- Email: [email protected]
A Regulated, NFκB-Assisted Import of Plasmid DNA into Mammalian Cell Nuclei
doi:10.1006/mthe.2001.0312, available online at http://www.idealibrary.com on IDEAL
ARTICLE
A Regulated, NFB-Assisted Import of Plasmid DNA into Mammalian Cell Nuclei Adi Mesika, Irina Grigoreva, Muriel Zohar,* and Ziv Reich1 Department of Biological Chemistry, Weizmann Institute of Science, Rehovot 76100, Israel *Research Division, Bio-Technology General, Rehovot 76100, Israel Received for publication January 2, 2001; accepted in revised form March 15, 2001; published online April 16, 2001
The success of synthetic DNA delivery systems in human gene therapy will be enhanced by increasing transfection efficiencies and by providing tighter control over targeting of the DNA into the nucleus. Here, we used DNA vectors that contain repetitive binding sites for the inducible transcription factor NFB, which is transported into the nucleus by the nuclear import machinery. Nuclear entry of the modified vectors was augmented 12-fold and was associated with corresponding increase in gene expression. Depending on their position, the binding sites could also function as transcriptional enhancers, increasing gene expression levels up to an additional 19-fold. Notably, nuclear targeting of the DNA and transgene transcription could both be regulated by exogenous stimulators which modulate the intracellular distribution of NFB. The approach provides a framework for the controlled targeting of constitutive or transcriptionally regulated synthetic vectors into mammalian cell nuclei. Key Words: synthetic gene therapy; nuclear pore complex; nuclear localization signal; nuclear factor B; B sites; tumor necrosis factor-␣.
INTRODUCTION Synthetic (nonviral) gene delivery systems are promising tools for gene therapy and DNA vaccination applications. Compared to viral-based systems, they possess several advantages, including excellent safety profiles, an essentially unlimited DNA carrying capacity, ease of production, and simplified QA/QC procedures (1– 4). However, transfection efficiency by these methods is unacceptably low (1–7), largely due to the inability of the DNA to effectively translocate through the nuclear pore complexes (NPCs) (8 –11). NPCs, which operate as selective conduits for nucleocytoplasmic exchange in eucaryotes, support two modes of transport. Particles smaller than 8 –9 nm (M r ⬃40 – 60 kDa) may pass by diffusion. Larger particles, on the other hand, are ushered selectively by soluble receptors of the karyopherin/importin  family, which recognize specific nuclear import (NLS) or export (NES) signal peptides displayed by the cargo (12–14). Under physiological ionic conditions, a typical plectonemically supercoiled DNA (fractional linking number || ⬃0.05) would have a diameter of ⬃10 nm (15–17). Smaller values, representing more relaxed plasmid con-
1 To whom correspondence and reprint requests should be addressed at the Department of Biological Chemistry, Weizmann Institute of Science, Rehovot 76100, Israel. Fax: ⫹972 8 947-4553. E-mail: [email protected].
MOLECULAR THERAPY Vol. 3, No. 5, May 2001, Part 1 of 2 Parts Copyright © The American Society of Gene Therapy 1525-0016/01 $35.00
figurations, will result in larger diameters. Passive diffusion of DNA plasmids through the NPC channels is therefore highly unlikely. The use of DNA as a pharmaceutical tool requires that control be provided over its targeting to the cell nucleus where it is transcribed, as well as over the timing and levels of expression of its encoded gene(s). While substantial progress in transcriptionally regulated vectors has been made, control over nuclear targeting is usually absent in current DNA delivery systems, synthetic or viral. Here, we have used a piggyback approach in which plasmid DNA was modified to include binding sites for the rapid response transcription factor nuclear factor B (NFB), allowing the latter to bind the DNA in the cytoplasm and transport it to the nucleus through the protein nuclear import machinery. Depending on their position relative to the transgene, the binding sites could also act as transcriptional enhancers, increasing further gene expression levels. Since nuclear translocation of NFB is an inducible process, nuclear uptake of plasmid DNA and gene transcription could both be regulated by exogenous stimulators that modulate the intracellular distribution of the protein. A system is therefore described for the targeted delivery of transcriptionally regulatable vectors into mammalian cell nuclei. When refined to include tissuespecific targeting elements, this system could be useful in therapeutic applications relying on gene transfer technologies.
653
ARTICLE MATERIALS
AND
METHODS
Plasmids. A fragment consisting of five tandem repeats of the Ig B motif, 5⬘-GGGGACTTTCC-3⬘, was amplified by PCR from pNFB-Luc (Stratagene) and cloned (BamHI) in pGL3 (Promega) downstream of the SV40 enhancer. To enable fluorescence labeling of pGL3, a segment containing peptide nucleic acid (PNA) recognition sequences was excised (BsaI/BglII) from pGeneGrip (Gene Therapy Systems) and cloned in the MCS region (BglII/SmaI) of the control and modified plasmids. Modifications were confirmed by double-stranded sequencing. Plasmid DNA was purified on Qiagen maxiprep columns, cleaned with phenol/chloroform, ethanol precipitated, and stored frozen at ⬃1 mg/ml. Transfections and luciferase activity measurements. Cells were grown in DMEM containing 10% FCS and antibiotics. Cells were serum-starved for 6 h before transfection. Transfections were made for 2 h, using SuperFect (Qiagen), with 0.25–1.0 g DNA, as specified in the figure legends. Transfections made with calcium phosphate gave similar results. Induction of NFB was by rhTNF-␣ (20 ng/ml; Pepro Tech EC) or TPA (100 ng/ml; Sigma). Unless noted otherwise, induction took place posttransfection. Luciferase activity was measured with the Luciferase Assay System (Promega) and was normalized to total protein (Pierce) or -galactosidase activity (Promega). Data shown represent the means derived from at least three independent measurements; SEM values were calculated using compound quantity formulation. DNA labeling and fluorescence microscopy. Labeling of plasmid DNA was made by site-specific hybridization to rhodamine-labeled PNA clamps (Gene Therapy Systems). PNA binding was done according to the manufacturer’s recommendations. Free PNA was removed by ethanol precipitation. Recovery of plasmids and PNA binding were determined spectrophotometrically and by gel electrophoresis, respectively. Fluorescence and DIC images were acquired simultaneously on fixed cells (4% paraformaldehyde/5% sucrose, in PBS) with an Olympus Fluoview 200 confocal laser scanning microscope (BX50WI-based), using a 60⫻ PlanApo oil immersion objective (NA 1.4). Data processing was performed using the NIH Image software. Figures shown are representative of ⬃200 images acquired from three independent experiments. Indirect immunofluorescence. Cells were fixed in PBS containing 4% paraformaldehyde (EM grade; Electron Microscopy Sciences) and 5% sucrose and permeabilized with 0.5% Triton X-100 in PBS. Primary antibodies (rabbit anti-hp50 and anti-hp65) were purchased from Santa Cruz Biotechnology. Secondary antibody was Cy3-conjugated goat anti-rabbit F(ab⬘)2 (minimal cross-reaction to human serum proteins) and was purchased from Jackson ImmunoResearch Laboratories. Images were acquired with a Nikon Eclipse TE300 microscope (60⫻ PlanApo oil immersion objective (NA 1.4)), equipped with high-resolution Hamamatsu Orca digital CCD camera.
RESULTS The transfer approach used in this study relies on the binding of endogenous NLS-proteins to specific DNA sequence elements within the vector (18 –20). As a mediator, we chose NFB (21). This choice was motivated by several factors. Binding affinity of NFB members to their DNA-binding sites (B sites) is high (K D ⬃10⫺10–10⫺13 M) and their translocation into the nucleus is rapid (minutes). The proteins are ubiquitously expressed, allowing various cell lines to be tested. Importantly, nuclear import of NFB is an inducible process; in the absence of stimulus, the proteins are predominantly cytoplasmic (22–25). Nuclear transfer of target DNA could therefore be controlled by extracellular addition of NFB activators. The structure of the two plasmids used in this study is shown in Fig. 1. The pNFB-Luc vector contains five direct repeats of the Ig B motif, 5⬘-GGGGACTTTCC-3⬘. This motif binds members of the NFB/Rel family (p49, p50,
654
FIG. 1. Plasmids used in this study. Commercially available pNFB-Luc was used without modifications. B-pGL3 is based on pGL3 and includes 5 B sites, cloned from pNFB-Luc, and 10 PNA binding sites used for site-specific fluorescence labeling. The PNA sites were also included in the control pGL3 plasmid.
and p65) with picomolar affinities (26, 27). The B sites are fused to a minimal promoter and are essential for transgene expression. The other plasmid, B-pGL3, has a well-developed constitutive expression cassette based on SV40 promoter and enhancer elements. In this plasmid, the B sites were cloned downstream of the SV40 enhancer such that they do not participate directly in the transcriptional regulation of the reporter gene. B-pGL3 also contains 10 PNA binding sequences that enable sitespecific labeling of the DNA. The PNA sites were also cloned into the parental plasmid, pGL3, which served as control. All human cell lines used in the study were responsive to known NFB stimuli, such as tumor necrosis factor ␣ (TNF-␣) and phorbol esters, as demonstrated by their ability to drive luciferase expression in pNFB-Luc-transfected cells (not shown and Fig. 4B) and from immunofluorescence microscopy studies, which showed stimulation-dependent nuclear accumulation of the NFB proteins p50 and p65 (not shown). The DNA was introduced into the cells chemically to faithfully mimic conditions relevant for DNA transfer in vivo. The transfection efficiencies of pGL3 and B-pGL3 were assayed using luciferase activity as a functional measure (Fig. 2). Serum-starved HeLa, Hek-293, Hep G2, and U373 cells were transfected with either plasmid in the presence or absence of stimulus and were harvested for analysis 18 h later. Measurements made on TNF-␣-activated cells revealed that the transfection activity of B-pGL3 was substantially higher, reaching up to 35 times that of pGL3. This effect was not exclusive to TNF-␣. Addition of the phorbol ester TPA, another activator of NFB, led to similar results (not shown). The two plasmids also differed in their responsiveness to stimulation. Application of TNF-␣ to B-pGL3-transfected cells resulted in a significant increase in gene activity. In contrast, only a minor effect could be seen upon stimulation of cells transfected with pGL3. The small increase observed for pGL3 may be due to the presence of an intrinsic B site in its SV40 enhancer region (nt 2332–2403). Stimulation of cells transfected with pCDNA3, which does not contain NFB binding elements, had no effect on gene activity (not shown). Interestingly, B-pGL3 performed better under MOLECULAR THERAPY Vol. 3, No. 5, May 2001, Part 1 of 2 Parts Copyright © The American Society of Gene Therapy
ARTICLE
FIG. 2. Transfection activity of pGL3 and B-pGL3 (0.25 g/105 cells) determined 18 h posttransfection. Cells were HeLa (A), Hek-293 (B), Hep G2 (C), and U373 (D). Data are presented as a fold increase in luciferase activity determined relative to unstimulated cells transfected with pGL3.
nonstimulatory conditions as well, with increases now ranging from 2.6- to 5.8-fold, depending on cell type. We interpret this enhancement in basal activity as a consequence of low-level migration of NFB molecules into the nucleus. This interpretation is consistent with immunofluorescence microscopy data, showing a weak, but significant p65 and p50 fluorescence in the nuclei of unstimulated cells (not shown and see, e.g., (22, 23)). The nuclear import of the pGL3 and B-pGL3 plasmids was studied directly by confocal fluorescence microscopy. Labeling of plasmids was done by site-specific hybridization to rhodamine-conjugated PNA clamps. This mode of labeling preserves the plasmids’ native supercoiled conformation and does not alter their nuclease sensitivity and intracellular distribution (28). Also, since binding is sequence-specific and highly stable, the fluorophore content on each plasmid molecule is identical, and undesired labeling of endogenous nucleic acids is avoided. Transfection of cells by the pGL3 plasmid resulted in punctate cytoplasmic fluorescence and low nuclear signal (Fig. 3A). Occasionally, a rim of fluorescent spots was observed around the nuclear envelope. Application of TNF-␣ to the cells did not alter significantly the fluorescence pattern (Fig. 3B), in agreement with the expression data presented in Fig. 2. Incubation of (unstimulated) cells with B-pGL3 gave rise to two populations. The minority population (⬍10%) exhibited a fluorescence pattern very similar to that revealed by the pGL3-transfected cells. The majority of the cells, however, revealed clear nuclear fluorescence with a diffuse appearance; the number of fluorescent spots localized to the cytoplasm was significantly reduced (Fig. 3C). When TNF-␣ was added, nuclear fluorescence increased further at the expense of the punctate cytoplasmic distribution (Fig. 3D). These results provide direct MOLECULAR THERAPY Vol. 3, No. 5, May 2001, Part 1 of 2 Parts Copyright © The American Society of Gene Therapy
evidence that B-pGL3 translocates to the nucleus more effectively than pGL3 and that it does so in a TNF-␣dependent manner. The data also support a model wherein the ability of the constituent molecules to migrate into the nucleus determines the relative abundance of the condensed (aggregated) and dispersed forms of the DNA. The next set of experiments aimed to quantify the contributions of the B sites to both the nuclear uptake of plasmid DNA and its subsequent transcription. A timecourse study was performed using control and modified pGL3, as well as pNFB-Luc (Fig. 4). HeLa cells were transfected with each of the plasmids and allowed to incubate for 2, 24, or 43 h. Unlike previous experiments, stimulation was not applied immediately posttransfection but was given at the end of incubation. Five hours later, cells were harvested and analyzed for luciferase activity. We first address the results obtained at 48 h after transfection. Within this period, about two cell divisions could take place, leading to similar levels of nuclear accumulation of the plasmids, due to nuclear envelope breakdown concomitant with mitosis. Any increase in gene activity, observed upon stimulation, should thus reflect primarily contributions arising from transcriptional enhancement by the B sites. For B-pGL3, such contributions were relatively small (1.6-fold) and, in fact, indistinguishable from those obtained for the control plasmid (1.3-fold). For pNFB-Luc, contributions were substantially larger (19-fold) due to direct transcriptional activation of the pNFB promoter (Fig. 4B). The aforementioned considerations suggest that the effect produced by the B sites on nuclear import of the DNA should be highest at relatively short posttransfection times. At longer times, a “dilution” effect comes into play due to ongoing mitotic activity. Indeed, while no significant change was observed over time in cells transfected with pGL3 (Fig. 4A) or with pCDNA3 (not shown), a 12and 11-fold decrease in fold induction occurred in BpGL3- (Fig. 4A) and pNFB-Luc transfected cells (Fig. 4B) after the first time interval. In both cases, no significant decrease was observed between the later time points, indicating that transcriptional activity, rather than nuclear translocation, now dominated expression levels. An estimate could therefore be made for the relative contribution of B-mediated DNA nuclear uptake to transgene expression simply by dividing the ratios observed at the short (7 h) times by those obtained at 48 h. For both B plasmids this contribution is found to be the same, around 12-fold. Similarly, a value of ⬃15-fold is derived for the enhancement in nuclear uptake of B-pGL3 relative to pGL3 for the data set presented in Fig. 2A, although incubation with the DNA was substantially longer. Additional experiments are planned in which p50⫺/p65⫺ cells will be transfected with plasmids encoding a p65 mutant lacking transcriptional activity, but retaining DNA binding ability. This system, which eliminates contributions arising from transcription-related enhancement, will be used to further corroborate these findings.
655
ARTICLE FIG. 3. Nuclear import of control and modified pGL3 as monitored by confocal fluorescence microscopy. HeLa cells (105) were incubated for 10 h with 1 g of rhodamine-labeled pGL3 (A, B) or B-pGL3 (C, D), in the absence (A, C) or presence (B, D) of TNF-␣. The deep blue color of the central nuclear regions depicted in (A) represents background fluorescence only slightly higher than that revealed by nontransfected cells.
DISCUSSION A system for controlled, facilitated import of plasmid DNA into mammalian cell nuclei is described. The system provides two mechanisms to increase transgene expression, the first of which operates by promoting the nuclear entry of the DNA. This mechanism increases transfection efficacy ⬃12-fold— comparable to that observed for synthetic NLS-(PNA)-DNA hybrids (8-fold (29)), and is insen-
FIG. 4. B-mediated nuclear translocation vs transcriptional enhancement: A time-course study. HeLa cells were transfected (0.5 g DNA/2 ⫻ 105 cells) with control and modified pGL3 (A) or pNFB-Luc (B) and were analyzed for luciferase activity at indicated time intervals after transfection. Stimulation was applied 5 h before harvesting the cells. Fold induction was determined separately for each plasmid at each time point.
656
sitive to the position of the B sites along the plasmid DNA. The second functions through transcriptional enhancement. As would be expected, the transcription enhancement is context-dependent, with contributions varying by a factor of 12 between B-pGL3 and NFB-Luc. Another feature provided by the system is control. This is achieved at two levels—nuclear uptake and transcription. In the present configuration, both levels are regulated in a similar manner. Control schemes that offer separate regulatory elements for each level can, however, be readily envisioned. Of the two levels, transcription can be regulated more tightly. Under optimal conditions (Fig. 2), the background noise associated with B-pGL3 is ⬃10% of maximal activity (determined by the ⫾TNF-␣ ratio). For NFB-Luc, on the other hand, this number drops to 0.4% (Fig. 4B, 7 h) due to an almost complete transcriptional blockage in the absence of stimulus. The low background activity of NFB-Luc may also be due to the fact it is devoid of massive promoter/enhancer elements, which can mediate nuclear transfer of the DNA by binding to ubiquitously expressed transcription factors. The SV40 enhancer, present in the pGL3 vectors, is an example of such elements (18). A minimal promoter configuration comes, however, with a price, which is a lower gene expression capacity. This is indeed the case for NFBLuc, whose stimulated gene expression levels were about fivefold lower than those of B-pGL3. Several mechanisms might be responsible for the obMOLECULAR THERAPY Vol. 3, No. 5, May 2001, Part 1 of 2 Parts Copyright © The American Society of Gene Therapy
ARTICLE served enhancement in nuclear uptake of the B DNA. The most obvious ones are targeting of the DNA to the NPC surface and subsequent triggering of the pore opening, both mediated by the NLS moiety of NFB. In addition, complex formation between NFB and the DNA may protect the DNA from reentry into the endocytic pathway, increasing its probability of reaching the nuclear transport machinery. In this case, DNA–NFB complexes would be advantageous over DNA molecules directly coupled to NLS peptides, as NLS receptors are localized predominantly to the nuclear envelope and do not penetrate deep into the cytoplasm. Another advantage provided by using a protein linker between the DNA and the tagging peptide is that the latter might be more accessible to the transport receptors. This is particularly true for classical NLS sequences, which are positively charged and, therefore, may interact strongly with the DNA backbone. This might explain why high NLS/DNA ratios are often required to effect efficient nuclear importation of the DNA when it is coupled to such signals (29, 30). Capped-linear DNA structures provide an exception to this rule, as they can be imported with high efficacy by singly attached classical NLS peptide (31). These structures, however, have an effective diameter of only ⬃3 nm, well below the diffusion limit of the NPC channels. Strategies for cell-specific DNA nuclear targeting and transgene expression have recently been reported. The smooth muscle ␥-actin promoter was used to preferentially target plasmid DNA into smooth muscle cell nuclei by a process mediated by the smooth muscle-specific transcription factor, SRF (20). In another study, selective transcriptional activation was achieved at low oxygen tension by vectors carrying hypoxia-responsive elements (32). Ultimately, one would like to combine these strategies with the one described in this work, by using binding elements for transcription factors whose nuclear trafficking can be affected by exogenous elicitors and whose expression is limited to certain tissues or pathological states. A further refinement can be achieved by using different control elements for DNA nuclear import and transgene transcription. ACKNOWLEDGMENTS The authors thank V. Brumfeld, E. Isacoff, Eric J. Messika, A. Minsky, E. Reuveny, and E. J. Wachtel for continuous advice and helpful discussions. Z.R. is an incumbent of the Abraham and Jennie Fialkow Career Development Chair. This work was supported by the Levine Foundation and the Kekst Family Foundation for Molecular Genetics.
REFERENCES 1 Luo, D., and Saltzman, W. M. (2000). Synthetic DNA delivery systems. Nat. Biotechnol. 18: 33–37.
MOLECULAR THERAPY Vol. 3, No. 5, May 2001, Part 1 of 2 Parts Copyright © The American Society of Gene Therapy
2 Rolland, A. P. (1998). From genes to gene medicines: Recent advances in nonviral gene delivery. Crit. Rev. Ther. Drug Carrier Syst. 15: 143–198. 3 Felgner, P. L. (1997). Nonviral strategies for gene therapy. Sci. Am. 276: 102–106. 4 Crystal, R. G. (1995). Transfer of genes to humans: Early lessons and obstacles to success. Science 270: 404 – 410. 5 Felgner, P. L. (1996). Improvements in cationic liposomes for in vivo gene transfer. Hum. Gene Ther. 7: 1791–1793. 6 Gao, X., and Huang, L. (1995). Cationic liposome-mediated gene transfer. Gene Ther. 2: 710 –722. 7 Behr, J. P. (1994). Gene transfer with synthetic cationic amphiphiles: Prospects for gene therapy. Bioconjugate Chem. 5: 382–389. 8 Brunner, S., Sauer, T., Carotta, S., Cotton, M., Saltik, M., and Wagner, E. (2000). Cell cycle dependence of gene transfer by lipoplex, polyplex and recombinant adenovirus. Gene Ther. 7: 401– 407. 9 Brisson, M., Tseng, W. C., Almonte, C., Watkins, S., and Huang, L. (1999). Subcellular trafficking of the cytoplasmic expression system. Hum. Gene Ther. 10: 2601–2613. 10 Labat-Moleur, F., et al. (1996). An electron microscopy study into the mechanism of gene transfer with lipopolyamines. Gene Ther. 3: 1010 –1017. 11 Zabner, J., Fasbender, A. J., Moninger, T., Poellinger, K. A., and Welsh, M. J. (1995). Cellular and molecular barriers to gene transfer by a cationic lipid. J. Biol. Chem. 270: 18997–19007. 12 Kiseleva, E., Goldberg, M. W., Cronshaw, J., and Allen, T. D. (2000). The nuclear pore complex: Structure, function, and dynamics. Crit. Rev. Eukaryotic Gene Expression 10: 101– 112. 13 Ohno, M., Fornerod, M., and Mattaj, I. W. (1998). Nucleocytoplasmic transport: The last 200 nanometers. Cell 92: 327–336. 14 Nigg, E. A. (1997). Nucleocytoplasmic transport: Signals, mechanisms and regulation. Nature 386: 779 –787. 15 Rybenkov, V. V., Vologodskii, A. V., and Cozzarelli, N. R. (1997). The effect of ionic conditions on the conformations of supercoiled DNA. I. Sedimentation analysis. J. Mol. Biol. 267: 299 –311. 16 Marko, J. F., and Siggia, E. D. (1994). Fluctuations and supercoiling of DNA. Science 265: 506 –508. 17 Boles, T. C., White, J. H., and Cozzarelli, N. R. (1990). Structure of plectonemically supercoiled DNA. J. Mol. Biol. 213: 931–951. 18 Dean, D. A. (1997). Import of plasmid DNA into the nucleus is sequence specific. Exp. Cell Res. 230: 293–302. 19 Langle-Rouault, F., et al. (1998). Up to 100-fold increase of apparent gene expression in the presence of Epstein–Barr virus oriP sequences and EBNA1: Implications of the nuclear import of plasmids. J. Virol. 7: 6181– 6185. 20 Vacik, J., Dean, B. S., Zimmer, W. E., and Dean, D. A. (1999). Cell-specific nuclear import of plasmid DNA. Gene Ther. 6: 1006 –1014. 21 Sen, R., and Baltimore, D. (1986). Multiple nuclear factors interact with the immunoglobulin enhancer sequence. Cell 46: 705–716. 22 Carlotti, F., Chapman, R., Dower, S. K., and Qwarnstrom, E. E. (1999). Activation of nuclear factor B in single living cells. J. Biol. Chem. 274: 37941–37949. 23 Ding, G. J., et al. (1998). Characterization and quantitation of NFB nuclear translocation induced by interleukin-1 and tumor necrosis factor-␣. J. Biol. Chem. 273: 28897– 28905. 24 May, M. J., and Ghosh, S. (1997). Rel/NF-kappa B and I kappa B proteins: An overview. Semin. Cancer Biol. 8: 63–73. 25 Grilli, M., Chiu, J. J., and Lenardo, M. J. (1993). NF-B and Rel: Participants in multiform transcriptional regulatory systems. Int. Rev. Cytol. 143: 1– 62. 26 Duckett, C. S., et al. (1993). Dimerization of NF-B2 with RelA(p65) regulates DNA binding, transcriptional activation, and inhibition by an IB-␣ (MAD-3). Mol. Cell Biol. 13: 1315–1322. 27 Fujita, T., Nolan, G. P., Ghosh, S., and Baltimore, D. (1992). Independent modes of transcriptional activation by the p50 and p65 subunits of NF-B. Genes Dev. 6: 775–787. 28 Zelphati, O., Liang, X., Hobart, P., and Felgner, P. L. (1999). Gene chemistry: Functionally and conformationally intact fluorescent plasmid DNA. Hum. Gene Ther. 10: 15–24. 29 Brande´n, L. J., Mohamad, A. J., and Smith, C. I. E. (1999). A peptide nucleic acid– nuclear localization signal fusion that mediates nuclear transport of DNA. Nat. Biotechnol. 17: 784 –787. 30 Sebestyen, M. G., et al. (1998). DNA vector chemistry: The covalent attachment of signal peptides to plasmid DNA. Nat. Biotechnol. 16: 80 – 85. 31 Zanta, M. A., Belguise-Valladier, P., and Behr, J. P. (1999). Gene delivery: A single nuclear localization signal peptide is sufficient to carry DNA to the cell nucleus. Proc. Natl. Acad. Sci. USA 96: 91–96. 32 Shibata, T., Giaccia, A. J., and Brown, J. M. (2000). Development of a hypoxiaresponsive vector for tumor-specific gene therapy. Gene Ther. 7: 493– 498.
657
ARTICLE
A Regulated, NFB-Assisted Import of Plasmid DNA into Mammalian Cell Nuclei Adi Mesika, Irina Grigoreva, Muriel Zohar,* and Ziv Reich1 Department of Biological Chemistry, Weizmann Institute of Science, Rehovot 76100, Israel *Research Division, Bio-Technology General, Rehovot 76100, Israel Received for publication January 2, 2001; accepted in revised form March 15, 2001; published online April 16, 2001
The success of synthetic DNA delivery systems in human gene therapy will be enhanced by increasing transfection efficiencies and by providing tighter control over targeting of the DNA into the nucleus. Here, we used DNA vectors that contain repetitive binding sites for the inducible transcription factor NFB, which is transported into the nucleus by the nuclear import machinery. Nuclear entry of the modified vectors was augmented 12-fold and was associated with corresponding increase in gene expression. Depending on their position, the binding sites could also function as transcriptional enhancers, increasing gene expression levels up to an additional 19-fold. Notably, nuclear targeting of the DNA and transgene transcription could both be regulated by exogenous stimulators which modulate the intracellular distribution of NFB. The approach provides a framework for the controlled targeting of constitutive or transcriptionally regulated synthetic vectors into mammalian cell nuclei. Key Words: synthetic gene therapy; nuclear pore complex; nuclear localization signal; nuclear factor B; B sites; tumor necrosis factor-␣.
INTRODUCTION Synthetic (nonviral) gene delivery systems are promising tools for gene therapy and DNA vaccination applications. Compared to viral-based systems, they possess several advantages, including excellent safety profiles, an essentially unlimited DNA carrying capacity, ease of production, and simplified QA/QC procedures (1– 4). However, transfection efficiency by these methods is unacceptably low (1–7), largely due to the inability of the DNA to effectively translocate through the nuclear pore complexes (NPCs) (8 –11). NPCs, which operate as selective conduits for nucleocytoplasmic exchange in eucaryotes, support two modes of transport. Particles smaller than 8 –9 nm (M r ⬃40 – 60 kDa) may pass by diffusion. Larger particles, on the other hand, are ushered selectively by soluble receptors of the karyopherin/importin  family, which recognize specific nuclear import (NLS) or export (NES) signal peptides displayed by the cargo (12–14). Under physiological ionic conditions, a typical plectonemically supercoiled DNA (fractional linking number || ⬃0.05) would have a diameter of ⬃10 nm (15–17). Smaller values, representing more relaxed plasmid con-
1 To whom correspondence and reprint requests should be addressed at the Department of Biological Chemistry, Weizmann Institute of Science, Rehovot 76100, Israel. Fax: ⫹972 8 947-4553. E-mail: [email protected].
MOLECULAR THERAPY Vol. 3, No. 5, May 2001, Part 1 of 2 Parts Copyright © The American Society of Gene Therapy 1525-0016/01 $35.00
figurations, will result in larger diameters. Passive diffusion of DNA plasmids through the NPC channels is therefore highly unlikely. The use of DNA as a pharmaceutical tool requires that control be provided over its targeting to the cell nucleus where it is transcribed, as well as over the timing and levels of expression of its encoded gene(s). While substantial progress in transcriptionally regulated vectors has been made, control over nuclear targeting is usually absent in current DNA delivery systems, synthetic or viral. Here, we have used a piggyback approach in which plasmid DNA was modified to include binding sites for the rapid response transcription factor nuclear factor B (NFB), allowing the latter to bind the DNA in the cytoplasm and transport it to the nucleus through the protein nuclear import machinery. Depending on their position relative to the transgene, the binding sites could also act as transcriptional enhancers, increasing further gene expression levels. Since nuclear translocation of NFB is an inducible process, nuclear uptake of plasmid DNA and gene transcription could both be regulated by exogenous stimulators that modulate the intracellular distribution of the protein. A system is therefore described for the targeted delivery of transcriptionally regulatable vectors into mammalian cell nuclei. When refined to include tissuespecific targeting elements, this system could be useful in therapeutic applications relying on gene transfer technologies.
653
ARTICLE MATERIALS
AND
METHODS
Plasmids. A fragment consisting of five tandem repeats of the Ig B motif, 5⬘-GGGGACTTTCC-3⬘, was amplified by PCR from pNFB-Luc (Stratagene) and cloned (BamHI) in pGL3 (Promega) downstream of the SV40 enhancer. To enable fluorescence labeling of pGL3, a segment containing peptide nucleic acid (PNA) recognition sequences was excised (BsaI/BglII) from pGeneGrip (Gene Therapy Systems) and cloned in the MCS region (BglII/SmaI) of the control and modified plasmids. Modifications were confirmed by double-stranded sequencing. Plasmid DNA was purified on Qiagen maxiprep columns, cleaned with phenol/chloroform, ethanol precipitated, and stored frozen at ⬃1 mg/ml. Transfections and luciferase activity measurements. Cells were grown in DMEM containing 10% FCS and antibiotics. Cells were serum-starved for 6 h before transfection. Transfections were made for 2 h, using SuperFect (Qiagen), with 0.25–1.0 g DNA, as specified in the figure legends. Transfections made with calcium phosphate gave similar results. Induction of NFB was by rhTNF-␣ (20 ng/ml; Pepro Tech EC) or TPA (100 ng/ml; Sigma). Unless noted otherwise, induction took place posttransfection. Luciferase activity was measured with the Luciferase Assay System (Promega) and was normalized to total protein (Pierce) or -galactosidase activity (Promega). Data shown represent the means derived from at least three independent measurements; SEM values were calculated using compound quantity formulation. DNA labeling and fluorescence microscopy. Labeling of plasmid DNA was made by site-specific hybridization to rhodamine-labeled PNA clamps (Gene Therapy Systems). PNA binding was done according to the manufacturer’s recommendations. Free PNA was removed by ethanol precipitation. Recovery of plasmids and PNA binding were determined spectrophotometrically and by gel electrophoresis, respectively. Fluorescence and DIC images were acquired simultaneously on fixed cells (4% paraformaldehyde/5% sucrose, in PBS) with an Olympus Fluoview 200 confocal laser scanning microscope (BX50WI-based), using a 60⫻ PlanApo oil immersion objective (NA 1.4). Data processing was performed using the NIH Image software. Figures shown are representative of ⬃200 images acquired from three independent experiments. Indirect immunofluorescence. Cells were fixed in PBS containing 4% paraformaldehyde (EM grade; Electron Microscopy Sciences) and 5% sucrose and permeabilized with 0.5% Triton X-100 in PBS. Primary antibodies (rabbit anti-hp50 and anti-hp65) were purchased from Santa Cruz Biotechnology. Secondary antibody was Cy3-conjugated goat anti-rabbit F(ab⬘)2 (minimal cross-reaction to human serum proteins) and was purchased from Jackson ImmunoResearch Laboratories. Images were acquired with a Nikon Eclipse TE300 microscope (60⫻ PlanApo oil immersion objective (NA 1.4)), equipped with high-resolution Hamamatsu Orca digital CCD camera.
RESULTS The transfer approach used in this study relies on the binding of endogenous NLS-proteins to specific DNA sequence elements within the vector (18 –20). As a mediator, we chose NFB (21). This choice was motivated by several factors. Binding affinity of NFB members to their DNA-binding sites (B sites) is high (K D ⬃10⫺10–10⫺13 M) and their translocation into the nucleus is rapid (minutes). The proteins are ubiquitously expressed, allowing various cell lines to be tested. Importantly, nuclear import of NFB is an inducible process; in the absence of stimulus, the proteins are predominantly cytoplasmic (22–25). Nuclear transfer of target DNA could therefore be controlled by extracellular addition of NFB activators. The structure of the two plasmids used in this study is shown in Fig. 1. The pNFB-Luc vector contains five direct repeats of the Ig B motif, 5⬘-GGGGACTTTCC-3⬘. This motif binds members of the NFB/Rel family (p49, p50,
654
FIG. 1. Plasmids used in this study. Commercially available pNFB-Luc was used without modifications. B-pGL3 is based on pGL3 and includes 5 B sites, cloned from pNFB-Luc, and 10 PNA binding sites used for site-specific fluorescence labeling. The PNA sites were also included in the control pGL3 plasmid.
and p65) with picomolar affinities (26, 27). The B sites are fused to a minimal promoter and are essential for transgene expression. The other plasmid, B-pGL3, has a well-developed constitutive expression cassette based on SV40 promoter and enhancer elements. In this plasmid, the B sites were cloned downstream of the SV40 enhancer such that they do not participate directly in the transcriptional regulation of the reporter gene. B-pGL3 also contains 10 PNA binding sequences that enable sitespecific labeling of the DNA. The PNA sites were also cloned into the parental plasmid, pGL3, which served as control. All human cell lines used in the study were responsive to known NFB stimuli, such as tumor necrosis factor ␣ (TNF-␣) and phorbol esters, as demonstrated by their ability to drive luciferase expression in pNFB-Luc-transfected cells (not shown and Fig. 4B) and from immunofluorescence microscopy studies, which showed stimulation-dependent nuclear accumulation of the NFB proteins p50 and p65 (not shown). The DNA was introduced into the cells chemically to faithfully mimic conditions relevant for DNA transfer in vivo. The transfection efficiencies of pGL3 and B-pGL3 were assayed using luciferase activity as a functional measure (Fig. 2). Serum-starved HeLa, Hek-293, Hep G2, and U373 cells were transfected with either plasmid in the presence or absence of stimulus and were harvested for analysis 18 h later. Measurements made on TNF-␣-activated cells revealed that the transfection activity of B-pGL3 was substantially higher, reaching up to 35 times that of pGL3. This effect was not exclusive to TNF-␣. Addition of the phorbol ester TPA, another activator of NFB, led to similar results (not shown). The two plasmids also differed in their responsiveness to stimulation. Application of TNF-␣ to B-pGL3-transfected cells resulted in a significant increase in gene activity. In contrast, only a minor effect could be seen upon stimulation of cells transfected with pGL3. The small increase observed for pGL3 may be due to the presence of an intrinsic B site in its SV40 enhancer region (nt 2332–2403). Stimulation of cells transfected with pCDNA3, which does not contain NFB binding elements, had no effect on gene activity (not shown). Interestingly, B-pGL3 performed better under MOLECULAR THERAPY Vol. 3, No. 5, May 2001, Part 1 of 2 Parts Copyright © The American Society of Gene Therapy
ARTICLE
FIG. 2. Transfection activity of pGL3 and B-pGL3 (0.25 g/105 cells) determined 18 h posttransfection. Cells were HeLa (A), Hek-293 (B), Hep G2 (C), and U373 (D). Data are presented as a fold increase in luciferase activity determined relative to unstimulated cells transfected with pGL3.
nonstimulatory conditions as well, with increases now ranging from 2.6- to 5.8-fold, depending on cell type. We interpret this enhancement in basal activity as a consequence of low-level migration of NFB molecules into the nucleus. This interpretation is consistent with immunofluorescence microscopy data, showing a weak, but significant p65 and p50 fluorescence in the nuclei of unstimulated cells (not shown and see, e.g., (22, 23)). The nuclear import of the pGL3 and B-pGL3 plasmids was studied directly by confocal fluorescence microscopy. Labeling of plasmids was done by site-specific hybridization to rhodamine-conjugated PNA clamps. This mode of labeling preserves the plasmids’ native supercoiled conformation and does not alter their nuclease sensitivity and intracellular distribution (28). Also, since binding is sequence-specific and highly stable, the fluorophore content on each plasmid molecule is identical, and undesired labeling of endogenous nucleic acids is avoided. Transfection of cells by the pGL3 plasmid resulted in punctate cytoplasmic fluorescence and low nuclear signal (Fig. 3A). Occasionally, a rim of fluorescent spots was observed around the nuclear envelope. Application of TNF-␣ to the cells did not alter significantly the fluorescence pattern (Fig. 3B), in agreement with the expression data presented in Fig. 2. Incubation of (unstimulated) cells with B-pGL3 gave rise to two populations. The minority population (⬍10%) exhibited a fluorescence pattern very similar to that revealed by the pGL3-transfected cells. The majority of the cells, however, revealed clear nuclear fluorescence with a diffuse appearance; the number of fluorescent spots localized to the cytoplasm was significantly reduced (Fig. 3C). When TNF-␣ was added, nuclear fluorescence increased further at the expense of the punctate cytoplasmic distribution (Fig. 3D). These results provide direct MOLECULAR THERAPY Vol. 3, No. 5, May 2001, Part 1 of 2 Parts Copyright © The American Society of Gene Therapy
evidence that B-pGL3 translocates to the nucleus more effectively than pGL3 and that it does so in a TNF-␣dependent manner. The data also support a model wherein the ability of the constituent molecules to migrate into the nucleus determines the relative abundance of the condensed (aggregated) and dispersed forms of the DNA. The next set of experiments aimed to quantify the contributions of the B sites to both the nuclear uptake of plasmid DNA and its subsequent transcription. A timecourse study was performed using control and modified pGL3, as well as pNFB-Luc (Fig. 4). HeLa cells were transfected with each of the plasmids and allowed to incubate for 2, 24, or 43 h. Unlike previous experiments, stimulation was not applied immediately posttransfection but was given at the end of incubation. Five hours later, cells were harvested and analyzed for luciferase activity. We first address the results obtained at 48 h after transfection. Within this period, about two cell divisions could take place, leading to similar levels of nuclear accumulation of the plasmids, due to nuclear envelope breakdown concomitant with mitosis. Any increase in gene activity, observed upon stimulation, should thus reflect primarily contributions arising from transcriptional enhancement by the B sites. For B-pGL3, such contributions were relatively small (1.6-fold) and, in fact, indistinguishable from those obtained for the control plasmid (1.3-fold). For pNFB-Luc, contributions were substantially larger (19-fold) due to direct transcriptional activation of the pNFB promoter (Fig. 4B). The aforementioned considerations suggest that the effect produced by the B sites on nuclear import of the DNA should be highest at relatively short posttransfection times. At longer times, a “dilution” effect comes into play due to ongoing mitotic activity. Indeed, while no significant change was observed over time in cells transfected with pGL3 (Fig. 4A) or with pCDNA3 (not shown), a 12and 11-fold decrease in fold induction occurred in BpGL3- (Fig. 4A) and pNFB-Luc transfected cells (Fig. 4B) after the first time interval. In both cases, no significant decrease was observed between the later time points, indicating that transcriptional activity, rather than nuclear translocation, now dominated expression levels. An estimate could therefore be made for the relative contribution of B-mediated DNA nuclear uptake to transgene expression simply by dividing the ratios observed at the short (7 h) times by those obtained at 48 h. For both B plasmids this contribution is found to be the same, around 12-fold. Similarly, a value of ⬃15-fold is derived for the enhancement in nuclear uptake of B-pGL3 relative to pGL3 for the data set presented in Fig. 2A, although incubation with the DNA was substantially longer. Additional experiments are planned in which p50⫺/p65⫺ cells will be transfected with plasmids encoding a p65 mutant lacking transcriptional activity, but retaining DNA binding ability. This system, which eliminates contributions arising from transcription-related enhancement, will be used to further corroborate these findings.
655
ARTICLE FIG. 3. Nuclear import of control and modified pGL3 as monitored by confocal fluorescence microscopy. HeLa cells (105) were incubated for 10 h with 1 g of rhodamine-labeled pGL3 (A, B) or B-pGL3 (C, D), in the absence (A, C) or presence (B, D) of TNF-␣. The deep blue color of the central nuclear regions depicted in (A) represents background fluorescence only slightly higher than that revealed by nontransfected cells.
DISCUSSION A system for controlled, facilitated import of plasmid DNA into mammalian cell nuclei is described. The system provides two mechanisms to increase transgene expression, the first of which operates by promoting the nuclear entry of the DNA. This mechanism increases transfection efficacy ⬃12-fold— comparable to that observed for synthetic NLS-(PNA)-DNA hybrids (8-fold (29)), and is insen-
FIG. 4. B-mediated nuclear translocation vs transcriptional enhancement: A time-course study. HeLa cells were transfected (0.5 g DNA/2 ⫻ 105 cells) with control and modified pGL3 (A) or pNFB-Luc (B) and were analyzed for luciferase activity at indicated time intervals after transfection. Stimulation was applied 5 h before harvesting the cells. Fold induction was determined separately for each plasmid at each time point.
656
sitive to the position of the B sites along the plasmid DNA. The second functions through transcriptional enhancement. As would be expected, the transcription enhancement is context-dependent, with contributions varying by a factor of 12 between B-pGL3 and NFB-Luc. Another feature provided by the system is control. This is achieved at two levels—nuclear uptake and transcription. In the present configuration, both levels are regulated in a similar manner. Control schemes that offer separate regulatory elements for each level can, however, be readily envisioned. Of the two levels, transcription can be regulated more tightly. Under optimal conditions (Fig. 2), the background noise associated with B-pGL3 is ⬃10% of maximal activity (determined by the ⫾TNF-␣ ratio). For NFB-Luc, on the other hand, this number drops to 0.4% (Fig. 4B, 7 h) due to an almost complete transcriptional blockage in the absence of stimulus. The low background activity of NFB-Luc may also be due to the fact it is devoid of massive promoter/enhancer elements, which can mediate nuclear transfer of the DNA by binding to ubiquitously expressed transcription factors. The SV40 enhancer, present in the pGL3 vectors, is an example of such elements (18). A minimal promoter configuration comes, however, with a price, which is a lower gene expression capacity. This is indeed the case for NFBLuc, whose stimulated gene expression levels were about fivefold lower than those of B-pGL3. Several mechanisms might be responsible for the obMOLECULAR THERAPY Vol. 3, No. 5, May 2001, Part 1 of 2 Parts Copyright © The American Society of Gene Therapy
ARTICLE served enhancement in nuclear uptake of the B DNA. The most obvious ones are targeting of the DNA to the NPC surface and subsequent triggering of the pore opening, both mediated by the NLS moiety of NFB. In addition, complex formation between NFB and the DNA may protect the DNA from reentry into the endocytic pathway, increasing its probability of reaching the nuclear transport machinery. In this case, DNA–NFB complexes would be advantageous over DNA molecules directly coupled to NLS peptides, as NLS receptors are localized predominantly to the nuclear envelope and do not penetrate deep into the cytoplasm. Another advantage provided by using a protein linker between the DNA and the tagging peptide is that the latter might be more accessible to the transport receptors. This is particularly true for classical NLS sequences, which are positively charged and, therefore, may interact strongly with the DNA backbone. This might explain why high NLS/DNA ratios are often required to effect efficient nuclear importation of the DNA when it is coupled to such signals (29, 30). Capped-linear DNA structures provide an exception to this rule, as they can be imported with high efficacy by singly attached classical NLS peptide (31). These structures, however, have an effective diameter of only ⬃3 nm, well below the diffusion limit of the NPC channels. Strategies for cell-specific DNA nuclear targeting and transgene expression have recently been reported. The smooth muscle ␥-actin promoter was used to preferentially target plasmid DNA into smooth muscle cell nuclei by a process mediated by the smooth muscle-specific transcription factor, SRF (20). In another study, selective transcriptional activation was achieved at low oxygen tension by vectors carrying hypoxia-responsive elements (32). Ultimately, one would like to combine these strategies with the one described in this work, by using binding elements for transcription factors whose nuclear trafficking can be affected by exogenous elicitors and whose expression is limited to certain tissues or pathological states. A further refinement can be achieved by using different control elements for DNA nuclear import and transgene transcription. ACKNOWLEDGMENTS The authors thank V. Brumfeld, E. Isacoff, Eric J. Messika, A. Minsky, E. Reuveny, and E. J. Wachtel for continuous advice and helpful discussions. Z.R. is an incumbent of the Abraham and Jennie Fialkow Career Development Chair. This work was supported by the Levine Foundation and the Kekst Family Foundation for Molecular Genetics.
REFERENCES 1 Luo, D., and Saltzman, W. M. (2000). Synthetic DNA delivery systems. Nat. Biotechnol. 18: 33–37.
MOLECULAR THERAPY Vol. 3, No. 5, May 2001, Part 1 of 2 Parts Copyright © The American Society of Gene Therapy
2 Rolland, A. P. (1998). From genes to gene medicines: Recent advances in nonviral gene delivery. Crit. Rev. Ther. Drug Carrier Syst. 15: 143–198. 3 Felgner, P. L. (1997). Nonviral strategies for gene therapy. Sci. Am. 276: 102–106. 4 Crystal, R. G. (1995). Transfer of genes to humans: Early lessons and obstacles to success. Science 270: 404 – 410. 5 Felgner, P. L. (1996). Improvements in cationic liposomes for in vivo gene transfer. Hum. Gene Ther. 7: 1791–1793. 6 Gao, X., and Huang, L. (1995). Cationic liposome-mediated gene transfer. Gene Ther. 2: 710 –722. 7 Behr, J. P. (1994). Gene transfer with synthetic cationic amphiphiles: Prospects for gene therapy. Bioconjugate Chem. 5: 382–389. 8 Brunner, S., Sauer, T., Carotta, S., Cotton, M., Saltik, M., and Wagner, E. (2000). Cell cycle dependence of gene transfer by lipoplex, polyplex and recombinant adenovirus. Gene Ther. 7: 401– 407. 9 Brisson, M., Tseng, W. C., Almonte, C., Watkins, S., and Huang, L. (1999). Subcellular trafficking of the cytoplasmic expression system. Hum. Gene Ther. 10: 2601–2613. 10 Labat-Moleur, F., et al. (1996). An electron microscopy study into the mechanism of gene transfer with lipopolyamines. Gene Ther. 3: 1010 –1017. 11 Zabner, J., Fasbender, A. J., Moninger, T., Poellinger, K. A., and Welsh, M. J. (1995). Cellular and molecular barriers to gene transfer by a cationic lipid. J. Biol. Chem. 270: 18997–19007. 12 Kiseleva, E., Goldberg, M. W., Cronshaw, J., and Allen, T. D. (2000). The nuclear pore complex: Structure, function, and dynamics. Crit. Rev. Eukaryotic Gene Expression 10: 101– 112. 13 Ohno, M., Fornerod, M., and Mattaj, I. W. (1998). Nucleocytoplasmic transport: The last 200 nanometers. Cell 92: 327–336. 14 Nigg, E. A. (1997). Nucleocytoplasmic transport: Signals, mechanisms and regulation. Nature 386: 779 –787. 15 Rybenkov, V. V., Vologodskii, A. V., and Cozzarelli, N. R. (1997). The effect of ionic conditions on the conformations of supercoiled DNA. I. Sedimentation analysis. J. Mol. Biol. 267: 299 –311. 16 Marko, J. F., and Siggia, E. D. (1994). Fluctuations and supercoiling of DNA. Science 265: 506 –508. 17 Boles, T. C., White, J. H., and Cozzarelli, N. R. (1990). Structure of plectonemically supercoiled DNA. J. Mol. Biol. 213: 931–951. 18 Dean, D. A. (1997). Import of plasmid DNA into the nucleus is sequence specific. Exp. Cell Res. 230: 293–302. 19 Langle-Rouault, F., et al. (1998). Up to 100-fold increase of apparent gene expression in the presence of Epstein–Barr virus oriP sequences and EBNA1: Implications of the nuclear import of plasmids. J. Virol. 7: 6181– 6185. 20 Vacik, J., Dean, B. S., Zimmer, W. E., and Dean, D. A. (1999). Cell-specific nuclear import of plasmid DNA. Gene Ther. 6: 1006 –1014. 21 Sen, R., and Baltimore, D. (1986). Multiple nuclear factors interact with the immunoglobulin enhancer sequence. Cell 46: 705–716. 22 Carlotti, F., Chapman, R., Dower, S. K., and Qwarnstrom, E. E. (1999). Activation of nuclear factor B in single living cells. J. Biol. Chem. 274: 37941–37949. 23 Ding, G. J., et al. (1998). Characterization and quantitation of NFB nuclear translocation induced by interleukin-1 and tumor necrosis factor-␣. J. Biol. Chem. 273: 28897– 28905. 24 May, M. J., and Ghosh, S. (1997). Rel/NF-kappa B and I kappa B proteins: An overview. Semin. Cancer Biol. 8: 63–73. 25 Grilli, M., Chiu, J. J., and Lenardo, M. J. (1993). NF-B and Rel: Participants in multiform transcriptional regulatory systems. Int. Rev. Cytol. 143: 1– 62. 26 Duckett, C. S., et al. (1993). Dimerization of NF-B2 with RelA(p65) regulates DNA binding, transcriptional activation, and inhibition by an IB-␣ (MAD-3). Mol. Cell Biol. 13: 1315–1322. 27 Fujita, T., Nolan, G. P., Ghosh, S., and Baltimore, D. (1992). Independent modes of transcriptional activation by the p50 and p65 subunits of NF-B. Genes Dev. 6: 775–787. 28 Zelphati, O., Liang, X., Hobart, P., and Felgner, P. L. (1999). Gene chemistry: Functionally and conformationally intact fluorescent plasmid DNA. Hum. Gene Ther. 10: 15–24. 29 Brande´n, L. J., Mohamad, A. J., and Smith, C. I. E. (1999). A peptide nucleic acid– nuclear localization signal fusion that mediates nuclear transport of DNA. Nat. Biotechnol. 17: 784 –787. 30 Sebestyen, M. G., et al. (1998). DNA vector chemistry: The covalent attachment of signal peptides to plasmid DNA. Nat. Biotechnol. 16: 80 – 85. 31 Zanta, M. A., Belguise-Valladier, P., and Behr, J. P. (1999). Gene delivery: A single nuclear localization signal peptide is sufficient to carry DNA to the cell nucleus. Proc. Natl. Acad. Sci. USA 96: 91–96. 32 Shibata, T., Giaccia, A. J., and Brown, J. M. (2000). Development of a hypoxiaresponsive vector for tumor-specific gene therapy. Gene Ther. 7: 493– 498.
657