Retroviral Vectors for the Transduction of Autoregulated, Bidirectional ...

ARTICLE

doi:10.1006/mthe.2001.0480, available online at http://www.idealibrary.com on IDEAL

Retroviral Vectors for the Transduction of Autoregulated, Bidirectional Expression Cassettes Jacqueline Unsinger, Andrea Kröger, Hansjörg Hauser,* and Dagmar Wirth Department of Gene Regulation and Differentiation, GBF-National Research Center for Biotechnology, D-38124 Braunschweig, Germany *To whom correspondence and reprint requests should be addressed. Fax: ++ 531-6181-262. E-mail: [email protected].

Regulated transgene expression is increasingly used in research but is also needed for certain therapies. Regulatory systems are usually composed of two expression units, one bearing the gene of interest under control of a regulatable promoter and the other, a constitutively expressed transactivator that modulates the activity of the regulatable promoter. Because the cotransfer of two independent elements is not efficient in primary cells, single transduction step vectors conferring regulatable gene expression cassettes would be helpful. We have developed retroviral vectors containing an autoregulatory bidirectional expression cassette that encodes all components necessary for regulated expression of a gene of interest. The influence of the orientation of the reporter gene with respect to the viral long terminal repeat (LTR) and the effect of transcriptionally inactive LTRs were investigated using mouse leukemia virus (MLV) and self-inactivating (SIN)-based retroviral vectors. Strict regulation was observed when the reporter was inserted in antisense orientation with respect to the LTR, whereas a sense arrangement of the reporter resulted in a loss of regulation capacity. Expression and regulation of the antisense-orientated reporter gene were homogenous in infected cell pools and investigated cell clones. Long-term observations of infected cells over a period of 30 passages revealed stable expression and regulation. These autoregulated, bidirectional retroviral vectors combine the advantages of singlestep transduction with strict regulation of the gene of interest in the infected target cells. Key words: gene therapy, LTRs, retroviral vectors, tet-off system, stable expression, regulated expression

INTRODUCTION Strictly regulated expression of genes in mammalian cells is currently achieved using nonmammalian regulatable expression systems. The prototype of these systems is the tetracycline (tet) dependent system [1]. It consists of two components: a transactivating protein (tTA or rtTA), the activity of which can be modulated by the addition of tet, and a promoter which is strictly dependent on the binding of the transactivating protein. Although regulated gene expression can be achieved in cell culture and in transgenic animals when using this system, regulated expression is still a problem for gene therapy purposes. This is due to two major limitations. First, the regulation potential of the system is mainly influenced by the site of chromosomal integration, which frequently reduces the regulation capacity of the integrated cassette. Only a limited number of integration sites support strict regulation. Thus, only a small percentage of cells provide the desired expression properties. Second, both the tTA gene and the gene of interest have to be transferred. Although the subsequent or simultaneous transfer of two vectors is feasible for many

484

permanent cell lines, limitations in the transduction efficiency make such approaches unpractical for primary cells and even more for in vivo treatments. For gene therapy applications, single-step transduction systems leading to homogeneous expression would be desirable. Autoregulatory systems represent a tool for single-step transduction. Moreover, their use should overcome the toxic side effects (like transcriptional squelching) mediated by high levels of transactivators like tTA [2–4]. In autoregulatory expression systems, the tTA gene is controlled by the tTA-dependent promoter, resulting in a positive feedback loop, and transactivator effects can be restricted to the time of induction [5–8]. Most of the autoregulated cassettes based on the tet system have so far been evaluated after nonviral gene transfer. Regulation factors achieved with these systems were usually in the range of 2 to 300 [5,6,8]. Retroviral transfer of a unidirectional autoregulated expression cassette resulted in a low regulation capacity of the infected population, and intensive screening and selection was required to obtain a homogeneously regulated cell

MOLECULAR THERAPY Vol. 4, No. 5, November 2001 Copyright © The American Society of Gene Therapy 1525-0016/01 $35.00


ARTICLE

A

positive selection marker, the thymidine kinase (T) as a negative selection marker, and enhanced green fluorescent protein (eGFP,G) for positive/negative selection and monitoring of expression. eGFP was chosen B because its analysis by FACS allows the determination of single-cell expression levels in whole populations. We integrated the autoregulated cassette into an MLVbased retroviral vector. In LTR-THTG the reporter gene HTG is inserted in an antisense orientation with respect to the LTR. In LTRGTHT the regulatory casFIG. 1. Regulation pattern of LTR-THTG and LTR-GTHT infected cells. (A) Schematic presentation of MLV-based retroviral vec- sette is arranged with the tors carrying the auto-bi expression cassette. The transactivator (tTA) encoding RNA is initiated from the internal tet-regulat- reporter in the same orienable bidirectional promoter. The second mRNA initiated from this promoter encodes for a triple fusion gene HTG. In the case tation as the viral LTR (Fig. of LTR-THTG, the reporter HTG is inserted in reverse orientation with respect to the retroviral LTR. A sense orientation of HTG and viral LTR is given in the vector LTR-GTHT. Bidirectional arrow, bidirectional promoter PbitTA; LTR, retroviral long terminal 1A). Both retroviral vectors repeats. (B) GFP expression analysis of retrovirally transduced auto-bi cassettes. NIH3T3 cells were infected with LTR-THTG were stably transfected into and LTR-GTHT. Pools of hygromycin-resistant cells were cultivated in medium either without or in the presence of Dox for 3 the packaging cell line d. Subsequently, GFP expression was determined by FACS. Thick line, GFP expression in the induced state; thin line, GFP PT67. To assess the regulaexpression in the repressed state; dotted line, mock transduced cells. tion capacity, we infected NIH3T3 cells with an MOI population [9]. However, such screenings are not feasible < 0.01 to ensure single-copy integrants. Pools of stably infected cells were isolated by hygromycin selection and for most therapeutic approaches. analyzed for GFP expression by FACS at repressed (+ Dox) Here, we have evaluated for the first time the potential of an autoregulatory, bidirectional cassette (auto-bi cas- or induced (– Dox) conditions (Fig. 1B). When the reporter gene HTG was positioned in reverse sette) after single-step transduction using retroviral vectors. orientation to the retroviral LTR, we found tight and The optimal orientation of the auto-bi cassette within an homogenous expression regulation of the reporter in the murine leukemia virus (MLV)-based retroviral vector with infected cell pool (Fig. 1B). If the reporter was arranged in respect to the viral long terminal repeat (LTR) was examsense with respect to the LTR, similar levels of GFP expresined. Also the influence of either a transcriptionally inacsion were obtained in the induced state. However, exprestive (self-inactivating (SIN) vector) or an active retroviral sion was not significantly repressed in the presence of Dox. LTR on the regulation capacity of the inserted bidirectional This demonstrates that the orientation of the inserted cassette was investigated. The results demonstrate that the best arrangement for regulated gene expression is the com- genes is crucial for regulation. To evaluate the influence of the enhancer sequence in bination of an active LTR with a reporter gene inserted in the retroviral LTR on reporter expression, we used a SIN the opposite orientation with respect to this LTR. retroviral vector [11] for transduction of the auto-bi cassette. The enhancer deletion leads to the inactivation of RESULTS the retroviral 5’ LTR and thus eliminates potential cis effects on the bidirectional promoter. For optimal regulaWe constructed autoregulated retroviral expression vection the reporter gene HTG was inserted in antisense oritors based on the bidirectional promoter [10] for regulated gene expression (Fig. 1A). The autoregulated, bicistronic entation with respect to the retroviral LTR (SIN-THTG). Regulation properties of the SIN- and the MLV-based vec(auto-bi) cassette contains one mRNA which is initiated by the bidirectional promoter and which encodes the trans- tor were compared (Fig. 2). Therefore, NIH3T3 cells were activator protein (tTA). The second mRNA results in infected with both virus types, hygromycin-resistant cell expression of the reporter and selection gene HTG. The pools and cell clones were isolated, and GFP expression was measured by FACS at repressed (+ Dox) and induced triple fusion gene HTG encodes a single protein with activities of the hygromycin-B-phosphotransferase (H) as a (– Dox) conditions, respectively.

MOLECULAR THERAPY Vol. 4, No. 5, November 2001 Copyright © The American Society of Gene Therapy

485

ARTICLE


A

FIG. 2. Regulation pattern of LTR-THTG and SIN-THTG infected cells. NIH3T3 cells were infected with LTR-THTG and SINTHTG. Hygromycin-resistant cell pools (A) and clones (B) were cultivated in the absence or presence of Dox for 3 d. GFP expression was measured by FACS. Thick line, GFP expression in the induced state; thin line, GFP expression in the repressed state; dotted line, mock transduced cells. Left column, LTR-THTG infected pools and cells; right column, SIN-THTG infected pools and cells.

B

SIN-vector, whereas the background expression (uninduced state) was lower in infectants with the full-length LTRs. Neither the expression level nor the regulation capacity in clones infected with the retroviral auto-bi vectors showed strong clone-to-clone variation. Compared with most constitutive [1,12,13] and autoregulatory systems [9], the retroviral infected bidirectional autoregulatory cassette showed more uniform and therefore more predictable results. To evaluate the repression pattern of LTR-THTG and SINTHTG infected cells, we cultivated both pools and two cell clones each in medium supplemented with increasing concentrations of Dox. Subsequently, eGFP expression was measured. Cell pools and cell clones demonstrated a very similar repression pattern (Fig. 3). A slight decrease of eGFPexpressing cells (to about 80% positive cells) was observed by increasing the concentration of Dox up to 0.01 ng/ml medium. At a concentration of 0.1 ng/ml a strong decrease in GFPexpressing cells down to 10% to 30% of cells was uniform in all samples. At 1 ng/ml Dox total repression was reached, supporting the consistent behavior of different clones and mixtures. We determined the stability of expression and regulation after retroviral transfer after long-term cultivation of one cell clone and one cell pool of either SIN-THTG or LTR-THTG infected cells. For this purpose we grew cells over 30 passages

Tight and homogeneous regulation was found in cells infected with either LTR-THTG or SIN-THTG. Homogenicity was confirmed by analysis of individual clones (five representative clones for each vector are shown in Fig. 2B). Nevertheless, little differences in expression level and regulation were observed for LTRTHTG and SIN-THTG in cell clones and in pools of infected cells (Figs. 2A and 2B). The level of expression in the induced state was higher in the infectants with the

486



ARTICLE

We investigated the influence of the reporter gene orientation with respect to the retroviral LTR by inserting the autoregulatory cassette in both orientations into an MLVbased retroviral vector. Following infection of NIH3T3 cells, the regulation potential of both orientations was evaluated in dependence of Dox by FACS analysis. If the reporter HTG is arranged in antisense orientation with respect to the LTR (Fig. 1A, LTR-THTG), strict eGFP regulation was observed. Even pools of infected cells were found to be homogeneously regulated. In the opposite arrangement, where the reporter is expressed in sense with respect to the viral LTR (in LTR-GTHT) slightly increased GFP expression levels were detected in the FIG. 3. Dose-response curve of GFP from cell pools and cell clones in dependence of Dox. induced state in comparison with the reverse Pools of LTR-THTG and SIN-THTG infected cells and two cell clones from each pool were culorientation. However, for the latter constellativated in the presence of the indicated concentrations of Dox for 3 d. Medium was changed daily to maintain the exact level of Dox. The percentage of GFP-expressing cells was calcu- tion we observed a reduced regulation capacity. GFP expression was scarcely repressible by lated after FACS analysis. the addition of Dox. This low performance of LTR-GTHT might be due to three effects. First, the active 5 LTR may lead to the expression of genomic in Dox containing medium. Dox was removed at passages viral RNA that encodes for all in-sense inserted genes. 6, 13, 22, and 30. At 7 days after removal the percentage of GFP-expressing cells was determined (Fig. 4). Independent of Possibly, this leads to the accidental translation of the reporter gene due to splicing events and thereby to tTAcultivation length under repressing conditions, in every cell population 40% to 60% of cells could be re-induced to independent expression. Second, an activating influence express GFP by removal of Dox. The percentage of express- of the retroviral LTR on the inserted tTA-dependent proing cells could be increased up to 90% if selection pressure moter cannot be excluded. The LTR promoter/enhancer elements could interfere with the transcriptional reguwas exerted simultaneously (Fig. 4). In a number of experiments we also investigated the lation of the tet-dependent promoter and might conexpression behavior of transduced cells during long-term tribute to the elevated expression levels in the repressed cultivation (30 passages or even longer) under induced state. However, such an activating influence should be conditions. No significant changes in regulation behavior independent from the orientation of the integrated of the induced cell clones and pools were observed over genes. Third, genomic viral RNA activated from the 5 time (data not shown). Although the growth rate of cells LTR might act as an antisense RNA and hybridize with the reversely orientated RNA which is initiated from the continuously expressing tTA was slightly reduced with respect to untransduced control, there were no obvious tTA-dependent promoter. This would result in an antitoxic side effects leading to cell death. These data confirm sense effect: the reduction of the translatable RNA and the stability of expression and regulation of the retrovirally thus reduced levels of the inversely orientated gene transduced auto-bi cassettes even after long periods of product. The artificial tet promoter belongs to the strongest promoters [14] and is probably even stronger repression. than the viral LTR. Thus the concentration of the reporter RNA initiated from the tTA-promoter reporter DISCUSSION RNA should be much higher than the concentration of The aim of this study was to establish a retroviral vector the genomic viral RNA. In consequence only a minor for regulated gene expression. This vector should provide part of the transcribed reporter RNA would be hybridized optimal regulation properties in infected cells. We focused and thus silenced by the antisense effect. Therefore, GFP on auto-bi expression cassettes based on the tet system expression is detectable. In contrast, in the repressed and created retroviral vectors encoding all components state where expression from the tTA-dependent pronecessary for regulated gene expression. For reduction of moter is low, the reporter RNA should be totally blocked the reported toxic side effect, the expression of the trans- by the genomic RNA. Reporter expression should not be activator itself was regulated by placing the tTA under the detectable. This would explain why the use of an autobi expression cassette with reporter genes orientated in control of the tTA-dependent promoter. As in all autoregan antisense manner shows superior regulation properulatory systems, high transactivator concentrations are ties by reduction of background expression. restricted to the time of gene induction.


487

ARTICLE


FIG. 4. GFP induction of cell pools and clones after long-term cultivation under repressing conditions (+ Dox). One hygromycin-resistant cell population and one cell clone of NIH3T3 cells infected with LTR-THTG and SIN-THTG, respectively, were cultivated for 30 passages in Dox containing medium (black bars). At the indicated passages the repressed cultures were washed, split, and cultivated in medium either with (white bars) and without (gray bars) selection pressure. The percentage of GFP-positive cells was calculated following FACS analysis. The bars at passage 2 indicate the expression in the induced and repressed states directly following hygromycin selection. GFP expression in the induced state at passage 2 was set to 100 percent and all other values were normalized to this.

It would be expected that the effect of the antisense RNA is less pronounced for the retroviral SIN vector. Because the retroviral enhancer elements are deleted transcriptional activity is strongly decreased. In fact, this is observed in cells infected with SIN-THTG. The GFP expression levels for both induced and repressed states are slightly increased if compared with LTR-THTG. Thus, loss of reduction of SIN-LTR derived antisense RNA results in an increase of reporter RNA levels, both in the induced and in the repressed state. The transactivator used in our system corresponds to the original tet-off transactivator [1]. This transactivator still comprises all domains that are responsible for interactions with cellular transcription factors and might lead to the so-called “transcriptional squelching.” This effect, the reduction of the cellular transcription level, is attributed to the fact that the potent transactivator recruits and thereby depletes cellular transcription factors. Although it was reported that toxic side effects were observed for an autoregulatory system [15], there are several examples for the successful use of autoregulatory systems in vitro and in vivo [5–8]. For the vectors used here, we observed no toxic effects in NIH3T3 cells leading to increased cell death, although the growth rate of the transduced cells slowed slightly, which might be a consequence of transactivator expression. Recently, the toxicity of the transactivator could be reduced by specific

488

deletion of domains interacting with cellular transcription factors [16]. Additionally, the transactivating properties of the reverse acting transactivator (rtTA) were significantly improved by enhanced binding activity to the tet-operator sequences and by reduced squelching activity [17]. The use of these better-tolerated transactivator mutants should further improve the versatility of this autoregulatory expression system. Thus, retroviral transfer of the auto-bi expression cassettes represents a novel tool for homogeneously regulatable expression, making further selection or screening superfluous. This approach offers new perspectives for achieving regulated expression in gene therapy purposes.

MATERIALS

AND

METHODS

Vector construction. Hygtk, a fusion gene of the selection markers hygromycin-B-phosphotransferase and thymidine kinase, was PCR amplified from pGET [18] and inserted via BamHI/XhoI into pEGFP-N1 (Clontech), 5 in-frame to eGFP, resulting in a triple fusion gene of hygromycin-B-phosphotransferase, thymidine kinase, and eGFP (pHGT). Functional assays revealed full activity of all components of the triple fusion protein. The tTA gene was isolated via EcoRI/BamHI out of pUHD15-1 [1] and inserted into the correspondingly restricted pRBT2, a vector containing the bidirectional promoter from pBI-4 [10], leading to pRBTtTA. HTG was inserted as a PmeI fragment into the SwaI restricted pRBTtTA, generating pTHTG. The auto-bi cassette of pTHTG was integrated via NotI into pSINN (based on pSIR (Clontech) with an expanded MCS) resulting in SIN-THTG.



LTR-THTG was constructed using three fragments: pMSCVneo [19] was restricted with NotI/HindIII and ligated with the NsiI/HindIII fragment of pTHTG and the NsiI/NotI fragment of pSIN-THTG leading to pLTR-THTG. The autoregulated cassette was isolated by restricting pTHTG with BsrGI blunting and subsequently NotI restriction. The 4625-bp fragment was inserted into the NotI/HindII opened pMSCVneo (LTR-GTHT). Cell culture and gene transduction. Murine fibroblast NIH3T3 cells (ATCC CRL-1658) and the murine retroviral packaging cell line PT67 (Clontech) were maintained in Dulbecco’s modified Eagle’s medium (DMEM). Medium was supplemented with 10% fetal calf serum, 2 mM L-glutamine, and penicillin (10 U/ml) and streptomycin (100 g/ml). For selection of NIH3T3 and PT67 cells, 200 U/ml hygromycin B was used. For retroviral infection, PT67 cells were transfected with 10 g of pLTRTHTG, pSIN-THTG, or pLTR-GTHT by calcium phosphate coprecipitation according to Spitzer and colleagues [20] and stable retroviral producer was selected by 200 units per ml of hygromycin B. NIH3T3 cells were infected with the supernatant of stably transfected cells (MOI < 0.01) in the presence of Polybrene (8 g/ml) [20] and selected for hygromycin resistance. Reporter gene expression. Cells were grown for 3 days on six-well plates with or without Dox (2 g/ml) for the indicated times before reporter gene analysis. Flow cytometry was carried out as described [20].

ACKNOWLEDGMENTS This work was supported by Grant 01GE9905 from the Federal Ministry of Education and Research (BMBF) of Germany. RECEIVED FOR PUBLICATION JANUARY 30; ACCEPTED SEPTEMBER 7, 2001.

REFERENCES 1. Gossen, M., and Bujard, H. (1992). Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc. Natl. Acad. Sci. USA 89: 5547–5551. 2. Sadowski, I., Ma, J., Triezenberg, S., and Ptashne, M. (1988). GAL4-VP16 is an unusually potent transcriptional activator. Nature 335: 563–564. 3. Triezenberg, S. J., Kingsbury, R. C., and McKnight, S. L. (1988). Functional dissection of VP16, the trans-activator of herpes simplex virus immediate early gene expression. Genes Dev. 2: 718–729. 4. Triezenberg, S. J., LaMarco, K. L., and McKnight, S. L. (1998). Evidence of DNA:protein


ARTICLE

interactions that mediate HSV-1 immediate early gene activation by VP16. Genes Dev. 2: 730–742. 5. Strathdee, C. A., McLeod, M. R., and Hall, J. R. (1999). Efficient control of tetracyclineresponsive gene expression from an autoregulated bi-directional expression vector. Gene 229: 21–29. 6. Mohammadi, S., and Hawkins, R. E. (1998). Efficient transgene regulation from a single tetracycline-controlled positive feedback regulatory system. Gene Ther. 5: 76–84. 7. Shockett, P., Difilippantonio, M., Hellman, N., and Schatz, D. (1995). A modified tetracycline-regulated system provides autoregulatory, inducible gene expression in cultured cells and transgenic mice. Proc. Natl. Acad. Sci. USA 92: 6522–6526. 8. Gould, D. J., et al. (2000). A novel doxycycline inducible autoregulatory plasmid which displays on/off regulation suited to gene therapy applications. Gene Ther. 7: 2061–2070. 9. Hofmann, A., Nolan, G. P., and Blau, H. M. (1996). Rapid retroviral delivery of tetracycline-inducible genes in a single autoregulatory cassette. Proc. Natl. Acad. Sci. USA 93: 5185–5190. 10. Baron, U., Freundlieb, S., Gossen, M., and Bujard, H. (1995). Co-regulation of two gene activities by tetracycline via a bidirectional promoter. Nucleic Acids Res. 23: 3605–3606. 11. Yu, S. F., et al. (1986). Self-inactivating retroviral vectors designed for transfer of whole genes into mammalian cells. Proc. Natl. Acad. Sci. USA 83: 3194–3198. 12. Fussenegger, M., Moser, S., Mazur, X., and Bailey, E. (1997). Autoregulated multicistronic expression vectors provide one-step cloning of regulated product gene expression in mammalian cells. Biotechnol. Prog. 13: 733–740. 13. Pitzer, C., Schindowski, K., Pomer, S., Wirth, T., and Zöller, M. (1999). In vivo manipulation of interleukin-2 expression by a retroviral tetracycline (tet)-regulated system. Cancer Gene Ther. 6: 139–146. 14. Liang, X., Hartikka, J., Sukhu, L., Manthorpe, M., and Hobart, P. (1996). Novel, high expressing and antibiotic-controlled plasmid vectors designed for use in gene therapy. Gene Ther. 3: 350–356. 15. Gallia, G. L., and Khalili, K. (1998). Evaluation of an autoregulatory tetracycline regulated system. Oncogene 16: 1879–1884. 16. Baron, U., Gossen, M., and Bujard, H. (1997). Tetracycline-controlled transcription in eukaryotes: novel transactivators with graded transactivation potential. Nucleic Acids Res. 25: 2723–2729. 17. Urlinger, S., et al. (2000). Exploring the sequence space for tetracycline-dependent transcriptional activators: Novel mutations yield expanded range and sensitivity. Proc. Natl. Acad. Sci. USA 97: 7963–7968. 18. Karreman, S., Hauser, H., and Karreman, C. (1996). On the use of double FLP recognition targets (FRTs) in the LTR of retroviruses for the construction of high producer cell lines. Nucleic Acids Res. 24: 1616–1624. 19. Hawley, R. G., Lieu, F. H. L., Fong, A. Z. C., and Hawley, T. S. (1994). Versatile retroviral vectors for potential use in gene therapy. Gene Ther. 1: 136–138. 20. Spitzer, D., Hauser, H., and Wirth, D. (1999). Complement-protected amphotropic retroviruses from murine packaging cells. Hum. Gene Ther. 10: 1893–1902.

489