Suppression of gene expression by a cell-permeable Tet repressor

Abstract

Engineered transcription factors designed to selectively activate or repress endogenous genes have great potential in medical and biotechnological applications. Ultimately, their success will depend on the development of efficient delivery systems. We show here that a chimeric tetracycline- controlled transcription factor, encompassing the Tet repressor (TetR) from the tetracycline-resistance operon (tet from Escherichia coli transposon Tn10) and a cell membrane transducing peptide, is able to regulate transcription from a tetracycline responsive promoter (pCMV2xtetO₂). When added directly to cultured cells, TetR fused to the full-length Antennapedia homeodomain (AntpHD) from Drosophila (TetRAntp), was able to selectively repress transcription in cells transiently transfected with a tetracycline-regulated reporter transcription unit. Moreover, TetRAntp could repress expression of a tetracycline responsive reporter transcription unit stably integrated into the genome of HeLa cells, demonstrating the possibility of manipulating endogenous gene expression by cell-permeable transcription factors.

INTRODUCTION

Engineered transcription factors are powerful tools for manipulating the expression of target genes (1–5). Regulation of gene expression at the level of transcription has great potential for basic research, drug target validation and therapy. Progress in the understanding of structural and functional relationships between DNA and primary amino acid sequences in zinc finger and leucine zipper transcription factors, enables the design of customised transcription factors with the ability to bind DNA in the proximity of a target gene. It is anticipated that these transcription factors could find applications in fields of great medical importance, such as in tumour therapy, inflammatory and autoimmune diseases, or in pathological conditions where there is a need to provide cytokines, hormones or growth factors in a temporal and spatial regulated manner.

Delivery of functional transcription regulators into cells and organisms represents a major bottleneck in exploiting transcription factor technology. Artificial transcription factors of limited molecular size can be delivered by electroporation (6) or by cationic liposomes (7), although these methods lead to high cell mortality and cytotoxicity (8). Cationic lipids are commonly used to deliver DNA and have been shown recently to introduce protein-based transcription factors directly into cells (9). Despite efficient delivery, the use of cationic lipids to deliver proteins may be restricted to in vitro applications due to large complex size and failure to work in the presence of serum (9).

Viral vectors have also been successfully employed to deliver engineered transcription factors both in vitro and in vivo (1,5). Viral vectors are very efficient and versatile; however, their use for gene delivery in vivo has generated some controversy due to their toxicity and immunogenicity (8), as well as the risks associated with recombination and integration within the host genome. Furthermore, constitutive regulation of a target gene is not always desirable and certain cells appear unable to tolerate the stable expression of potent, unregulated transcription factors (4). These issues have been addressed by placing the expression of the transcription factor under chemical control (1) or by directly regulating transcription factor activity using small-molecule inducers (4).

As an alternative to current systems, transcription factors could also be delivered into target cells as recombinant proteins linked to a protein transduction domain (PTD). Protein transduction technology is based on initial observations that the Tat protein from HIV-1 (10,11) and the Antennapedia homeodomain (AntpHD) from Drosophila (12) can enter cells when added to the culture media. A number of peptide sequences that translocate across cell membranes have since been identified, many of which were shown to deliver a wide range of peptide and protein cargoes in vitro and in vivo (13,14). For example, both a PTD sequence derived from the Kaposi fibroblast growth factor and the Tat PTD can mediate delivery of Cre recombinase, inducing recombination of LoxP-flanked target sites in a temporally regulated manner (15,16). The herpes simplex virus protein VP22 has been reported to deliver T7 RNA polymerase (T7 RNAP) to cultured cells, inducing dose-dependent expression of a luciferase reporter gene under the control of the T7 promoter (17).

A major advantage of PTDs is their apparent versatility with respect to the range of cell types that can be transduced (14). This includes certain populations of fragile mammalian nerve cells and quiescent primary lymphocytes, which were refractory to electroporation (18,19).

Here we describe the use of the AntpHD to deliver the tetracycline repressor (TetR) to regulate a tetracycline responsive transcription unit in a temporally regulated and reversible manner. In the absence of tetracycline, TetR represses transcription from a hybrid human cytomegalovirus (hCMV) major immediate-early promoter engineered to contain two tet operator sites (20). Transcription is then derepressed when the inducer, tetracycline, is added to the cells. Using this system we show regulation of transcription of transiently expressed and stably integrated transgenes.

MATERIALS AND METHODS

Plasmid construction

To generate pcDNA6/TRAntp, the antp homeobox was amplified by PCR from the template pSECTAG2 (C.J.Ingham, unpublished) and cloned into the expression vector pcDNA6/TR (Invitrogen) using an EcoRI restriction site at the C-terminus of the tetR coding sequence. To create pGL3B/TO, a fragment containing pCMV2xtetO₂ was excised from pcDNA4/TO (Invitrogen) using MluI and XhoI restriction sites and inserted into pGL3-Basic (Promega).

The pDS56/TR expression vector was derived from pDS56*-6xHis (a gift from H.Mueller). The tetR sequence was amplified by PCR using the template pcDNA6/TR, and inserted into BamHI and SalI restriction sites. To create pDS56/TRAntp, an AntpHD-encoding PCR fragment was amplified from the template pSECTAG2 and cloned at the C-terminus of tetR. All PCR generated fragments were confirmed by sequencing.

To create the transformation vector pHR’CMVTO-eGFP, the pCMV promoter was excised from pHR’CMV-eGFP (a gift from A.Thrasher) by digestion with ClaI and BamHI restriction enzymes. pCMV2xtetO₂ was amplified by PCR using pcDNA4/TO as a template, and cloned into pHR’CMV-eGFP. To generate pHR’ CMVTO-LacZ, the eGFP coding sequence was excised from pHR’CMVTO-eGFP by digestion with BamHI and XhoI, and replaced with a LacZ-encoding fragment from pcDNA4/TOLacZ (Invitrogen).

Expression and purification of recombinant TetR proteins

Escherichia coli M15 cells harbour the pUHA1 plasmid, coding for LacI, to allow isopropyl β-d-thiogalactoside (IPTG)-inducible expression of His-tagged proteins. The vector pDS56*-6xHis allowed expression of recombinant TetR proteins containing a six histidine tag at their C-termini. The presence of the histidine tag facilitates purification of recombinant proteins by nickel chelate chromatography (21). Bacterial cultures were grown in 2YT medium supplemented with 200 µg/ml ampicillin and 100 µg/ml kanamycin, at 37°C until the OD₆₀₀ reached 0.9. Expression of recombinant proteins was induced by the addition of IPTG to the culture at a final concentration of 1 mM. To purify TetRAntp under denaturing conditions, the bacterial pellet was re-suspended in 40 ml of denaturation buffer (6 M GuHCl, 0.5 M NaCl, 20 mM Tris pH 8.0). Samples were sonicated and the lysate centrifuged at room temperature (RT) for 20 min at 4000 g. The cleared supernatant was collected and left at room temperature overnight before being applied directly to the column. After an equilibration step with elution buffer (8 M urea, 100 mM NaH₂PO₄, pH 8), the proteins were eluted by lowering the pH of the 8 M urea solution stepwise to pH 2.

TetR was purified under native conditions as follows: the bacterial pellet was re-suspended in 40 ml of Talon lysis buffer (20 mM Tris–HCl, pH 7.9, 100 mM NaCl). Samples were sonicated and the lysate centrifuged at 4°C for 20 min at 4000 g. Cleared lysate was applied directly to the column before washing with Talon lysis buffer followed by 10 mM imidazole in Talon lysis buffer. The proteins were eluted in 50 mM imidazole in Talon lysis buffer. Fractions containing recombinant protein were pooled and dialysed with 20 mM Tris–HCl pH 7.9, 100 mM NaCl, 0.1% Tween 20. Purified proteins were quantified using the Micro BCA™ Protein Assay Reagent Kit (Pierce).

Cell culture

HeLa and 293T cells were cultured at 37°C, 10% CO₂ in Dulbecco’s modified Eagle’s medium (DMEM) (Invitrogen), containing 10% foetal calf serum (FCS, Harlan), 100 U/ml penicillin and 100 µg/ml streptomycin.

Transfection and transduction of HeLa cells

Co-transfection experiments. HeLa cells were plated at a density of 5 × 10⁴ cells/well (24 well plates). Cells were transfected with 6 µg of pcDNA6/TR, 1 µg of pGL3B/TO and 0.3 µg of pRL-SV40 (Promega) using Lipofectin (Invitrogen) according to the manufacturer’s instructions. The plasmid pRL-SV40, containing the SV40 enhancer/promoter region, was used as an internal control. Expression of Luciferase and Renilla was determined using the Dual Luciferase Reporter Assay kit (Promega) according to the manufacturer’s instructions.

Protein transduction into HeLa cells followed by transfection of a reporter plasmid. Cells were plated at a density of 1.5 × 10⁴ cells/well in glass chamber slides. Cells were treated for 8 h with 50 µl of dialysis buffer, containing 0–5 µg of protein, added directly to the culture medium. After transduction, cells were washed three times with PBS and transfected with 0.5 µg of pGL3B/TO and 0.1 µg of pRL-SV40. After 3 h, fresh media was added and cells were incubated for a further 3.5 h in the presence or absence of tetracycline (1 µg/ml).

Transfection of stably transformed HeLa cells. For time-course and northern blot analysis, 5 × 10⁴ cells/well were plated on a 24 well plate and transfected with 6 µg of pcDNA6/TR.

Transduction of stably transformed HeLa cells. For protein transduction experiments, transformed HeLa cells (pCMVTO-LacZ) were plated at 3 × 10⁴ cells/well (24 well plates) and grown in the absence or presence of tetracycline (1 µg/ml). Three protein treatments were carried out at 16-h intervals using 12 µg of TetRAntp. Prior to the addition of protein to the culture medium, cells were washed and fresh media added containing tetracycline (1 µg/ml) as indicated. After 48 h, cells were harvested and β-galactosidase activity measured using the β-Gal Reporter Gene Assay Kit (Roche) according to the manufacturer’s protocol. For northern blots, transformed HeLa cells (pCMVTO-eGFP) were seeded at 1 × 10⁵ cells/well on 6-well plates and grown in the absence or presence of tetracycline (1 µg/ml). Cells were treated twice at 5 h intervals with 150 µl of dialysis buffer alone, or with 30 µg of TetR or TetRAntp protein diluted in 150 µl dialysis buffer.

Immunoblot analysis

Samples were separated by 10% SDS–PAGE and transferred onto nitrocellulose membrane (Amersham). Membranes were blocked for 1 h in 4% dry milk in 2× TBST (20 mM Tris–HCl pH 8, 300 mM NaCl, 0.1% Tween 20) at room temperature, and then probed using anti-TetR serum (1:3200) in 2× TBST for 1 h. After washing with 2× TBST, bound antibodies were detected using a secondary goat anti-mouse immunoglobulin (H+L) (Promega) conjugated to alkaline phosphatase. Phosphatase activity was detected by incubating the membranes with 0.3 mg/ml NBT (Nitro-blue Tetrazolium) and 0.15 mg/ml BCIP (5-bromo-4-chloro-3-inolyl phosphate) in 100 mM Tris–HCl pH 9.5, 100 mM NaCl, 5 mM MgCl₂.

Anti-TetR serum

Purified TetR protein was used to immunise Balb/c mice to generate specific antisera. Mice were immunised intraperitoneally three times with 50 µg of TetR in complete Freund’s adjuvant (first immunisation) or in incomplete Freund’s adjuvant. Five days after the final immunisation, 1 ml of blood was collected from the animals by cardiac puncture.

Electrophoretic mobility shift assay

To make a double stranded DNA probe the 2xtetO₂ operator sequence from pcDNA4/TO was amplified by PCR to include EcoRI sites at the 5′ and 3′ ends. 10 pmol of probe were radiolabelled in 10 µl labelling buffer containing 10 mM Tris–HCl pH 8.3, 50 mM KCl, 1.5 mM MgCl₂, 5 µM dCTP and dGTP, 3 µl [α-³²P]dATP and [α-³²P]dTTP (10 µCi/µl, 1000 Ci/mmol, Amersham) and 1 U of Taq DNA polymerase. Mixtures were incubated for 1 h at 37°C. Labelled probes were purified from the non-incorporated labelled nucleotides using ProbeQuant G-50 Micro Columns (Pharmacia Biotech) according to the manufacturer’s instructions.

The binding of protein to DNA was carried out in 20 µl of binding buffer containing 0.4 pmol radiolabelled probe, 10 mM Tris–HCl, 10 mM KCl, 10 mM MgCl₂, 125 mM NaCl, 0.1 mM EDTA, 1 mM dithiothreitol (DTT), 10% glycerol and 0.5 µg salmon sperm DNA. 0.75 µg of purified TetR or TetRAntp was added to the reaction mixture and incubated at room temperature for 20 min. Binding reactions were blocked on ice and loaded onto a 5% polyacrylamide non-denaturing 1× TBE gel (0.09 M Tris–borate, 2 mM EDTA) and run in 1× TBE at 100 V for 5 h at 10°C. To confirm that the amount of bound probe was reduced in the presence of tetracycline, samples were quantified using Scion Image software (version 4.0.2).

Virus production and stable transduction of HeLa cells

The packaging vector pCMVΔR8.9 and the envelope coding plasmid pMDG were a gift from A. Thrasher and have been reported elsewhere (22). Viruses were produced as previously described (23). Infectious virus particles were added drop-wise to semi-confluent HeLa cells grown in 6-well plates. A homogenous population of HeLa cells expressing high levels of eGFP were obtained by sorting cells through a Coulter Epics Elite Cell Sorter.

Northern blotting

Total RNA was isolated using TRI-Reagent (Helena Biosciences) according to the manufacturers instructions. For northern blot analysis, 4 µg of total RNA was loaded on a 1% agarose-formaldehyde gel and transferred onto a nylon HybondN+ membrane (Amersham). Pre-hybridisation, hybridisation and washes of the membrane were performed using the HybondN+ protocol. The eGFP and GAPDH probes were labelled with [α-³²P]dATP (Amersham) using the High Prime DNA Labelling Kit (Roche) according to the manufacturers instructions. Labelled probes were purified using ProbeQuant G-50 Micro Columns. Relative levels of RNA were calculated using Scion Image software (version 4.0.2).

Immunofluorescence

Cells were plated at a density of 1.5 × 10⁴ cells/well in glass chamber slides and treated for 8 h with purified protein (TetRAntp or TetR) added directly to the culture medium. Following protein treatment, slides were washed and left at room temperature to dry. Slides were placed at –20°C for 10 min before fixing the cells in 3% (v/v) formaldehyde in 1× PBS for 15 min followed by blocking in 1% BSA in PBS for 20 min. After washing with 0.1% BSA in PBS, cells were incubated with a primary antibody diluted in 0.1% BSA in PBS for 1 h. Cells were then incubated with a secondary FITC-conjugated goat anti-mouse Immunoglobulin (Becton Dickinson) for 45 min. Slides were mounted using mounting medium for fluorescence (Vector Laboratories) and viewed using a confocal microsocpe.

RESULTS

TetRAntp represses transcription of a target gene

To assess the ability of TetRAntp to repress transcription from a tetracycline-responsive promoter, HeLa cells were cotransfected with a plasmid expressing either TetR (pcDNA6/TR), TetRAntp (pcDNA6/TRAntp) or an empty vector (pCMV-control) together with a Firefly Luciferase reporter plasmid containing the pCMV2xtetO₂ promoter (pGL3B/TO). For standardisation of transfection, cells were also cotransfected with a control plasmid coding for Renilla Luciferase (pRL-SV40). Levels of Firefly Luciferase and Renilla Luciferase were determined using the Dual Reporter Assay System (Promega). In the absence of tetracycline, cotransfection of plasmids encoding for TetR or TetRAntp repressed the tetracycline-regulated promoter 6.4- and 5.7-fold respectively (Fig. 1). In cells cotransfected with the empty pCMV-control plasmid, the ratio Firefly Luciferase to Renilla Luciferase was comparable in the absence or presence of tetracycline. These results demonstrate that in the absence of tetracycline, the ability of TetR to repress a transcription unit containing the tetO element was not affected by the presence of the AntpHD sequence fused to the C-terminus of TetR.

Figure 1.

AntpHD does not disrupt the activity of TetR. HeLa cells were cotransfected with plasmids encoding for TetRAntp (pcDNA6/TRAntp), TetR (pcDNA6/TR) or an empty vector (pCMV-control), and with a Firefly Luciferase reporter (pGL3B/TO) and a Renilla Luciferase control (pRL-SV40). Twenty-four hours post-transfection, fresh medium was added to the cells in the presence (grey bars) or absence (white bars) of Tc (1 µg/ml). Forty-eight hours post-transfection, cells were harvested and Luciferase expression determined using the Dual-Luciferase Reporter Assay System (Promega). These results show one experiment carried out in triplicates and are representative of three independent experiments. Error bars represent standard error of the mean (SEM) (n = 3).

Recombinant TetRAntp binds to tetO sequences

Recombinant TetR and TetRAntp proteins, designed to include a C-terminal 6x histidine tag within their sequence, were expressed in bacteria and purified using a Ni-NTA chelating column. A schematic representation of the proteins is shown in Figure 2A. Purified recombinant proteins were analysed by immunoblotting (Fig. 2B) with anti-TetR serum. The results indicated that both TetR and TetRAntp migrated according to their predicted molecular weights of 25 and 31 kDa, respectively.

Figure 2.

Binding of Purified TetR and TetRAntp to tetO. (A) A schematic representation of the proteins used in this study, the numbers indicate amino acids. (B) Western blot of purified TetR and TetRAntp. Proteins were separated by SDS–PAGE followed by immunoblotting with TetR anti-serum and detected using a goat anti-mouse alkaline phosphatase conjugate. (C) The tetO sequence used as a probe. (D) Retardation of ³²P-labelled probe. Labelled DNA only (lane 1), labelled DNA was incubated with a non- specific protein (lane 2), purified TetR (lanes 3 and 4) or purified TetRAntp (lanes 5 and 6). Binding reactions were carried out in the absence or presence of tetracycline (Tc) (1 µg/ml) as indicated.

The ability of purified TetR and TetRAntp to bind tetO sequences was evaluated by electromobility shift assay. Radioactively labelled DNA containing tetO sequences from pcDNA4/TO (Fig. 2C) was incubated with purified recombinant TetR or TetRAntp. Our results show that the electrophoretic mobility of labelled oligonucleotide probes containing tetO sequences was retarded in the presence of purified TetR (Fig. 2D, lanes 3 and 4) and TetRAntp (Fig. 2D, lanes 5 and 6) in a tetracycline-dependent manner. Binding of TetR or TetRAntp to tetO sequences was diminished by 58.4 and 35.85%, respectively, in the presence of tetracycline, confirming the selectivity of the interaction between TetR and operator DNA. In the absence of tetracycline, the amount of probe bound by TetR was greater than that of TetRAntp (Fig. 2D, lanes 4 and 6). This may be due to the denaturing and refolding steps involved in the purification of TetRAntp, resulting in loss in biological activity in a fraction of the purified product.

Two stoichiometric complexes were formed by TetR-DNA binding, possibly corresponding to the occupation of one or two tet operator sites by the TetR dimer. The TetRAntp–DNA complex migrated slower than the TetR–DNA complexes through the non-denaturing gel (Fig. 2D). TetRAntp has a higher estimated isoelectric point (8.75) than TetR (6.02), which may explain the low migration and resolution of the TetRAntp–DNA complex.

Regulation of transcription by transduced TetRAntp

As a preliminary step to determine whether TetRAntp could translocate across the cell membrane, the uptake of purified TetRAntp or TetR into HeLa cells was determined by indirect immunofluorescence using confocal microscopy. TetRAntp, but not TetR was taken up by a high percentage of cells and localised to both cytoplasm and nucleus (Fig. 3).

Figure 3.

Transduction of recombinant TetRAntp into HeLa cells. HeLa cells were incubated with 1 µg of purified TetRAntp (A and B), TetR (C and D) or no protein (E and F). After 8 h, cells were washed and stained with TetR anti-serum followed by a FITC-labelled goat anti-mouse secondary antibody. Cells were observed using a confocal microscope.

Next, purified recombinant TetRAntp or TetR were added directly to the cell culture medium to assess their ability to transduce across the membrane of living cells and regulate a transcription unit containing tetO regulatory sites. After 8 h of incubation, cells were washed and cotransfected with a reporter construct, containing the luciferase reporter gene under the control of the pCMV2xtetO₂ promoter (pGL3B/TO), together with a plasmid containing the Renilla luciferase reporter gene (pRL-SV40) as an internal control. Levels of Firefly Luciferase and Renilla Luciferase expression were determined using the Dual-Luciferase Reporter Assay System. Levels of Firefly Luciferase were significantly reduced (P > 0.05) when cells were treated with 5 µg/well of TetRAntp (Fig. 4). Firefly Luciferase expression was repressed by 42.4% in the absence of tetracycline when compared to tetracycline-treated cells. The incubation of cultured cells with a lower concentration of TetRAntp protein (1 µg/well) did not repress Firefly Luciferase expression. TetR had no effect on Firefly Luciferase levels, even at the highest protein concentration demonstrating that TetRAntp, but not TetR, repressed transcription of a transiently expressed target gene. Accordingly, the AntHD should have conferred to the repressor the ability to cross the cell membrane.

Figure 4.

Repression of transcription by TetRAntp. HeLa cells were incubated with buffer alone, or with 1 or 5 µg of protein (TetR or TetRAntp). Washed cells were cotransfected with a Firefly Luciferase reporter plasmid (pGL3B/TO) and a Renilla Luciferase control plasmid (pRL-SV40), and grown in the presence (grey bars) or absence (white bars) of tetracycline (Tc) (1 µg/ml). Luciferase expression was determined using the Dual Luciferase Reporter Assay System. For cells grown in the absence or presence of Tc and treated with buffer alone, the Luciferase/Renilla (L/R) ratio was assigned an arbitrary value of 1. The expression ratio was calculated by dividing the L/R ratio for TetR or TetRAntp treated cells, by the L/R ratio of their respective buffer controls. Results show the average of three independent experiments carried out in duplicate. Error bars represent standard error of the mean (SEM) (n = 3). Statistical analysis was carried out using Student’s t-test, *P > 0.05.

To examine whether TetRAntp could repress a stably expressed gene, two HeLa cell lines were established by lentiviral infection. These cell lines were designed to stably express either the green fluorescent protein eGFP or β-galactosidase reporter genes from the tetracycline-responsive promoter pCMV2xtetO₂. To assess the ability of TetR to repress the stably integrated reporter transcription units, cells expressing eGFP (pCMVTO-eGFP) or β-galactosidase (pCMVTO-LacZ) from the pCMV2xtetO₂ promoter were transfected with a plasmid encoding for TetR (pcDNA6/TR) and grown in the presence or absence of tetracycline. The relative amounts of eGFP transcripts were quantified by northern blot analysis, which showed a 51% reduction in transcription 60 h post-transfection in the absence of tetracycline relative to tetracycline treated cells (Fig. 5A). A β-galactosidase reporter assay was used to quantify β-galactosidase expression 48 h post-transfection in tetracycline treated and non-treated cells. After 48 h, β-galactosidase activity was repressed by 46.6% in the absence of tetracycline when compared to tetracycline-treated cells (not shown).

Figure 5.

TetRAntp can repress transcription of a stable transgene. (A) Northern blot analysis of HeLa cells stably transduced with the pCMV2xtetO₂-eGFP cassette. Cells were grown in the presence (+) or absence (–) of tetracycline and treated with TetRAntp or buffer alone for 10 h. Cells transfected with a plasmid coding for TetR (pcDNA6/TR) were grown in the presence (+) or absence (–) of tetracycline for 60 h. Membranes were probed with a 0.7 kb eGFP probe. For normalisation a 1 kb GAPDH probe was used, and eGFP to GAPDH ratios were calculated. To calculate the transcription ratio, an arbitrary value of 1.00 was assigned to treatments carried out in the presence (+) of tetracycline. The relative levels of transcription in the absence (–) of tetracycline were then calculated. (B) HeLa cells stably transduced with the pCMV2xtetO₂-lacZ cassette were treated over a 48 h period with TetRAntp in the presence (grey bar) or absence (white bar) of tetracycline (Tc) 1 µg/ml. β-Galactosidase activity was measured using the β-Gal Reporter Gene Assay kit (Roche). Northern blot analysis is representative of two independent experiments. β-Galactosidase expression shows one experiment carried out in triplicates and is representative of two independent experiments using different protein preparations. Error bars represent standard error of the mean (SEM) (n = 3). Statistical analysis was carried out using Student’s t-test, **P > 0.01.

The ability of TetRAntp to repress transcription of stably integrated transgenes in a tetracycline-dependent manner was assessed in both cell lines. HeLa cells stably expressing eGFP (pCMVTO-eGFP) were treated with purified TetRAntp leading to a 31% reduction in eGFP expression in the absence of tetracycline, relative to tetracycline-treated cells (Fig. 5A). No significant reduction in eGFP transcription was detected in control cells treated with purified TetR (not shown) or buffer alone (6 and 5% respectively). To test the effect of TetRAntp on β-galactosidase expression, pCMVTO-LacZ cells were grown in the absence or presence of tetracycline and treated three times over a 48 h time period with purified TetRAntp. After incubation with TetRAntp, levels of β-galactosidase activity were repressed by 40.9% (P > 0.01) in cells grown without tetracycline when compared to tetracycline treated cells (Fig. 5B). Taken together, these results demonstrate that the addition of recombinant TetRAntp to cultured cells specifically repressed the activity of the target transcription unit.

DISCUSSION

We have assessed whether the fusion of the tet repressor sequence to a cell-permeable peptide would allow the development of a novel generation of transcription factors able to regulate promoter activity when added exogenously to target cells. Our data show that TetRAntp could regulate gene expression in a tetracycline dependent manner; reporter gene activity was repressed upon addition of recombinant TetRAntp and derepressed in the presence of tetracycline. This suggests that the suppression of reporter gene expression was the result of the specific activity of TetRAntp on the tetracycline-responsive promoter. Furthermore, transcription was not repressed when purified TetR was added to the cells. All together, these findings indicate that TetRAntp crossed the surface membrane of healthy metabolically active cells, and that changes in transcriptional activity were not the result of non-specific protein uptake by dying cells or cells with compromised membrane structures. The internalisation of a chimeric transactivator lacking a specific PTD sequence has been reported (9). Internalisation of this protein occurred in a concentration-dependent manner and activated the expression of Luciferase from a transfected reporter plasmid (9). We cannot rule out the possibility that in our system non-specific cellular uptake of TetR may have occurred. If that were the case, however, the intracellular concentration of TetR was insufficient to repress transcription. Importantly, our data suggest that intracellular concentrations of TetRAntp reached sufficient levels to suppress the transcription of stably integrated transgenes.

While Penetratin, the 16 amino acid peptide corresponding to the third helix of the AntpHD (24,25) has been widely used as a PTD, only a handful of reports describe intracellular delivery using the 60 amino acid AntpHD (18,26). To date, delivery by Penetratin is generally restricted to cargo of 100 amino acids (27) or less (14,25). In the work described here, the AntpHD was used to transduce a functional transcription factor encompassing 207 amino acids.

In general, the success of PTD-mediated protein delivery is difficult to predict and may depend on the nature of the protein cargo. This is clearly illustrated in the case of VP22, which has been shown to mediate intercellular trafficking of large proteins such as P53 (28) and Flp recombinase (17), although attempts to deliver GFP (29,30) or a chimeric transactivator (9) were unsuccessful. Additionally, cell-type may also play a role in the outcome of PTD delivery, for example the internalisation of the AntpHD is most effective in cells expressing alpha-2,8-polysialic acid on their surface (31,32). Conflicting reports suggest that the internalisation of HIV-1 Tat protein may or may not require cell surface heparin sulfate (HS) proteoglycans (33,34), while the low-density lipoprotein receptor-related protein (LPR) has been shown to promote efficient uptake of Tat into neurons (35).

In vivo protein transduction has been reported using other PTDs (15,36). Whether the AntpHD can mediate delivery of biologically active proteins without undesirable effects in vivo remains to be determined. Furthermore, it will be important to determine whether this technology can be applied to the delivery of other classes of transcription factors, such as zinc fingers. The Tat PTD has been shown to enhance cellular uptake of a chimeric transactivator comprised of the Gal4 DNA binding domain and the VP16 transactivation domain (9). However, Tat-mediated delivery of the chimeric transcription factor was shown to be less efficient than commercial reagents for protein transfection (9). Ye et al. (9) speculated that the hypothetical protein denaturation and refolding steps involved in Tat-mediated transduction (37) may be a problem in the delivery of highly ordered transcriptional regulators.

The direct delivery of transcriptionally active proteins into cells using PTD technology has a vast number of applications, in both experimental and therapeutic settings. The publication of the human genome, combined with recent advances in transcription factor engineering (2,5,38), has opened up the possibility of creating designer transcription factors capable of regulating a given gene with high specificity. However, the delivery of artificial transcription factors into cells and organisms is a considerable problem (39,40). PTD technology may offer an alternative approach to deliver protein-based transcription factors in vitro and in vivo.