Skip to main content
NCBI home page
RNA logoLink to RNA
. 2010 Jul;16(7):1328–1339. doi: 10.1261/rna.1887910

Titers of lentiviral vectors encoding shRNAs and miRNAs are reduced by different mechanisms that require distinct repair strategies

Ying Poi Liu 1, Monique A Vink 1, Jan-Tinus Westerink 1, Eva Ramirez de Arellano 1, Pavlina Konstantinova 1, Olivier Ter Brake 1, Ben Berkhout 1
PMCID: PMC2885682  PMID: 20498457

Abstract

RNAi-based gene therapy is a powerful approach to treat viral infections because of its high efficiency and sequence specificity. The HIV-1-based lentiviral vector system is suitable for the delivery of RNAi inducers to HIV-1 susceptible cells due to its ability to transduce nondividing cells, including hematopoietic stem cells, and its ability for stable transgene delivery into the host cell genome. However, the presence of anti-HIV short hairpin RNA (shRNA) and microRNA (miRNA) cassettes can negatively affect the lentiviral vector titers. We show that shRNAs, which target the vector genomic RNA, strongly reduced lentiviral vector titers but inhibition of the RNAi pathway via saturation could rescue vector production. The presence of miRNAs in the vector RNA genome (sense orientation) results in a minor titer reduction due to Drosha processing. A major cause for titer reduction of miRNA vectors is due to incompatibility of the cytomegalovirus promoter with the lentiviral vector system. Replacement of this promoter with an inducible promoter resulted in an almost complete restoration of the vector titer. We also showed that antisense poly(A) signal sequences can have a dramatic effect on the vector titer. These results show that not all sequences are compatible with the lentiviral vector system and that care should be taken in the design of lentiviral vectors encoding RNAi inducers.

Keywords: lentiviral vector, RNAi, HIV-1, gene therapy, antiviral

INTRODUCTION

RNAi-mediated gene silencing is a powerful therapeutic approach to target disease-associated mRNAs and transcripts encoded by pathogenic viruses because of its high efficiency and sequence-specificity (Haasnoot and Berkhout 2006; Haasnoot et al. 2007b; Kim and Rossi 2007). In mammalian cells, stable RNAi can be obtained by intracellular expression of short hairpin RNAs (shRNAs) (Brummelkamp et al. 2002; Paddison et al. 2002). These transcripts are transported to the cytoplasm by Exportin-5 and processed by Dicer into small interfering RNAs (siRNAs) of ∼21 base pairs (bp) with 2-nucleotide (nt) 3′ overhangs. The siRNA duplex is incorporated into the RNA-induced silencing complex (RISC). The passenger strand of the siRNA is degraded and the guide strand of the siRNA programs RISC to cleave the perfectly complementary mRNA. Another vector-based RNAi approach is the use of artificial microRNAs (miRNAs) that closely resemble cellular miRNAs. These inhibitors are expressed as primary miRNAs (pri-miRNAs) that are cleaved by the RNase III-like endonuclease Drosha and its cofactor DGCR8 into precursor miRNAs (pre-miRNAs) (Han et al. 2004). Pre-miRNAs are hairpin RNAs of ∼70 nt in size that are transported to the cytoplasm by Exportin-5 and further processed by Dicer into an imperfect ∼22-nt miRNA duplex (Yi et al. 2003; Lund et al. 2004). The single-stranded mature miRNA directs RISC to complementary mRNA sequences to cause mRNA cleavage or translational repression, depending on the complementarity between the miRNA and the mRNA target (Bartel 2004; Filipowicz 2005).

We and others previously demonstrated that potent HIV-1 inhibition can be obtained using antiviral shRNAs and designed miRNAs (Banerjea et al. 2003; Boden et al. 2004a,b; Das et al. 2004a; Westerhout et al. 2005; Ter Brake et al. 2006, 2008; Liu et al. 2008) However, HIV-1 can escape through the selection of a single nucleotide substitution within the target sequence (Boden et al. 2003; Das et al. 2004a; Westerhout et al. 2005). For a durable RNAi-based gene therapy against HIV-1, a combinatorial attack is required in which multiple viral sequences are targeted simultaneously (Ter Brake et al. 2006, 2008). Therefore, we previously generated extended shRNA constructs that encode two or three active siRNAs (e2- or e3-shRNA) and an antiviral miRNA polycistron that encodes four active miRNAs (Liu et al. 2007, 2008, 2009b).

For the delivery of RNAi-based antivirals to HIV-1 susceptible cells, the lentiviral vector system is very attractive because it is highly efficient in transducing nondividing cells, including hematopoietic stem cells (Van den Haute et al. 2003; Gimeno et al. 2004; Harper et al. 2005; Ralph et al. 2005). In a durable HIV-1 treatment one can transduce CD34+ hematopoietic stem cells ex vivo with a lentiviral vector encoding HIV-1 specific shRNAs or miRNAs. The transduced cells should stably express the inhibitors and thus become resistant to HIV-1. Subsequently, the transduced stem cells can be engrafted back into the patient, where they will give rise to an HIV-1 resistant myeloid and lymphoid cell population. However, the presence of shRNA and miRNA cassettes can negatively affect the lentiviral vector titers, which may hamper clinical applications (Poluri and Sutton 2007; Ter Brake and Berkhout 2007). It is therefore important to study the mechanisms by which RNAi inducers affect the vector titer in order to propose methods that could restore titers.

There are several possible causes for reduction in titers. These inhibitory possibilities on lentiviral vector titers are illustrated in Figure 1, for miRNA vectors on the left and shRNA vectors on the right. An obvious problem for anti-HIV shRNAs and miRNAs can be that the target sequence is also present in the HIV-based vector genome. This vector targeting problem is marked as mechanism 1 and can simply be avoided by selecting antivirals that do not target the lentiviral vector. The shRNA or miRNA can possibly target its own sequence as part of the lentiviral vector genome (self-targeting) (Fig. 1, mechanism 2). Vector RNA genomes with a miRNA cassette face the additional problem that Drosha cleavage in the nucleus may inactivate the vector genome, either in the producer or transduced cell (Fig. 1, mechanism 3). It is also possible that transcription of the miRNA unit by polymerase II could interfere with transcription of the lentiviral vector (Fig. 1, mechanism 4). In general, expression of shRNAs or miRNAs may cause a titer reduction due to aspecific toxicity, e.g., by an unwanted off-target effect. In addition, stable RNA hairpin structures introduced by the RNAi cassettes may affect vector titers by affecting RNA nuclear export, genomic RNA packaging, or reverse transcription in the target cells (Suo and Johnson 1997; Klasens et al. 1999).

FIGURE 1.

FIGURE 1.

Open in a new tab

Possible causes for reduced titers of lentiviral vectors encoding miRNAs and shRNAs. The different mechanisms that can account for titer reduction in miRNA vectors (left panel) and shRNA vectors (right panel) are illustrated. First, anti-HIV-1 miRNA sequences can target similar sequences in the HIV-1-based lentiviral vector (mechanism 1, vector targeting). Second, mature miRNAs can target their own sequence as part of the lentiviral vector genome (mechanism 2, self-targeting). If the miRNA expression cassette is cloned in the sense orientation into the lentiviral vector, the target as part of the vector genome is imperfectly complementary. When the miRNA expression cassette is cloned in the antisense orientation, a perfectly complementary target sequence will be present in the vector genome. Third, miRNAs in the vector RNA genome can be processed by Drosha and thus lead to destruction of the vector genome (mechanism 3). Fourth, CMV-driven polymerase II transcription of the miRNA may interfere with RSV-driven polymerase II transcription of the vector RNA genome (mechanism 4). The situation for anti-HIV-1 shRNA expressing vectors seems less complex. The shRNAs can target HIV-1 sequences in the lentiviral vector (mechanism 1, vector targeting). Also, self-targeting of the shRNA sequence as part of the lentiviral vector genome can cause titer reduction (mechanism 2). However, this scenario is unlikely because the target is protected in a tight hairpin structure and thus becomes inaccessible to RISC.

Here, we studied the effect of different RNAi inducers: single shRNA or multiplex e3-shRNA cassettes and single miRNA or multiplex miRNA expression cassettes on the titer of the lentiviral vector. We demonstrate that the titers were drastically reduced by up to 1000-fold for some vectors compared to the control vector. The titer reduction of the shRNA and miRNA vectors is caused by different mechanisms. Based on this insight, we tested specific countermeasures that resulted in significantly improved titers. These insights are important for the clinical development of lentiviral vectors that induce RNAi to treat diseases.

RESULTS

Reduced titer of lentiviral vectors encoding RNAi inducers against HIV-1

We previously constructed several RNAi inducers against HIV-1, including shRNA and miRNA molecules (Das et al. 2004a; Ter Brake et al. 2006; Liu et al. 2008, 2009b). These inhibitors and the corresponding target sites in the HIV-1 RNA genome are indicated in Figure 2A. For combinatorial RNAi approaches, we also generated constructs that target multiple sites in the HIV-1 RNA genome: the e3-shRNA and the antiviral 4-miRNA polycistron (Liu et al. 2008, 2009b). The e3-shRNA encodes three siRNAs that target the nef19, pol1, and rev/tat (r/t) regions (Fig. 2A). The antiviral 4-miRNA polycistron encodes four miRNAs that target pol47, gag, r/t, and leader (ldr) sequences. The shRNA and e3-shRNA expression cassettes were cloned in the lentiviral vector JS1, which encodes the eGFP marker gene from an independent transcription unit driven by the PGK promoter (Fig. 2B). We used the H1 RNA polymerase III promoter to drive shRNA/e3-shRNA expression and these cassettes are cloned in the antisense (AS) orientation. The orientation of the shRNA expression cassette within the vector genome does not influence the titer (Ter Brake and Berkhout 2007). The expression of the miRNA units are controlled by the constitutive immediate early promoter of cytomegalovirus (CMV). These expression units were inserted into JS1 in the sense (S) orientation to avoid interference with transcription of the lentiviral RNA genome.

FIGURE 2.

FIGURE 2.

Open in a new tab

The RNAi inducers used in this study. (A) Structure of the shRNA, miRNA, e3-shRNA, and the antiviral 4-miRNA polycistron and the target positions within the HIV-1 genome are indicated by color coding. (B) The empty JS1 lentiviral vector is shown at the top with the RNAi-inducing gene cassettes that were introduced into the multiple cloning site (MCS). The shRNA and e3-shRNA are driven by the polymerase III H1 promoter and were cloned in antisense (AS) orientation. The miRNAs are expressed from the polymerase II CMV promoter and were cloned in the sense (S) orientation.

To test whether shRNAs and miRNAs against HIV-1 affect the lentiviral vector in terms of production of viral particles and titers, we produced vesicular stomatitis virus (VSV) G-pseudotyped vector particles by transient co-transfection of HEK 293T cells. We studied whether the production of the lentiviral vector particles was affected by determining the capsid (CA-p24) level in the culture supernatant after 2 d. As a control, we included the empty JS1 lentiviral vector. No gross differences in capsid level were measured for the different lentiviral vectors (Fig. 3A). To determine the titer, we subsequently transduced SupT1 T cells with a dilution series of the produced vectors. The percentage of eGFP positive cells was measured by FACS to determine the transduction units at 3 d post-transduction. Similar titers were measured for the empty JS1 vector and the shNef vector (Fig. 3B). For the shLdr vector we measured a dramatic 188-fold reduction of titer. This reduction is caused by targeting of HIV-1 Ldr sequences that are also present in the vector backbone (Fig. 2B; Ter Brake and Berkhout 2007). Thus, shLdr serves as a control for the impact of a direct RNAi attack on the lentiviral vector. The same mechanism may explain the drop in titer for e3-shRNA, which encodes the siNef19 inhibitor that attacks the lentiviral genome in the 3′ untranslated region (Fig. 2B). A profound titer reduction of 56- and 70-fold was observed for the miRNA and 4-miRNA vectors, respectively (Fig. 3B).

FIGURE 3.

FIGURE 3.

Open in a new tab

Lentiviral vector particle production and titers of the vectors encoding the RNAi inducers. (A) Lentiviral vector particle production was determined by measuring the CA-p24 levels of the lentiviral vector stocks. (B) Lentiviral vector titers (transducing units/mL [TU/mL]) were determined by measuring the percentage of eGFP positive cells at 3 d post-transduction. The mean values and standard deviations were shown from four independent transfections that were performed in duplo.

Inhibition of the RNAi pathway to increase the lentiviral vector titer

The shLdr and e3-shRNA constructs encode siRNAs that directly target vector RNA sequences (Fig. 2B). We therefore expected that inhibition of the RNAi pathway during vector production would lead to a (partial) repair of the titers. To test this, we used different approaches to abort the RNAi mechanism by co-transfection of the following: an excess luciferase reporter with the corresponding shRNA target sequence (Luc) as RNAi target decoy; an excess shRNAs (a 5xshRNA plasmid encoding five shRNAs) to saturate the RNAi machinery; a plasmid encoding VA RNA of Adenovirus as Dicer inhibitor (Andersson et al. 2005); and a plasmid encoding the RNAi suppressor protein VP35 of Ebola virus (Haasnoot et al. 2007a; de Vries et al. 2008), siRNAs against Dicer (Paddison et al. 2002), or an shRNA against Drosha. Furthermore, we tested whether overexpression of the chromosome region maintenance 1 (CRM1) protein (Popa et al. 2002), which is involved in the nuclear export of unspliced and partially spliced viral RNAs, could improve production of viral particles and thereby the titer (Wodrich and Krausslich 2001; Rawlinson et al. 2009). We co-transfected the different RNAi inhibitors with the lentiviral vector and the standard set of packaging plasmids and subsequently determined the titer. The empty JS1 lentiviral vector was included as control and its titer is not significantly affected by the different effector molecules (Fig. 4A). Co-transfection of the RNAi inhibitors did not affect the titer of the shNef vector either (Fig. 4B, left). The titer of the shLdr vector is extremely low due to vector targeting, but was significantly improved by an excess shRNAs, siRNA against Dicer, and the shRNA against Drosha (Fig. 4B, right).

FIGURE 4.

FIGURE 4.

Open in a new tab

Inhibition of the RNAi pathway to improve the lentiviral vector titers. (A) Titers of the empty JS1 lentiviral vector and the effect of saturation of the RNAi pathway. The plasmid encoding Bluescript (pBS) was used as a control. Decoy RNAi targets encoded in luciferase reporters, excess shRNA (p5xshRNA), VA RNA (pVA RNA), VP35 (pVP35), siRNA against Dicer (siDicer), and shRNA against Drosha (pshDrosha) were used to saturate the RNAi pathway. CRM1 (pCRM1) was used to increase nuclear export of the vector RNA genome. The titers were determined by measuring the percentage of eGFP positive cells at 3 d post-transduction. The same variations were tested for JS1 vectors with the following RNAi cassettes: (B) H1-shNef and H1-shLdr; (C) H1-e3-shRNA; and (D) CMV-miRNA and CMV-4-miRNA. The mean values and standard deviations were shown that were based on four independent experiments.

The fact that knockdown of Drosha could affect the shRNA pathway may seem odd. However, we used an shRNA to reduce the Drosha expression level. This shRNA needs to be processed by Dicer and is subsequently incorporated into RISC to cause Drosha knockdown. Thus, we think it is not the Drosha knockdown itself that caused an improved titer, but rather the saturation of RISC by siRNAs against Drosha. Similarly, overexpression of five irrelevant shRNAs (5xshRNAs) could augment the titer, indicating that any siRNA can cause this effect. The titer of the shLdr vector can be improved to the level that is only twofold lower than that of the shNef vector. A very similar pattern of improvement was apparent for the e3-shRNA vector, although the initial defect due to vector targeting and thus, the magnitude of repair, is more modest than observed for the shLdr vector (Fig. 4C). However, inhibition of the RNAi pathway did not significantly improve the titer of the two miRNA vectors (Fig. 4D). These results suggest a different cause of the titer reduction in the miRNA vectors.

Titer reduction of miRNA vectors is partially Drosha dependent

The low titer observed for the miRNA vectors is likely caused by recognition and processing of the miRNAs as part of the vector RNA genome by the endonuclease Drosha (Fig. 1, mechanism 3). However, knocking down Drosha during vector production did not affect the titer of the miRNA vectors (Fig. 4D). This may be due to a long intracellular half-life of the Drosha enzyme. To intensify Drosha knockdown in the producer cells, we performed another experiment in which we transfected the shDrosha plasmid 2 and 1 d prior to vector production. Drosha knockdown of ∼60% in the producer cells was measured by Western blot analysis (results not shown). Similar viral particle production was measured for the JS1 control and the two miRNA vectors in the presence of shDrosha or the control shLuc construct (Fig. 5A). The miRNA vectors showed the expected drop in titer. Interestingly, titers of both CMV-miRNA and CMV-4-miRNA showed a significant improvement upon intensified Drosha knockdown (Fig. 5B). This up-regulation was not observed for the JS1 control, indicating that the titer reduction is at least partially Drosha dependent.

FIGURE 5.

FIGURE 5.

Open in a new tab

Titer decrease of miRNA vectors is partially Drosha dependent. (A) miRNA expressing vectors were produced in cells where Drosha was knocked down using intensified shRNA treatment. As a negative control shRNAs against Luc (shLuc) was used. Capsid p24 levels were measured by ELISA for the JS1, CMV-miRNA, and CMV-4-miRNA vectors. (B) Titers of these vectors that were produced in cells with or without Drosha knockdown. Averages and standard deviations were derived from three independent experiments.

Removal of the CMV promoter from miRNA vectors restores the titers

To study the titer problem of the CMV-4-miRNA vector in more detail, we made a new construct in which the miRNA unit is placed in the antisense orientation (Fig. 6A, CMV-4-miRNA AS). This means that the complement of the miRNA sequence will end up in the vector RNA genome, thus avoiding Drosha recognition. We measured only a modest titer improvement compared to the sense vector (Fig. 6B). The slight difference in results in Figure 5B versus Figure 6B may be due to differential experimental setups, e.g., the triple versus single transfection. Combined with the results of the Drosha knockdown experiment (Fig. 5B), it seems that Drosha-mediated cleavage of the vector genome (Fig. 1, mechanism 3) contributes only slightly to the observed titer reduction.

FIGURE 6.

FIGURE 6.

Open in a new tab

Absence of miRNA expression during vector production restores titers. (A) Schematic of the JS1 vector and the miRNA gene cassettes that were introduced into this vector. The antiviral 4-miRNA polycistron under the control of the CMV promoter was cloned in the sense or antisense orientation (CMV-4-miRNAs S and AS) into the JS1 vector. 4-miRNA S and 4-miRNA AS were generated by removal of the CMV promoter. (B) Titers of the vectors described in A were determined. Averages and standard deviations were derived from three independent experiments.

Thus, other explanations are required to explain the low titer of the CMV-4-miRNA AS vector. To test whether miRNA expression is the direct cause of the titer problem, we modified both the sense and antisense constructs by deletion of the CMV promoter (Fig. 6A, 4-miRNA S, 4-miRNA AS). By removal of the CMV promoter, the titer of both miRNA vectors was improved approximately 400-fold (Fig. 6B). These results indicate that either miRNA expression or the CMV promoter (its sequence or activity) caused the titer reduction. Notably, the titer of the promoterless miRNA vector in antisense orientation was restored to the level obtained with the empty JS1 lentiviral vector. The titer of the miRNA vector in sense orientation still showed a minor reduction, which is in fact expected due to Drosha processing of the miRNA segment as part of the vector RNA genome (Fig. 1, mechanism 3).

These results suggest that there are at least two causes for the titer reduction in miRNA vectors. Expression of the miRNA seems to have a major impact on the titer, and the sense constructs additionally encounter the problem of Drosha-mediated processing of the miRNA-containing vector genome. In addition, CMV-driven transcription of the miRNA unit may have a negative impact on the titer. We therefore decided to measure the level of mature miRNA expression from the lentiviral vectors. For this, we used four luciferase reporters with the appropriate HIV-1 targets. We observed very similar results for all reporters (Fig. 7). Potent inhibition of luciferase expression was measured for the miRNA constructs in the sense orientation, but to our surprise not for those in the antisense orientation (Fig. 7). These results indicate that both sense constructs express the miRNA. This result is expected for the CMV-4-miRNA construct, but may seem strange for the CMV-less construct due to the lack of a promoter. In the latter case the miRNA is in fact processed from the vector genomic transcript during vector production. This also means that these miRNAs will not be produced in the transduced cells because we used a self-inactivating (SIN) lentiviral vector that removes the upstream promoter during transduction. In contrast, the antisense CMV-4-miRNA construct did not express active miRNAs, which is likely due to the absence of a polyadenylation [poly(A)] signal for the primary miRNA transcript, which may reduce the transcript stability. The sense constructs will produce pri-miRNA transcripts that use the poly(A) signal within the 3′LTR of the vector genome. The combined results indicate that it is not the expression of the miRNAs that causes the severe titer reduction. Instead, the CMV promoter (its sequence or activity), either in sense or antisense orientation, is causing the titer reduction.

FIGURE 7.

FIGURE 7.

Open in a new tab

Luciferase reporter knockdown by the lentiviral constructs encoding the 4-miRNA polycistron. To test for miRNA expression and activity, we co-transfected the miRNA constructs with corresponding luciferase reporter constructs and a plasmid encoding Renilla luciferase. The normalized luciferase expression in the presence of the JS1 vector was set at 100% for each luciferase reporter. An shRNA construct against the specific target was used as positive control. The mean values and standard deviations were shown from two independent transfections that were performed in duplo.

Excluding miRNA-mediated titer reduction

Our results suggest that expression of antiviral miRNAs in the producer cells is not the cause of the poor vector titers. To confirm this we co-transfected a miRNA expression plasmid with the CMV-4-miRNA cassette (Liu et al. 2008) during lentiviral vector production. The plasmid Bluescript (pBS) was used as a negative control. The titer of the control JS1 vector was not affected by the expression of the antiviral 4-miRNA polycistron in trans (Fig. 8), which confirms the observation that expression of the miRNAs does not cause the decrease in vector titer. Likewise, the titer of the sense miRNA vectors was not significantly affected. These results suggest that the expressed mature miRNAs are not targeting the partially complementary sequences that are part of the lentiviral vector genome (Fig. 1 mechanism 2). However, when the miRNA cassette is present in the antisense orientation, a perfect complementary target will reside in the lentiviral vector genome for the miRNAs. Indeed, we observed a ∼2 log titer reduction for the antisense miRNA vectors when the 4-miRNA construct was provided in trans during vector production. Similar results were obtained in co-transfections with another 4-miRNA construct that expresses the same set of siRNAs, except for the vector-targeting siLdr (results not shown).

FIGURE 8.

FIGURE 8.

Open in a new tab

Titer reduction is not caused by miRNA expression. Titers of the set of the 4-miRNAs vectors that were produced in cells with or without 4-miRNA overexpression in trans. pBS was used as a negative control.

Titers of lentiviral vectors containing inducible promoters for miRNA expression

To further examine whether the titer reduction of miRNA vector was due to the presence or activity of the CMV promoter, we constructed a new set of lentiviral vectors that express the four miRNAs from a tetracycline inducible (tetO) promoter (Fig. 9A). Expression of the miRNAs could be induced by addition of doxycycline in the presence of rtTA transactivator protein. The advantage of using a conditionally regulated promoter is that we could also determine whether miRNA expression has a negative effect on the lentiviral vector titer. As controls, we constructed lentiviral vectors encoding only the tetO promoter and the poly(A) signal, both in the sense and antisense orientations. The tetO-4-miRNA vectors were cloned with or without the poly(A) signal in both the sense and antisense orientation (Fig. 9A).

FIGURE 9.

FIGURE 9.

Open in a new tab

Titer of lentiviral vectors with inducible miRNA cassettes. (A) Schematic of the JS1 lentiviral vector and the tetracycline (tetO) inducible 4-miRNA gene cassettes that were inserted. The tetO promoter and the poly(A) signal were cloned in the lentiviral vector as controls. The antiviral 4-miRNA polycistron under the control of the tetO-promoter was cloned in the sense or antisense orientation without a poly(A) signal (tetO-4-miRNA S or tetO-4-miRNA AS) or with a poly(A) signal (tetO-4-miRNA-pA S or tetO-4-miRNA-pA AS). (B) The titer of the vectors described in A was determined by measuring the percentage of eGFP positive SupT1 cells 3 d post-transduction. Averages and standard deviations were derived from four independent experiments.

To produce lentiviral vector particles, we co-transfected the newly constructed lentiviral vectors with the packaging plasmids and an additional plasmid expressing the rtTA protein (Das et al. 2004b). To test the impact of promoter activity and miRNA expression on the titer, we added 0, 10, 100, or 1000 ng/mL doxycycline during lentiviral vector production. The produced lentiviral vector particles were subsequently used to transduce target cells in the absence of doxycycline. We observed high lentiviral vector titers for the vectors in the sense orientation. The titers were increased ∼2 log compared to the CMV vectors. When the SV40 poly(A) signal was introduced an approximate 0.5 log reduction of the titer was observed (tetO-4-miRNA-pA). Strikingly, we measured an increase in vector titers upon induction of miRNA expression with doxycycline. Addition of more doxycycline during lentiviral vector production resulted in higher vector titers. Another picture was observed for the antisense constructs, especially the tetO-pA AS construct showed a very low titer. However, the tetO-4-miRNA AS vector showed an almost 2 log improvement of titer compared to tetO-pA-AS. Inclusion of the poly(A) signal in tetO-4-miRNA-pA AS also reduced the titer by 100-fold. These results suggest that the presence of the poly(A) signal in the antisense orientation caused a dramatic reduction in vector titer. Another interesting observation is that the induction of miRNA expression in the antisense vectors does not result in a drop in titer due to self-targeting (Fig. 1, mechanism 2).

These combined results suggest that the titer reduction of miRNA-containing lentiviral vectors is due to a multiplicity of causes, including Drosha cleavage of the vector RNA genome (sense only), the CMV promoter sequence or activity (sense and antisense) and the presence of poly(A) signal sequences in the antisense orientation. Our data stress the importance of a careful design of miRNA-encoding lentiviral vectors for clinical applications.

DISCUSSION

Lentiviral vectors encoding shRNAs or miRNAs are useful tools to specifically knock down disease-associated mRNAs and some are now being considered for clinical applications (Castanotto and Rossi 2009). Furthermore, these vectors are also powerful laboratory tools to study gene functions. Although relatively low vector titers are sufficient for many in vitro applications, high vector titers are important for clinical applications. In this study, we set out to test the negative impact of anti-HIV RNAi inducers, either shRNAs or miRNAs, on the lentiviral vector system that is well-suited for gene therapy applications in AIDS patients. We found that the insertion of a single shRNA expression cassette did not reduce the titer of the lentiviral vector. However, the lentiviral vector titer is severely reduced when the antiviral shRNA targets sequences of the vector system (Fig. 1, mechanism 1; Ter Brake and Berkhout 2007).

We could repair the shRNA vector titer by inhibition of the RNAi pathway using excess shRNAs, siRNA against Dicer, or shRNA against Drosha. The impact of the latter two reagents can in part be attributed to knockdown of Dicer and Drosha, but it is likely that also in these cases RNAi is predominantly inhibited via saturation of RISC, since irrelevant siRNAs derived from shRNAs have the same effect. Other RNAi inhibitory factors including VA RNA, VP35, and luciferase-based mRNA targets did not improve the titers. For VP35 and the luciferase targets, the expression level may be too low to compete with the siRNAs that target the lentiviral vector. This is less likely for VA RNA because it is expressed from the same promoter as the shRNAs. An explanation could be that shRNAs are more efficiently processed by Dicer and incorporated into RISC than VA RNA. We also did not observe an increase in vector titer when CRM1 was overexpressed during vector production. As CRM1 is involved in nuclear RNA export we attempted to increase vector RNA genome expression. This result suggests that the titer reduction is not due to nuclear retention of viral RNA genomes. To analyze the titer defect in more detail, it is important to determine whether full-length vector RNA is made in the producer cells and whether these are correctly packaged into lentiviral particles.

A more complex situation was apparent for miRNA-expressing lentiviral vectors, which showed dramatically reduced titers. Inhibition of the RNAi pathway by saturation or the use of RNAi suppressors did not restore the titer. The decline in the titer could be caused by Drosha recognition and processing of the pri-miRNA as part of the vector RNA genome (Fig. 1, mechanism 3, sense orientation of miRNA unit). Indeed, intensified shRNA-mediated knockdown of Drosha resulted in a significant improvement of the vector titer, but the titer levels are still much reduced compared to the titer of the empty vector. Consistent with this idea is the observation that the titer is slightly improved when the miRNA cassette is placed in the antisense orientation to avoid Drosha processing of the vector RNA genome.

Vectors containing antisense miRNA cassettes encounter another problem because the expressed mature miRNA may attack the fully complementary target sequence in the vector RNA genome in the producer cell, thus causing a drop in titer (Fig. 1, mechanism 2). We showed that this self-targeting by co-transfection of a miRNA vector in trans caused a 2-log reduction in vector titers. However, miRNA expression in cis may affect the titers differently. In this respect, miRNA vectors differ from shRNA expressing vectors that are not subjected to self-targeting because the target sequence is folded in a stable RNA structure that occludes the RNAi target (Fig, 1, right panel, mechanism 2; Westerhout et al. 2005; Ter Brake and Berkhout 2007; Westerhout and Berkhout 2007). Similarly, Zhou et al. (2009) showed that shRNA expressing lentiviral vector do not have reduced titers caused by self-targeting. In contrast, Poluri and Sutton (2007) showed that lentiviral vectors encoding shRNAs have reduced titers because the vector RNA serves as a target for the expressed shRNA. These discrepancies likely originate from the fact that they used a different lentiviral vector system. Similar to our findings, Poluri and Sutton (2007) showed that the titer reduction can be alleviated by inhibition of the RNAi pathway.

In addition to these orientation-specific vector problems, we also observed a strong negative effect by the presence of the CMV promoter of the miRNA cassette, and this effect was apparent in both orientations. Thus, either the sequence or the activity of the CMV promoter affects the titer. This titer reduction is only observed when the CMV promoter is present in cis in the lentiviral vector and not when provided on another plasmid in trans; thus excluding transcription factor squelching effects. This cis effect could be due to competition or interference of the internal CMV promoter with the RSV promoter that drives vector expression. It has previously been reported that the human CMV promoter is more active than the RSV promoter (Lee et al. 1997; Xu et al. 2001) and transcription of the RNAi cassette may be favored over expression of full-length vector RNA during vector production. In gammaretroviral SIN vectors, the use of a strong internal promoter has indeed been associated with low vector titers because of abundant transcription of the transgene rather than the full-length RNA genome in the producer cells (Schambach et al. 2006).

It has recently been reported that bidirectional promoter interference can occur between two promoters in lentiviral vector constructs (Curtin et al. 2008). This promoter interference caused a marked reduction of transgene expression from the adjacent transcription unit. To avoid any form of transcriptional interference, expression of the miRNA cassette during vector production should be prevented. This condition could be met by the use of a conditionally regulated promoter, which should be inactive during lentiviral vector production. Lentiviral vectors encoding shRNAs do not have reduced titers due to transcriptional interference. This is likely because shRNA transcription by RNA polymerase III does not interfere with transcription of the vector genome by RNA polymerase II.

Our combined results suggest that the main titer reduction of miRNA vectors is not RNAi related, but caused by the activity or sequence of the CMV promoter. To test this, we replaced the constitutive CMV promoter with a tetracycline-inducible (tetO) promoter. Indeed, the lentiviral vector titers are increased by 100-fold, which are only five- to 10-fold lower than that of the empty vector. The advantage of the inducible promoter is that we could test the impact of miRNA expression on vector production. Interestingly, we observed a twofold increase in vector titer upon induction of miRNA expression. This effect was only observed in the sense vectors and not in the antisense vectors. We initially reasoned that induction of miRNA expression could augment the titers because Drosha-mediated cleavage of the vector RNA genome was avoided due to saturation of Drosha by the miRNAs. However, this titer improvement was also observed in the negative control in which only the tetO promoter and a poly(A) signal are present. These results indicate that there is another underlying cause for the vector improvement. An explanation could be that activation of the transgene promoter can have a positive effect on the promoter of the full-length vector RNA genome.

For the vectors with the antisense miRNA cassette, we anticipated that the titer will decrease upon miRNA expression because the expressed miRNAs can target the perfectly complementary antisense miRNA sequence in the vector RNA genome by self-targeting (Fig. 1, mechanism 2). In contrast to the results of our previous experiment, in which we mimicked self-targeting by co-transfection of the miRNA expression plasmid, we now observed equal vector titers when the miRNAs were expressed. Thus, these results suggest that titer reduction due to self-targeting in antisense miRNA vectors is negligible.

In this new set of lentiviral vectors with inducible promoters, we also included a poly(A) signal downstream from the miRNAs to test its effect on the lentiviral vector titer. Generally, internal transcription units are inserted in the lentiviral vector without poly(A) signal to avoid truncation of the full-length viral genome (Shimotohno and Temin 1981), although inclusion of an internal poly(A) signal can enhance transgene expression in virus-infected target cells (Shimotohno and Temin 1981; Maxwell et al. 1991; Narita et al. 2000; Hager et al. 2008; Tian and Andreadis 2009). In accordance with the findings of Hager et al. (2008), we found that an internal poly(A) signal reduced the transduction titer with ∼1 log. In addition, we demonstrated that the presence of the miRNA cassette in the antisense orientation resulted in lower titers than the sense constructs. This finding is consistent with a report that shows a reduction in vector titers when the transgene cassette is cloned in the antisense orientation (Mitta et al. 2005). In addition, we showed that the presence of antisense poly(A) signal sequences reduced the vector titer even further.

Taken together, we showed that both anti-HIV shRNA and miRNA cassettes can reduce the lentiviral vector titer. Titer decrease was observed for antiviral shRNAs that target the vector RNA sequences, but the titer can be restored by inhibition of the RNAi pathway. In contrast, the major cause for the titer reduction of miRNA vectors is not RNAi-related. Drosha processing of the vector RNA genome does only cause a minor reduction in vector titer when the miRNA expression cassette is present in the sense orientation (Fig. 1, mechanism 3). We demonstrate that miRNA expression from the sense and antisense miRNA vectors does not trigger self-targeting of the vector RNA genome (Fig. 1, mechanism 2). Most importantly, the presence of the CMV promoter, in both orientations, is detrimental to the vector titers (Fig. 1, mechanism 4). Replacement of the CMV promoter with an inducible promoter almost completely restored the vector titer. These results suggest that the commonly used CMV promoter (activity or sequence) is not compatible with the lentiviral vector system. Besides the CMV promoter, we also observed that antisense poly(A) signal sequences can lead to a severe titer reduction. These findings indicate that a careful design of lentiviral vectors is necessary to develop a lentiviral vector-based gene therapy for clinical use.

MATERIALS AND METHODS

DNA constructs

Lentiviral vector plasmids are derived from the construct (pRRLcpptpgkgfppreSsin) (Seppen et al. 2002), which we renamed JS1. The JS1-based plasmids H1-shNef, H1-shLdr, and H1-e3-shRNA were obtained by cloning of the H1-shRNA cassette from the original pSUPER construct (digestion with XhoI and PstI) into the corresponding sites of JS1. The JS1 plasmids CMV-miRNA S and CMV-4-miRNA S were obtained by digestion of the original pcDNA6.2 construct (Liu et al. 2008) with NruI and XhoI. The fragments were treated with Klenow enzyme according to the manufacturer's protocol (Roche) and inserted into the EcoRV site of JS1. The JS1 plasmids 4-miRNA S and 4-miRNA AS were obtained by PCR amplification of the miRNA polycistron from the original pcDNA6.2 construct with the following primers: XhoF: CGCTCGAGGAGGTGTTAATTCTAATTATCTATTTCA and PstR: CATCTGCAGGCATTGCAACCGATCCCAACCTGTGTA; and PstF: CATCTGCAGGAGGTGTTAATTCTAATTATCTATT and XhoR: CGCTCGAGGCATTGCAACCGATCCCAACCTGTGT. The PCR fragments were digested with PstI and XhoI and inserted into the corresponding sites of JS1.

To construct the tetracycline inducible lentiviral vectors, the original pcDNA6.2 4-miRNA construct was digested with BamHI and XhoI and inserted in the BamHI and SalI sites of pTRE-Tight vector (Clontech), named pTRE-Tight-4-miRNA. JS-1-tetO-pA S and AS were generated by digesting the empty pTRE-Tight vector with XhoI and inserting the fragment into the corresponding site in JS-1. The JS1 plasmids, tetO-4-miRNA S and AS, were obtained by digestion of the pTRE-Tight-4-miRNA construct with XhoI and EcoRV. The fragment was treated with Klenow enzyme according to the manufacturer's protocol (Roche) and inserted into the XhoI-digested and Klenow-treated JS1 vector. The JS1 plasmids tetO-4-miRNA-pA S and AS were obtained by digesting the pTRE-Tight-4-miRNA construct with XhoI and the fragments were inserted into the corresponding sites of JS1.

Luciferase reporters Luc-Pol, Luc-Gag, Luc-R/T, Luc-Nef, and Luc-Ldr were described previously (Ter Brake et al. 2006, 2008; Westerhout et al. 2006). The packaging plasmids pSYNGP (Kotsopoulou et al. 2000), pRSV-rev (Dull et al. 1998), pVSVg (Zufferey et al. 1998), and the construct expressing 5xshRNA (Gag5, Pol1, Pol6, Pol9, Pol47) from repeated H1 promoters were constructed as described previously (Ter Brake and Berkhout 2008). The Luc-ACDE reporter with four target sequences of the antiviral 4-miRNA polycistron was described elsewhere (Liu et al. 2009a). This reporter was also used to produce an RNAi decoy for CMV-miRNA S because it also encodes a target for this miRNA. The plasmid expressing the Adenovirus VA RNAs (pVA RNAs) (Andersson et al. 2005), the Ebola VP35 protein (Haasnoot et al. 2007a; de Vries et al. 2008), the CRM1 co-factor (Popa et al. 2002), the siRNA against Dicer, and the shRNA against Luciferase (Paddison et al. 2002) have been described elsewhere.

All DNA constructs were sequence-verified using the BigDye Terminator Cycle Sequencing kit (ABI). Hairpin RNA constructs were sequenced using a sample denaturation temperature of 98°C and upon addition of 1 M Betaine.

Cell culture and transfections

Human embryonic kidney (HEK) 293T adherent cells were grown in DMEM (Gibco-BRL) supplemented with 10% fetal calf serum (FCS) (Hybond), minimal essential medium nonessential amino acids, penicillin (100 U/mL) and streptomycin (100 μg/mL) at 37°C and 5% CO2. SupT1 suspension T cells were grown in advanced RPMI supplemented with 1% FCS, penicillin (30 U/mL) and streptomycin (30 μg/mL) and L-glutamine at 37°C and 5% CO2.

To knock down Drosha in the producer cells, 3.0 × 105 HEK 293T cells were seeded in six-well plates. The next two consecutive days, the cells were transfected with 250 ng of shLuc or shDrosha using Lipofectamine 2000 reagent according to the manufacturer's protocol (Invitrogen). The third day, lentiviral vector production was initiated by co-transfection with the lentiviral vector and packaging constructs as described below.

Lentiviral vector production and transduction

For production of the lentiviral vector, 6.0 × 105 HEK 293T cells were seeded per well in six-well plates in 2 mL of DMEM/10% FCS without antibiotics. The next day, the medium was replaced with 0.4 mL DMEM/10% FCS without antibiotics. Subsequently, the empty JS1 vector (0.95 μg), JS1-shNef, JS1-shLdr, JS1-e3-shRNA, JS1-CMV-miRNA S, JS1-CMV-4-miRNA, and JS1-4-miRNA in sense or antisense orientation were co-transfected with packaging plasmids pSYNGP (0.6 μg) (Kotsopoulou et al. 2000), pRSV-Rev (0.25 μg) (Dull et al. 1998), and pVSVg (0.33 μg) (Zufferey et al. 1998) with Lipofectamine 2000 reagent as suggested by the manufacturer (Invitrogen). When indicated, competitor or suppressor plasmids (2.9 μg) were added to the co-transfection. The siRNA against Dicer was used at 100 nM final concentration. For the tetO vectors, we co-transfected 20 ng of plasmid encoding the rtTA V1 transactivator (rtTA F86Y A209T) (Das et al. 2004b) and added 0, 10, 100, or 1000 ng/mL of doxycycline to the medium. The medium was replaced with 2 mL Opti-MEM with antibiotics on the second day. For the tetO vectors, we added doxycycline to the medium. On the third day, the medium containing lentiviral vector was harvested. Cellular debris was removed by centrifugation for 5 min at 1200 rpm. Production of lentiviral vector particles was determined by CA-p24 ELISA as previously described (Jeeninga et al. 2005). Capsid values were corrected for between-session variation (Ruijter et al. 2006).

SupT1 cells were transduced with a dilution series of the lentiviral vector stocks to determine the titer. Three days post-transduction, cells were analyzed with FACS to detect eGFP positive cells. The titer is expressed as transducing units per milliliter vector stock, and values were corrected for between-session variation (Ruijter et al. 2006).

Luciferase assay

HEK 293T cells were plated one day before transfection in 24-well plates at a density of 1.3 × 105 cells per well in 500 μL DMEM/10% FCS without antibiotics and transfected using Lipofectamine 2000 reagents according to the manufacturer's protocol (Invitrogen). Cells were co-transfected with 100 ng of the firefly luciferase expression plasmid, 1 ng of renilla luciferase expression plasmid (pRL-CMV), and 100 ng of the JS1 constructs. Two days post-transfection, firefly and renilla luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer's instructions. Relative luciferase activities were calculated from the ratio between firefly and renilla luciferase activities.

ACKNOWLEDGMENTS

We thank Stephan Heynen for performing the CA-p24 ELISA, Joost Haasnoot for helpful discussions and critical reading of the manuscript, and Atze Das for useful discussions. We thank Mustafa Ceylan for the generous gift of the 5 shRNA-expressing construct p5xshRNA. The pCRM1 plasmid was a kind gift from Tom Hope (Northwestern University) and pshDrosha was obtained from Bryan Cullen (Duke University). The pTRE-Tight vector and the rtTA V1 transactivator (rtTA F86Y A209T) were generously provided by Atze Das (University of Amsterdam). This work was sponsored by NWO-CW (TOP grant).

Footnotes

Article published online ahead of print. Article and publication date are at http://www.rnajournal.org/cgi/doi/10.1261/rna.1887910.

REFERENCES

  1. Andersson MG, Haasnoot PCJ, Xu N, Berenjian S, Berkhout B, Akusjarvi G 2005. Suppression of RNA interference by adenovirus virus-associated RNA. J Virol 79: 9556–9565 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Banerjea A, Li MJ, Bauer G, Remling L, Lee NS, Rossi J, Akkina R 2003. Inhibition of HIV-1 by lentiviral vector-transduced siRNAs in T lymphocytes differentiated in SCID-hu mice and CD34+ progenitor cell-derived macrophages. Mol Ther 8: 62–71 [DOI] [PubMed] [Google Scholar]
  3. Bartel DP 2004. MicroRNAs: Genomics, biogenesis, mechanism, and function. Cell 116: 281–297 [DOI] [PubMed] [Google Scholar]
  4. Boden D, Pusch O, Lee F, Tucker L, Ramratnam B 2003. Human immunodeficiency virus type 1 escape from RNA interference. J Virol 77: 11531–11535 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Boden D, Pusch O, Lee F, Tucker L, Ramratnam B 2004a. Efficient gene transfer of HIV-1-specific short hairpin RNA into human lymphocytic cells using recombinant adeno-associated virus vectors. Mol Ther 9: 396–402 [DOI] [PubMed] [Google Scholar]
  6. Boden D, Pusch O, Silbermann R, Lee F, Tucker L, Ramratnam B 2004b. Enhanced gene silencing of HIV-1 specific siRNA using microRNA designed hairpins. Nucleic Acids Res 32: 1154–1158 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Brummelkamp TR, Bernards R, Agami R 2002. A system for stable expression of short interfering RNAs in mammalian cells. Science 296: 550–553 [DOI] [PubMed] [Google Scholar]
  8. Castanotto D, Rossi JJ 2009. The promises and pitfalls of RNA-interference-based therapeutics. Nature 457: 426–433 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Curtin JA, Dane AP, Swanson A, Alexander IE, Ginn SL 2008. Bidirectional promoter interference between two widely used internal heterologous promoters in a late-generation lentiviral construct. Gene Ther 15: 384–390 [DOI] [PubMed] [Google Scholar]
  10. Das AT, Brummelkamp TR, Westerhout EM, Vink M, Madiredjo M, Bernards R, Berkhout B 2004a. Human immunodeficiency virus type 1 escapes from RNA interference-mediated inhibition. J Virol 78: 2601–2605 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Das AT, Zhou X, Vink M, Klaver B, Verhoef K, Marzio G, Berkhout B 2004b. Viral evolution as a tool to improve the tetracycline-regulated gene expression system. J Biol Chem 279: 18776–18782 [DOI] [PubMed] [Google Scholar]
  12. de Vries W, Haasnoot J, van der Velden J, van Montfort T, Zorgdrager F, Paxton W, Cornelissen M, van Kuppeveld F, de Haan P, Berkhout B 2008. Increased virus replication in mammalian cells by blocking intracellular innate defense responses. Gene Ther 15: 545–552 [DOI] [PubMed] [Google Scholar]
  13. Dull T, Zufferey R, Kelly M, Mandel RJ, Nguyen M, Trono D, Naldini L 1998. A third-generation lentivirus vector with a conditional packaging system. J Virol 72: 8463–8471 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Filipowicz W 2005. RNAi: The nuts and bolts of the RISC machine. Cell 122: 17–20 [DOI] [PubMed] [Google Scholar]
  15. Gimeno R, Weijer K, Voordouw A, Uittenbogaart CH, Legrand N, Alves NL, Wijnands E, Blom B, Spits H 2004. Monitoring the effect of gene silencing by RNA interference in human CD34+ cells injected into newborn RAG2−/− γc−/− mice: functional inactivation of p53 in developing T cells. Blood 104: 3886–3893 [DOI] [PubMed] [Google Scholar]
  16. Haasnoot PCJ, Berkhout B 2006. RNA interference: Its use as antiviral therapy. Handbook of experimental pharmacology, pp. 117–150 Springer, Berlin: [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Haasnoot J, de Vries W, Geutjes EJ, Prins M, de Haan P, Berkhout B 2007a. The Ebola virus VP35 protein is a suppressor of RNA silencing. PLoS Pathog 3: e86 doi: 10.1371/journal.ppat.0030086 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Haasnoot J, Westerhout EM, Berkhout B 2007b. RNA interference against viruses: Strike and counterstrike. Nat Biotechnol 25: 1435–1443 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hager S, Frame FM, Collins AT, Burns JE, Maitland NJ 2008. An internal polyadenylation signal substantially increases expression levels of lentivirus-delivered transgenes but has the potential to reduce viral titer in a promoter-dependent manner. Hum Gene Ther 19: 840–850 [DOI] [PubMed] [Google Scholar]
  20. Han J, Lee Y, Yeom KH, Kim YK, Jin H, Kim VN 2004. The Drosha–DGCR8 complex in primary microRNA processing. Genes Dev 18: 3016–3027 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Harper SQ, Staber PD, He X, Eliason SL, Martins IH, Mao Q, Yang L, Kotin RM, Paulson HL, Davidson BL 2005. RNA interference improves motor and neuropathological abnormalities in a Huntington's disease mouse model. Proc Natl Acad Sci 102: 5820–5825 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Jeeninga RE, Jan B, Van der Linden B, Van den Berg H, Berkhout B 2005. Construction of a minimal HIV-1 variant that selectively replicates in leukemic derived T-cell lines: Toward a new virotherapy approach. Cancer Res 65: 3347–3355 [DOI] [PubMed] [Google Scholar]
  23. Kim DH, Rossi JJ 2007. Strategies for silencing human disease using RNA interference. Nat Rev Genet 8: 173–184 [DOI] [PubMed] [Google Scholar]
  24. Klasens BIF, Huthoff HT, Das AT, Jeeninga RE, Berkhout B 1999. The effect of template RNA structure on elongation by HIV-1 reverse transcriptase. Biochim Biophys Acta 1444: 355–370 [DOI] [PubMed] [Google Scholar]
  25. Kotsopoulou E, Kim VN, Kingsman AJ, Kingsman SM, Mitrophanous KA 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 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lee AH, Suh YS, Sung JH, Yang SH, Sung YC 1997. Comparison of various expression plasmids for the induction of immune response by DNA immunization. Mol Cells 7: 495–501 [PubMed] [Google Scholar]
  27. Liu YP, Haasnoot J, Berkhout B 2007. Design of extended short hairpin RNAs for HIV-1 inhibition. Nucleic Acids Res 35: 5683–5693 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Liu YP, Haasnoot J, Ter Brake O, Berkhout B, Konstantinova P 2008. Inhibition of HIV-1 by multiple siRNAs expressed from a single microRNA polycistron. Nucleic Acids Res 36: 2811–2824 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Liu YP, Gruber J, Haasnoot J, Konstantinova P, Berkhout B 2009a. RNAi-mediated inhibition of HIV-1 by targeting partially complementary viral sequences. Nucleic Acids Res 37: 6194–6204 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Liu YP, von Eije KJ, Schopman NC, Westerink JT, Brake O, Haasnoot J, Berkhout B 2009b. Combinatorial RNAi against HIV-1 using extended short hairpin RNAs. Mol Ther 17: 1712–1723 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lund E, Guttinger S, Calado A, Dahlberg JE, Kutay U 2004. Nuclear export of microRNA precursors. Science 303: 95–98 [DOI] [PubMed] [Google Scholar]
  32. Maxwell IH, Brown JL, Maxwell F 1991. Inefficiency of expression of luciferase reporter from transfected murine leukaemia proviral DNA may be partially overcome by providing a strong polyadenylation signal. J Gen Virol 72: 1721–1724 [DOI] [PubMed] [Google Scholar]
  33. Mitta B, Rimann M, Fussenegger M 2005. Detailed design and comparative analysis of protocols for optimized production of high-performance HIV-1-derived lentiviral particles. Metab Eng 7: 426–436 [DOI] [PubMed] [Google Scholar]
  34. Narita M, Takanaga K, Yoshida Y, Kadomatsu K, Muramatsu T, Matsubara S, Hamada H, Goto S, Saisho H, Sakiyama S, et al. 2000. Polyadenylation signal facilitates the expression of foreign gene that is driven by an internal promoter located in the reverse orientation to long terminal repeat of retrovirus. Anticancer Res 20: 279–282 [PubMed] [Google Scholar]
  35. Paddison PJ, Caudy AA, Bernstein E, Hannon GJ, Conklin DS 2002. Short hairpin RNAs (shRNAs) induce sequence-specific silencing in mammalian cells. Genes Dev 16: 948–958 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Poluri A, Sutton RE 2007. Titers of HIV-based vectors encoding shRNAs are reduced by a Dicer-dependent mechanism. Mol Ther 16: 378–386 [DOI] [PubMed] [Google Scholar]
  37. Popa I, Harris ME, Donello JE, Hope TJ 2002. CRM1-dependent function of a cis-acting RNA export element. Mol Cell Biol 22: 2057–2067 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Ralph GS, Radcliffe PA, Day DM, Carthy JM, Leroux MA, Lee DC, Wong LF, Bilsland LG, Greensmith L, Kingsman SM, et al. 2005. Silencing mutant SOD1 using RNAi protects against neurodegeneration and extends survival in an ALS model. Nat Med 11: 429–433 [DOI] [PubMed] [Google Scholar]
  39. Rawlinson SM, Pryor MJ, Wright PJ, Jans DA 2009. CRM1-mediated nuclear export of dengue virus RNA polymerase NS5 modulates interleukin-8 induction and virus production. J Biol Chem 284: 15589–15597 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Ruijter JM, Thygesen HH, Schoneveld OJ, Das AT, Berkhout B, Lamers WH 2006. Factor correction as a tool to eliminate between-session variation in replicate experiments: Application to molecular biology and retrovirology. Retrovirology 3: 1–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Schambach A, Mueller D, Galla M, Verstegen MM, Wagemaker G, Loew R, Baum C, Bohne J 2006. Overcoming promoter competition in packaging cells improves production of self-inactivating retroviral vectors. Gene Ther 13: 1524–1533 [DOI] [PubMed] [Google Scholar]
  42. Seppen J, Rijnberg M, Cooreman MP, Oude Elferink RP 2002. Lentiviral vectors for efficient transduction of isolated primary quiescent hepatocytes. J Hepatol 36: 459–465 [DOI] [PubMed] [Google Scholar]
  43. Shimotohno K, Temin HM 1981. Formation of infectious progeny virus after insertion of herpes simplex thymidine kinase gene into DNA of an avian retrovirus. Cell 26: 67–77 [DOI] [PubMed] [Google Scholar]
  44. Suo Z, Johnson KA 1997. RNA secondary structure switching during DNA synthesis catalyzed by HIV-1 reverse transcriptase. Biochemistry 36: 14778–14785 [DOI] [PubMed] [Google Scholar]
  45. Ter Brake O, Berkhout B 2007. Lentiviral vectors that carry anti-HIV shRNAs: Problems and solutions. J Gene Med 9: 743–750 [DOI] [PubMed] [Google Scholar]
  46. Ter Brake O, Berkhout B 2008. Development of an RNAi-based gene therapy against HIV-1. In Therapeutic oligonucleotides (ed. Kurreck J), pp. 296–311 RSC Publishing, Cambridge, UK [Google Scholar]
  47. Ter Brake O, Konstantinova P, Ceylan M, Berkhout B 2006. Silencing of HIV-1 with RNA interference: A multiple shRNA approach. Mol Ther 14: 883–892 [DOI] [PubMed] [Google Scholar]
  48. Ter Brake O, ‘t Hooft K, Liu YP, Centlivre M, von Eije KJ, Berkhout B 2008. Lentiviral vector design for multiple shRNA expression and durable HIV-1 inhibition. Mol Ther 16: 557–564 [DOI] [PubMed] [Google Scholar]
  49. Tian J, Andreadis ST 2009. Independent and high-level dual-gene expression in adult stem-progenitor cells from a single lentiviral vector. Gene Ther 16: 874–884 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Van den Haute C, Eggermont K, Nuttin B, Debyser Z, Baekelandt V 2003. Lentiviral vector-mediated delivery of short hairpin RNA results in persistent knockdown of gene expression in mouse brain. Hum Gene Ther 14: 1799–1807 [DOI] [PubMed] [Google Scholar]
  51. Westerhout EM, Berkhout B 2007. A systematic analysis of the effect of target RNA structure on RNA interference. Nucleic Acids Res 35: 4322–4330 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Westerhout EM, Ooms M, Vink M, Das AT, Berkhout B 2005. HIV-1 can escape from RNA interference by evolving an alternative structure in its RNA genome. Nucleic Acids Res 33: 796–804 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Westerhout EM, Ter Brake O, Berkhout B 2006. The virion-associated incoming HIV-1 RNA genome is not targeted by RNA interference. Retrovirology 3: 57–65 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Wodrich H, Krausslich HG 2001. Nucleocytoplasmic RNA transport in retroviral replication. Results Probl Cell Differ 34: 197–217 [DOI] [PubMed] [Google Scholar]
  55. Xu ZL, Mizuguchi H, Ishii-Watabe A, Uchida E, Mayumi T, Hayakawa T 2001. Optimization of transcriptional regulatory elements for constructing plasmid vectors. Gene 272: 149–156 [DOI] [PubMed] [Google Scholar]
  56. Yi R, Qin Y, Macara IG, Cullen BR 2003. Exportin-5 mediates the nuclear export of pre-microRNAs and short hairpin RNAs. Genes Dev 17: 3011–3016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Zhou D, Zhang J, Wang C, Bliesath JR, He Q, Yu D, Li-He Z, Wong-Staal F 2009. A method for detecting and preventing negative RNA interference in preparation of lentiviral vectors for siRNA delivery. RNA 15: 732–740 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Zufferey R, Dull T, Mandel RJ, Bukovsky A, Quiroz D, Naldini L, Trono D 1998. Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J Virol 72: 9873–9880 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from RNA are provided here courtesy of The RNA Society

RESOURCES