Skip to main content
NCBI home page
Molecular Therapy logoLink to Molecular Therapy
. 2014 Jun 3;22(8):1460–1471. doi: 10.1038/mt.2014.82

Efficient Transient Genetic Manipulation In Vitro and In Vivo by Prototype Foamy Virus-mediated Nonviral RNA Transfer

Martin V Hamann 1,2, Nicole Stanke 1,2, Erik Müllers 1,2, Kristin Stirnnagel 1,2, Sylvia Hütter 1,2, Benedetta Artegiani 2, Sara Bragado Alonso 2, Federico Calegari 2, Dirk Lindemann 1,2,*
PMCID: PMC4435593  PMID: 24814152

Abstract

Vector systems based on different retroviruses are widely used to achieve stable integration and expression of transgenes. More recently, transient genetic manipulation systems were developed that are based on integration- or reverse transcription-deficient retroviruses. Lack of viral genome integration is desirable not only for reducing tumorigenic potential but also for applications requiring transient transgene expression such as reprogramming or genome editing. However, all existing transient retroviral vector systems rely on virus-encoded encapsidation sequences for the transfer of heterologous genetic material. We discovered that the transient transgene expression observed in target cells transduced by reverse transcriptase-deficient foamy virus (FV) vectors is the consequence of subgenomic RNA encapsidation into FV particles. Based on this initial observation, we describe here the establishment of FV vectors that enable the efficient transient expression of various transgenes by packaging, transfer, and de novo translation of nonviral RNAs both in vitro and in vivo. Transient transgene expression levels were comparable to integrase-deficient vectors but, unlike the latter, declined to background levels within a few days. Our results show that this new FV vector system provides a useful, novel tool for efficient transient genetic manipulation of target tissues by transfer of nonviral RNAs.

Introduction

Retroviral vector systems have been used for several decades, both in basic research and in clinical trials for the treatment of various diseases.1,2 Common retroviral vectors enable a long-term transgene expression due to vector genome integration into host cell chromosomes. However, this feature bears the risk of detrimental mutagenesis, which became apparent in several clinical trials where vector genome integration led to oncogene activation.3 Many applications, such as reprogramming by exogenous transcription factors or genome editing by recombinases or site-specific nucleases, do not require long-term modification of target tissues. Therefore, a whole toolbox of retroviral vector systems for transient delivery of nucleic acids or proteins was developed in recent years.4,5 Transient genetic manipulation systems include retroviral vectors blocked either at the step of integration or reverse transcription. Integrase-deficient (inactive integrase, iIN) retroviruses display a transient transgene expression which is derived from episomal viral DNA. However, they retain a low level of residual integration capacity due to cell-mediated nonhomologous recombination of viral and cellular sequences.5 In contrast, transgene expression of reverse transcriptase-deficient (inactive reverse transcriptase, iRT) retroviruses, which is derived from translation of two viral genomic RNA (vgRNA) copies per particle delivered into the cytoplasm of target tissues, is generally lower in comparison to iIN vectors.4 Hence, unlike iIN vectors, iRT vectors have no residual integration capacity and transgene expression is completely transient.

Common to all, current retroviral nucleic acid transfer systems is the requirement of residual viral sequences for encapsidation and transfer of the heterologous genetic material. This is due to the evolutionary need of retroviruses to achieve a preferential encapsidation of vgRNA out of a large pool of cellular RNAs, many of which are similarly structured with 5′ cap and 3′ poly A tail. Our current mechanistic understanding of the retroviral genome encapsidation process is far from complete.6,7 It is well established though, that the retroviral capsid protein Gag has both specific and nonspecific nucleic acid–binding features. Nonspecific interactions are mediated mainly by the negatively charged nucleic acid chain backbone and positively charged amino acids located in different domains of Gag. These are important for correct assembly and release of retroviral particles, and viral or cellular nucleic acids are thought to act as a scaffold for these processes. The highly specific interaction of C-terminal Gag domains (mainly the nucleocapsid and cysteine–histidine (Cys-His) motives therein) and the so-called genomic packaging signal (Psi) results in an enrichment of viral genome dimers into viral particles. Psi constitutes an elaborate secondary RNA structure, which is primarily located in the 5′ region of the viral genome and is unique to full-length vgRNA. However, particles of different retroviral origin were shown to also contain various cellular RNAs in addition to the vgRNA.8 These RNAs are copackaged during assembly by different mechanisms and can constitute up to 50% of the total nucleic acid mass found in released virions. Whether they are functionally active in newly infected host cells and thereby influence viral replication is not known.

Foamy viruses (FVs) represent a subgroup of retroviruses with unique replication features and have been successfully used as gene transfer tools for stable genetic modification of various target tissues in vitro and in vivo in recent years.9,10 FVs are characterized by an extremely broad tropism, which is mainly conveyed by the FV glycoprotein interaction with heparan sulfate and additional unknown receptor molecules on the surface of target cells.11,12 Although productive transduction by FV vectors requires host cell mitosis, a long-lasting latent survival in interphase cells due to highly stable capsids,13 allows for efficient gene transfer also into slowly dividing target tissues such as hematopoietic stem cells.14

Unlike in other retroviruses, FV Gag proteins lack Cys-His motifs. Instead, their C-terminus contains glycine-arginine-rich (GR) motifs that serve essential functions in viral replication, including viral genome encapsidation.15 We recently found that FV particles, like other retroviruses, copackage cellular RNAs and that the packaging of viral and cellular RNAs requires a concerted function of positively charged residues in the C-terminus of Gag (Müllers et al., Unpublished Data). Hence, in the current study, we investigated whether this general retroviral feature of nonviral RNA encapsidation allows the delivery of nonviral RNAs and leads to their translation upon transduction of target tissues with recombinant vector particles. Comparison to several common vector systems of different retroviral origin revealed that prototype FV (PFV) particles are the most efficient in this respect. We exploited this feature to transfer and transiently translate heterologous nonviral RNA molecules encoding different transgenes in various target tissues in vitro and in vivo, thereby establishing a new PFV vector system for efficient and transient genetic manipulation without any residual retroviral sequences.

Results

Transduction with PFV vector particles containing different enzymatically inactive Pol proteins results in transient transgene expression

In the course of characterizing the importance of the PFV-encoded protease during viral replication,16 we observed a transient fluorescence signal by flow cytometry analysis of target cells transduced with green fluorescent protein (GFP)-encoding PFV vectors containing various types of enzymatically inactive Pol proteins (Figure 1). The fluorescence intensity of cells transduced with PFV particles containing an inactive integrase (iIN) was largely transient and declined over a time period of 14 days post transduction (p.t.), as also observed in similar PFV and HIV vector systems (Figure 1; wt, iIN)5,17. Transduction with iIN-PFV resulted in an initial high fluorescence signal 24 hours p.t., at a level similar to wild-type particle-transduced cells, probably as the result of transgene expression from nonintegrated episomal proviral DNA. Subsequently, target cell fluorescence declined drastically within 1 week but did not vanish completely. Instead iIN-PFV–transduced cells showed a small fraction (~3%) of stably expressing target cells at the last time point analyzed (30-fold reduced compared to wt; Figure 1, compare wt, iIN day 14). Most likely, these GFP-positive cells resulted from a low level nonviral integrase-mediated insertion of vector genomes.

Figure 1.

Figure 1

Open in a new tab

Transient target cell fluorescence induced by transduction with PFV vectors with inactive enzymatic functions. (a) Dot blot profiles of HT1080 target cells transduced with undiluted, cell-free viral supernatants harvested from 293T transiently transfected with 4-component PFV vector system expression constructs comprising a GFP-encoding standard transfer vector (puc2MD9), Gag (pcoPG4) and Env (pcoPE) packaging plasmids as well as different variants of Pol packaging plasmids (pcoPP (wt), pcoPP iPR (iPR), pcoPP iRT (iRT), pcoPP iIN (iIN)) as indicated on top. Transduced and uninfected control (mock) HT1080 target cells were trypsinized at different time points p.t. as indicated on the left, and single cell suspensions were analyzed by flow cytometry. Shown are FL1 (GFP) / FSC dot blots of the individual samples of a representative experiment (n=6). Gates for low (low) and high (high) fluorescence signals were set in the GFP channel and the percentage of cells in the individual gates are indicated in the dot blot. (b) Time course analysis of mean fluorescence intensity profiles in the GFP channel of the same transduced HT1080 target cell populations shown in (a).

Surprisingly, a transient GFP signal was also observed in target cells upon transduction with PFV particles containing an inactive protease (iPR) or an inactive reverse transcriptase (iRT) (Figure 1, iPR, iRT). In contrast to cells transduced with either wt Pol or iIN Pol containing particles, the transient GFP signal in iPR- and iRT-PFV vector-transduced samples was significantly weaker at 24 hours p.t. and declined to background levels within 4–7 days p.t. (Figure 1). As both types of viruses fail to reverse transcribe their packaged viral genome (vgRNA),16 we decided to investigate the basis for the transient expression phenotype of iPR- or iRT-PFV and reasoned that these vectors could constitute the basis for a vector system for transient transgene expression.

Transient transgene expression by PFV vectors deficient in reverse transcription is mediated by transfer of subgenomic viral RNA

Transfection of packaging cells with the PFV transfer vector (pMD9-GFP) used in the experiments described above allows for the transcription of three types of mRNAs (Figure 2a). First, a CMV promoter-driven unspliced, full-length vgRNA (vgRNA) containing all cis active viral sequences (CAS) that are essential for viral genome encapsidation, assembly of infectious particles and productive, stable transduction of target cells.18 Second, a CMV promoter-driven single-spliced vgRNA (svgRNA), and third, an unspliced, subgenomic RNA (subRNA) originating from the internal promoter. The latter two mRNAs lack large parts of the viral CAS-I and CAS-II sequences needed for efficient genome encapsidation. Furthermore, the two mRNAs originating from the CMV promoter (vgRNA and svgRNA) harbor extensive stretches of 5′ noncoding sequences upstream from the gfp ORF. Hence, we assumed that only the subRNA driven by the internal promoter is efficiently translated. This implies that the transient GFP fluorescence observed in iPR- and iRT-PFV vector-transduced cells (Figure 1) derives mainly from putatively copackaged subRNA, that is transferred into and translated in the transduced target cells. To explore this possibility, we generated PFV vector particles, where the standard PFV transfer vector (MD9-GFP) was replaced by a mammalian expression vector containing a similar SFFV U3 driven GFP cassette (SFFV-GFP), but lacking any PFV derived sequences (Figure 2a). Subsequently, we compared the expression profile of SFFV-GFP-derived vector particles (SFFV-GFP wt) to those of wild-type (MD9-GFP wt) and enzymatically deficient PFV particles (MD9-GFP iRT, iIN) generated with the standard PFV transfer vector after target cell transduction.

Figure 2.

Figure 2

Open in a new tab

Mean fluorescence intensity profiles of cells transduced with different PFV vector particles. (a) Schematic illustration of the structural organization of different transfer vectors (transfer vectors) and packaging plasmids (packaging components) used for generation of the PFV vector particles employed in this study. The RNAs derived from the individual transfer vectors are indicated as dashed arrows with coding regions marked as solid lines below their schematic DNA structural organization. The different variants of the individual packaging plasmids used encoding PFV Gag, Pol or Env are indicated in brackets. Their features are explained in detail in the Materials and Methods section. CAS: cis acting sequences I to III required for productive stable transduction. CMV: cytomegalovirus immediate early promoter, R: long terminal repeat (LTR) repeat region; U5: LTR unique 5' region; ▵U3: enhancer – promoter deleted LTR unique 3' region; partial coding sequences of PFV Gag, Pol and Env overlapping the CAS sequences are indicated by dashed boxes and marked with ▵. SFFV U3: spleen focus forming virus U3 enhancer promoter; EGFP: enhanced green fluorescent protein ORF; CRE: Cre ORF; SD: splice donor; SA: splice acceptor; bGH pA: bovine growth hormone poly A signal: p68: Gag p68 subunit; p3: Gag p3 subunit; PR: Pol protease domain; RT: Pol reverse transcriptase domain; IN: Pol integrase domain; LP: Env leader peptide subunit; SU: Env surface subunit; TM: Env transmembrane subunit. (b) Time course analysis of mean fluorescence intensity (MFI) profiles in the GFP channel of HT1080 target cells transduced with undiluted viral supernatants by spinoculation. Shown are the results of representative experiments (n=3). Details of transduction procedures for short-term (left graph) and long-term (right graph) time courses are given in the M&M section. Viral supernatants used were generated using wild-type Env and Gag packaging plasmids but varied in their transfer vector (RNA; MD9-GFP: puc2MD9 EGFP, SFFV-GFP: pSFFVU3 EGFP, CMV-GFP: pcziEGFP) and Pol protein (Pol; wt: pcoPP, iRT: pcoPP2, iIN: pcoPP3) composition as indicated in the legend to the right. Mock: uninfected cells.

Target cells were transduced with plain vector supernatants containing the different PFV vector particles by a spinoculation procedure at low temperature in order to synchronize virus uptake and infection. This enabled flow cytometric analysis of the target cell fluorescence at very early time points after initiating particle uptake. The analysis of the mean fluorescence intensities (MFI) in the GFP channel of the entire target cell populations over a time period of 16 days p.t. is summarized in Figure 2b. All transduced target cell samples showed MFI values above background (mock) as early as 1 hour p.t. but displayed significant differences in their MFI profiles over time. The MFI of wild-type PFV vector particle-transduced cell populations (MD9-GFP wt) increased to a high level for up to 3–5 days p.t. and then stayed constant for the whole examination period. In contrast, the values of standard vector particles with inactive IN (MD9-GFP iIN) or inactive RT (MD9-GFP iRT) plateaued between 16 and 24 hours p.t. before declining from 3 days p.t. onwards. Furthermore, in comparison with wild type, the maximum MFI values obtained were 10- and 100-fold reduced for iIN and iRT-transduced cell populations, respectively. Whereas the MFI of iRT-transduced cell populations returned uniformly to background levels at day 7 it remained clearly above background in iIN-transduced cell populations (Figure 2b, right panel). This is due to the presence of a small fraction of cells (~3%) constitutively expressing GFP in iIN-transduced samples at day 7 or later (Figure 1a, iIN column). Strikingly, cells transduced with vector particles generated with the nonviral pSFFV U3 expression vector (SFFV-GFP wt) showed an MFI expression profile over time that was nearly identical to that of iRT-transduced cells (MD9-GFP iRT) (Figure 2b). This strongly suggested that the GFP fluorescence observed in iRT and iPR vector-transduced cells is mainly the result of a translation of copackaged subRNA.

Dose-dependent RNA transfer and transgene expression

Next, we evaluated whether the promoter strength of the transgene expressing vector influenced transgene expression in transduced target cells. To this end, we compared the SFFV U3 promoter expression vector (SFFV), which produces nonspliced, GFP-encoding mRNAs, to that of a “RNA transfer” vector construct (CMV), which generates spliced messages derived from a strong CMV immediate early enhancer-promoter-intron A expression cassette (Figure 2a). Viral supernatants were produced with identical amounts of both vectors and packaging components. As depicted in Figure 2b, the CMV-GFP expression vector (CMV-GFP wt) led to a 10-fold higher maximal MFI in transduced target cells compared to SFFV vector (SFFV-GFP wt), and at a level similar to that of IN-deficient standard PFV vector (MD9-GFP iIN). However, unlike in iIN-transduced cells, the fluorescence signal of the CMV-GFP transduced cells returned to background within 7 days p.t. (Figure 2b right panel, compare MD9-GFP iIN and CMV-GFP wt graphs).

Furthermore, we examined the influence of transgene encoding RNA levels in the packaging cell on particle-associated RNA copy numbers and transgene expression in transduced target cells. Vector particles were produced keeping the amount of packaging component constant, but using decreasing amounts of transgene encoding CMV expression vector. Quantitative PCR and flow cytometry analysis demonstrated a good correlation of the copy numbers of GFP mRNA in the packaging cell to the particle-associated RNA copy numbers, and the GFP expression level observed upon target cell transduction (see Supplementary Figure S1).

Superior transfer and translation of nonviral mRNAs by PFV vectors

Various types of retroviruses are known to encapsidate cellular RNAs in addition to their viral genomes (Müllers et al., Unpublished Data).8 Therefore, we compared the efficiency of PFV vector particles at encapsidating and transferring nonviral RNAs derived from the CMV-driven expression vector to those of murine leukemia virus (MLV) and human immunodeficiency virus type 1 (HIV-1)-based vectors. Viral vector supernatants were generated using the canonical Gag/Pol packaging constructs and either a corresponding GFP-encoding standard viral transfer vector for stable (S) transgene insertion or the CMV-driven nonviral GFP mRNA transfer vector for transient (T) expression. To ensure an identical cellular uptake in target cells, all vector particles were pseudotyped with PFV Env. Flow cytometry titration analysis of the plain viral vector supernatants for stable transgene expression (S) revealed an approximately 8-fold higher infectious titer for MLV-based particles compared with PFV and HIV-1 (Figure 3a). The level of transient transgene expression induced by the plain supernatants for RNA transfer (T) was examined by determining the MFI of target cells 48h p.t. We observed signals clearly above background for MLV and HIV 1-derived vectors, indicating that both systems also allow the transfer and translation of nonviral RNA.(Figure 3b). However, they were less efficient than the PFV vectors, as the PFV-transduced cells showed an approximately 10-fold higher MFI signal (Figure 3b). Of note, the viral particle-associated GFP mRNA copy numbers of transient (T) virus supernatants determined by qPCR analysis varied only twofold among the different virus types (Figure 3c). In pelleted supernatants from cells transfected only with the RNA-transfer vector (T) and no packaging components (Figure 3c, bar 7), GFP mRNA copy numbers were at least 60-fold lower compared to all viral particle samples (Figure 3c). This result supports the notion that the detected RNAs in all virus samples are almost entirely associated with virions and do not result from cellular vesicles or debris.

Figure 3.

Figure 3

Open in a new tab

Comparison of retroviral vectors of different viral origin. (a) Viral titers of different retroviral supernatants generated by transient transfection of 293T with, i) the respective transfer vector – packaging plasmid combinations of different retroviral origin for stable (S) or transient (T) expression, as indicated (bar 1-6); ii) only the transient CMV-GFP transfer vector (T) without packaging components (bar 7); or iii) only pUC19 (mock, bar 8); as described in M&M. Shown are titers of an individual representative experiment (n=5) determined by transduction of HT1080 target cells with serial dilutions of plain viral supernatants and flow cytometric analysis 48 h p.t.. Mean values and standard deviations for each viral supernatant were determined from values of two to three target cell samples transduced with different virus dilutions as described in M&M. (b) GFP fluorescence intensity profiles of HT1080 cells 48 h p.t. with undiluted plain viral supernatants and mock transduced cells as indicated. The MFI values of the individual samples are given in the legend. (c) Viral particle-associated RNA copy numbers of different vector preparations as indicated, determined by qPCR analysis using EGFP-specific primer – probe sets as described in M&M. Shown are mean values and standard deviations of duplicate qPCR samples from the same individual, representative experiment (n=2) as shown in (a). Values are expressed as copies per ml pelleted viral supernatant.

Taken together, these data indicate that target cell transfer of nonviral RNAs is a general retroviral vector feature. However, PFV vectors appear to be superior in enabling their translation in transduced cells by at least 10-fold.

Transient transgene expression by PFV vectors requires viral capsid and glycoprotein functions, but not enzyme activity

Transduction of target cells and stable, high level transgene expression by standard PFV vectors (MD9-GFP) requires wild-type Gag, Pol and Env packaging components (Figure 2b). To determine the minimal essential packaging components and their functional features for transient transgene expression, we generated virus supernatants using the standard PFV vector (MD9-GFP) or the RNA-transfer vector (CMV-GFP) in combination with different combinations of wild-type and mutant Gag, Pol, and Env packaging constructs (Figure 2a). The analysis of the target cell MFI 48 hours after transduction with the different viral vector supernatants revealed a set of features and requirements essential for transient expression (Figure 4). First, PFV Pol and its enzymatic functions (iRT, iIN) were dispensable (Figure 4, bar 1-6). Second, transient gene expression was only observed with the combined use of Env and either the wt Gag precursor or its large processing product p68 (Figure 4, bar 5-13). Third, expression was strongly reduced by inactivation of Gag RNA binding motifs (iNAB) (Figure 4, bar 1, 4, 5, 6, 8, 9, 14). Finally, transgene expression also required uptake by a fusion-competent glycoprotein (wt) into target cells since incubation of target cells with fusion-deficient glycoprotein (iFuse) containing supernatants resulted in no fluorescence signal (Figure 4, bar 1, 4, 15, 16). Thus, a minimal vector system consisting of a nonviral RNA expression vector and packaging constructs encoding wild-type PFV Gag and Env is sufficient for transfer and translation of transgenes in target cells.

Figure 4.

Figure 4

Open in a new tab

Identification of structural components and their functions essential for PFV-mediated RNA transfer. PFV vector supernatants were generated by transient transfection of 293T cells with the different packaging component combinations indicated on the x-axis and either puc2MD9 (gray bars, MD9-GFP) or pcziEGFP (black bars, CMV-GFP) for stable and transient GFP expression, respectively. The MFI of HT1080 cells transduced with equal amounts of undiluted cell-free viral supernatant was determined 48 h p.t.. Shown are the mean MFI values and standard deviations of four independent experiments. wt: wild type; iRT: inactive RT; iIN: inactive IN; p68: Gag p68; iNAB: nucleic acid-binding-deficient Gag; iFuse: fusion-deficient Env.

Virus-encoded protein presence in transduced cells is primarily a result of de novo protein synthesis from transferred RNA

The steady increase of target cell fluorescence within 24 hours p.t. from PFV particles containing GFP-encoding nonviral RNAs (Figure 2b, CMV-GFP) and the strongly diminished signal induced by mutant particles deficient in nucleic acid binding (Figure 4, iNAB Gag) strongly suggested that a de novo translation of transferred GFP mRNA is the main cause for the transient fluorescence signal observed. However, these data do not formally exclude a contribution from particle-associated GFP proteins that could be encapsidated during vector particle production and then transferred to the target cells. To assess the influence of these potentially transferred proteins, the MFI of target cells transduced with different types of viral vector supernatants was monitored over a time period of 24 hours in the absence or presence of the translation inhibitor cycloheximide (Figure 5).

Figure 5.

Figure 5

Open in a new tab

Protein versus RNA transfer and de novo translation. Time course analysis of mean fluorescence intensity profiles in the GFP channel of HT1080 target cells transduced with undiluted viral supernatants by spinoculation. Shown are the results of representative experiments (n=3). Details of transduction procedures by spinoculation and cycloheximide treatment for target cell translational suppression are given in the M&M section. Viral supernatants were generated using the 4-component PFV vector system consisting of the components indicated in the legends. Dashed lines mark samples treated with cycloheximide (Cyc. +) and solid lines mark untreated samples (Cyc. -). (a) RNA transfer: Viral particles were generated using wild-type Env and Pol packaging plasmids and the combinations of Gag packaging plasmids (Gag) and transfer vector (RNA) as indicated in the legend. wt: wild type; iNAB: nucleic acid-binding-deficient Gag; MD9-GFP: puc2MD9; CMV-GFP: pcziEGFP; -: no packaging components; mock: uninfected cells. (b) Protein transfer: Viral particles were generated using Pol packaging plasmids and the combinations of Gag (Gag), Env (Env) packaging plasmids and transfer vector (RNA) as indicated in the legend. wt: wild type; GFP: C-terminal GFP-tagged p68 Gag; iFuse: fusion-deficient Env; CMV-GFP: pcziEGFP; CMV: empty transfer vector pczi; mock: uninfected cells.

Inhibition of protein translation in target cells incubated with standard PFV vectors (MD9-GFP wt) resulted in an MFI profile with values within a 2-fold range during the whole observation period (Figure 5a, black square, dashed line). In contrast, target cells transduced in parallel that were not blocked in translation showed MFI signals that steadily increased up to 150-fold within 24 hours p.t. (Figure 5a, black square, solid line). Similar MFI profiles were observed for target cells incubated with wild-type RNA-transfer vector particles (CMV-GFP wt) (Figure 5, black triangle). In translationally active cells, MFI values increased up to 50-fold within 24 hours p.t., whereas in translationally blocked cells MFI values remained low and stable within a twofold range. This indicated that de novo translation is the major source of the signal increase observed in both types of samples.

Analysis of RNA transfer vector supernatants harboring a nucleic acid-binding-deficient Gag variant (CMV-GFP iNAB) further supported this notion (Figure 5a, white triangle). Here the low MFI values, which were similar to those induced by RNA transfer vector particles containing wt Gag at translation blocked state (CMV-GFP wt; Figure 5a black triangle dashed line), stayed within a twofold range regardless of the translational status of the transduced host cell (Figure 5a, compare white and black triangles). A similar MFI profile, but at an even lower level, only slightly above background, was obtained for target cells incubated with supernatants from 293T cells transfected with the RNA transfer vectors alone (CMV-GFP -), omitting the packaging components (Figure 5a, gray triangle). Furthermore, if cells were treated with CMV-GFP particles having the wild-type glycoprotein replaced by a fusion-deficient Env variant (CMV-GFP wt/iFuse), a similar initial fluorescence signal as for wild-type Env containing particles (CMV-GFP wt/wt) was observed (Figure 5b, compare black and gray triangles at 0 hour p.t.). However, unlike the wild type (Figure 5b, black triangle), the signal of iFuse particle treated cells (Figure 5b, gray triangle) rapidly declined to background levels within 4 hour p.t. regardless of the translational status of the cells. These samples indicate that a low level protein transfer takes place either by vesicles or viral particles released from the packaging cells, which is diminished in the case of nonfusogenic viral particles, probably by lysosomal degradation.

This concept was further supported by analysis of viral particles generated from a noncoding transfer vector in combination with packaging constructs for a GFP-tagged Gag as well as wild-type Pol and Env (CMV-GFP/wt) that mimic a strong fluorescence signal induced by GFP protein transfer (Figure 5b, black circle). The initial fluorescence values of these cells were 12-fold higher than those incubated with wild-type RNA transfer vector particles (Figure 5b, compare black circle and black triangle at 0 hour p.t.). Independent from translation inhibition the CMV-GFP/wt cells displayed an MFI profile with slightly decreasing values (twofold) at early time points that remained constant from 4 hours onwards (Figure 5b, black circle). Analogous Gag-GFP particles with a fusion-deficient Env (CMV-GFP/iFuse), however, induced a target cell MFI profile characterized by a rapidly and substantially decreasing signal (10-fold) within the first 4 hours independent from translation inhibition (Figure 5b, gray circles). At later time points the fluorescence signal stayed constant in cycloheximide treated cells, whereas it further declined to nearly background in untreated cells. Notably, none of the samples transduced with GFP-tagged Gag viral particles displayed the characteristic increase in the target cell MFI over time seen in viral particles containing GFP-encoding RNAs.

In summary, these data demonstrate that the fluorescence signal in target cells transduced with wild-type PFV particles containing GFP-encoding nonviral RNAs is primarily due to de novo translation upon mRNA delivery into the target cell cytoplasm, despite a low level transfer of GFP encapsidated into released particles during vector production.

PFV-mediated transfer of Cre-encoding RNAs allows the efficient editing of the host cell genome

Having developed the PFV RNA transfer system, we next evaluated its potential to convey an enzymatic activity into target cells rather than a marker gene signal. To this end different types of PFV particles encapsidating mRNAs coding for an expression-optimized Cre recombinase were produced. Plain vector supernatants were used to infect mouse embryonic fibroblasts (MEF) cells containing a genomic yellow fluorescent protein (YFP) expression cassette, which harbors a floxed translational stop sequence in front of the eyfp ORF (Figure 6a).19 Subsequently, induction of YFP expression upon excision of the translational stop sequence by Cre-mediated recombination was monitored by flow cytometry over time (Figure 6b). Infection of reporter MEF cells with a standard PFV vector supernatant (MD9-Cre wt/wt, black square) as well as the wild-type RNA transfer vector supernatant (CMV-Cre wt/wt, black triangle) induced YFP expression in the majority (75–85%) of target cells (Figure 6b). Of note, the number of YFP positive cells at early time points p.t. was higher for RNA transfer vector-transduced cells than for cells infected with standard PFV vector supernatants (Figure 6b, d1–d3). Similar to the results obtained with the GFP mRNA transfer (Figure 5), cell culture supernatants generated with Cre mRNA in the absence of packaging components (Cre -/-), or supernatants containing particles composed of Cre mRNA, wild-type PFV Gag and the fusion-deficient PFV Env (CMV-Cre wt/iFuse), did not induce any significant YFP expression in reporter MEF cells (Figure 6b, light and dark gray triangle). However, virus supernatants produced in combination with the nucleic acid-binding-deficient PFV Gag packaging construct and wild-type Env (CMV-Cre iNAB/wt) did induce a YFP expression in a small fraction (~1%) of reporter MEFs (Figure 6b, white triangle). The latter is probably the result of low level Cre protein transfer, encapsidated during vector particle production.

Figure 6.

Figure 6

Open in a new tab

Cre-mediated genomic recombination in transduced primary mouse embryonic fibroblasts. Time course analysis of Cre-induced YFP expression in EYFP reporter MEF cells transduced with different retroviral vector supernatants. (a) Schematic outline of the ROSA26 locus with the integrated EYFP indicator cassette in the native and recombined state (adapted from ref. 19). SA: adenovirus splice acceptor; loxP: loxP recognition sequence; PGK: phosphoglycerate kinase promoter; Neo: neomycin resistance gene; tSV40 pA: triple SV40 poly A site; EYFP: enhanced yellow fluorescent protein ORF; bGH pA: bovine growth hormone poly A site. (b) MEF cells were transduced with undiluted viral vector supernatants generated by transient transfection of 293T cells with different combinations of the 4-component PFV vector system as indicated in the legend. The percentage of YFP positive cells of duplicate samples was determined every other day. Shown are data of an individual representative experiment (n=2). Mock: uninfected cells.

Taken together, these data demonstrate that the PFV RNA transfer vector system enables the recombination of genomic elements in primary target cell populations as efficiently as stable integrating vector systems. Furthermore, Cre-mediated genome editing is predominantly induced by de novo translation of transferred Cre-encoding mRNA.

Transient transgene expression by PFV-mediated RNA transfer in vivo

Stereotaxic injection of standard HIV-1 vector particles pseudotyped with VSV-G has been used to achieve stable transgene integration into both postmitotic as well as dividing neural stem cells during murine embryonic brain development and adult hippocampal neurogenesis.20,21 Therefore, we chose the latter system to evaluate the potential of the PFV RNA transfer system for transient genetic manipulation in vivo. Brains were coinjected with concentrated PFV GFP mRNA particles and, as positive control, with mCherry-encoding HIV-1 particles pseudotyped with VSV-G (Figure 7a). Three or ten days postinjection, the brains were isolated and tissue sections were examined for GFP and mCherry expression (Figure 7b). Analysis of the dentate gyrus (Figure 7c) showed persistent mCherry expression both in the subgranular zone (sgz), where neural stem cells are located, as well as in the granular layer (gl) containing postmitotic neurons (Figure 7d). The mCherry signal was revealed to persist at 10 days to comparable levels to those observed at 3 days. In contrast, PFV-driven GFP was observed primarily in the sgz and only at 3 days, but not 10 days p.i. (Figure 7d). This demonstrates that an efficient transient transgene expression in vivo can be achieved by the PFV vector-mediated transfer of nonviral RNAs.

Figure 7.

Figure 7

Open in a new tab

Transient transgene expression by PFV-mediated in vivo RNA transfer. (a) Schematic representation of the stereotaxic viral injection. (b) Experiment layout, with mice sacrificed 3 or 10 days post injection. (c) Fluorescence 4′,6-diamidino-2-phenylindole (DAPI) picture of a 40 μm coronal section of the dentate gyrus of the hippocampus with the dotted box indicating the magnified details shown in D. (d) Fluorescence picture of the dentate gyrus coinfected with VSV-G pseudotyped HIV-1 vector particles for stable mCherry expression and PFV vector particles containing EGFP encoding nonviral RNAs, at 3 dpi (top) and 10 dpi (bottom). n=3 (2 hippocampi analyzed per mouse). Scale bar: 100 μm. gl: granular layer; sgz: subgranular zone.

Discussion

Although retroviral particle assembly is associated with selective packaging of the viral genome, leading to its enrichment in released particles, it is estimated that up to 50% of the particle-associated nucleic acid mass is of nonviral origin.8 Potential functions of these cellular RNAs for viral replication in newly infected cells are not known. Here, we report, to our knowledge for the first time, the characterization and optimization of a novel method that allows the transient expression of transgenes in a dose-dependent manner by exploiting the physiological tendency of retroviruses to encapsidate cellular RNAs. Astonishingly, our data show that the efficiency of PFV vector particles in transferring and/or translating these nonviral RNAs appears to be superior compared with particles of MLV- or HIV-origin. The reason for this apparent difference remains to be elucidated. Despite having similar GFP-encoding mRNA copy numbers in particle preparations (within twofold range), the PFV-transduced cells showed 10-fold higher MFI, suggesting a more efficient uptake or disassembly compared with HIV or MLV. Therefore, an increased encapsidation of cellular RNAs into FV particles appears unlikely, although a more careful comparison of different retroviral vector particles, including quantitative data of physical particle release and nucleic acid composition are required to characterize this phenotype in further detail. Whether this feature of PFV poses a safety concern to conventional FV vectors by transduction and translation of cellular RNAs expressed in the packaging cells, such as SV40 T-antigen encoding RNA, remains to be seen and requires a large-scale analysis of the cellular RNA content in different retroviral vector particles.

We utilized this outstanding feature of PFV vector particles to develop a system for transient genetic manipulation of target cells by the transfer of nonviral RNAs. Unlike all currently available retroviral systems for transient transgene expression4,5 our system does not rely on viral sequences to be present in the encapsidated and transferred RNA. Furthermore, the Pol protein was completely dispensable for transgene expression, indicating that PFV particle disassembly and RNA release do not require virus-encoded enzymatic functions. This is in line with a recent report that PR activity is not absolutely essential for PFV entry.16 Although minimal protein transfer could be detected, several lines of evidence support the notion that the observed phenotypes (GFP-induced fluorescence, Cre-mediated recombination) are mainly the result of a de novo translation of the transferred RNA. First, blocking de novo protein synthesis in target cells prevents the steep increase in the GFP MFI profiles. Second, neither PFV-mutant particles deficient in RNA encapsidation nor FV particles with fusion-incompetent glycoproteins induced an efficient Cre-mediated recombination or GFP-dependent increase in target cell fluorescence. The proof-of-principle experiment using Cre-encoding mRNAs demonstrates the feasibility of the PFV RNA-transfer system. It allows rapid genome modification by taking the advantage of the broad host range of FVs, which enables efficient targeting of almost all cell types including primary tissues, and excludes the drawbacks of a long-lasting, potential genotoxic expression of the delivered enzyme due to integration or episomal expression cassettes.22,23 Furthermore, stereotaxical injection into mouse brains demonstrated that these vector particles can also be readily used for transient in vivo manipulation of target tissues. HIV vectors in the adult hippocampus were previously shown to efficiently target both postmitotic neurons and neural stem cells that coexist within the sgz. 21 Hence, the observation that the majority of cherry-encoding HIV vector-transduced cells in the sgz also transiently express GFP (Figure 7d) suggests that PFV vectors are similarly efficient in targeting both cell types. This is consistent with in vitro data using target cells arrested in the cell cycle by aphidicolin treatment suggesting that unlike stable integration of PFV proviral DNA genomes, transient transgene expression by PFV particle-mediated nonviral RNA transfer is readily observed also in nondividing cells (data not shown).24,25

So far, we could demonstrate efficient transfer and translation of protein-encoding mRNAs of 1.1–4.9 kb in length by this new PFV vector system. It remains to be seen if these vectors can also be utilized for functional transfer of other types of RNAs, for example, small noncoding RNA, such as shRNAs or miRNAs, and thereby be used for gene inactivation studies. Furthermore, we currently lack information on the copy numbers of specific nonviral RNAs in individual FV particles. It is therefore unknown whether overexpressed nonviral RNAs are encapsidated at a higher level than the two copies of viral genome thought to be present in natural PFV particles and found in other retroviral particles.

In summary, this new PFV-based vector system for efficient transfer of nonviral RNAs into target cells is a promising addition to the retroviral toolbox allowing transient manipulation of different target tissues in vitro and in vivo. Further potential applications include the transient expression of cellular transcription factors for reprogramming or trans-differentiation as well as transient expression of site-specific nucleases in various target tissues.

Materials and Methods

Cell lines and culture conditions. The human kidney cell line 293T,26 the human fibrosarcoma cell line HT108027 and d14.5 MEF from R26R-EYFP reporter mice19 were cultivated at 37°C and 5% CO2 in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% heat-inactivated fetal calf serum and antibiotics.

Recombinant DNAs and plasmid expression constructs. The expression-optimized PFV 4-component vector system consisting of the packaging plasmids pcoPG4 (Gag wt), pcoPP (Pol wt), pcoPE (Env wt) and the standard transfer vector puc2MD9 (MD9) enabling constitutive GFP expression upon host cell genome integration used for production of PFV wt particles was described previously.28,29 The transfer vector puc2MD9 Cre was generated by replacing the egfp ORF of puc2MD9 with an expression-optimized ORF encoding Cre (hCre) excised from pCAGGs-Cre-IRES-Puro.30

For some experiments, the variant expression-optimized PFV Pol expression constructs pcoPP1 (with inactive PR domain (Pol iPR), D24A mutation,16,31,32) pcoPP2 (with an inactive RT domain (Pol iRT), YVDD312–315GAAA mutation28,33), or pcoPP3 (with inactive IN domain (Pol iIN), D936A mutation16,34) were used. Furthermore, some experiments included use of the variant expression-optimized PFV Gag expression constructs pcoPG4 p68 (with deleted C-terminal p3 domain, aa 622-648 (p68)16), its C-terminal GFP-tagged variant pcoPG4-CeG p68 (Gag p68-GFP), or pcoPG4 ▵GR (nucleic acid-binding-deficient due to glycine-arginine-rich sequence I to III deletion (Gag iNAB) (Müllers et al., Unpublished Data)). Finally, the fusion-deficient variant glycoprotein (Env) packaging construct pcoPE32 (R571T mutation, (Env iFuse))35 was used in some cases. The nonviral RNA transfer vector pczi (CMV) contains a cytomegalovirus enhancer-promoter-intron A expression cassette with multiple cloning site in a pCDNA3.1+ zeo backbone whereas pSFFVU3 (SFFV) harbors a spleen focus forming virus U3 promoter expression cassette in a pCDNA3 backbone. A schematic outline of the most important features of these constructs is given in Figure 2a.

Lentiviral PFV Env pseudotypes were generated using the mutant PFV Env packaging vector pcoPE01,36 the HIV-1 Gag-Pol expressing packaging vector pCD/NL-BH37 and the HIV-1 based transfer vector p6NST60. This lentiviral transfer vector is a derivative of p6NST50 and contains a spleen focus forming virus U3 promoter-driven GFP expression cassette.38 MLV PFV Env pseudotypes were produced using pcoPE01, the MLV Gag-Pol expressing packaging vector pHIT6039 and the MLV-based transfer vector pczCFG2 fEGFPf, which is a variant of the pczCFG2 fEGN32 having the EGFP-neomycin resistance fusion gene replaced by a egfp ORF.

Viral supernatant production and viral particle concentration. Retroviral vector particles were produced as previously described.28,32,40 Briefly, the individual plasmids of the different retroviral vector systems were transiently cotransfected into 293T cells using polyethyleneimine (PEI) in 10 cm dishes using 16 µg total DNA. For PFV supernatants transfer vector as well as Gag, Pol and Env packaging components were used at a ratio of 15-26:4:2:1. Lentiviral and MLV vector supernatants were produced using transfer vector to Gag/Pol and Env packaging construct ratios of 7:7:1. If necessary, empty pUC19 plasmid was used in all experiments as stuffer to keep the total DNA amount of transfected DNA constant. At 26–30 hours posttransfection, the medium was exchanged for fresh culture medium (for in vivo injections FCS was omitted) and cell-free viral vector supernatant was harvested 48 h by using 0.45 μm sterile filters. For in vivo injections viral particles were concentrated by ultracentrifugation of cell-free viral supernatant at 25,000 rpm (82,705 g), 4°C for 3 hours in SW32Ti rotors (Beckman). Subsequently, pelleted particles were resuspended in PBS to obtain 200× volume concentrated viral supernatants. In a second centrifugation step concentrator columns (Millipore, Amicon Ultra 0.5 ml, 100 kDa MWCO) were used according to the manufacturer's instructions (10 min at 14,000 g, room temperature) to further concentrate viral particle stocks. By this 3000× volume concentrated virus stocks were generated that were snap-frozen in aliquots and stored at −80°C until use.

Analysis of viral infectivity and transgene expression. Infectivity of viral supernatants produced was determined by a flow cytometric based GFP marker gene transfer assay as described previously.29,40 HT1080 cells were seeded at a density of 2 × 104 cells/well in 12-well plates 16–24 hours before transduction. Target cells were incubated with 1 ml of cell-free viral supernatant generated by transient transfection as described above and 10-fold serial dilutions thereof for 4–6 hours before replacement with normal growth medium. Determination of the percentage of GFP-expressing cells by flow cytometry analysis was performed 48–72 hours posttransduction (p.t.). The values of the percentage of GFP-positive cells were used for titer determination as previously described.40 For an analysis of transient gene expression the mean fluorescent intensity (MFI) in the GFP channel of cultures transduced with undiluted viral supernatants was determined from all living cells as gated by their FSC/SSC profile. For synchronized infections a modified spinoculation protocol was used that involved preincubation of target cell (2 × 104 to 2 × 105 cells/well plated in 6-well dishes one day in advance) at 10°C for 10 min. Subsequently the growth medium was replaced with 2 ml cold virus supernatant with the plates kept on cool pads. Next the viral supernatant containing tissue culture plates were centrifuged for 30 min at 1,100× g at 10°C in a tissue culture centrifuge for loading of viral particles onto the target cells, but preventing their endocytic uptake or glycoprotein-mediated membrane fusion. Uptake and infection were then initiated by replacing the viral supernatant containing all nonadsorbed vector particles with fresh warm growth medium and incubation at 37°C, 5% CO2 for the time periods indicated. For short-term (up to 72 hours) time courses duplicate target cell wells plated at the same time were consecutively transduced by spinoculation in different time intervals and all cells of all samples were harvested for flow cytometry analysis at the same time. For long-term time courses (up to 21 days) duplicate target cell wells were transduced simultaneously. At the indicated time point post transduction cells were trypsinized, a fraction of the cell suspension used for flow cytometry analysis and the remainder replated. At the next time point, the same cultures were treated in a similar fashion. In experiments involving target cell translational suppression, cycloheximide was added to the target cell cultures at 100 µg/ml final concentration 10 min prior to addition of viral supernatant that was also supplemented with cycloheximide at 100 µg/ml. In all subsequent media changes cycloheximide was supplemented as well to maintain a continuous translational suppression.

Quantitative PCR analysis. Preparation of particle samples for qPCR analysis was performed as previously described.28,41 Briefly, viral particles concentrated by ultracentrifugation were DNase I digested to reduce plasmid contamination prior to nucleic acid extraction using a QIAamp viral RNA Mini kit (QIAGEN). Furthermore, cellular nucleic acids were extracted from the cell pellet of one 100-mm dish of transiently transfected 293T cells using the RNeasy Mini kit (QIAGEN) according to the manufacturer's protocol. Following a DNase I digest post extraction RNA samples (1/12 of viral nucleic acids, 250 ng total cellular RNA) were reverse transcribed using oligo dT30 primer. Subsequently, aliquots were used in duplicates for qPCR analysis employing a EGFP-specific primer-probe set, Maxima Probe qPCR Master Mix including ROX dye (Thermo Scientific) and a StepOnePlus (Applied Biosystems) quantitative PCR machine. All values obtained were referred to a standard curve consisting of 10-fold serial dilutions of a reference plasmid (puc2MD9) containing the target sequences. All sample values included were in the linear range of the standard curves with a span from 10 to 109 copies per qPCR reaction. Viral qPCR values are expressed as copy numbers corresponding to 1 ml pelleted plain supernatant calculated from the amount of supernatant corresponding to the fraction of extracted and reverse transcribed nucleic acids used in each qPCR reaction. Determined qPCR values of cellular samples were calculated as copies/ng total RNA.

In vivo stereotaxic injection and processing of brain tissue. A mixture of lentiviral vector particles encoding for mCherry and PFV particles carrying nonviral RNAs encoding GFP were stereotaxically coinjected (ratio 1:3) into 8–10-wk-old, isoflurane-anesthetized, C57/Bl6 female mice using an injector (Nanoliter 2000; World Precision Instruments) and a stereotaxic frame (model 900; Kopf Instruments) at ±1.6 mm mediolateral, –1.9 anterior–posterior, and –1.9 mm dorsoventral from bregma, as already reported.20,21 Three animals were used for each experimental condition. Brains were collected 3 or 10 days postinjection, fixed in 4% paraformaldehyde in PBS and cut with a sliding vibrotome (Leica). Coronal, 40-µm thick vibratome sections were used to perform immunohistochemistry for GFP and mCherry, as already described.21 As primary antibodies goat antiGFP (1:800, Rockland) and rabbit antiRFP (1:2000, Rockland) were used. Fluorescence composite pictures of the hippocampus were acquired using an automated microscope (ApoTome; Carl Zeiss). Animal experiments were approved by local authorities (24D-9168.11-1/2008-16 and 2007–2).

SUPPLEMENTARY MATERIAL Figure S1. Correlation of packaging cell and particle-associated RNA levels to target cell transgene expression PFV vector particles were generated by transient transfection of 293T cells with constant amounts of packaging plasmids for Gag, Pol and Env as well as decreasing amounts of the GFP-encoding RNA transfer vector (pcziEGFP (CMV-EGFP)) as indicated on the x-axis.

Acknowledgments

We thank K. Anastassiadis for providing the hCRE ORF, R. Behrendt for providing R26R-EYFP MEFs, and Michael Thomson for manuscript editing. M.V.H. and E.M. were partially supported by a DIGS-BB fellowship. This work was partially supported by grants from the DFG (LI621/3-3 and SPP1230 LI621/6-1), and CRTD (seed grant 2011) to D.L. A.B., S.B.A. and F.C. were supported by the CRTD, the TU-Dresden and the DFG Collaborative Research Center SFB655 (subproject A20). We declare that there are no conflicts of interest with the published work.

Supplementary Material

Supplementary Figure S1

Correlation of packaging cell and particle-associated RNA levels to target cell transgene expression PFV vector particles were generated by transient transfection of 293T cells with constant amounts of packaging plasmids for Gag, Pol and Env as well as decreasing amounts of the GFP-encoding RNA transfer vector (pcziEGFP (CMV-EGFP)) as indicated on the x-axis.

Click here for additional data file. (897.5KB, tiff)

References

  1. Suerth JD, Schambach A, Baum C. Genetic modification of lymphocytes by retrovirus-based vectors. Curr Opin Immunol. 2012;24:598–608. doi: 10.1016/j.coi.2012.08.007. [DOI] [PubMed] [Google Scholar]
  2. Naldini L. Ex vivo gene transfer and correction for cell-based therapies. Nat Rev Genet. 2011;12:301–315. doi: 10.1038/nrg2985. [DOI] [PubMed] [Google Scholar]
  3. Cavazza A, Moiani A, Mavilio F. Mechanisms of retroviral integration and mutagenesis. Hum Gene Ther. 2013;24:119–131. doi: 10.1089/hum.2012.203. [DOI] [PubMed] [Google Scholar]
  4. Schott JW, Galla M, Godinho T, Baum C, Schambach A. Viral and non-viral approaches for transient delivery of mRNA and proteins. Curr Gene Ther. 2011;11:382–398. doi: 10.2174/156652311797415872. [DOI] [PubMed] [Google Scholar]
  5. Banasik MB, McCray PB. Integrase-defective lentiviral vectors: progress and applications. Gene Ther. 2010;17:150–157. doi: 10.1038/gt.2009.135. [DOI] [PubMed] [Google Scholar]
  6. Johnson SF, Telesnitsky A. Retroviral RNA dimerization and packaging: the what, how, when, where, and why. PLoS Pathog. 2010;6:e1001007. doi: 10.1371/journal.ppat.1001007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Rein A, Datta SA, Jones CP, Musier-Forsyth K. Diverse interactions of retroviral Gag proteins with RNAs. Trends Biochem Sci. 2011;36:373–380. doi: 10.1016/j.tibs.2011.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Rulli SJ, Jr, Hibbert CS, Mirro J, Pederson T, Biswal S, Rein A. Selective and nonselective packaging of cellular RNAs in retrovirus particles. J Virol. 2007;81:6623–6631. doi: 10.1128/JVI.02833-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Rethwilm A, Lindemann D. Knipe DM, Howley PM. Fields Virology, 6th edn., vol. 2. Chapter 52. Lippincott Williams & Wikins, a Wolters Kluwer business; Philadelphia, PA; 2013. Foamy viruses. pp. 1613–1632. [Google Scholar]
  10. Verhoeyen E. Advances in foamy virus vector technology and disease correction could speed the path to clinical application. Mol Ther. 2012;20:1105–1107. doi: 10.1038/mt.2012.97. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Nasimuzzaman M, Persons DA. Cell Membrane-associated heparan sulfate is a receptor for prototype foamy virus in human, monkey, and rodent cells. Mol Ther. 2012;20:1158–1166. doi: 10.1038/mt.2012.41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Plochmann K, Horn A, Gschmack E, Armbruster N, Krieg J, Wiktorowicz T, et al. Heparan sulfate is an attachment factor for foamy virus entry. J Virol. 2012;86:10028–10035. doi: 10.1128/JVI.00051-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Lehmann-Che J, Renault N, Giron ML, Roingeard P, Clave E, Tobaly-Tapiero J, et al. Centrosomal latency of incoming foamy viruses in resting cells. PLoS Pathog. 2007;3:e74. doi: 10.1371/journal.ppat.0030074. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Trobridge GD, Horn PA, Beard BC, Kiem HP. Large animal models for foamy virus vector gene therapy. Viruses. 2012;4:3572–3588. doi: 10.3390/v4123572. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Müllers E. The foamy virus Gag proteins: what makes them different. Viruses. 2013;5:1023–1041. doi: 10.3390/v5041023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hütter S, Müllers E, Stanke N, Reh J, Lindemann D. Prototype foamy virus protease activity is essential for intraparticle reverse transcription initiation but not absolutely required for uncoating upon host cell entry. J Virol. 2013;87:3163–3176. doi: 10.1128/JVI.02323-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Deyle DR, Li Y, Olson EM, Russell DW. Nonintegrating foamy virus vectors. J Virol. 2010;84:9341–9349. doi: 10.1128/JVI.00394-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lindemann D, Rethwilm A. Foamy virus biology and its application for vector development. Viruses. 2011;3:561–585. doi: 10.3390/v3050561. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Srinivas S, Watanabe T, Lin CS, William CM, Tanabe Y, Jessell TM, et al. Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev Biol. 2001;1:4. doi: 10.1186/1471-213X-1-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Artegiani B, Lange C, Calegari F. Expansion of embryonic and adult neural stem cells by in utero electroporation or viral stereotaxic injection. J Vis Exp. 2012;6:pii. doi: 10.3791/4093. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Artegiani B, Lindemann D, Calegari F. Overexpression of cdk4 and cyclinD1 triggers greater expansion of neural stem cells in the adult mouse brain. J Exp Med. 2011;208:937–948. doi: 10.1084/jem.20102167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Galla M, Schambach A, Falk CS, Maetzig T, Kuehle J, Lange K, et al. Avoiding cytotoxicity of transposases by dose-controlled mRNA delivery. Nucleic Acids Res. 2011;39:7147–7160. doi: 10.1093/nar/gkr384. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Galla M, Will E, Kraunus J, Chen L, Baum C. Retroviral pseudotransduction for targeted cell manipulation. Mol Cell. 2004;16:309–315. doi: 10.1016/j.molcel.2004.09.023. [DOI] [PubMed] [Google Scholar]
  24. Bieniasz PD, Weiss RA, McClure MO. Cell cycle dependence of foamy retrovirus infection. J Virol. 1995;69:7295–7299. doi: 10.1128/jvi.69.11.7295-7299.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Müllers E, Stirnnagel K, Kaulfuss S, Lindemann D. Prototype foamy virus gag nuclear localization: a novel pathway among retroviruses. J Virol. 2011;85:9276–9285. doi: 10.1128/JVI.00663-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. DuBridge RB, Tang P, Hsia HC, Leong PM, Miller JH, Calos MP. Analysis of mutation in human cells by using an Epstein-Barr virus shuttle system. Mol Cell Biol. 1987;7:379–387. doi: 10.1128/mcb.7.1.379. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Rasheed S, Nelson-Rees WA, Toth EM, Arnstein P, Gardner MB. Characterization of a newly derived human sarcoma cell line (HT-1080) Cancer. 1974;33:1027–1033. doi: 10.1002/1097-0142(197404)33:4<1027::aid-cncr2820330419>3.0.co;2-z. [DOI] [PubMed] [Google Scholar]
  28. Müllers E, Uhlig T, Stirnnagel K, Fiebig U, Zentgraf H, Lindemann D. Novel functions of prototype foamy virus Gag glycine- arginine-rich boxes in reverse transcription and particle morphogenesis. J Virol. 2011;85:1452–1463. doi: 10.1128/JVI.01731-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Stirnnagel K, Lüftenegger D, Stange A, Swiersy A, Müllers E, Reh J, et al. Analysis of prototype foamy virus particle-host cell interaction with autofluorescent retroviral particles. Retrovirology. 2010;7:45. doi: 10.1186/1742-4690-7-45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Anastassiadis K, Fu J, Patsch C, Hu S, Weidlich S, Duerschke K, et al. Dre recombinase, like Cre, is a highly efficient site-specific recombinase in E. coli, mammalian cells and mice. Dis Model Mech. 2009;2:508–515. doi: 10.1242/dmm.003087. [DOI] [PubMed] [Google Scholar]
  31. Konvalinka J, Löchelt M, Zentgraf H, Flügel RM, Kräusslich HG. Active foamy virus proteinase is essential for virus infectivity but not for formation of a Pol polyprotein. J Virol. 1995;69:7264–7268. doi: 10.1128/jvi.69.11.7264-7268.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Lindemann D, Pietschmann T, Picard-Maureau M, Berg A, Heinkelein M, Thurow J, et al. A particle-associated glycoprotein signal peptide essential for virus maturation and infectivity. J Virol. 2001;75:5762–5771. doi: 10.1128/JVI.75.13.5762-5771.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Moebes A, Enssle J, Bieniasz PD, Heinkelein M, Lindemann D, Bock M, et al. Human foamy virus reverse transcription that occurs late in the viral replication cycle. J Virol. 1997;71:7305–7311. doi: 10.1128/jvi.71.10.7305-7311.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Enssle J, Moebes A, Heinkelein M, Panhuysen M, Mauer B, Schweizer M, et al. An active foamy virus integrase is required for virus replication. J Gen Virol. 1999;80 (Pt 6):1445–1452. doi: 10.1099/0022-1317-80-6-1445. [DOI] [PubMed] [Google Scholar]
  35. Stirnnagel K, Schupp D, Dupont A, Kudryavtsev V, Reh J, Müllers E, et al. Differential pH-dependent cellular uptake pathways among foamy viruses elucidated using dual-colored fluorescent particles. Retrovirology. 2012;9:71. doi: 10.1186/1742-4690-9-71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lindemann D, Stirnnagel K, Lüftenegger D.2013Foamy viral envelope genes. In: European Patent Office, EP (ed.). Bulletin 2013/33 55
  37. Mochizuki H, Schwartz JP, Tanaka K, Brady RO, Reiser J. High-titer human immunodeficiency virus type 1-based vector systems for gene delivery into nondividing cells. J Virol. 1998;72:8873–8883. doi: 10.1128/jvi.72.11.8873-8883.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Amer DA, Kretzschmar G, Müller N, Stanke N, Lindemann D, Vollmer G. Activation of transgenic estrogen receptor-beta by selected phytoestrogens in a stably transduced rat serotonergic cell line. J Steroid Biochem Mol Biol. 2010;120:208–217. doi: 10.1016/j.jsbmb.2010.04.018. [DOI] [PubMed] [Google Scholar]
  39. Soneoka Y, Cannon PM, Ramsdale EE, Griffiths JC, Romano G, Kingsman SM, et al. A transient three-plasmid expression system for the production of high titer retroviral vectors. Nucleic Acids Res. 1995;23:628–633. doi: 10.1093/nar/23.4.628. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Ho YP, Schnabel V, Swiersy A, Stirnnagel K, Lindemann D. A small-molecule-controlled system for efficient pseudotyping of prototype foamy virus vectors. Mol Ther. 2012;20:1167–1176. doi: 10.1038/mt.2012.61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Mannigel I, Stange A, Zentgraf H, Lindemann D. Correct capsid assembly mediated by a conserved YXXLGL motif in prototype foamy virus Gag is essential for infectivity and reverse transcription of the viral genome. J Virol. 2007;81:3317–3326. doi: 10.1128/JVI.01866-06. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Figure S1

Correlation of packaging cell and particle-associated RNA levels to target cell transgene expression PFV vector particles were generated by transient transfection of 293T cells with constant amounts of packaging plasmids for Gag, Pol and Env as well as decreasing amounts of the GFP-encoding RNA transfer vector (pcziEGFP (CMV-EGFP)) as indicated on the x-axis.

Click here for additional data file. (897.5KB, tiff)

Articles from Molecular Therapy are provided here courtesy of The American Society of Gene & Cell Therapy

RESOURCES