Retroviral and transposon-based tet-regulated all-in-one vectors with reduced background expression and improved dynamic range

HUMAN GENE THERAPY 22:166–176 (February 2011)ª Mary Ann Liebert, Inc.
DOI: 10.1089/hum.2010.099 Retroviral and Transposon-Based Tet-Regulated All-In-One Vectors with Reduced Background Expression and Improved Dynamic Range Niels Heinz,1 Axel Schambach,1 Melanie Galla,1 Tobias Maetzig,1 Christopher Baum,1 Rainer Loew,2 and Bernhard Schiedlmeier1 The regulated expression of therapeutic genes may become crucial in gene therapy when their constitutiveexpression interferes with cell fate in vivo. The efficient regulation of transgene expression requires tightly con-trolled inducible promoters, as shown for the tetracycline regulatory system (tet-system). However, its applicationrequires the introduction of two components into the target cell genome: the tet-responsive transactivator and theregulated expression cassette. In order to facilitate the usage of the tet-system for approaches in gene therapy, bothcomponents have to be transferred by a single vector, thus eliminating the preselection of transactivator positivecells. Published ‘‘all-in-one'' vectors for regulated transgene expression display a relatively low signal-to-noiseratio, resulting in regulatory windows of around 500-fold even in selected clones. In this study, we show that amodified vector architecture combined with the introduction of new tet-responsive promoters, Ptet, improved thedynamic range of such all-in-one vectors to levels up to 14,000-fold for viral and 25,000-fold for nonviral transfervectors in nonclonal human cell lines, and up to 2,800-fold in murine hematopoietic cell lines. This improvedregulation was the result of a strong reduction of background expression in the off-state, even if cells weretransduced at high multiplicity of infection, while induction remained at high levels. In addition, the resultsindicated that successful regulation of gene expression in different target cells depended on vector architecture aswell as the choice of the Ptet-promoter.
In the past, diverse modifications were introduced to improve the tet-regulated transgene expression system.
Drug-inducible controlofgeneexpression isawidely These modifications have included improvements of the tet- used strategy for investigation of gene function in many responsive promoter, where a reduction of the tet-operator basic areas of genetic and cellular biology. Regulated trans- sequences (tetO) toward 36-nucleotide (nt) spacing from core gene expression is also a desired safety feature in gene ther- to core improved the regulatory behavior (Agha-Mohammadi apy applications, allowing tightly controlled reversible gene et al., 2004). The transactivator has also been subject to various expression (Toniatti et al., 2004; Goverdhana et al., 2005). The refinements (Baron et al., 1997; Urlinger et al., 2000; Das et al., tetracycline regulatory system (tet-system) developed by 2004). The original transactivator binds only in the absence Gossen and Bujard (1992) has become a valuable tool for re- of tetracycline (TetOff), whereas modifications have led to peated transgene induction. It is based on the incorporation of the reverse function (Gossen et al., 1995), i.e., the reverse two components into the same target cell. The first ensures transactivator binds only in the presence of doxycycline, a the presence of sufficient amounts of the transactivator, a tetracycline derivative. This so-called TetOn system has been fusion protein of the tet-repressor and VP16 proteins from preferentially used since its invention, particularly in vivo, as herpes simplex virus, and the second consists of the inducible it does not require permanent administration of doxycycline transgene-expressing unit. The latter contains repeats of the to switch off transgene expression. Furthermore, the transac- tet-operator sequences (tetO) to which the transactivator can tivator has been optimized by replacing the original VP16 bind to activate a minimal promoter.
transactivation domain by a so-called minimal domain.
1Experimental Hematology, Hannover Medical School, 30625 Hannover, Germany.
2EUFETS GmbH, 55743 Idar-Oberstein, Germany.
IMPROVED TET-REGULATED ALL-IN-ONE VECTORS Additionally, the codon usage has been optimized for higher Ptet-T8 promoter was generated by fusion of the MMTV-U3 expression levels and stability in eukaryotes. One of the most core promoter fragment (88/þ121) (Loew et al., 2006) to promising transactivator is the TetOn variant rtTA2s-M2 a tet-operator heptamer with 36-nt center-to-center spaced (Urlinger et al., 2000), which is used in this study.
operators. Details of the lmg* dual reporter gene, coupling Effective and stable transfer of the tet-system components luciferase, and enhanced green fluorescence protein (eGFP) into a wide range of different cell types can be achieved via will be published elsewhere (Loew et al., 2010b) and are viral delivery systems (Kenny et al., 2002; Vigna et al., 2002; available on request.
Chtarto et al., 2003; Barde et al., 2006; Loew et al., 2006). Thecomponents of the tet-system can be delivered either on Cultivation of cells separate vectors (the so-called two-vector system) or in the HT1080 cells were cultured in Dulbecco's modified Eagle's more advanced approach by a single vector (the all-in- medium (Biochrom, Berlin, Germany) supplemented with one vector system). The major hindrances for the successful 10% fetal calf serum (FCS) (PAA, Pasching, Austria) and use of the tet-system are reliable introduction of both com- 0.1 mg/ml sodium pyruvate (PAA). BaF3 and 32D cells were ponents into the same cell and a high regulatory window, cultivated in RPMI 1640 medium (PAA) supplemented with i.e., high induction of the transgene and tight regulation in 10% FCS, 0.1 mg/ml sodium pyruvate and 5 ng of mIL-3/ml the absence of induction. Both approaches, therefore, face (Peprotech, London, UK). All cells were incubated at 378C and some problems. The two-vector system often relies on pre- selection for the integration of the transactivator-expressing 2. For induction of tet-regulated transgene expression, cells were cultivated in the presence of doxycycline (Sigma) unit (Gopalkrishnan et al., 1999; Qu et al., 2004). In the next at 1 mg/ml for 4 days. Medium containing doxycyline was step, the second vector containing the inducible cassette is refreshed every 2 days. Transduced cells were sorted ac- introduced, followed by a second selection to identify clones cording to eGFP fluorescence in the presence of doxycyline on displaying tightly regulated transgene expression. The all-in- a FACSAria (BD Bioscience, Heidelberg, Germany). Cells one vector system circumvents the need for preselection, were subsequently cultivated for 12 days in the absence of because a single transduction is sufficient for introducing all doxycyline before the background activity was measured.
components needed for tet-regulated transgene expression(Paulus et al., 1996). The major drawback of this system is a Preparation of viral supernatant, transduction, typically much higher background activity compared with the two-vector system. The all-in-one vector approach,therefore, showed regulatory windows of maximum 500-fold, For preparation of viral supernatant, 293T cells were seeded even after clonal selection (Kafri et al., 2000; Haack et al., 2004; onto 10-cm dishes and cotransfected (calcium phosphate) Barde et al., 2006). Recently, transposon based systems have with plasmids coding for the viral vector (10 mg/dish), gag/ attracted interest for the stable transfer of expression cassettes pol (10 mg), and VSV-G (0.5 mg) envelope protein in the pres- into target cells (Ivics et al., 1997; Wilson et al., 2007; Ma´te´s et al., ence of 20 mM HEPES (PAA) and 25 mM chloroquine (Sigma).
2009). So far, however, like for viral one-vector systems, the Every 12 h, the medium of the transfected producer cells was tet-regulation suffered from high background activity, even replaced by 8 ml of fresh culture medium including 20 mM after clonal selection, after transposon-mediated transfer into HEPES. The supernatant was harvested 36 h after transfec- target cells (Saridey et al., 2009).
tion, sterile-filtrated, and stored in aliquots at 808C. Titration In this study, we describe a novel all-in-one vector system, of viral supernatants was performed on SC-1 cells in the allowing tight control of transgene expression without loss presence of protamine sulfate (Sigma; 4 mg/ml) and doxycy- of inducibility. The central refinement of our vectors consists cline (Sigma; 1 mg/ml). Cells (1105) were seeded on six-well of modifications of the regulatory cassettes, employing new plates 24 h prior to infection. The transduction was performed Ptet-promoters. Combined with small alterations of vector using serial dilutions in 1 ml final volume, and fresh medium architecture, the modifications improved transgene regula- including doxycycline was added 24 h after transduction. The tion in murine and human bulk cell cultures after retroviral cells were FACS-analyzed 6 days after transduction. The titers and nonviral gene transfer, which greatly simplifies the ap- were compared on the basis of transducing units per millili- plication of tet-regulated expression in basic biology and so- ter of supernatant (TU/ml). For analyzing the tet-regulated matic gene therapy.
transgene expression, 1105 HT1080 cells were seeded on 24-well plates 24 h prior to transduction. Supernatants of known Materials and Methods titer were added at a multiplicity of infection (MOI) of 0.1 or 3,as described in Results, in the presence of protamine sulfate and doxycyline. Populations obtained with MOI 0.1 (1–3% All components needed for tet-regulated transgene ex- positive cells) were enriched by one round of sorting, whereas pression were subsequently introduced into plasmids con- MOI 3 populations were not enriched. For comparison, the taining either the basic ES.1-g-retroviral backbone (Loew et al., luciferase values were normalized to 100% eGFP-positive 2010a) or the terminal repeats for Sleeping Beauty transpo- sase, kindly provided by Z. Ivics (Ma´te´s et al., 2009). The Ptet- Cotransfection of HT1080 cells with plasmids for transient promoter variants Ptet-T2, -T6, and -T11 were generated by expression of Sleeping Beauty transposase (2 mg of pCMV- PCR with overlapping oligos by standard procedures as re- SB100x; Ma´te´s et al., 2009) and the transposable element commended by the suppliers (Phire-Taq, Biozym, Oldendorf, (2 mg) was performed using polyethylenimine (PEI; 25 kDa Germany), respectively. Structure and sequence details are linear; Polyscience, Niles, IL). Cells were seeded on six-well given in Supplementary Figs. S1 and S2 (Supplementary Data plates. One hundred microliters of PEI (0.1 g/L in 150 mM are available online at www.liebertonline.com/hum). The NaCl; pH 5.5) reagent and 100 ml of DNA solution (5 mg of HEINZ ET AL.
DNA in 150 mM NaCl; pH 5.5) were added to 1.8 ml of MOV-scT2 and MOV-hcT2 (Fig. 1A). Based on the Ptet-1 medium. Transfection reagent was replaced with normal (Gossen and Bujard, 1992), the minimal promoter Ptet-T2 medium including doxycyline after *12 h.
(Supplementary Fig. S1) was developed, which contains con-sensus sequences for the TATA box (cTATA) and the binding Reporter gene analysis site of the transcription factor IIB (cTFIIB) (Lagrange et al.,1998). In the context of the monocistronic vectors, these eGFP fluorescence was measured with a FACSCalibur (BD modifications increased the regulatory window by *10-fold, Bioscience) and analyzed using FlowJo software (Tree Star due to increased inducible activity and a reduced background Inc., Ashland, OR). Dead cells were excluded for analysis activity in the off-state of the system (Loew et al., 2010b).
through exclusion of propidium iodide (Sigma; 1 mg/ml)- The basic experimental design is shown in Fig. 1B. The positive cells. Luciferase activity was measured as previously regulatory characteristics of the vectors were investigated described (Loew et al., 2010a). In brief, cells were centrifuged by the expression of the lmg* dual reporter gene, consisting and resuspended in lysis buffer. Typically 1–10 ml of the cell of luciferase and eGFP, that allowed the simultaneous mea- lysate was measured at room temperature on a tube lumin- surement of both gene activities (Loew et al., 2010b). Human ometer (LB 9507; Berthold Technologies, Bad Wildbad, Ger- HT1080 cells were transduced to an efficiency of a maximum many). Values were normalized against the total protein 20% eGFP-positive cells after induction, and highly purified content of the cell lysate. Protein determination was per- populations of transduced cells (>90%) were obtained by formed using the Coomassie Plus (Bradford) Assay kit (Pierce FACS. The vector copy number was determined via qPCR Biotechnology, Rockford, IL). One microliter of cell lysate was (Table 1), revealing a similar average of vector integrates for treated according to the manufacturer's instruction in tripli- all tested populations. After cultivation of these sorted pop- cate. Analysis was performed on a microplate reader and ulations in either the presence (i.e., ‘‘on-state'') or absence analyzed with SoftMax Pro 4.0 software (Molecular Devices, (‘‘off-state'') of doxycycline, the cells were analyzed for eGFP Sunnyvale, CA). For a standard curve, different amounts of fluorescence and luciferase activity.
bovine serum albumin (0.5–8 mg) were used. Statistical anal-ysis was performed using Student's t test.
Exploring effects of vector elements on the regulatory properties of the MOV vector Mean copy number was determined via quantitative PCR We determined the tet-inducible transgene expression of (qPCR) using primers detecting the wPRE (woodchuck the vectors MOV-scT2 and MOV-hcT2, in HT1080 cells. The posttranscriptional element; forward, 50-GAGGAGTTGTGG dynamic range of gene regulation based on the induced eGFP CCCGTTGT-30; reverse, 50-TGACAGGTGGTGGCAATGCC-30) expression and the luciferase activity was similar (*220-fold) (Modlich et al., 2006) and PTBP2 (polypyrimidine tract bind- for both constructs (Fig. 1C). Thus, MOV-hcT2 was chosen as ing protein 2; tm_PTBP2_optimized2_FW: 50-TCTCCA an initial reference for further experiments. Next, we tested whether the removal of the splice acceptor from MOV-hcT2 GTTCCCGCAGAATGGTGAGGTG-30) as internal reference would lead to higher vector titers. For this purpose, we re- for human and mouse cells (Rahman et al., 2004). qPCR was moved the pol/env border fragment containing the native performed using FAST SYBR Green reagent (Stratagene, splice acceptor (SA) site, resulting in the vector MOV.1-hcT2 Santa Clara, CA) on a StepOnePlus cycler equipped with (Fig. 1D). As shown in Fig. 2, no significant differences in StepOne Software v2.0 (Applied Biosystems, Carlsbad, CA).
the retroviral titers were observed, with titers in the range of For quantification, a plasmid standard was used containing 1–3106 TU/ml. However, comparison of the luciferase ac- the sequences for wPRE and PTBP2.
tivities determined in the enriched cell populations (Fig. 1D,right panel) revealed a fivefold reduction of background ex- pression in the off-state in populations transduced by MOV.1-hcT2, whereas the level of induction remained unchanged.
Retroviral ‘‘all-in-one'' vector: design Thus, the dynamic range of gene regulation was increased and experimental outline three- to fourfold (Fig. 1D, left panel) by this modification of We constructed bidirectional tetracycline-inducible all-in- the vector backbone.
one vectors, integrating all components required for tet-regulated transgene expression into the ES.1-g-retroviral SIN Introduction of improved Ptet-promoters vector backbone (Loew et al., 2010a). The reverse (TetOn) further reduces background expression transactivator rtTA2S-M2 (Urlinger et al., 2000) was expressedunder the control of the human phosphoglycerate kinase To further increase the regulatory window, new Ptet- promoter (hPGK), which is constitutively active in a wide promoters, Ptet-T6, Ptet-T8, and Ptet-T11 (Supplementary range of mammalian cells. The inducible unit was inserted as Figs. S1 and S2) were integrated into the MOV.1-hcT2 an antisense expression cassette (30–50relative to the viral ge- backbone, resulting in the vectors MOV.1-hcT6, MOV.
nome), which consisted of an optimized tet-responsive pro- 1-hcT8, and MOV.1-hcT11, respectively (Fig. 1D). The intro- moter (Ptet-T2) containing a tet-operator heptamer fused to a duction of the new Ptet-promoters resulted in an improved minimal promoter followed by the cDNA of choice, a con- signal-to-noise ratio in transduced HT1080 cell populations, stitutive transport element from Mason Pfizer Monkey Virus i.e., the background expression in the off-state was reduced (CTE) (Schambach et al., 2000) and polyA signals from either while high induction levels were maintained, clearly dem- the SV40 (SV40pA) or the human growth hormone (hghpA).
onstrating the advanced regulatory properties of these novel These initial MoMLV-based bidirectional vectors were termed Ptet-promoters (Fig. 3A, left panel). Based on the luciferase

IMPROVED TET-REGULATED ALL-IN-ONE VECTORS Experimental outline and ba- sic all-in-one vectors. (A) ES.1 based g-retroviral vector with SIN-LTR (DU3),extended packaging region (C/Cþ)and pol/env (p/e) border fragmentincluding the native splice acceptorsite (SA). All components needed fortet-regulated transgene expression wereinserted as a bidirectional expressioncassette. The reverse tet-responsivetransactivator variant M2 is constitu-tively expressed by the human PGKpromoter (þstrand), followed by thewoodchuck transcriptional regulatory element (PRE).
In the antisense direction (strand), thetet-inducible expression cassette is in-serted along with the constitutivetransport element (CTE), from MasonPfizer Monkey Virus, and either SV40late (SV40pA) or human growth hor-mone (hghpA) polyadenylation signal,resulting in the vectors MOV-scT2 andMOV-hcT2, respectively. Lmg* is a fu-sion protein of luciferase and eGFP. (B)In the basic experimental design,HT1080 cells were infected to obtain aprimary infection rate of <20%, andcells positive for eGFP were sorted toanalyze the regulation, based on eGFPfluorescence and luciferase activity inmass cultures. (C) Left panel: Fluores-cence signals of HT1080 cell populationstransduced by MOV-scT2 (dotted line)or MOV-hcT2 (solid line). Right panel:Luciferase activities determined fromthe identical populations (means  SD;n ¼ 3). (D) Schematic outline of the ret-roviral backbone modification. Basedon the vector MOV-hcT2, the pol/envborder fragment containing the nativesplice acceptor site was removed, re-sulting in the vector MOV.1-hcT2. Leftpanel: Representative FACS analysisdetermined for HT1080 populationstransduced by MOV-hcT2 (dotted line)or MOV.1-hcT2 (solid line). Right panel:Luciferase activity of the identical pop-ulations (means  SD; n ¼ 3; *P < 0.5).
Introduction of Ptet-T6, Ptet-T8, andPtet-T11 resulted in the vectors MOV.1-hcT6, MOV.1-hcT8, and MOV.1-hcT11,respectively.
measurements, the introduction of Ptet-T6 led to an ap- activity upon induction was reduced only *1.4- to 1.9-fold proximately threefold reduction of background activity rel- when compared with Ptet-T2. Moreover, the introduction of ative to the Ptet-T2 promoter, whereas induction was only all modified minimal promoters into the MOV.1 backbone moderately decreased (*15% decline). The introduction of resulted in approximately twofold higher vector titers com- Ptet-T8 and Ptet-T11 promoters (for details, see Materials pared with the Ptet-T2 variant (Fig. 2). Taken together, the and Methods) further reduced background expression.
introduction of the novel Ptet-promoters into the bidirec- Measurement of luciferase revealed an *13- to 35-fold de- tional viral vector context increased the dynamic range of crease of background activity in the off-state, whereas the gene regulation in transduced HT1080 cell populations from HEINZ ET AL.
Table 1. Comparison of Mean Copy Numbers and Standard Deviations (SD) of the Sorted Populations Analyzed in this Study Comparison of titers. Titers were determined for the vectors used in this study. Titration was based on eGFPfluorescence and was performed on SC-1 cells. Values aregiven as transducing units per milliliter (TU/ml). Data are *700-fold for Ptet-T2 up to *2,000-, 7,000-, and 14,000-fold shown as means  SD (n ¼ 3). n.s., data not significantly dif- for the Ptet-T6, Ptet-T8, and Ptet-T11 promoter, respectively ferent (P > 0.05); *P < 0.05; **P < 0.01.
(Fig. 3A, right panel), combined with the additional advan-tage of increased vector titers.
vector production. Thus, we assumed that the inversion ofthe regulatory unit, preventing the generation of antisense Impact of high gene transfer rates transcripts in packaging cells, might further improve vector on the tet-induced gene regulation titers. Indeed, the inversion of the regulatory unit, result- In many gene therapy trials, multiple transduction cycles ing in the unidirectional vectors MOV.1-senseT6, MOV.1- were performed to ensure sufficient gene transfer (30–60%), senseT8, and MOV.1-senseT11 as illustrated in Fig. 4A, led potentially resulting in two to four vector copies per cell in to two- to threefold higher viral titer (*1–2107 TU/ml; >30% of the transduced cells (Kustikova et al., 2003; Fehse Fig. 2). However, determination of luciferase activity in and Roeder, 2008). Thus, we further explored whether the transduced HT1080 cells (Fig. 4B), revealed a 1.3- to 6-fold improved regulatory performance of our novel all-in-one reduced dynamic range of transgene regulation for these vectors is maintained in cells containing more than one vector unidirectional vectors when compared with their bidirec- insertion. We compared the luciferase activity in the on- and tional counterparts (Fig. 4C). Nevertheless, the unidirec- off-state between HT1080 cell pools that were transduced at tional vectors still mediated a 1,000- to 10,000-fold gene an MOI of either 0.1 or 3. Transduction of HT1080 cells at an MOI of 3 resulted in gene transfer rates ranging from 47% to69% for the different MOV.1 vectors, whereas at an MOI of 0.1 Properties of the unidirectional retroviral the gene transfer efficiencies were below 3%. For all three vectors in murine hematopoietic cell lines MOV.1 vectors, the range of transgene regulation observed in To test whether the unidirectional all-in-one vectors ensure HT1080 cells transduced at an MOI of 3 was only slightly also a robust high-level transgene regulation in other tissue lower (1.6- to 2.9-fold) than that in cell pools transduced at an culture models, we examined their regulatory properties in MOI of 0.1; thus, the dynamic window of gene regulation was murine hematopoietic cell lines. BaF3 (a pro B cell line) and maintained at levels above 1,000-fold (Fig. 3B). Furthermore, 32D cells (a bone marrow-derived myeloid cell line), both as summarized in Fig. 3C, the improved regulatory window of widely used for signaling studies, were transduced and en- our novel all-in-one vectors at high MOIs was also retained in riched by FACS as described for HT1080 cells. In contrast to four other established lines derived from a variety of tissues.
the induction levels achieved in HT1080 cells, transgene ex-pression levels in both hematopoietic cell lines were highly Comparison of bidirectional vs. unidirectional reduced (four- to 30-fold) in the on-state for both the bidi- vector architecture rectional and the unidirectional vectors (Fig. 5A). However, For gene therapeutic approaches, the vector titers that unidirectional vectors showed up to sixfold and 16-fold re- can be achieved for a certain construct are important. Titers duced background expression in BaF3 and 32D cells, in a side- generated by transient transfection of packaging cells with by-side comparison with the bidirectional vectors (Fig. 5A).
the bidirectional vectors MOV.1-hcT6 and MOV.1-hcT8 Overall, the unidirectional vector MOV.1-senseT11 conferred were about 5–6106 TU/ml. The increased titers after Ptet- the highest dynamic range of gene regulation (*1,000-fold) T2 was replaced by Ptet-T6, Ptet-T8, or Ptet-T11 within the in BaF3 cells, whereas in 32D cells the best regulatory win- bidirectional MOV.1 vectors (Fig. 2) suggested that the new dow (2,800-fold) was achieved with the unidirectional vector Ptet-promoters generated less antisense transcripts during MOV.1-senseT6 (Fig. 5B).
IMPROVED TET-REGULATED ALL-IN-ONE VECTORS C 1.0×108
242 1059 787
Novel Ptet-promoters improved the dynamic range of gene regulation after viral transfer. (A) Comparison of the vectors transferring the novel Ptet promoters. Left panel: Luciferase activities determined in HT1080 cells in the presence andabsence of doxycycline (means  SD; n ¼ 3; *P < 0.5; ***P < 0.0001). Right panel: Regulation factors resulting from luciferaseactivity in the on- and off-state. (B) HT1080 cells were transduced with the vectors MOV.1-hcT6, MOV.1-hcT8, and MOV.1-hcT11 at low or high MOI (MOI 0.1 and 3, respectively) and analyzed for their dynamic range. Populations transduced at anMOI of 3 were analyzed without enrichment. Shown are luciferase data (left panel; means  SD; n ¼ 3) normalized accordingto their transduction efficiency (%pos) and the resulting fold regulation (right panel). (C) The vectors MOV.1-hcT6, MOV.1-hcT8, and MOV.1-hcT11 were used to transduce different cell lines at an MOI of 3. Luciferase values were normalizedaccording to their transduction efficiency (%pos).
HEINZ ET AL.
we tested the unidirectional all-in-one vector concept in a transposon-based, nonviral vector system. The two compo- nents of the tet-system were integrated as distinct expression tetO7 MP lmg* hPGK M2 PRE units into the Sleeping Beauty transposable element resulting in TOV-T6, TOV-T8, and TOV-T11 vectors (Fig. 6A). In con-trast to the viral vectors, the terminal repeats of the transpo- son do not contain a polyA signal; thus, both units received a polyA signal (SV40pA for the tet-inducible and HGHpA forthe constitutive expressing cassette). HT1080 cells were co- transfected with TOV-T6, TOV-T8, and TOV-T11 and an ex-pression plasmid encoding the hyperactive Sleeping Beauty transposase (SB100x) (Ma´te´s et al., 2009). After stable inte- B 1.0×107
gration, eGFP-positive cells were enriched by FACS and cul- tivated as described for the viral approach. Determination of luciferase activity of TOV-T6, TOV-T8, and TOV-T11 trans- duced cells revealed similar levels of transgene induction, as ei 1.0×105
was observed in HT1080 populations transduced by the viral vector counterparts (Fig. 6B). In comparison with the viral approach, nonviral gene delivery conferred an even stronger reduction of luciferase activity in the off-state (Fig. 6B, left panel). Compared with the viral vectors, the background ac- [r 1.0×102
tivity was reduced four- and sixfold for Ptet-T6 and Ptet-T8controlled gene expression and, as a consequence, the dy- namic range of gene regulation increased to 18,000- and 25,000-fold. In contrast to the findings obtained with the viral vectors in transduced HT1080 cells, the introduction of Ptet- T11 showed no further improvement compared with Ptet-T8.
In summary, we showed that improved tet-regulated all- in-one vectors, based on the incorporation of novel minimalpromoters and modification of the vector architecture, re- sulted in highly improved regulatory windows in hemato- poietic and nonhematopoietic cell lines. In particular, our novel all-in-one vectors display very low background ex- pression in the off-state, which is of major interest for many approaches in gene therapy and cell biology.
The development of vectors, viral or nonviral, able to transfer both components of the tet-system simultaneously into a target cell, has been the subject of many studies. Inter- estingly, almost all research on this topic was done by col- leagues engaged in the development of gene therapy, which isa consequence of the work with primary cells that do not Comparison of bi- and unidirectional vectors in allow extended ex vivo handling, e.g., preselection. However, HT1080 cells. (A) Schematic outline of the modification in the the tight regulation of gene expression in cell popula- vector architecture. The tet-inducible cassette was inverted tions transduced by such all-in-one vectors was hampered within the MOV.1-hc backbone; therefore, expression of both by the interference of the Ptet-promoters (mostly Ptet-1) and genes is unidirectional. Resulting vectors were termed MOV the promoters (e.g., CMVie-promoter) used to express the tet- .1-senseT6, -senseT8, and -senseT11, respectively. (B) Luci- transactivators. Therefore, in most all-in-one vectors de- ferase activity determined in enriched HT1080 cell popula- scribed so far, the induction of gene expression did not exceed tions transduced by the indicated vectors. (C) The regulatorywindow based on luciferase activity determined in the on- 100- to 500-fold and displayed relatively low signal-to-noise and off-state.
ratios, even in single-cell clones (Gopalkrishnan et al., 1999;Johansen et al., 2002; Vigna et al., 2002).
In this study, we described tet-inducible all-in-one vec- tor systems optimized for tightly regulated transgene ex- A nonviral approach results in highly reduced pression in nonclonal cell populations, as required for gene activity in the uninduced state therapy approaches. Optimization was achieved by alter- DNA transposon-mediated gene delivery systems have ations of the retroviral architecture and incorporation recently been developed as an alternative tool for retroviral of novel tet-responsive promoters, together leading to gene transfer because they offer important advantages improved regulatory windows greater than 10,000-fold in over viral vectors (VandenDriessche et al., 2009). Therefore, nonclonal cell populations. Remarkably, we were able to IMPROVED TET-REGULATED ALL-IN-ONE VECTORS further improve the regulatory performance two- to eight- fold by incorporation of distinct expression cassettes for the tet-dependent transactivator and the regulatory unit into a Sleeping Beauty transposable element, resulting in up to The tet-responsive promoters Ptet-T6, Ptet-T8, and Ptet- T11 (Supplementary Figs. S1 and S2), introduced into the all- in-one vector systems, displayed reduced backgroundexpression in the off-state of transduced cell populations when compared with Ptet-T2, a tet-responsive promoterclosely related to Ptet-1 (Gossen and Bujard, 1992). The mg protein]
structure and functional properties of Ptet-T2 and Ptet-T6 promoters have been described in detail in a recent study (Loew et al., 2010b). The background expression as well as the induction level of Ptet-T8 was reduced comparedwith those of Ptet-T6, thus confirming earlier results with this minimal promoter (Loew et al., 2006). Interestingly, the MMTV-core promoter, as designed by evolution, contains several cis-elements that were assumed to be involved in its native regulation of hormone-responsive activity. In addition senseT6 se
to one GRE (glucocorticoid-receptor element), NF-1 and Oct-1 (Fox-AI) binding sites especially were of importance. Their coordinate binding is proposed to be responsible for nucle- osome repositioning during activation of transcription by the steroid hormone (Belikov et al., 2004; Holmqvist et al., 2005).
Thus, the tightly controlled expression of Ptet-T8 may be related to a particular promoter structure during the off-state of tet-controlled gene expression. This finding prompted usto modify the Ptet-T6 promoter accordingly. All residual CMVie-promoter sequences (50/–36 and 23/þ16) in-cluding the initiator element (Inr) were replaced by MMTV sequences. The resulting promoter, Ptet-T11, indeed showed mg protein]
superior background control in HT1080 cell populations (Fig.
3). However, further experiments to investigate whether re- moval of the Inr element and/or the introduction of the Oct- 1 (Fox-AI) binding site contributed to the enhanced control ofbackground expression in the off-state, when transferred by otherwise identical viral vectors, were beyond the scope of this study. Importantly, the regulatory properties of the vectors were fairly maintained even at high MOI, a test done to mimic the situation in gene therapy trials (Kustikova et al., senseT6 se
2003; Fehse and Roeder, 2008). Similar investigations for the performance of one-vector systems so far have been rarely included in publications (Barde et al., 2006).
The transiently produced vector titers differed significantly depending on the Ptet-promoter and vector architecture.
Increased titers were observed for bidirectional vectors, trans- ferring Ptet-promoters displaying tightly controlled back-ground expression in the off-state. The decreased backgroundactivity may generate less antisense transcripts, able to interfere Regulated gene expression in hematopoietic cell lines. (A) The vectors MOV.1-hcT6, MOV.1-senseT6, MOV.1-senseT8, MOV.1-hcT11, and MOV.1-senseT11 were used totransduce the hematopoietic cell lines BaF3 (upper panel) and 32D (lower panel). Luciferase activity was determined from enriched cell populations (means  SD; n ¼ 3; *P < 0.5; **P < 0.01; ***P < 0.0001). (B) The regulatory window based on luciferase activity determined in the on- and off-state.
HEINZ ET AL.
transposable element tio 20000
mg protein]
Tet-inducible Sleeping Beauty transposase system. (A) Schematic outline of the transposable element containing all components needed for tet-regulated transgene expression. The distinct tet-inducible and the constitutive transactivatorexpression cassettes were introduced for unidirectional expression, flanked by the terminal repeats (ITR) of Sleeping Beautytransposon. (B) Luciferase activity (left panel; means  SD; n ¼ 3) and background activity for all vectors determined afterenrichment of stably transformed HT1080 cell populations. The regulatory window for these vectors is shown (right panel).
with homodimer formation of the viral genome during viral Recently, transposon-based vector systems were used for particle formation and/or eliciting a cellular response against the delivery of expression cassettes (Ivics et al., 1997; Ding et al., double-stranded RNA (Maetzig et al., 2009). The same mech- 2005; Ma´te´s et al., 2009). Using the tet-system in the context of anism would also explain the approximately threefold increase a transposon based DNA vector, only low regulatory win- of titers obtained with the unidirectional vectors.
dows have been reported (Saridey et al., 2009). We assumed Interestingly, although in HT1080 cell populations the bi- that the novel Ptet-promoters should also mediate tightly directional design of the viral vectors resulted in a higher controlled gene expression in a transposon-based approach.
dynamic range compared with the unidirectional expression Indeed, nonclonal HT1080 cultures stably transformed by system, the latter was superior in hematopoietic cell lines.
Sleeping Beauty displayed very low background activity in This cell type-dependent shift of the dynamic range of tet- the off-state, resulting in an outstanding high regulation of regulation toward lower activity might be explained by dif- gene expression (>25,000-fold) in bulk cultures. The presence ferent compositions of the basal transcription machinery in or absence of viral components, e.g., the pol/env region (Fig.
these cells, as demonstrated for myoblast and terminally 1D), or the altered integration preferences offer possible ex- differentiated myotubes (Deato and Tjian, 2007). The indi- planations. Preferential integrations of the g-retroviral vectors vidual composition of the transcription machineries may in the vicinity of transcriptional start sites or CpG islands also explain preferences for regulatory units observed in might lead to stronger interactions of the minimal promoter standard cell lines, e.g., NIH3T3, CHO, HeLa, or 293T, after with nearby cellular enhancers (Lewinski et al., 2006; Beard introduction of Ptet-promoters by viral vectors (Fig. 3C).
et al., 2007). Accordingly, the more random integration of Other studies have also found different regulatory windows Sleeping Beauty transposon throughout the genome (Yant in different cell types, affecting both inducibility and back- et al., 2005) should reduce interference with cellular enhancers, ground activity (Haack et al., 2004; Markusic et al., 2005).
resulting in lower background activity of the minimal pro- Overall, we concluded that a strong interdependence be- moter in the off-state. Studies using clonal analysis clearly tween the vector architecture, the Ptet-promoters, and the showed that different clones behave differently in terms of cell type exists, underlining the need to optimize the setting their regulation, indicating an influence of the integration site for specific applications.
on regulatory behavior (Saridey et al., 2009). However, the IMPROVED TET-REGULATED ALL-IN-ONE VECTORS examination of basic mechanisms involved in chromosomal Tetracycline-inducible transgene expression mediated by a interference with the Ptet-promoters was beyond the scope of single AAV vector. Gene Ther. 10, 84–94.
this study, which aimed to provide suitable transfer vectors Das, A.T., Zhou, X., Vink, M., Klaver, B., Verhoef, K., Marzio, G., for regulated gene expression in bulk cultures.
and Berkhout, B. (2004). Viral evolution as a tool to improve the Together, the new vectors described here will likely im- tetracycline-regulated gene expression system. J. Biol. Chem.
prove the utility of the TetOn system in gene therapy ap- 279, 18776–18782.
proaches. The regulatory properties of the Ptet-promoters Deato, M.D., and Tjian, R. (2007). Switching of the core tran- were target cell-specific, affected by the vector architecture scription machinery during myogenesis. Genes Dev. 21, 2137– and greatly improved when a transposon-based transfer sys- tem was used. Because all data were obtained on cell popu- Ding, S., Wu, X., Li, G., Han, M., Zhuang, Y., and Xu, T. (2005).
Efficient transposition of the piggyBac (PB) transposon in lations, misleading results that might arise from influences of mammalian cells and mice. Cell 122, 473–483.
the chromosomal context in clonal analysis were avoided.
Fehse, B., and Roeder, I. (2008). Insertional mutagenesis and clonal dominance: biological and statistical considerations.
Gene Ther. 15, 143–153.
This work was supported by grants of the German min- Gopalkrishnan, R.V., Christiansen, K.A., Goldstein, N.I., istry for Research and Education (CB-Hermes, 01GN0930), Depinho, R.A., and Fisher, P.B. (1999). Use of the human EF- the Deutsche Forschungsgemeinschaft (KL 1311/4-1 and 1alpha promoter for expression can significantly increase suc- Cluster of Excellence REBIRTH Exc 62/1), and the European cess in establishing stable cell lines with consistent expression: a Union (PERSIST, HEALTH-F5-2009-222878). The Sleeping study using the tetracycline-inducible system in human cancercells. Nucleic Acids Res. 27, 4775–4782.
Beauty Transposase System was kindly provided by Zoltan Gossen, M., and Bujard, H. (1992). Tight control of gene ex- Ivics, Zsuzsanna Izsva´k, and Lajos Ma´te´s (MDC Berlin). We pression in mammalian cells by tetracycline-responsive pro- would like to acknowledge the assistance of the Cell Sorting moters. Proc. Natl. Acad. Sci. U.S.A. 89, 5547–5551.
Core Facility of the Hannover Medical School supported in Gossen, M., Freundlieb, S., Bender, G., Muller, G., Hillen, W., part by Braukmann-Wittenberg-Herz-Stiftung and Deutsche and Bujard, H. (1995). Transcriptional activation by tetracy- Forschungsgemeinschaft. The authors thank Tamaryin clines in mammalian cells. Science 268, 1766–1769.
Godinho for critical reading of the manuscript.
Goverdhana, S., Puntel, M., Xiong, W., Zirger, J.M., Barcia, C., Curtin, J.F., Soffer, E.B., Mondkar, S., King, G.D., Hu, J., Author Disclosure Statement Sciascia, S.A., Candolfi, M., Greengold, D.S., Lowenstein, P.R., A competing financial interest exists: A patent application and Castro, M.G. (2005). Regulatable gene expression systemsfor gene therapy applications: progress and future challenges.
for the novel Ptet promoters has been filed by Rainer Loew Mol. Ther. 12, 189–211.
and Herrmann Bujard.
Haack, K., Cockrell, A.S., Ma, H., Israeli, D., Ho, S.N., McCown, T.J., and Kafri, T. (2004). Transactivator and structurally op- timized inducible lentiviral vectors. Mol. Ther. 10, 585–596.
Agha-Mohammadi, S., O'Malley, M., Etemad, A., Wang, Z., Xiao, Holmqvist, P.H., Belikov, S., Zaret, K.S., and Wrange, O. (2005).
X., and Lotze, M.T. (2004). Second-generation tetracycline- FoxA1 binding to the MMTV LTR modulates chromatin regulatable promoter: repositioned tet operator elements opti- structure and transcription. Exp. Cell Res. 304, 593–603.
mize transactivator synergy while shorter minimal promoter Ivics, Z., Hackett, P.B., Plasterk, R.H., and Izsvak, Z. (1997).
offers tight basal leakiness. J. Gene Med. 6, 817–828.
Molecular reconstruction of Sleeping Beauty, a Tc1-like trans- Barde, I., Zanta-Boussif, M.A., Paisant, S., Leboeuf, M., Rameau, poson from fish, and its transposition in human cells. Cell 91, P., Delenda, C., and Danos, O. (2006). Efficient control of gene expression in the hematopoietic system using a single Tet-on Johansen, J., Rosenblad, C., Andsberg, K., Moller, A., Lundberg, inducible lentiviral vector. Mol. Ther. 13, 382–390.
C., Bjorlund, A., and Johansen, T.E. (2002). Evaluation of Tet- Baron, U., Gossen, M., and Bujard, H. (1997). Tetracycline- on system to avoid transgene down-regulation in ex vivo gene controlled transcription in eukaryotes: novel transactivators transfer to the CNS. Gene Ther. 9, 1291–1301.
with graded transactivation potential. Nucleic Acids Res. 25, Kafri, T., Van Praag, H., Gage, F.H., and Verma, I.M. (2000).
Lentiviral vectors: regulated gene expression. Mol. Ther. 1, Baum, C., Schambach, A., Bohne, J., and Galla, M. (2006). Ret- rovirus vectors: toward the plentivirus? Mol. Ther. 13, 1050– Kenny, P.A., Enver, T., and Ashworth, A. (2002). Retroviral vectors for establishing tetracycline-regulated gene expression Beard, B.C., Keyser, K.A., Trobridge, G.D., Peterson, L.J., Miller, in an otherwise recalcitrant cell line. BMC Mol. Biol. 3, 13.
D.G., Jacobs, M., Kaul, R., and Kiem, H.P. (2007). Unique in- Kustikova, O.S., Wahlers, A., Kuhlcke, K., Stahle, B., Zander, tegration profiles in a canine model of long-term repopulating A.R., Baum, C., and Fehse, B. (2003). Dose finding with ret- cells transduced with gammaretrovirus, lentivirus, or foamy roviral vectors: correlation of retroviral vector copy numbers virus. Hum. Gene Ther. 18, 423–434.
in single cells with gene transfer efficiency in a cell population.
Belikov, S., Holmqvist, P.H., Astrand, C., and Wrange, O. (2004).
Blood 102, 3934–3937.
Nuclear factor 1 and octamer transcription factor 1 binding Lagrange, T., Kapanidis, A.N., Tang, H., Reinberg, D., and Eb- preset the chromatin structure of the mouse mammary tumor right, R.H. (1998). New core promoter element in RNA poly- virus promoter for hormone induction. J. Biol. Chem. 279, merase II-dependent transcription: sequence-specific DNA binding by transcription factor IIB. Genes Dev. 12, 34–44.
Chtarto, A., Bender, H.U., Hanemann, C.O., Kemp, T., Lehtonen, Lewinski, M.K., Yamashita, M., Emerman, M., Ciuffi, A., E., Levivier, M., Brotchi, J., Velu, T., and Tenenbaum, L. (2003).
Marshall, H., Crawford, G., Collins, F., Shinn, P., Leipzig, J., HEINZ ET AL.
Hannenhalli, S., Berry, C.C., Ecker, J.R., and Bushman, F.D.
Saridey, S.K., Liu, L., Doherty, J.E., Kaja, A., Galvan, D.L., (2006). Retroviral DNA integration: viral and cellular deter- Fletcher, B.S., and Wilson, M.H. (2009). PiggyBac transposon- minants of target-site selection. PLoS Pathog. 2, e60.
based inducible gene expression in vivo after somatic cell gene Loew, R., Vigna, E., Lindemann, D., Naldini, L., and Bujard, H.
transfer. Mol. Ther. 17, 2115–2120.
(2006). Retroviral vectors containing Tet-controlled bidirec- Schambach, A., Wodrich, H., Hildinger, M., Bohne, J., Krausslich, tional transcription units for simultaneous regulation of two H.G., and Baum, C. (2000). Context dependence of different gene activities. J. Mol. Genet. Med. 2, 107–118.
modules for posttranscriptional enhancement of gene expres- Loew, R., Meyer, Y., Kuehlcke, K., Gama-Norton, L., Wirth, D., sion from retroviral vectors. Mol. Ther. 2, 435–445.
Hauser, H., Stein, S., Grez, M., Thornhill, S., Thrasher, A., Toniatti, C., Bujard, H., Cortese, R., and Ciliberto, G. (2004).
Baum, C., and Schambach, A. (2010a). A new PG13-based Gene therapy progress and prospects: transcription regulatory packaging cell line for stable production of clinical-grade self- systems. Gene Ther. 11, 649–657.
inactivating gamma-retroviral vectors using targeted integra- Urlinger, S., Baron, U., Thellmann, M., Hasan, M.T., Bujard, H., tion. Gene Ther. 17, 272–280.
and Hillen, W. (2000). Exploring the sequence space for Loew, R., Heinz, N., Hampf, M., Bujard, H., and Gossen, M.
tetracycline-dependent transcriptional activators: novel mu- (2010b). Improved Tet-responsive promoters with minimized tations yield expanded range and sensitivity. Proc. Natl. Acad.
background expression. BMC Biotechnol. 10, 81 (November Sci. U.S.A. 97, 7963–7968.
VandenDriessche, T., Ivics, Z., Izsva´k, Z., and Chuah, M.K. (2009).
Maetzig, T., Galla, M., Brugman, M.H., Loew, R., Baum, C., and Emerging potential of transposons for gene therapy and gen- Schambach, A. (2010). Mechanisms controlling titer and eration of induced pluripotent stem cells. Blood 114, 1461–1468.
expression of bidirectional lentiviral and gammaretroviral Vigna, E., Cavalieri, S., Ailles, L., Geuna, M., Loew, R., Bujard, vectors. Gene Ther. 17, 400–411.
H., and Naldini, L. (2002). Robust and efficient regulation of Markusic, D., Oude-Elferink, R., Das, A.T., Berkhout, B., and transgene expression in vivo by improved tetracycline- Seppen, J. (2005). Comparison of single regulated lentiviral dependent lentiviral vectors. Mol. Ther. 5, 252–261.
vectors with rtTA expression driven by an autoregulatory Wilson, M.H., Coates, C.J., and George, A.L., Jr. (2007). PiggyBac loop or a constitutive promoter. Nucleic Acids Res. 33, e63.
transposon-mediated gene transfer in human cells. Mol. Ther.
Ma´te´s, L., Chuah, M.K., Belay, E., Jerchow, B., Manoj, N., Acosta- 15, 139–145.
Sanchez, A., Grzela, D.P., Schmitt, A., Becker, K., Matrai, J., Ma, Yant, S.R., Wu, X., Huang, Y., Garrison, B., Burgess, S.M., and Kay, L., Samara-Kuko, E., Gysemans, C., Pryputniewicz, D., Miskey, M.A. (2005). High-resolution genome-wide mapping of trans- C., Fletcher, B., VandenDriessche, T., Ivics, Z., and Izsva´k, Z.
poson integration in mammals. Mol. Cell. Biol. 25, 2085–2094.
(2009). Molecular evolution of a novel hyperactive SleepingBeauty transposase enables robust stable gene transfer in ver-tebrates. Nat. Genet. 41, 753–761.
Address correspondence to: Modlich, U., Bohne, J., Schmidt, M., von Kalle, C., Knoess, S., Dr. Bernhard Schiedlmeier Schambach, A., and Baum, C. (2006). Cell-culture assays re- Experimental Hematology veal the importance of retroviral vector design for insertional Hannover Medical School mutagenesis. Blood 108, 2545–2553.
Carl-Neuberg-Straße 1 Paulus, W., Baur, I., Boyce, F.M., Breakefield, X.O., and Reeves, 30625 Hannover, Germany S.A. (1996). Self-contained, tetracycline-regulated retroviral vector system for gene delivery to mammalian cells. J. Virol.
70, 62–67.
Qu, Z., Thottassery, J.V., Van Ginkel, S., Manuvakhova, M., Westbrook, L., Roland-Lazenby, C., Hays, S., and Kern, F.G.
(2004). Homogeneity and long-term stability of tetracycline- 55743 Idar-Oberstein, Germany regulated gene expression with low basal activity by using thertTA2S-M2 transactivator and insulator-flanked reporter vec- tors. Gene 327, 61–73.
Rahman, L., Bliskovski, V., Kaye, J.K., and Zajac-Kaye, M.
Received for publication May 14, 2010; (2004). Evolutionary conservation of a 2-kb intronic sequence accepted after revision September 3, 2010.
flanking a tissue-specific alternative exon in the PTBP2 gene.
Genomics 83, 76–84.
Published online: December 7, 2010.

Source: http://ozyme.fr/gammes/clo/pdf/clo_tet-one_publi_heinz.pdf