PCR using primers lo-cated within gag (s1, nt 1172–1191; a1, nt 1399–1418). Northern blotanalysis was used to detect IAP retroelement transcripts using a probespanning the 3′ end of the transcript (nt 5673–7096). The splicedtranscript was identified by RT-PCR and sequencing.

Bioinformatics analyses

The mouse retroviral LTRs and the complete genomic mouseretroviral sequences were collected from RepeatMasker (genome.washington.edu; Repbase v 8.6) andGenbank (NCBI). They included 72complete retroviral genomic sequences (musretrovir (n=49) fromRepBase and MouseRetroviruses (n=23) from GenBank) and of 281solitary retroviral LTRs (musltr (n=215) from RepBase and MouseLTR(n=66) from GenBank).

For genomic back-mapping and the final extraction of the “retroviralcandidate” retroviral candidate” sequences,we used the completemousegenome sequence assembly from the !?tlsb-.05pt?>ENSEMBL database(ENSEMBL_golden-path: ENSEMBL_DB:mus_musculus_core_19_30,NCBIM30). The back-mapping of the previously described (Smulevitchet al., 2005) RTE of groups A, B, C and D and the mouse LTR/mouseretroviral genome sequences was performed with the Sequence Searchand Alignment by Hashing Algorithm (SSAHA; Ning, Cox, and Mullikin,2001) and a local installation of the ENSEMBL database (Birney et al.,2004).

Multiple sequence alignments and phylogenetic analyses wereperformed using ClustalW tools at European Bioinformatics Insti-tute (http://www.ebi.ac.uk). Back-mapping of Repbase retroelementprototypes and the known LTR-retrotransposon sequences onto mousegenomewas performed using BLAT utility in UCSC Genome Browser. RNAfoldings were performed as described (Zolotukhin et al., 2001).

Acknowledgments

We thank S. Lindtner for comments, our Werner H. Kirsten StudentIntern program recipients L. Kotani, T. Hudzik, and S. Sadtler for theircontributions, K. Lueders for reagents anddiscussions, T. Heidmann,H.G.Krausslich, V. Kulkarni, and M. Rosati for reagents, G-M. Zhang fortechnical assistance, and T. Jones for editorial assistance.

References

Bannert, N., Kurth, R., 2004. Retroelements and the human genome: new perspectiveson an old relation. Proc. Natl. Acad. Sci. U. S. A. 101 (Suppl 2), 14572–14579.

Bannert, N., Kurth, R., 2006. The evolutionary dynamics of human endogenousretroviral families. Annu. Rev. Genomics Hum. Genet. 7, 149–173.

Bar-Sinai, A., Bassa, N., Fischette,M., Gottesman,M.M., Love,D.C., Hanover, J.A.,Hochman, J.,2005. Mouse mammary tumor virus Env-derived peptide associates with nucleolar

targets in lymphoma, mammary carcinoma, and human breast cancer. Cancer Res. 65,7223–7230.

Bear, J., Tan, W., Zolotukhin, A.S., Tabernero, C., Hudson, E.A., Felber, B.K., 1999.Identification of novel import and export signals of human TAP, the protein thatbinds to the constitutive transport element of the type D retrovirus mRNAs.Mol. Cell Biol. 19, 6306–6317.

Birney, E., Andrews, T.D., Bevan, P., Caccamo, M., Chen, Y., Clarke, L., Coates, G., Cuff, J.,Curwen, V., Cutts, T., Down, T., Eyras, E., Fernandez-Suarez, X.M., Gane, P., Gibbins, B.,Gilbert, J., Hammond, M., Hotz, H.R., Iyer, V., Jekosch, K., Kahari, A., Kasprzyk, A.,Keefe, D., Keenan, S., Lehvaslaiho, H., McVicker, G., Melsopp, C., Meidl, P., Mongin, E.,Pettett, R., Potter, S., Proctor, G., Rae, M., Searle, S., Slater, G., Smedley, D., Smith, J.,Spooner, W., Stabenau, A., Stalker, J., Storey, R., Ureta-Vidal, A., Woodwark, K.C.,Cameron, G., Durbin, R., Cox, A., Hubbard, T., Clamp, M., 2004. An overview ofEnsembl. Genome Res. 14, 925–928.

Bogerd, H.P., Wiegand, H.L., Yang, J., Cullen, B.R., 2000. Mutational definition of functionaldomainswithin the Revhomolog encoded byhuman endogenous retrovirus K. J. Virol.74, 9353–9361.

Bray, M., Prasad, S., Dubay, J.W., Hunter, E., Jeang, K.-T., Rekosh, D., Hammarskjold, M.-L.,1994. A small element from the Mason–Pfizer monkey virus genomemakes humanimmunodeficiency virus type 1 expression and replication Rev-independent. Proc.Natl. Acad. Sci. U. S. A. 91, 1256–1260.

Cochrane, A., 2004. Controlling HIV-1 Rev function. Curr. Drug Targets Immune Endocr.Metabol. Disord. 4, 287–295.

Cochrane, A.W., McNally, M.T., Mouland, A.J., 2006. The retrovirus RNA traffickinggranule: from birth to maturity. Retrovirology 3, 18–34.

Coffin, J.M., Hughes, S.H., Varmus, H., 1997. “Retroviruses”. Cold Spring Harbor LaboratoryPress, Plainview, N.Y.

Cullen, B.R., 2003. NuclearmRNA export: insights fromvirology. Trends Biochem. Sci. 28,419–424.

Dewannieux, M., Dupressoir, A., Harper, F., Pierron, G., Heidmann, T., 2004. Identifica-tion of autonomous IAP LTR retrotransposons mobile in mammalian cells. Nat.Genet. 36, 534–539.

Ernst, R.K., Bray, M., Rekosh, D., Hammarskjold, M.-L., 1997. A structured retroviral RNAelement that mediates nucleocytoplasmic export of intron-containing RNA. Mol.Cell Biol. 17, 135–144.

Feinberg, M.B., Jarrett, R.F., Aldovini, A., Gallo, R.C., Wong-Staal, F., 1986. HTLV-IIIexpression and production involve complex regulation at the levels of splicing andtranslation of viral RNA. Cell 46, 807–817.

Felber, B.K., Hadzopoulou-Cladaras, M., Cladaras, C., Copeland, T., Pavlakis, G.N., 1989.rev protein of human immunodeficiency virus type 1 affects the stability andtransport of the viral mRNA. Proc. Natl. Acad. Sci. U. S. A. 86, 1495–1499.

Felber, B.K., Zolotukhin, A.S., Pavlakis, G.N., 2007. Posttranscriptional control ofHIV-1 and other retroviruses and its practical applications. Adv. Pharmacol. 55,161–197.

Grüter, P., Tabernero, C., von Kobbe, C., Schmitt, C., Saavedra, C., Bachi, A., Wilm, M.,Felber, B.K., Izaurralde, E., 1998. TAP, the human homolog of Mex67p, mediates CTE-dependent RNA export from the nucleus. Mol. Cell 1, 649–659.

Hadzopoulou-Cladaras, M., Felber, B.K., Cladaras, C., Athanassopoulos, A., Tse, A.,Pavlakis, G.N., 1989. The rev (trs/art) protein of human immunodeficiency virustype 1 affects viral mRNA and protein expression via a cis">cis-acting sequence inthe env region. J. Virol. 63, 1265–1274.

Hammarskjold, M.L., 2001. Constitutive transport element-mediated nuclear export.Curr. Top. Microbiol. Immunol. 259, 77–93.

Herold, A., Klymenko, T., Izaurralde, E., 2001. NXF1/p15 heterodimers are essential formRNA nuclear export in Drosophila. RNA 7, 1768–1780.

Hidaka, M., Inoue, J., Yoshida, M., Seiki, M., 1988. Post-transcriptional regulator (rex) ofHTLV-1 initiates expression of viral structural proteins but suppresses expression ofregulatory proteins. EMBO J. 7, 519–523.

Huang, X.J., Hope, T.J., Bond, B.L., McDonald, D., Grahl, K., Parslow, T.G., 1991. MinimalRev-response element for type 1 human immunodeficiency virus. J. Virol. 65,2131–2134.

Indik, S., Gunzburg, W.H., Salmons, B., Rouault, F., 2005. A novel, mouse mammarytumor virus encoded protein with Rev-like properties. Virology 337, 1–6.

Inoue, J., Seiki, M., Yoshida, M., 1986. The second pX product p27 chi-III of HTLV-1 isrequired for gag gene expression. FEBS Lett. 209, 187–190.

Inoue, J., Yoshida, M., Seiki, M., 1987. Transcriptional (p40X) and post-transcriptional(p27X-III) regulators are required for the expression and replication of human T-cellleukemia virus type I genes. Proc. Natl. Acad. Sci. U. S. A. 84, 3653–3657.

Ishihara, H., Tanaka, I., Wan, H., Nojima, K., Yoshida, K., 2004. Retrotransposition oflimited deletion type of intracisternal A-particle elements in the myeloid leukemiaClls of C3H/He mice. J. Radiat. Res. (Tokyo) 45 (1), 25–32.

Kutay, U., Guttinger, S., 2005. Leucine-rich nuclear-export signals: born to be weak.Trends Cell Biol. 15, 121–124.

Lindtner, S.L., Zolotukhin, A.S., Uranishi, H., Bear, J., Kulkarni, V., Smulevitch, S.,Samiotaki, M., Panayotou, G., Pavlakis, G.N., Felber, B.K., 2006. RNA-binding motifprotein 15 binds to the RNA transport element RTE and provides a direct link to theNXF1 export pathway. J. Biol. Chem. 281, 36915–36928.

Lower, R., Tonjes, R.R., Korbmacher, C., Kurth, R., Lower, J., 1995. Identification of a Rev-related protein by analysis of spliced transcripts of the human endogenousretroviruses HTDV/HERV-K. J. Virol. 69, 141–149.

Magin, C., Hesse, J., Lower, J., Lower, R., 2000. Corf, the Rev/Rex homologue of HTDV/HERV-K, encodes an arginine-rich nuclear localization signal that exerts a trans-dominant phenotype when mutated. Virology 274, 11–16.

Magin, C., Lower, R., Lower, J., 1999. cORF and RcRE, the Rev/Rex and RRE/RxRE ho-mologues of the human endogenous retrovirus family HTDV/HERV-K. J. Virol. 73,9496–9507.

99A.S. Zolotukhin et al. / Virology 377 (2008) 88–99

Maksakova, I.A., Romanish, M.T., Gagnier, L., Dunn, C.A., van de Lagemaat, L.N., Mager, D.L.,2006. Retroviral elements and their hosts: insertionalmutagenesis in themouse germline. PLoS Genet. 2, e2.

McCarthy, E.M., McDonald, J.F., 2004. Long terminal repeat retrotransposons of Musmusculus. Genome Biol. 5, R14.

Mertz, J.A., Simper, M.S., Lozano, M.M., Payne, S.M., Dudley, J.P., 2005. Mouse mammarytumor virus encodes a self-regulatory RNA export protein and is a complex retrovirus.J. Virol. 79, 14737–14747.

Mietz, J.A., Grossman, Z., Lueders, K.K., Kuff, E.L., 1987. Nucleotide sequence of acompletemouse intracisternal A-particle genome: relationship to known aspects ofparticle assembly and function. J. Virol. 61, 3020–3029.

Nappi, F., Schneider, R., Zolotukhin, A., Smulevitch, S., Michalowski, D., Bear, J., Felber, B.,Pavlakis, G., 2001. Identification of a novel posttranscriptional regulatory elementusing a rev and RRE mutated HIV-1 DNA proviral clone as a molecular trap. J. Virol.75, 4558–4569.

Ning, Z., Cox, A.J., Mullikin, J.C., 2001. SSAHA: a fast search method for large DNAdatabases. Genome Res. 11, 1725–1729.

Rosati, M., von Gegerfelt, A., Roth, P., Alicea, C., Valentin, A., Robert-Guroff, M., Venzon,D., Montefiori, D.C., Markham, P., Felber, B.K., Pavlakis, G.N., 2005. DNA vaccinesexpressing different forms of simian immunodeficiency virus antigens decreaseviremia upon SIVmac251 challenge. J. Virol. 79, 8480–8492.

Segref, A., Sharma, K., Doye, V., Hellwig, A., Huber, J., Luhrmann, R., Hurt, E., 1997.Mex67p, a novel factor for nuclear mRNA export, binds to both poly(A)+ RNA andnuclear pores. EMBO J. 16, 3256–3271.

Seiki, M., Inoue, J., Hidaka, M., Yoshida, M., 1988. Two cis">cis-acting elementsresponsible for posttranscriptional trans-regulation of gene expression of humanT-cell leukemia virus type I. Proc. Natl. Acad. Sci. U. S. A. 85, 7124–7128.

Smulevitch, S., Michalowski, D., Zolotukhin, A.S., Schneider, R., Bear, J., Roth, P., Pavlakis,G.N., Felber, B.K., 2005. Structural and functional analysis of the RNA transportelement, a member of an extensive family present in themouse genome. J. Virol. 79,2356–2365.

Smulevitch, S., Bear, J., Alicea, C., Rosati, M., Jalah, R., Zolotukhin, A.S., von Gegerfelt, A.S.,Michalowski, D., Moroni, C., Pavlakis, G.N., Felber, B.K., 2006. RTE and CTE mRNAexport elements synergistically increase expression of unstable, Rev-dependentHIV and SIV mRNAs. Retrovirology 3, 6–14.

Sodroski, J., Goh, W.C., Rosen, C., Dayton, A., Terwilliger, E., Haseltine, W., 1986. A secondpost-transcriptional trans-activator gene required for HTLV-III replication. Nature321, 412–417.

Stauber, R.H., Horie, K., Carney, P., Hudson, E.A., Tarasova, N.I., Gaitanaris, G.A., Pavlakis,G.N., 1998. Development and applications of enhanced green fluorescent proteinmutants. Biotechniques 24, 462–466 468–471.

Stoltzfus, C.M., Madsen, J.M., 2006. Role of viral splicing elements and cellular RNA bindingproteins in regulation of HIV-1 alternative RNA splicing. Curr. HIV Res. 4, 43–55.

Tabernero, C., Zolotukhin, A.S., Valentin, A., Pavlakis, G.N., Felber, B.K., 1996. Theposttranscriptional control element of the simian retrovirus type 1 forms anextensive RNA secondary structure necessary for its function. J. Virol. 70, 5998–6011.

Tabernero, C., Zolotukhin, A.S., Bear, J., Schneider, R., Karsenty, G., Felber, B.K., 1997.Identification of an RNA sequence within an intracisternal-A particle element ableto replace Rev-mediated posttranscriptional regulation of human immunodefi-ciency virus type 1. J. Virol. 71, 95–101.

Tan, W., Zolotukhin, A.S., Bear, J., Patenaude, D.J., Felber, B.K., 2000. The mRNA export inC. elegans is mediated by Ce-NXF-1, an ortholog of human TAP and S cerevisiaeMex67p. RNA 6, 1762–1772.

von Gegerfelt, A.S., Alicea, C., Valentin, A., Morrow, M., van Rompay, K.K., Ayash-Rashkovsky, M., Markham, P., Else, J.G., Marthas, M.L., Pavlakis, G.N., Ruprecht,R.M., Felber, B.K., 2006. Long lasting control and lack of pathogenicity of theattenuated Rev-independent SIV in rhesus macaques. AIDS Res. Hum. Retro-viruses 22, 516–528.

Waterston, R.H., Lindblad-Toh, K., Birney, E., Rogers, J., Abril, J.F., Agarwal, P., Agarwala,R., Ainscough, R., Alexandersson, M., An, P., Antonarakis, S.E., Attwood, J., Baertsch,R., Bailey, J., Barlow, K., Beck, S., Berry, E., Birren, B., Bloom, T., Bork, P., Botcherby, M.,Bray, N., Brent, M.R., Brown, D.G., Brown, S.D., Bult, C., Burton, J., Butler, J., Campbell,R.D., Carninci, P., Cawley, S., Chiaromonte, F., Chinwalla, A.T., Church, D.M., Clamp,M., Clee, C., Collins, F.S., Cook, L.L., Copley, R.R., Coulson, A., Couronne, O., Cuff, J.,Curwen, V., Cutts, T., Daly, M., David, R., Davies, J., Delehaunty, K.D., Deri, J.,Dermitzakis, E.T., Dewey, C., Dickens, N.J., Diekhans, M., Dodge, S., Dubchak, I., Dunn,D.M., Eddy, S.R., Elnitski, L., Emes, R.D., Eswara, P., Eyras, E., Felsenfeld, A., Fewell, G.A.,Flicek, P., Foley, K., Frankel, W.N., Fulton, L.A., Fulton, R.S., Furey, T.S., Gage, D.,Gibbs, R.A., Glusman, G., Gnerre, S., Goldman, N., Goodstadt, L., Grafham, D.,Graves, T.A., Green, E.D., Gregory, S., Guigo, R., Guyer, M., Hardison, R.C., Haussler,D., Hayashizaki, Y., Hillier, L.W., Hinrichs, A., Hlavina, W., Holzer, T., Hsu, F., Hua,A., Hubbard, T., Hunt, A., Jackson, I., Jaffe, D.B., Johnson, L.S., Jones, M., Jones, T.A.,Joy, A., Kamal, M., Karlsson, E.K., Karolchik, D., Kasprzyk, A., Kawai, J., Keibler, E.,Kells, C., Kent, W.J., Kirby, A., Kolbe, D.L., Korf, I., Kucherlapati, R.S., Kulbokas, E.J.,Kulp, D., Landers, T., Leger, J.P., Leonard, S., Letunic, I., Levine, R., Li, J., Li, M.,Lloyd, C., Lucas, S., Ma, B., Maglott, D.R., Mardis, E.R., Matthews, L., Mauceli, E.,Mayer, J.H., McCarthy,M., McCombie,W.R., McLaren, S., McLay, K., McPherson, J.D.,Meldrim, J., Meredith, B., Mesirov, J.P., Miller, W., Miner, T.L., Mongin, E.,Montgomery, K.T., Morgan, M., Mott, R., Mullikin, J.C., Muzny, D.M., Nash, W.E.,Nelson, J.O., Nhan, M.N., Nicol, R., Ning, Z., Nusbaum, C., O'Connor, M.J., Okazaki, Y.,Oliver, K., Overton-Larty, E., Pachter, L., Parra, G., Pepin, K.H., Peterson, J., Pevzner,P., Plumb, R., Pohl, C.S., Poliakov, A., Ponce, T.C., Ponting, C.P., Potter, S., Quail, M.,Reymond, A., Roe, B.A., Roskin, K.M., Rubin, E.M., Rust, A.G., Santos, R., Sapojnikov,V., Schultz, B., Schultz, J., Schwartz, M.S., Schwartz, S., Scott, C., Seaman, S., Searle,S., Sharpe, T., Sheridan, A., Shownkeen, R., Sims, S., Singer, J.B., Slater, G., Smit, A.,Smith, D.R., Spencer, B., Stabenau, A., Stange-Thomann, N., Sugnet, C., Suyama, M.,Tesler, G., Thompson, J., Torrents, D., Trevaskis, E., Tromp, J., Ucla, C., Ureta-Vidal,A., Vinson, J.P., Von Niederhausern, A.C., Wade, C.M., Wall, M., Weber, R.J., Weiss, R.B.,Wendl, M.C., West, A.P., Wetterstrand, K., Wheeler, R., Whelan, S., Wierzbowski, J.,Willey, D.,Williams, S.,Wilson, R.K.,Winter, E.,Worley, K.C.,Wyman, D., Yang, S., Yang,S.P., Zdobnov, E.M., Zody, M.C., Lander, E.S., 2002. Initial sequencing and comparativeanalysis of the mouse genome. Nature 420, 520–562.

Wilkie, G.S., Zimyanin, V., Kirby, R., Korey, C., Francis-Lang, H., Van Vactor, D., Davis, I.,2001. Small bristles, the Drosophila ortholog of NXF-1, is essential for mRNA exportthroughout development. RNA 7, 1781–1792.

Wodrich, H., Schambach, A., Krausslich, H.G., 2000. Multiple copies of the Mason–Pfizermonkey virus constitutive RNA transport element lead to enhanced HIV-1 Gagexpression in a context-dependent manner. Nucleic Acids Res. 28, 901–910.

Wodrich, H., Bohne, J., Gumz, E., Welker, R., Krausslich, H.G., 2001. A new RNA elementlocated in the coding region of a murine endogenous retrovirus can functionallyreplace the Rev/Rev-responsive element system in human immunodeficiency virustype 1 Gag expression. J. Virol. 75, 10670–10682.

Yang, J., Bogerd, H., Le, S.Y., Cullen, B.R., 2000. The human endogenous retrovirus K Revresponse element coincides with a predicted RNA folding region. RNA 6,1551–1564.

Yang, J., Bogerd, H.P., Peng, S., Wiegand, H., Truant, R., Cullen, B.R., 1999. An ancientfamily of human endogenous retroviruses encodes a functional homolog of theHIV-1 Rev protein. Proc. Natl. Acad. Sci. U. S. A. 96, 13404–13408.

Zolotukhin, A.S., Valentin, A., Pavlakis, G.N., Felber, B.K.,1994. Continuous propagation ofRRE(−) and Rev(−)RRE(−) human immunodeficiency virus type 1 molecular clonescontaining a cis">cis-acting element of simian retrovirus type 1 in humanperipheral blood lymphocytes. J. Virol. 68, 7944–7952.

Zolotukhin, A.S., Michalowski, D., Smulevitch, S., Felber, B.K., 2001. Retroviralconstitutive transport element evolved from cellular TAP(NXF1)-binding sequences.J. Virol. 75, 5567–5575.

Zolotukhin, A.S., Tan, W., Bear, J., Smulevitch, S., Felber, B.K., 2002. U2AF participates inthe binding of TAP (NXF1) to mRNA. J. Biol. Chem. 277, 3935–3942.