Faster, easier screening for difficult transfections when using the complete control kit
Peter Vaillancourt Katherine A. Felts
We describe the vector pFB-ERV, an MMLV-based replication-defective retroviral vector for delivery of the ecdysone receptor proteins RXR and VgEcR. In stable cell lines infected with an estimated single copy of the receptor cassette, induction ratios of greater than 200-fold are attainable in transient reporter transfection assays. The neo-resistance marker is expressed as the third open reading frame (ORF) in a tricistronic CMV expression cassette (the two receptors comprise the first and second ORFs); cell lines harboring single-copy integrants are resistant to as high as 1 mg/ml G418. The CMV promoter ### is flanked by unique restriction sites and, thus, can be replaced with a cell-type specific promoter of interest.
DNA vector-based systems that allow precise control of gene expression in vivo have become invaluable for the study of gene function in a variety of organisms, particularly when applied to the study of developmental and other biological processes for which the timing or dosage of gene expression is critical to gene function. Such systems have also been successfully used to overexpress toxic or disease-causing genes, to induce gene targeting, and to express antisense RNA. Inducible systems are currently being used by pharmaceutical companies to facilitate screening for inhibitors of clinically relevant biological pathways, and potential applications for gene therapy are being explored.1
Stratagenes complete control inducible system is based on the insect molting hormone ecdysone, which can stimulate transcriptional activation in mammalian cells harboring the ecdysone receptor protein from Drosophila melanogaster.2,3 The ecdysone-inducible system has a number of advantages over alternative systems. Firstly, the lipophilic nature and short in vivo half-life of the ecdysone analog ponasterone A (ponA) allow efficient penetrance into all tissues including brain, resulting in rapid and potent inductions and rapid clearance. Secondly, ecdysteroids are not known, nor are they expected, to affect mammalian physiology in any measurable way. Thirdly, the heterodimeric ponA responsive receptor and receptor DNA recognition element have been genetically altered such that trans-activation of endogenous genes by the ecdysone receptor, or of the ponA-responsive expression cassette by endogenuos transcription factors, is extremely unlikely.
In addition, it has been found that in the absence of inducer the heterodimer remains bound at the promoter in a complex with corepressors and histone deacetylase, and is, thus, tightly repressed until ligand binding, at which time high-level transcriptional activation occurs (i.e., the heterodimer is converted from a tight repressor to a trans-activator). Using the Complete Control plasmid-based system, induction ratios of greater than 1,000-fold have been achieved in both transient transfections and in stable cell lines.3
Using plasmid-based vectors for controlled gene expression is limiting because many cell types that are of academic, industrial, or clinical interest are difficult or virtually impossible to transfect using current transfection methods. In particular, primary human cell lines and lymphocytes are best transduced using viral delivery systems. The most popular and user-friendly of these are the retroviral vectors.4,5 Infection with retroviruses often yields transduction efficiencies close to 100%, and the proviral copy number can be easily controlled by varying the multiplicity of infection (MOI). This latter feature is particularly important for inducible systems for which low basal expression and high induction ratios are affected by copy number. Hence, viral infection (at an optimal MOI) of the target cell should yield a high frequency of clones capable of mediating desirable expression profiles without exhaustive colony screening.
We describe the vector pFB-ERV, an MMLV-based replication-defective retroviral vector for delivery of the ecdysone receptor proteins RXR and VgEcR.
The vector pFB-ERV contains a tricistronic message transcribed from the CMV promoter (Figure 1). The receptor proteins VgEcR and RXR are expressed from the first and second open reading frames (ORF), respectively,
and the neomycin-resistance gene is expressed from the third ORF. Translation of the RXR and Neo ORFs is mediated by the EMCV-IRES (both IRESs are identical). In this context, G418 selection and maintenance of the expression cassette in stably infected cells ensures that the receptor-encoding mRNA is transcribed. The CMV expression cassette was built into a self-inactivating (SIN) vector backbone,6,7 in which the retroviral promoter within the U3 region of the 3 LTR was deleted. In SIN vectors, viral genomic RNA is expressed from the 5 viral LTR in packaging cells; however, upon infection the virus replicates in such a way that the (inactive) 3 U3 promoter sequences are transferred to the 5 LTR, and the proviral 5 promoter is then inactive in infected cells. The rationale for this construction is the following: The CMV promoter is stronger than the MMLV LTR and is persistently active in a wider range of cell types, and inactivation of the 5 LTR obviates potential interference with the CMV promoter; the CMV promoter can be readily replaced with a cell-type promoter of interest using the unique EcoR I and Fse I sites without concern of ubiquitous read-through from the 5 LTR; spurious activation/transcription of 3-flanking endogenous genes from the promoter within the 3 LTR will not occur in the SIN vector; and inactivation of the proviral LTRs protects against mobilization of proviral pFB-ERV derivatives by endogenous retroviral structural proteins.
Vector titer was determined by G418-resistant colony formation. Amphotropic virus was produced by transient transfection using the producer line HW293-A (unpublished data), and viral supernatants were used to infect NIH3T3 cells. In Table 1, the titers for both experiments are on the order of 105 colony forming units (cfu)/ml. Even at high dilutions of supernatant (1:103), which likely give rise to single-copy infected cells in accordance with the Poisson kinetics of viral transduction, colonies are resistant to as high as 1 mg/ml G418 without substantial loss of titer, indicating efficient expression of the Neo gene from the third ORF.
NIH3T3 cells were infected with various dilutions of viral supernatant, and one day following infection the cells were transfected with the ecdysone-inducible reporter vector pEGSH-luc. The following day, cells were induced for 20 hours with ponA, then assayed for luciferase activity. In Figure 2 (at an MOI of 1.0), a strong induction is achieved, which is reproducible in separate infected populations.
NIH3T3 cells were infected with pFB-ERV supernatants and selected with 600 g/ml G418. Resistant colonies were picked and expanded. In an initial screen, 24 colonies were transfected with reporter vector, and induced with 10 M ponA or an equivalent volume of vehicle. All 24 of the infected lines showed a pon A- dependent induction to some degree (Figure 3). One clone, A610-20 (clone #20, Figure 3) gave an induction of greater than 200-fold in the initial screen. A retest of this cell line gave an induction of approximately 250-fold (Figure 3B ). This clone was produced by infection at an MOI of 0.1, and, therefore, the colony theoretically has a single integrated copy of the receptor expression cassette.
The 200-fold induction seen by transient reporter transfection of the A610-20 receptor line is likely to be an under-representation of the induction profile to be expected in double-stable lines made by selecting stably integrated inducible vectors. This is due to the tight repression that occurs for chromosomally integrated ecdysone-responsive promoters, compared with the relatively high background normally seen in transient transfection assays in which the reporter is free in the nucleus. The stable cell line ER-CHO, which was made using the receptor-expressing plasmid pERV3, shows at best 50-fold induction ratios by transient reporter transfection but the derivative double-stable line consistently gives induction ratios of 700- greater than 1,000-fold.3 Accordingly, we expect that the A610-20 receptor line will give rise to double-stable lines with induction ratios well in excess of 1,000-fold.
Retroviral delivery of the ecdysone receptors should significantly decrease the time and labor required to screen for stable clones that express the receptors at optimal levels, particularly for cell types that are normally difficult to transfect. However, for difficult-to-transfect cell types, delivery of the receptors with the pFB-ERV virus only solves half the problem. Currently, the ecdysone-inducible vectors need to be delivered by transfection. At this writing, several ecdysone-inducible retroviral constructions are being evaluated for efficient delivery of inducible expression cassettes that allow fine control of expression and high-level induction of the gene of interest. The use of a two-virus system for delivery of the edysone-inducible expression system will further expand the range of cell types in which ecdysone-regulated expression can be achieved, while further reducing the time, cost, and labor of screening for clones that show optimal expression characteristics.
Schockett, P.E. and Schatz, D.G. (1996) Proc. Natl. Acad. Sci. 93: 5173-5176.
No, D., et. al. (1996) Proc. Natl. Acad. Sci. 93: 3346-3351.
Wyborski, D. and Vaillancourt, P. (1999) Strategies in Mol. Biol. 12(1): 1-4.
Miller, A.D. (1997) In Retroviruses (Eds., Coffin, J.M., Hughes, S.H., and Varmus, H.E.) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. pp. 437-473.
Felts, K.A., et. al. (1999) Strategies in Molec. Biol. 12(2): 74-77.
Yu, S.-F., et. al. (1986) Proc. Natl. Acad. Sci. 83: 3194-3198.
Yee, J.-K. (1987) Proc. Natl. Acad. Sci. 84: 5197-5201.