A quasi-lentiviral green fluorescent protein reporter exhibits nuclear export features of late human immunodeficiency virus type 1 transcripts

Virology 352 (2006) 295 – 305 www.elsevier.com/locate/yviro

A quasi-lentiviral green fluorescent protein reporter exhibits nuclear export features of late human immunodeficiency virus type 1 transcripts Marcus Graf 1 , Christine Ludwig, Sylvia Kehlenbeck 2 , Kerstin Jungert 3 , Ralf Wagner ⁎ Institute of Medical Microbiology and Hygiene, Molecular Microbiology and Gene Therapy, University of Regensburg, 93053 Regensburg, Germany Received 29 December 2005; returned to author for revision 23 February 2006; accepted 2 May 2006 Available online 13 June 2006

Abstract We have previously shown that Rev-dependent expression of HIV-1 Gag from CMV immediate early promoter critically depends on the AUrich codon bias of the gag gene. Here, we demonstrate that adaptation of the green fluorescent protein (GFP) reporter gene to HIV codon bias is sufficient to turn this hivGFP RNA into a quasi-lentiviral message following the rules of late lentiviral gene expression. Accordingly, GFP expression was significantly decreased in transfected cells strictly correlating with reduced RNA levels. In the presence of the HIV 5′ major splice donor, the hivGFP RNAs were stabilized in the nucleus and efficiently exported to the cytoplasm following fusion of the 3′ Rev-responsive element (RRE) and coexpression of HIV-1 Rev. This Rev-dependent translocation was specifically inhibited by leptomycin B suggesting export via the CRM1-dependent pathway used by late lentiviral transcripts. In conclusion, this quasi-lentiviral reporter system may provide a new platform for developing sensitive Rev screening assays. © 2006 Elsevier Inc. All rights reserved. Keywords: HIV-1; GFP; Codon usage; Rev; RRE; RNA export; CTE; CRM1

Introduction Viruses evolved individual strategies to control coordinated gene expression and selectively promote their mRNA export by recruiting cellular key players of general trafficking pathways (reviewed in Cullen, 2003). The retroviral and in particular the HIV-1 nuclear export system significantly contributed to our present understanding of mRNA export. Arising from a complicated pattern of alternative splicing events, more than 40 viral mRNAs are made from a single genome-length transcript and are basically classified into three species of various length (Schwartz et al., 1990). The ∼2-kb short mRNAs encoding the regulatory proteins Tat, Rev, and Nef are the first to be expressed after viral infection followed ⁎ Corresponding author. Fax: +49 941 944 6484. E-mail address: [email protected] (R. Wagner). 1 Geneart AG, Biopark, Josef-Engert-Straβe 11, D-93053 Regensburg, Germany. 2 SERVIER Deutschland GmbH, Westendstraβe 170, D-80686 München, Germany. 3 University of Ulm, Innere Medizin 1, Robert-Koch-Str.8, D-89081 Ulm, Germany. 0042-6822/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.virol.2006.05.001

by the late ∼4.3-kb intermediate and ∼9.2-kb unspliced mRNAs coding for structural and accessory proteins (Kim et al., 1989). This temporal pattern of protein expression strictly depends on the presence of the trans-active Rev protein promoting the nuclear export of late HIV-1 RNAs via interaction with a complex 351-nt RNA stem-loop structure, the so-called Rev-responsive element (RRE) (Olsen et al., 1990; reviewed in Cullen, 2003). The Rev protein shuttles back into the nucleus where it binds to this cis-acting target sequence contained in all late incompletely spliced HIV transcripts directing them into the CRM1-dependent export pathway (Fornerod et al., 1997; Neville et al., 1997). This pathway is generally utilized by cellular NES-containing proteins and U-rich small nuclear and 5S ribosomal RNAs (Fischer et al., 1995). Although Rev/RRE interaction is a prerequisite for HIV-1 late gene expression (Emerman et al., 1989; Fischer et al., 1994; Malim et al., 1989b, 1990), the critical contribution of different cis-acting elements to nuclear retention of incompletely spliced transcripts is still not fully clarified. Usually, intron-containing cellular RNAs are recognized by nuclear splicing commitment factors that keep the message from leaving the nucleus until

296

M. Graf et al. / Virology 352 (2006) 295–305

splicing is completed (Mikaelian et al., 1996; Chang and Sharp, 1989; Legrain and Rosbash, 1989). Since Rev, by promoting export of unspliced RNAs overcomes the export restrictions imposed by the splicing factors, the mutual influence of Rev function and the cellular splicing machinery has been investigated extensively (Mikaelian et al., 1996, Lu et al., 1990). Based on these studies, it was suggested that Rev directly interferes with splicing and rescues the export of RNAs entrapped within the splicing machinery by active disassembly of the spliceosomes (Chang and Sharp, 1989; Kjems et al., 1991) that comprise ∼300 splicing-related factors (reviewed in Jurica and Moore, 2003). However, a current model favors a rather indirect mechanism where Rev expression promotes nuclear RNA export thereby reducing the amount of substrates in the nucleus available for splicing (reviewed in Hope, 1999). Whereas the 1st and the 4th 5′ splice donor sites used in nearly all spliced transcripts have been shown to be used very efficiently, further investigations descried suboptimal splice acceptor (SA) sites as assignable cause for the relatively inefficient splicing of late HIV-1 transcripts and thus, the resulting variety of alternatively spliced products (O'Reilly et al., 1995; Staffa and Cochrane, 1994). However, Rev function is not solely determined by inefficient splicing but also depends on the presence of a strong 5′ splice site as has been demonstrated for successful expression of env mRNAs (Lu et al., 1990). Assembly of the spliceosome is initiated by binding of U1 snRNP to the 5′ splice site via complementary base pairing. Additionally, the 5′ splice donor site is necessary to protect the viral pre-mRNAs from nuclear degradation. This stabilizing function of the splice donor (SD) could be strictly separated from the splicing function (Kammler et al., 2001). Furthermore, cis-acting inhibitory sequences referred to as exonic splicing silencer (ESS) and shown to derogate splicing efficiency have been found adjacent to or within tat and rev exons (Amendt et al., 1994; Si et al., 1998). However, the sole impact on splicing-related events appeared insufficient to explain Rev's line of action as late HIV-1 transcripts remain Rev-dependent even in the absence of functional splice sites (Nasioulas et al., 1994). A further role in negative regulation of late HI-viral gene expression has been assigned to less characterized AU-rich RNA sequence elements (Maldarelli et al., 1991; Olsen et al., 1992). Indeed, several cis-acting repressive (CRS) or inhibitory (INS) sequences have been located in the coding regions of p17MA, p24CA, pol, and env including the RRE (Brighty and Rosenberg, 1994; Cochrane et al., 1991; Nasioulas et al., 1994; Schwartz et al., 1992a, 1992b). Accordingly, mutational inactivation of the corresponding sequences within gag allowed for Gag expression in the absence of Rev (Schneider et al., 1997; Schwartz et al., 1992a, 1992b). Moreover, we and others could previously demonstrate that Rev dependence can be overcome completely by adapting the codon usage of late HIV-1 transcripts to the codon preference of mammalian genes (Graf et al., 2000; Haas et al., 1996; Kotsopoulou et al., 2000; Wagner et al., 2000). These

modifications allowed for a constitutive nuclear export of the corresponding messages even in the absence of a 5′ splice donor and a 3′ RRE (Graf et al., 2000). Taken together, these findings suggest that Rev dependence of late lentiviral transcripts relies on (i) a functional 5′ splice donor in the UTR, (ii) AU-rich codon usage, and (iii) the presence of weak splice acceptor signals. To prove this hypothesis, these signals thought to be important for regulating late lentiviral gene expression were conferred to gfp gene. This study clearly demonstrated that a randomly chosen codon-optimized gene can be subjected to late lentiviral gene regulation exhibiting typical stability features of HI-viral nuclear mRNAs and strict Rev dependence. Results Design of quasi-lentiviral GFP reporter plasmids Lately, we have shown by using a subgenomic Gag reporter that the strict regulation of late HI-viral gene expression mainly determined by strong splice donor and weak acceptor signals, AU-rich codon bias and Rev dependence can be overcome by conferring mammalian codon usage on the HIV-1-derived gag gene (Graf et al., 2000). In order to determine whether these characteristics of late HI-viral gene regulation can in return be transferred to any non-viral transcript, a cell culture system based on a common reporter gene was established. For this purpose, a synthetic version of a GFP-encoding gene was generated, exhibiting the codon bias of HIV-1 gag. Accordingly, sequence homology between the resulting synthetic sequence, referred to as hivGFP and the original codon-optimized and thus, humanized variant huGFP was reduced to 68%, whereas the amino acid composition of the corresponding gene products remained unaltered (Fig. 1). In order to mimic the natural shape of unspliced HIV-1 RNAs, the GFP variants were provided with cis-active elements known to contribute to temporally regulated HIV-1 gene expression. Thus, the 103-bp 5′ untranslated region (UTR) comprising the major splice donor (SD) with a match to the consensus sequence C/AAG/GUA/GAGU was cloned upstream of the GFP encoding genes. In addition, a 861-bp fragment representing the highly structured cis-acting Revresponsive element (RRE) was inserted downstream of the GFP encoding sequences to allow for Rev binding and Revmediated RNA export. Additionally, this fragment harbors the terminal splice acceptor site of the HIV-1 genome known to be used very inefficiently (O'Reilly et al., 1995; Staffa and Cochrane, 1994). Alternatively, the RRE region was replaced by the Mason-Pfizer monkey virus (MPMV) constitutive transport element (CTE)—a functional analog of the Rev/ RRE nuclear translocation system (Bray et al., 1994). Finally, all gfp sequences were put under transcriptional control of the CMV immediate early promoter enhancer unit in order to circumvent the anti-repressor Tat/TAR regulation of the LTR promoter. An overview of all used reporter constructs is given in Fig. 2.

M. Graf et al. / Virology 352 (2006) 295–305

297

Fig. 1. Sequence alignment of the humanized (upper sequence) and HIV-adapted (lower sequence) gfp genes with corresponding GFP amino acids indicated below the second base of each triplett. The original AT-content of huGFP (37%) was elevated to 69% in hivGFP yielding a final homology of 68% for both sequences.

HIV-like codon bias inhibits high level expression of GFP reporter genes in mammalian cells

constructs indicating that the AU-rich codon usage of the hivGFP gene had no major impact on translation.

The influence of HIV-like codon bias upon GFP expression was analyzed by transient transfection of H1299 cells with the various reporter constructs and subsequent Western blot analysis (Fig. 3A). Whereas GFP expression from all humanized reporter genes proved to be comparably efficient, no specific signal was detected in hivGFP transfectants using a GFP-specific antibody, irrespective of 5′ or 3′ cis-acting elements. In order to exclude the possibility that inefficient expression of hivGFP is due to compromised translation efficiency, all GFP-constructs were subjected to a coupled transcription/translation assay. As depicted in Fig. 3B, the in vitro approach yielded almost equal amounts of GFP for all

Efficient expression of hivGFP requires a functional Rev/RRE system and the 5′ major SD within the upstream untranslated region In order to assess the Rev responsiveness of the various GFP encoding genes, human H1299 cells were transiently transfected with the GFP reporter constructs in the absence or presence of a Rev expression plasmid. GFP expression was analyzed by Western blot using a GFP-specific antibody. As shown in Fig. 4A, high level GFP expression driven by the humanized reporter gene (huGFP) was completely independent of a functional Rev/RRE system and remained unaltered when

298

M. Graf et al. / Virology 352 (2006) 295–305

Fig. 2. Schematic representation of HIV-adapted (hiv) and humanized (hu) GFP expression plasmids. Black boxes represent HIV-adapted GFP encoding genes (hivGFP), whereas open boxes represent humanized GFP encoding genes (huGFP). cis-Acting elements like the 5′ untranslated region (UTR), the 3′ HIV-1 Rev-responsive element (RRE) or the 3′ MPMV-derived CTE were fused to the GFP reading frames as indicated.

the 5′ located major SD was provided. In contrast, GFP expression from the HIV-adapted hivGFP-RRE construct was hardly detectable neither in the absence nor in the presence of Rev. Addition of 5′ UTR provided prerequisites for rescuing of GFP expression by Rev, which did, however, not reach the levels achieved by huGFP. Comparable results were obtained with different cell lines (COS-7, HeLa) basically ruling out a contribution of cell-type-specific factors to the observed effects

Fig. 3. Expression analysis of the humanized (hu) and quasi-lentiviral (hiv)GFP reporter genes. H1299 cells were transiently transfected with 15 μg of the indicated GFP expression plasmids, respectively. Cells were harvested 48 h posttransfection and 50 μg of cell lysates was subjected to Western blot analysis following separation by 12.5% SDS-PAGE. GFP expression was analyzed using a GFP-specific antibody. Positions of molecular weight markers are indicated on the left (A). 500 ng of each reporter plasmid was subjected to coupled in vitro transcription/translation using a rabbit reticulocyte lysate. [35S]-labeled proteins were separated by SDS-PAGE and visualized by autoradiography (B).

Fig. 4. Analysis of Rev-dependent GFP expression. H1299 cells were transiently transfected with 15 μg of the indicated GFP expression plasmids, respectively. To determine the Rev dependence of the GFP expression constructs, 15 μg of either pcDNA3.1 (−) or a Rev expression vector (+) were cotransfected as depicted, and cells were harvested 48 h posttransfection. For Western blot analysis 50 μg of cell lysates was separated by 12.5% SDS-PAGE, and GFP expression was analyzed using a GFP-specific antibody. Positions of molecular weight markers are indicated on the left (A). To quantify GFP expression, transfected cells were subjected to FACS (black bars) or fluorimeter (grey bars) analysis (B). The means obtained from two independent FACS analyses are given as percentage of huGFP-derived fluorescence.

(data not shown). Thus, in accordance with our previous findings (Graf et al., 2000), these data strongly support the requirement of a functional major SD for successful expression of quasi-HIV-1 gene products. For quantitative evaluation of GFP expression, 293T cells transfected with the various reporter constructs in presence or absence of Rev were subjected to FACS analysis (Fig. 4B). To quantify GFP expression derived from two independent transfections, fluorescence measured in huGFP transfectants was set to 100% and values obtained with all other constructs were calculated accordingly (black scale bars). Cells transfected with UTR-huGFP-RRE showed a slightly diminished fluorescence (∼80%) compared to huGFP-transfected cells and GFP expression was further reduced following cotransfection of pcRev. As expected, cells transfected with hivGFP-RRE hardly showed any green fluorescence even when Rev was present. GFP expression was elevated to a small extent when the 5′ major SD was provided in cis and was finally increased to ∼40% in the presence of Rev, which corresponds to the 10-fold transaction rate reported for the subgenomic Gag reporter (Graf et al., 2000) and is only slightly below the value measured for UTR-huGFP-RRE and Rev. These results were nicely reflected when GFP production from transfected cells was quantified in a fluorimeter (grey scale bars).

M. Graf et al. / Virology 352 (2006) 295–305

Taken together, these data substantiate our previous findings using a codon-optimized gag gene (Graf et al., 2000) and suggest that the presence of cis-active elements located up- and downstream of the GFP encoding region have only a minor effect upon expression of the humanized gene variant, whereas these elements proved to be essential to drive Rev-dependent expression of the HIV-adapted reporter. However, although this fully equipped HIV-like gene did not reach huGFP expression levels, coexpression of UTR-huGFP and UTR-hivGFP together with RRE resulted only in minor differences, indicating that these HIV-derived cis-active elements prove to be highly functional also in this artificial system. Expression of HIV-adapted GFP depends on the nuclear export mechanism of late HIV-1 genes To further evaluate the significance of Rev function for RNA export of the GFP reporter transcripts, H1299 cells were transfected with UTR-huGFP-RRE or UTR-hivGFPRRE and cotransfected with either (i) pc-Rev or (ii) pcRevM10 encoding a transdominant negative Rev mutant (Malim et al., 1989a). Additionally, cells were cotransfected with HX10 in order to examine the efficacy of wild-type Rev derived from a proviral background (Fig. 5A). Western blot analysis of the transfected cells revealed that neither functional nor deficient Rev proteins significantly influenced high level expression of huGFP (Fig. 5A, lanes 1–4). In contrast, expression of hivGFP depended on the presence of wild-type Rev, irrespective of its origin (Fig. 5A, lanes 5–7). As

299

expected, the RevM10 mutant was not able to rescue hivGFP transcripts, whereas endogenous levels of HX10-derived Rev were sufficient to support nuclear export of the HIV-like reporter RNAs. These results clearly demonstrate that successful expression of the quasi-lentiviral reporter construct UTRhivGFP-RRE critically depends on the presence of a functional Rev protein. To determine the extent, to which Rev-mediated expression of hivGFP might be influenced by RevM10, H1299 cells were cotransfected with constant amounts of UTR-hivGFPRRE plus wild-type Rev and increasing amounts of RevM10, and GFP expression was quantified in a fluorimeter. As summarized in Fig. 5B, Rev-mediated GFP expression was inhibited by RevM10 in a dose-dependent manner. The Streptomyces-derived metabolite leptomycin B (LMB) is another well-known inhibitor of Rev function based on its ability to directly interfere with the interaction of Rev's nuclear export signal with CRM1 (Fornerod et al., 1997; Kudo et al., 1998). Since LMB has been shown to specifically block Rev function in the nanomolar range (2–10 nM) (Fischer et al., 1995; Wolff et al., 1997), its influence on the quasi-lentiviral reporter gene was examined. Indeed, UTR-hivGFP-RRE transcripts proved to be susceptible to LMB treatment in the presence of Rev. This LMB-mediated inhibition of hivGFP expression was shown to be concentration-dependent in FACS (Fig. 5C) and Western blot analysis (data not shown), suggesting that Rev-mediated nuclear translocation of the UTR-hivGFP-RRE mRNAs is guided specifically via the CRM1-dependent pathway used by late lentiviral HIV-1 transcripts.

Fig. 5. Influence of functional and export-deficient Rev proteins on GFP expression. H1299 cells were cotransfected with 15 μg of either UTR-huGFP-RRE or UTRhivGFP-RRE and 15 μg of the indicated Rev expression plasmids. Cells were harvested 48 h posttransfection, and 50 μg of total protein was subjected to Western blot analysis (A). GFP expression was detected using a GFP-specific antibody. Positions of molecular weight markers are indicated on the left. For further analysis of RevM10 effects, H1299 cells were transiently cotransfected with 7 μg of UTR-hivGFP-RRE and 3.5 μg of pc-Rev together with increasing amounts of RevM10 expression plasmid or pcDNA3.1 as indicated. Cells were harvested 48 h posttransfection, and 50 μg of total protein was subjected to fluorimeter analysis (B). Fluorescence from cells lacking the RevM10 plasmid was set to 100%. Dose-dependent inhibition of Rev-dependent GFP expression by LMB was analyzed upon transient cotransfection of H1299 cells with 5 μg of UTR-hivGFP-RRE and 5 μg Rev expression vector in the presence of increasing amounts of LMB as indicated. Cells were harvested 48 h posttransfection and subjected to FACS analysis (C). Fluorescence from cells lacking LMB was set to 100%.

300

M. Graf et al. / Virology 352 (2006) 295–305

The HIV-1 5′UTR/major SD is required to protect hivGFP mRNAs from rapid nuclear degradation To further study the contribution of RNA stability and nucleo-cytoplasmic transport of reporter transcripts to differential expression of the various gfp genes, RNAs purified from nuclear and cytoplasmic compartments of reporter-transfected H1299 cells were subjected to Northern blot analysis. Radiolabeled RRE antisense riboprobes were used to quantify both wild-type and HIV-adapted GFP encoding transcripts from nuclear and cytoplasmic fractions. As demonstrated in Fig. 6B, the cytoplasmic levels of GFP-encoding RNAs nicely correlated with GFP expression patterns obtained from Western blot or FACS analyses (Fig. 4) confirming that differential GFP expression was not caused by attenuated translational efficiencies. However, an in depth analysis of the corresponding nuclear and cytoplasmic fractions revealed that rather (i) stability of nuclear RNAs, (ii) alternative splicing events and (iii) codon usage itself account for the observed differences in GFP production. These principles have also been proposed to mainly drive differential expression of wild-type and humanized Gag in former studies (Graf et al., 2000). Accordingly, hivGFP mRNAs lacking the 5′ major SD appeared to be degraded in the nucleus irrespective of Rev being present and were therefore not detectable neither in the nucleus nor in the cytoplasm (Fig. 6). Although small amounts of hivGFP mRNAs comprising the 5′ UTR were visible in the cytoplasmic cell fraction in the absence of Rev, the majority of UTR-hivGFP-RRE transcripts was only efficiently translocated to the cytoplasm when Rev was provided in trans leading to a >10-fold increase in cytoplasmic hivGFP mRNAs. An additional shorter RNA variant was detected in the cytoplasmic fractions and to a lesser extent in the nuclear fractions of cells transfected with UTRhivGFP-RRE, suggesting that the 5′ major SD promotes partial splicing of hivGFP transcripts. In analogy to the multiply spliced lentiviral RNAs, these shorter transcripts were continuously exported from the nucleus even in the absence of Rev

Fig. 6. Cellular distribution of hivGFP and huGFP mRNAs. H1299 cells were transfected with the indicated GFP reporter constructs in the absence (−) or presence (+) of a Rev expression plasmid. Cells were harvested 48 h posttransfection and polyadenylated mRNAs isolated from nuclear (A) and cytoplasmic (B) cell fractions were subjected to Northern blot analysis. GFP encoding transcripts were detected using a radiolabeled RRE antisense riboprobe (upper panel). Beta-actin RNA was used for standardization (lower panel).

and thus, dominated over the full-length RNAs in the cytoplasm. Expectedly, cotransfection of Rev caused an increase in nucleo-cytoplasmic export of the unspliced RNA species by promoting usage of the CRM-1-mediated pathway, which resulted in significantly enhanced levels of full-length hivGFP RNAs detectable in the cytoplasm. Presence of the 5′ major SD also induced partial splicing of the huGFP RNAs, all of which were efficiently exported to the cytoplasm in a Rev-independent manner. Accordingly, the spliced huGFP RNAs are not supposed to contribute to quantitative GFP expression and might thus explain the observed decrease in huGFP expression in the presence of the 5′ UTR (Fig. 4B). Nonetheless, compared to the hiv-like GFP transcripts the full-length humanized GFP variants still reached higher levels of total RNA in both the nuclear and cytoplasmic fractions of transfected cells. These results clearly argue for the critical contribution of nucleic acid composition to RNA stability, nucleo-cytoplasmic RNA export and, thus, expression efficiency of the corresponding reporter RNAs. The MPMV-derived constitutive nuclear transport element partially rescues quasi-lentiviral mRNA export The highly specific lentiviral Rev/RRE-system was demonstrated to be functionally replaced by a cis-acting element present in incompletely spliced RNAs of less complex retroviruses like the Mason-Pfizer monkey virus (MPMV). Unlike the two-component Rev/RRE system, this MPMV-derived constitutive transport element (CTE) directly binds to the cellular mRNA export receptor TAP (Grüter et al., 1998), and unspliced RNAs are subsequently translocated through the nuclear pore by a Ran-independent mechanism (Braun et al., 1999; Kang et al., 2000; Kang and Cullen, 1999). As CTE has been shown to substitute for Rev/RRE function in the context of late HIV-1 transcripts (Bray et al., 1994), the capability of a 3′ located CTE to promote nuclear export of the HIV-adapted GFP RNA in cis was analyzed. As demonstrated by Western blot analysis of transfected 293T cells, expression of the humanized GFP variant was not visibly affected by the 3′ CTE (Fig. 7). In contrast, CTE was capable of rescuing hivGFP mRNAs depending, however, on the presence of the 5′ UTR. These results clearly indicate that CTE-mediated mRNA translocation is independent of additional viral cofactors, whereas nuclear export of RRE-containing mRNAs absolutely depends on Rev function. Nevertheless, Rev/RRE interaction resulted in higher GFP expression than the CTE-based mechanism, suggesting that CTE can only partially replace the established Rev/RRE system in the examined cell line. However, CTE-mediated GFP expression was comparable to expression levels obtained via Rev/RRE in H1299 cells (Fig. 7B) confirming the former observation that efficiency of CTE function depended on the used cell line (Bray et al., 1994). Moreover, we could demonstrate that unlike the Rev/RRE-dependent UTR-hivGFP variant, expression of the UTR-hivGFP-CTE reporter was not significantly influenced by LMB (Fig. 7B) supporting the assumption that both RNA species were translocated via different export pathways.

M. Graf et al. / Virology 352 (2006) 295–305

301

applicability of our GFP reporter for analysis of differential retroviral gene expression. Discussion

Fig. 7. Comparison of Rev/RRE- and CTE-mediated GFP expression. 293T cells were transiently transfected with indicated plasmids, harvested 48 h posttransfection, and 50 μg of total protein was subjected to Western blot analysis with a GFP-specific antibody (A). To quantify the influence of LMB on GFP expression, H1299 cells were transfected with indicated reporter genes and were grown for 48 h in the presence (black bars) or absence (grey bars) of 5 nM LMB prior to FACS analysis (B). The means obtained from three independent transfections are given as percentage of huGFP-derived fluorescence. To analyze effects of exogenous Sam68 on GFP expression, 293T cells were cotransfected with respective GFP reporter genes in the presence (+) or absence (−) of an HA-tagged Sam68 expression construct, and 50 μg of total protein was analyzed by Western blot with GFP- and Sam68-specific antibodies (C). Positions of molecular weight markers are denoted on the left, and GFP- and Sam68-specific bands are indicated by an arrow on the right.

The deficient expression of a CTE-dependent gag-pol reporter in 293T cells was recently shown to be boosted in the presence of exogenous Sam68, a cellular RNA binding protein (Coyle et al., 2003). To test whether coexpressed Sam68 was also capable of enhancing CTE-mediated expression of the UTR-hivGFP message, we cotransfected 293T cells with the various GFP reporters together with an HA-tagged Sam68 expression construct. Indeed, Western blot analysis demonstrated that CTE-directed GFP expression was significantly improved in 293T cells upon overexpression of Sam68 (Fig. 7C). However, exogenous Sam68 did not substantially elevate huGFP levels, and enhancement of Rev/RRE-dependent expression was comparably moderate. Similar results were obtained by FACS analysis, where expression of UTRhivGPF-CTE was increased by more than 50% in the presence of HA-Sam68 (data not shown). These observations are in accordance with recent reports by Coyle et al. (2003) indicating that the Sam68 effect upon CTE function was highly specific. In sum, these data further approve the

Success of late lentiviral gene expression strictly depends on the concerted action of various cis- and trans-acting viral and cellular factors promoting nuclear export of non-terminally spliced viral messages. We and others have previously demonstrated that nuclear export restrictions basically imposed by proposed AU-rich inhibitory sequence elements (Maldarelli et al., 1991; Schwartz et al., 1992a, 1992b), inefficient splicing, and Rev/RRE interaction (reviewed in Cullen, 2003) can be overcome by transferring the codon usage of highly expressed mammalian genes to late HIV-1 transcripts (Graf et al., 2000; Haas et al., 1996; Kotsopoulou et al., 2000; Nguyen et al., 2004). These GC-rich messages have been shown to be exported via a CRM1-independent pathway in the absence of RRE and Rev allowing for high level protein expression which cannot be attributed to enhanced translational efficiencies of the codon-optimized genes (this study, Gao et al., 2003). However, the critical contribution of distinct cis-acting elements towards conferring Rev dependence and directing mRNAs into the CRM1-mediated export pathway is still controversially discussed. Here, we have shown that adaptation of the huGFP sequence to the HIV codon bias is a sufficient prerequisite to render GFP expression Rev-dependent. Accordingly, we could demonstrate that hiv-like GFP mRNAs (i) need to be stabilized by the 5′ major SD in the nucleus, (ii) require a functional 3′ RRE for Rev response, (iii) depend on the nuclear CRM1-mediated export pathway of late lentiviral genes, and (iv) can partly be rescued by the MPMV-derived CTE. These findings suggest that any randomly chosen gene can be subdued to quasi-lentiviral gene regulation by applying codon bias of late HIV-1 transcripts together with the corresponding splicing signals and the Rev/RRE export machinery. Besides, Northern blot analyses revealed that spliced hivGFP RNA coproducts nicely mimic multiply spliced HIV-1 mRNA species that are rapidly translocated from the nucleus to the cytoplasm in a Rev-independent manner. Splicing of the hivGFP RNAs was induced in the presence of the 5′ major SD resulting in a ∼300-bp mRNA species that chiefly correlates with the calculated length of an RNA utilizing the 5′ major SD site and two closely adjacent suboptimal SA sites located in the 3′ part of the RRE (O'Reilly et al., 1995; Staffa and Cochrane, 1994). These two sites were also confirmed in silico by using a splice site prediction program (http://genes.mit.edu/burgelab/ maxent/Xmaxentscan_scoreseq_acc.html). Whereas this algorithm did not identify any additional SD or SA sites within hivGFP, a second acceptor was predicted to map in the 5′ part of the huGFP sequence, which may account for the larger RNA splice product found in huGFP-transfected cells (Fig. 6). However, whereas huGFP-derived RNA variants appeared to be stable per se, hivGFP RNAs lacking the 5′ major SD were rapidly degraded in the nucleus. Thus, a functional 5′ splice site may contribute to increased gene expression prior to splicing by providing binding sites for stabilizing factors as has been shown

302

M. Graf et al. / Virology 352 (2006) 295–305

for the U1 snRNA (Kammler et al., 2001). Indeed, related experiments by Nguyen and colleagues with native and codonoptimized versions of the accessory vpu and vif genes have clearly demonstrated that the lack of accumulation of the corresponding wild-type transcripts in the cytoplasm was not due to compromised transcription but rather a consequence of rapid degradation of the AU-rich messages withheld in the nucleus (Nguyen et al., 2004). Thus, codon bias seems to be the major determinant of RNA instability restricting expression of late lentiviral genes and, thus, expression of hiv-like GFP. RNA stability is basically determined by secondary structures and enthalpy which are in turn defined by the GC and AU content of the message. Apart from thermodynamic conditions, specific AU-rich binding motifs have been assigned a role in promoting RNA degradation. In particular adenylate/uridylate-rich elements (AREs) located in the 3′ untranslated region of many cellular mRNAs have been shown to direct rapid nuclear RNA degradation thereby contributing to a controlled expression of proto-oncogenes (reviewed in Chen and Shyu, 1995). During the last years, increased attention has been drawn to vaguely defined AU-rich RNA sequences referred to as INS or CRS, which have been demonstrated to play a decisive role in export restrictions of late lentiviral transcripts (Schneider et al., 1997; Schwartz et al., 1992a, 1992b). Recently, Wolff and colleagues have developed a valid algorithm for INS prediction based on a combined approach using bioinformatics and Rev-dependent reporter gene expression (Wolff et al., 2003). Indeed, this method helped to identify so far unknown INS regions within pol, env, and nef sequences that might provide binding sites for regulatory proteins. In this regard, various cellular factors have been reported to bind specifically to AU-rich sequence elements thereby modulating RNA metabolism, stability, nuclear, and cytoplasmic trafficking or translational efficiency of non-terminally spliced mRNAs. Among those proteins, several specific INS binders potentially contributing to instability, nuclear retainment, and Rev dependence of late lentiviral RNAs were identified such as the polyadenylate-binding protein PABP1 or the splicing component hnRNP A1, both interacting with INS-1 in p17gag (Afonina et al., 1997; Najera et al., 1999). Another splicing component, hnRNP C, has been shown to bind to a cis-acting repressive sequence (CRS) located in the pol-gene (Olsen et al., 1992), and recently, the heterodimeric factor PSF/p54nrb has been demonstrated to modulate expression of INS-containing gag and env RNAs (Zolotukhin et al., 2003). Furthermore, some cellular proteins have been suggested to assist in Rev-dependent lentiviral RNA export by bridging or strengthening interactions of Rev with CRM1 or the nuclear pore complex like the eukaryotic initiation factor 5A (Elfgang et al., 1999 and references therein) or the recently described DEAD box RNA helicase DDX3 (Yedavalli et al., 2004). Yet other Rev cofactors like the nucleoporin hRIP, the kinesin-like protein REBP or hnRNP A2 have been implicated in nuclear or cytoplasmic trafficking or sub-compartmental localization of non-terminally spliced viral RNAs (Beriault et al., 2004; Venkatesh et al., 2003). Besides, the cellular RNA binding protein and proposed

Rev interactor Sam68 has been assigned a role in promoting CRM1-mediated Rev nuclear export, HIV-1 RNA 3′end processing or cytoplasmic utilization of intron-containing RNAs (Coyle et al., 2003; Li et al., 2002; McLaren et al., 2004). In our experimental setting, overexpression of Sam68 had a clear positive effect on CTE function in 293T cells, whereas a formerly reported enhancement of Rev/RREmediated reporter expression could not be confirmed (Reddy et al., 1999). This might be explained by a restricted support of CTE-mediated reporter expression by endogenous Sam68, described for this specific cell-line (Coyle et al., 2003). However, distinct RNA binding motifs have only been precisely described for some of the abovementioned proteins. The critical contribution of corresponding cis-active elements possibly created by codon adaptation, to the expression profile of the HIV-like GFP reporter will have to be carefully considered and further experiments are currently underway to clarify these aspects. As green fluorescent protein is a widely established and easily evaluable reporter system, the Rev-dependent UTRhivGFP-RRE construct provides various possibilities for the development of screening assays allowing Rev-dependent upscaling of reporter gene expression and thus quantifying the responsiveness to mutated or attenuated Rev proteins or new Rev-targeted antiviral inhibitors. In this context, it should be mentioned that an established chloramphenicol acetyltransferase (CAT) assay has widely been used to analyze Rev function. The corresponding reporter pDM128 provides a cat gene inserted into an HIV-1 intron that contains the RRE, thus rendering CAT expression Rev-dependent. CAT expression has been shown to be rescued in the presence of Rev that kinetically competes with splicing (Hope et al., 1990). Although transaction rates obtained with our reporter (10-fold) were basically lower than reported for the CAT system (20- to 100-fold—depending on the experimental setting), our results nicely coincide with the formerly described subgenomic Gag reporter (Graf et al., 2000) that appears to more closely reflect the natural HIV context. As described for the CAT reporter, Rev-mediated reduction in splicing was also observed for both hivGFP and huGPF variants harboring the 5′ SD and the 3′ SA downstream of the RRE sequence. However, in the absence of Rev, translocation of unspliced huGFP transcripts was still much more efficient than export of the quasi-lentiviral RNAs which points to a partly splicing-independent mechanism due to intrinsic instability or active retention of the AU-rich hivGPF messages in the nucleus by yet unknown cellular cofactors. Notably, a potential contribution of nuclear retention signals within the CAT encoding sequence to Rev dependence has never been analyzed. It is therefore worth mentioning that the cat gene is quite AT-rich (55%), closely resembling HIV-1 gag composition (56%). In the light of AU-rich elements involved in nuclear export regulation, it might be interesting to compare expression profiles of a codonoptimized versus an HIV-adapted cat variant in the established reporter context. Apart from the specific quasi-lentiviral features noted above, the GFP reporter also provides some technical advantages over the rather laborious isotopic CAT assay, above all non-isotopic

M. Graf et al. / Virology 352 (2006) 295–305

303

Material and methods

The corresponding PCR product was cloned into the BamHI/ EcoRI-linearized eukaryotic expression vector pcDNA3-HA providing inserted sequences with a 5′-HA-tag. Generation of the plasmid pc-ERR comprising an RRE variant cloned in inverse orientation has been described previously (Graf et al., 2000). Sequences were aligned using the DNAMAN software (Lynnon biosoft, version 5.2.9).

Plasmid DNA construction

Cell culture and transfection

For generation of the eukaryotic expression vector pc-huGFP a codon-optimized synthetic gene encoding the green fluorescent protein of Aequorea victoria (huGFP; Genbank accession no. AX511259) was used as template for PCR along with oligonucleotides 5′-GATCGAATTCAACCATGGTGAGCAAGGGCGAGGAG-3′ and 5′-GATCCTCGAGAAGGATCCTTTACTTGTACAGCTCGTC-3′. The resulting fragment was digested with EcoRI and XhoI and inserted into pcDNA3.1 (Invitrogen). For construction of the pc-hivGFP expression plasmid, a matrix displaying the codon usage of HIV-1IIIB Gag protein was generated using the gcg software (Wisconsin genetics computer group) and was used to backtranslate the amino acid sequence of GFP into an artificial hivGFP sequence. This synthetic hivGFP gene exhibiting the codon usage of late HIV-1 genes was generated by Geneart AG via stepwise PCR amplification. Briefly, three subfragments – each constructed with three pairs of overlapping oligonucleotides – were amplified by PCR as described previously by Graf et al. (2000). Using appropriate terminal oligonucleotides, EcoRI and NcoI restriction sites were introduced 5′ and BamHI and XhoI sites 3′ of the coding sequence. Following ligation of the subfragments, the final hivGFP fragment was inserted into pcDNA3.1 via EcoRI and XhoI restriction resulting in pc-hivGFP. All subgenomic HIV-1-derived sequences were generated by PCR amplification with provirus HX10 as template (GenBank accession no. M15654). The untranslated region (UTR) located upstream of the gag sequence was amplified by PCR using oligonucleotides 5′-ATCGAATTCCGACGCAGGACTCGGCTTGC-3′ and 5′-GATCCCATGGCTCTCTCCTTCAGCCTCCG-3′. Following digestion with EcoRI and NcoI, the resulting 103-bp fragment was inserted upstream of the GFP encoding regions of pc-huGFP and pc-hivGFP resulting in pc-UTRhuGFP and pc-UTR-hivGFP, respectively. The Rev-responsive element (RRE) was amplified using oligonucleotides 5′GATCGGATCCGAGATCTTCAGACCTGGAGGAG-3′ and 5′-GATCCTCGAGGTTCACTAATCGAATGGATCTG-3′. Following digestion with BamHI and XhoI the corresponding fragment was inserted downstream of the GFP encoding regions of pc-huGFP, pc-hivGFP, pc-UTR-huGFP, and pc-UTRhivGFP resulting in pc-huGFP-RRE, pc-hivGFP-RRE, pcUTR-huGFP-RRE, and pc-UTR-hivGFP-RRE, respectively. For synthesis of the eukaryotic Sam68 expression construct, Sam68 coding cDNA was generated via RT-PCR from human B-lymphocyte-derived cytoplasmic RNA with oligonucleotides 5′-ATCATCTGATCAATGCAGCGCCGGGACGACCC-3′ and 5′-GATGAATTCTCATTAATAACGTCCATATGGGTGCTCT-3′ comprising BclI and EcoRI restriction sites, respectively.

The adherent human lung carcinoma cells H1299, the African green monkey kidney COS-7 cells and the adherent human kidney epithelial cells 293T were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum. For transfections 2.5 × 105 or 1 × 106 cells were plated on 6-well or 10-cm dishes and transfected 24 h later with the given amounts of plasmid DNA, respectively using the calcium phosphate precipitation technique. Leptomycin B (LMB) was added to cell culture supernatants 24 h prior to cell harvest.

and quick read-outs via FACS, fluorimeter or microscopy of living cells. We further hope that this quasi-lentiviral model system may help to extend our present understanding of nuclear retention mechanisms thereby providing new insights in Revdependent export regulation of unspliced lentiviral transcripts.

Western blot Cells were harvested 48 h posttransfection, washed two times in PBS, lysed in 0.5% Triton X-100, 100 mM Tris/HCl (pH 7.4), and subjected to repeated freeze/thaw cycles. Unsoluble components were removed by centrifugation (20,800 × g, 5 min), and the total amount of protein was determined using the Bio-Rad Protein Assay (Bio-Rad Laboratories, Germany) following the manufacturer's instructions. Fifty micrograms of total protein was separated by sodium dodecyl sulfate (SDS)-12.5% polyacrylamide gel electrophoresis (PAGE) and transferred to nitrocellulose. Expression of GFP was detected with a GFP-specific monoclonal antibody (Clontech) and visualized by chromogenic staining or chemiluminescence. In vitro transcription/translation Using a TNT coupled reticulocyte lysate system (Promega), 500 ng of purified plasmid-DNA providing a T7 promoter was transcribed and translated in the presence of [35S]methionine for 1 h at 30 °C. Proteins were then subjected to SDS-PAGE and visualized by autoradiography. FACS and fluorimeter analysis For FACS analysis transfected cells were harvested 48 h posttransfection and resuspended in 0.1% PBS/NaH3. Fluorescent-activated cell sorting was performed with a FACS Calibur from Becton Dickinson. Alternatively, harvested cells were treated as described for Western blot analysis, and 50 μg of total protein diluted in 2.5 ml PBS was subjected to a fluorimeter. Northern blot Nuclear and cytoplasmic RNA species were prepared using RNeasy-Kit (Qiagen) following the manufacturer's instructions.

304

M. Graf et al. / Virology 352 (2006) 295–305

Plasma membranes of 107 transfected cells were lysed using 175 μl lysis buffer (50 mM Tris, 140 mM NaCl, 1.5 mM MgCl2, 0.5% NP-40, pH 8.0), and nuclei were separated from the cytoplasm by centrifugation (300 × g, 2 min). Nuclear and cytoplasmic RNAs were prepared from the corresponding fractions using RNeasy spin columns, and polyadenylated mRNA was isolated using the Oligotex mRNA Kit from Qiagen following the manufacturer's instructions. For Northern blot analysis, RNA preparations were separated on a 1% agarose gel (40 mM morpholinepropanesulfonic acid, 10 mM NaAc, 1 mM EDTA, 6.5% formaldehyde, pH 7.0) and capillary-blotted onto a positively charged nylon membrane. The membrane was hybridized with a radiolabeled probe at 50°C over night. Hybridization was visualized by exposure to a phosphor imager screen and analyzed using Molecular Analyst Software (BioRad). An RRE-specific radiolabeled riboprobe was generated by in vitro run-off transcription of XhoI-linearized pCMV-ERR using Riboprobe in Vitro Transcription Systems (Promega) according to the manufacturer's instructions. A β-actin-specific radiolabeled DNA probe was synthesized by random priming and elongated with Ladderman Labeling Kit (Takara Biochemical) following the manufacturer's instructions. The β-actin cDNA was released by digestion of pGEM-β-actin with Hind III and EcoRI and gel purified. Acknowledgments This work was supported by DFG grant WA873/1-4. We thank Joachim Hauber and Roland Stauber for kindly providing the eukaryotic Rev and RevM10 expression plasmids, respectively. The human β-actin cDNA was obtained from Thomas Dobner, and leptomycin B was a generous gift from Barbara Wolff. MG and CL contributed equally to this work. References Afonina, E., Neumann, M., Pavlakis, G.N., 1997. Preferential binding of poly (A)-binding protein 1 to an inhibitory RNA element in the human immunodeficiency virus type 1 gag mRNA. J. Biol. Chem. 272, 2307–2311. Amendt, B.A., Hesslein, D., Chang, L.J., Stoltzfus, M., 1994. Presence of negative and positive cis-acting RNA splicing elements within and flanking the first tat coding exon of HIV-1. Mol. Cell. Biol. 14, 3960–3970. Beriault, V., Clement, J.F., Levesque, K., Lebel, C., Yong, X., Chabot, B., Cohen, E.A., Cochrane, A.W., Rigby, W.F., Mouland, A.J., 2004. A late role for the association of hnRNP A2 with the HIV-1 hnRNP A2 response elements in genomic RNA, Gag, and Vpr localization. J. Biol. Chem. 279, 44141–44153. Braun, I.C., Rohrbach, E., Schmitt, C., Izaurralde, E., 1999. TAP binds to the constitutive transport element (CTE) through a novel RNA-binding motif that is sufficient to promote CTE-dependent RNA export from the nucleus. EMBO J. 18, 1953–1965. 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 genome makes human immunodeficiency virus type 1 expression and replication Rev-independent. Proc. Natl. Acad. Sci. U.S.A. 91, 1256–1260. Brighty, D.W., Rosenberg, M., 1994. A cis-acting repressive sequence that overlaps the Rev-responsive element of human immunodeficiency virus type 1 regulates nuclear retention of env mRNAs independently of known splice signals. Proc. Natl. Acad. Sci. U.S.A. 91, 8314–8318.

Chang, D.D., Sharp, P.A., 1989. Regulation by HIV Rev depends upon recognition of splice sites. Cell 59, 789–795. Chen, C.-Y., Shyu, A.-B., 1995. AU-rich elements: characterization and importance in mRNA degradation. Trends Biochem. Sci. 20, 465–470. Cochrane, A.W., Jones, K.S., Beidas, S., Dillon, P.J., Skalka, A.M., Rosen, C.A., 1991. Identification and characterization of intragenic sequences which repress HIV structural gene expression. J. Virol. 65, 5305–5313. Coyle, J.H., Guzik, B.W., Bor, Y.-C., Jin, L., Eisner-Smerage, L., Taylor, S.J., Rekosh, D., Hammarskjold, M.L., 2003. Sam68 enhances the cytoplasmic utilization of intron-containing RNA and is functionally regulated by the nuclear kinase Sik/BRK. Mol. Cell. Biol. 23, 92–103. Cullen, B.R., 2003. Nuclear mRNA export: insights from virology. Trends Biochem. Sci. 28, 419–424. Elfgang, C., Rosorius, O., Hofer, L., Jaksche, H., Hauber, J., Bevec, D., 1999. Evidence for specific nucleocytoplasmic transport pathways used by leucine-rich nuclear export signals. Proc. Natl. Acad. Sci. U.S.A. 96, 6229–6234. Emerman, M., Vazeux, R., Peden, K., 1989. The rev gene product of the human immunodeficiency virus affects envelope-specific RNA localization. Cell 57, 1155–1165. Fischer, U., Meyer, S., Teufel, M., Heckel, C., Luhrmann, R., Rautmann, G., 1994. Evidence that HIV-1 Rev directly promotes the nuclear export of unspliced RNA. EMBO J. 13, 4104–4112. Fischer, U., Huber, J., Boelens, W.C., Mattaj, I.W., Luhrmann, R., 1995. The HIV-1 Rev activation domain is a nuclear export signal that accesses an export pathway used by specific cellular RNAs. Cell 82, 475–483. Fornerod, M., Ohno, M., Yoshida, M., Mattaj, I.W., 1997. CRM1 is an export receptor for leucine-rich nuclear export signals. Cell 90, 967–970. Gao, F., Li, Y., Decker, J.M., Peyerl, F.W., Bibollet-Ruche, F., Rodenburg, C.M., Chen, Y., Shaw, D.R., Allen, S., Musonda, R., Shaw, G.M., Zajac, A.J., Letvin, N., Hahn, B.H., 2003. Codon usage optimization of HIV type 1 subtype C gag, pol, env, and nef genes: in vitro expression and immune responses in DNA-vaccinated mice. AIDS Res. Hum. Retroviruses 19, 817–823. Graf, M., Bojak, A., Deml, L., Bieler, K., Wolf, H., Wagner, R., 2000. Concerted action of multiple cis-acting sequences is required for Rev dependence of late human immunodeficiency virus type 1 gene. J. Virol. 74, 10822–10826. Grüter, P., Tabemero, 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. Haas, J., Park, E.-C., Seed, B., 1996. Codon usage limitation in the expression of HIV-1 envelope glycoprotein. Curr. Biol. 6, 315–324. Hope, T.J., 1999. The ins and outs of HIV Rev. Arch. Biochem. Biophys. 365, 186–191. Hope, T.J., Huang, X., McDonald, D., Parslow, T.G., 1990. Steroid-receptor fusion of the human immunodeficiency virus type 1 Rev transactivator: mapping cryptic functions of the arginine-rich motif. Proc. Natl. Acad. Sci. U.S.A. 87, 7787–7791. Jurica, M.S., Moore, M.J., 2003. Pre-mRNA splicing—Awash in a sea of proteins. Mol. Cell 12, 5–14. Kammler, S., Leurs, C., Freund, M., Krummheuer, J., Seidel, K., Tange, T.O., Lund, M.K., Kjems, J., Scheid, A., Schaal, H., 2001. The sequence complementarity between HIV-1 5′ splice site SD4 and U1 snRNA determines the steady-state level of an unstable env pre-mRNA. RNA 7, 421–434. Kang, Y., Cullen, B.R., 1999. The human Tap protein is a nuclear mRNA export factor that contains novel RNA-binding and nucleocytoplasmic transport sequences. Genes Dev. 13, 1126–1139. Kang, Y., Bogerd, H.P., Cullen, B.R., 2000. Analysis of cellular factors that mediate nuclear export of RNAs bearing the Mason-Pfizer monkey virus constitutive transport element. J. Virol. 74, 5863–5871. Kim, S.Y., Byrn, R., Groopman, J., Baltimore, D., 1989. Temporal aspects of DNA and RNA synthesis during human immunodeficiency virus infection: evidence for differential gene expression. J. Virol. 63, 3708–3713. Kjems, J., Frankel, A.D., Sharp, P.A., 1991. Specific regulation of mRNA splicing in vitro by a peptide from HIV-1 Rev. Cell 67, 169–178.

M. Graf et al. / Virology 352 (2006) 295–305 Kotsopoulou, E., NarryKim, V., Kingsman, A.J., Kingsman, S.M., Mitrophanous, K.A., 2000. A Rev-independent human immunodeficiency virus type 1 (HIV-1)-based vector that exploits a codon-optimized HIV-1 gag-pol gene. J. Virol. 74, 4839–4852. Kudo, N., Wolff, B., Sekimoto, T., Schreiner, E.P., Yoneda, Y., Yanagida, M., Horinouchi, S., Yoshida, M., 1998. Leptomycin B inhibition of signalmediated nuclear export by direct binding to CRM1. Exp. Cell Res. 242, 540–547. Legrain, P., Rosbash, M., 1989. Some cis- and trans-acting mutants for splicing target pre-mRNA to the cytoplasm. Cell 57, 573–583. Li, J., Liu, Y., Kim, B.O., He, J.J., 2002. Direct participation of Sam68, the 68-kilodalton Src-associated protein in mitosis, in the Crm1-mediated Rev nuclear export pathway. J. Virol. 76, 8374–8382. Lu, X.B., Heimer, J., Rekosh, D., Hammarskjold, M.L., 1990. U1 small nuclear RNA plays a direct role in the formation of a rev-regulated human immunodeficiency virus env mRNA that remains unspliced. Proc. Natl. Acad. Sci. U.S.A. 87, 7598–7602. Maldarelli, F., Martin, M.A., Strebel, K., 1991. Identification of posttranscriptionally active inhibitory sequences in human immunodeficiency virus type 1 RNA: novel level of gene regulation. J. Virol. 65, 5732–5743. Malim, M.H., Tiley, L.S., McCarn, D.F., Rusche, J.R., Hauber, J., Cullen, B.R., 1990. HIV-1 structural gene expression requires binding of the Rev transactivator to its RNA target sequence. Cell 60, 675–683. Malim, M.H., Bohnlein, S., Hauber, J., Cullen, B.R., 1989a. Functional dissection of the HIV-1 Rev trans-activator-derivation of a trans-dominant repressor of Rev function. Cell 58, 205–214. Malim, M.H., Hauber, J., Le, S.Y., Maizel, J.V., Cullen, B.R., 1989b. The HIV-1 rev trans-activator acts through a structured target sequence to activate nuclear export of unspliced viral mRNA. Nature 338, 254–257. McLaren, M., Asai, K., Cochrane, A., 2004. A novel function for Sam68: enhancement of HIV-1 RNA 3 (processing. RNA 10, 1119–1129. Mikaelian, I., Krieg, M., Gait, M.J., Karn, J., 1996. Interactions of INS (CRS) elements and the splicing machinery regulate the production of Revresponsive mRNAs. J. Mol. Biol. 257, 246–264. Najera, I., Krieg, M., Karn, J., 1999. Synergistic stimulation of HIV-1 Revdependent export of unspliced mRNA to the cytoplasm by hnRNP A1. J. Mol. Biol. 285, 1951–1964. Nasioulas, G., Zolotukhin, A.S., Tabernero, C., Solomin, L., Cunningham, C.P., Pavlakis, G.N., Felber, B.K., 1994. Elements distinct from HIV-1 splice sites are responsible for the Rev dependence of env mRNA. J. Virol. 68, 2986–2993. Neville, M., Stutz, F., Lee, L., Davis, L.I., Rosbash, M., 1997. The importin-β family member CRM1p bridges the interaction between Rev and the nuclear pore complex during nuclear export. Curr. Biol. 7, 767–775. Nguyen, K.L., Llano, M., Akari, H., Miyagi, E., Poeschla, E.M., Strebel, K., Bour, S., 2004. Codon optimization of the HIV-1 vpu and vif genes stabilizes their mRNA and allows for highly efficient Rev-independent expression. Virology 319, 163–175. Olsen, H.S., Nelbock, P., Cochrane, A.W., Rosen, C.A., 1990. Secondary structure is the major determinant for interaction of HIV rev protein with RNA. Science 247, 845–848. Olsen, H.S., Cochrane, A.W., Rosen, C., 1992. Interaction of cellular factors with intragenic cis-acting repressive sequences within the HIV genome. Virology 191, 709–715.

305

O'Reilly, M.M., McNally, M.T., Beemon, K.L., 1995. Two strong 5′ splice sites and competing, suboptimal 3′ splice sites involved in alternative splicing of human immunodeficiency virus type 1. Virology 213, 373–385. Reddy, T.R., Xu, W., Mau, J.K., Goodwin, C.D., Suhasini, M., Tang, H., Frimpong, K., Rose, D.W., Wong-Staal, F., 1999. Inhibition of HIV replication by dominant negative mutants of Sam68, a functional homolog of HIV-1. Rev. Nat. Med. 5, 635–642. Schneider, R., Campbell, M., Nasioulas, G., Felber, B.K., Pavlakis, G.N., 1997. Inactivation of the human immunodeficiency virus type 1 inhibitory elements allows Rev-independent expression of Gag and Gag/protease and particle formation. J. Virol. 71, 4892–4903. Schwartz, S., Felber, B.K., Benko, D.M., Fenyo, E.-M., Pavlakis, G.N., 1990. Cloning and functional analysis of multiply spliced mRNA species of human immunodeficiency virus type 1. J. Virol. 64, 2519–2529. Schwartz, S., Campbell, M., Nasioulas, G., Harrison, J., Felber, B.K., Pavlakis, G.N., 1992a. Mutational inactivation of an inhibitory sequence in human immunodeficiency virus type 1 results in Rev-independent gag expression. J. Virol. 66, 7176–7182. Schwartz, S., Felber, B.K., Pavlakis, G.N., 1992b. Distinct RNA sequences in the gag region of human immunodeficiency virus type 1 decrease RNA stability and inhibit expression in the absence of Rev protein. J. Virol. 66, 150–159. Si, Z.H., Rauch, D., Stoltzfus, C.M., 1998. The exon splicing silencer in human immunodeficiency virus type 1 Tat exon 3 is bipartite and acts early in spliceosome assembly. Mol. Cell. Biol. 18, 5404–5413. Staffa, A., Cochrane, A., 1994. The tat/rev intron of human immunodeficiency virus type 1 is inefficiently spliced because of suboptimal signals in the 3′ splice site. J. Virol. 68, 3071–3079. Venkatesh, L.K., Gettermeier, T., Chinnadurai, G., 2003. A nuclear kinesin-like protein interacts with and stimulates the activity of the leucine-rich nuclear export signal of the human immunodeficiency virus type 1 Rev protein. J. Virol. 77, 7236–7243. Wagner, R., Graf, M., Bieler, K., Wolf, H., Grunwald, T., Foley, P., Überla, K., 2000. Rev-independent expression of synthetic gag-pol genes of human immunodeficiency virus type 1 and simian immunodeficiency virus: implications for the safety of lentiviral vectors. Hum. Gene Ther. 11, 2403–2413. Wolff, B., Sanglier, J.J., Wang, Y., 1997. Leptomycin B is an inhibitor of nuclear export: inhibition of nucleo-cytoplasmic translocation of the human immunodeficiency virus type 1 (HIV-1) Rev protein and Rev-dependent mRNA. Chem. Biol. 4, 139–147. Wolff, H., Brack-Werner, R., Neumann, M., Werner, T., Schneider, R., 2003. Integrated functional and bioinformatics approach for the identification and experimental verification of RNA signals: application to HIV-1 INS. Nucleic Acids Res. 31, 2839–2851. Yedavalli, V.S., Neuveut, C., Chi, Y.H., Kleiman, L., Jeang, K.T., 2004. Requirement of DDX3 DEAD box RNA helicase for HIV-1 Rev-RRE export function. Cell 119, 381–392. Zolotukhin, A.S., Michalowski, D., Bear, J., Smulevitch, S.V., Traish, A.M., Peng, R., Patton, J., Shatsky, I.N., Felber, B.K., 2003. PSF acts through the human immunodeficiency virus type 1 mRNA instability elements to regulate virus expression. Mol. Cell. Biol. 23, 6618–6630.