Research Article | Published:

Enhanced cationic liposome-mediated transfection using the DNA-binding peptide μ (mu) from the adenovirus core

Abstract

Promising advances in nonviral gene transfer have been made as a result of the production of cationic liposomes formulated with synthetic cationic lipids (cytofectins) that are able to transfect cells. However few cationic liposome systems have been examined for their ability to transfect CNS cells. Building upon our earlier use of cationic liposomes formulated from 3β-[N-(N′,N′-dimethylaminoethane)carbamoyl] cholesterol (DC-Chol) and dioleoyl-L-α-phosphatidyl-ethanolamine (DOPE), we describe studies using two cationic viral peptides, μ (mu) and Vp1, as potential enhancers for cationic liposome-mediated transfection. Mu is derived from the condensed core of the adenovirus and was selected to be a powerful nucleic acid charge neutralising and condensing agent. Vp1 derives from the polyomavirus and harbours a classical nuclear localisation signal (NLS). Vp1 proved disappointing but lipopolyplex mixtures formulated from pCMVβ plasmid, mu peptide and DC-Chol/DOPE cationic liposomes were able to transfect an undifferentiated neuronal ND7 cell line with β-galactosidase reporter gene five-fold more effectively than lipoplex mixtures prepared from pCMVβ plasmid and DC-Chol/DOPE cationic liposomes. Mu was found to give an identical enhancement to cationic liposome-mediated transfection of ND7 cells as poly-L-lysine (pLL) or protamine sulfate (PA). The enhancing effects of mu were found to be even greater (six- to 10-fold) when differentiated ND7 cells were transfected with mu-containing lipopolyplex mixtures. Differentiated ND7 cells represent a simple ex vivo-like post-mitotic CNS cell system. Successful transfection of these cells bodes well for transfection of primary neurons and CNS cells in vivo. These findings have implications for experimental and therapeutic uses of cationic liposome-mediated delivery of nucleic acids to CNS cells.

Introduction

The ability to transfect neuronal cells efficiently and safely could provide a powerful tool for the elucidation of neuronal function and may lead to novel treatments for neurological disorders. Gene therapy for the CNS has been hampered by the lack of efficient means for transducing post-mitotic neurons. Most studies have utilised viral vectors for gene delivery. However, many viral vectors are plagued by problems of immunity and cytotoxicity and are not easily manipulated by nonvirologists.123 Nonviral vectors are now emerging as an alternative method of cellular transduction. The most promising advances in nonviral gene transfer have been in the production of synthetic liposomes formulated with cationic lipids (cytofectins) able to transfect cells. Such cationic liposomes are relatively easy to use, have a broad applicability and lack cytotoxicity.4

Novel cationic liposome formulations are constantly being developed.5 However, few of these complexes have been examined for their ability to transduce cells within the CNS efficiently.6789 Cationic liposomes act via electrostatic interactions with negatively charged nucleic acids and subsequently are taken across the cell membrane by a process of slow endocytosis.61011 Once inside, a proportion of the bound nucleic acids dissociates and escapes from early endosomes into the cytoplasm either to perform a function there in the case of mRNA, or else traffic into the nucleus as in the case of DNA. Frequently, cationic liposomes are formulated with the neutral lipid dioleoyl-L-α-phosphatidylethanolamine (DOPE) that is thought to assist in endosome disruption.812 Indeed, one of the main cationic liposome systems shown to mediate successful transfection in the CNS is formulated from DOPE and the cytofectin N-[1-(2,3-dioleyloxy)propyl]-N,N,N trimethyl ammonium chloride (DOTMA).13141516

Recently, there have been attempts to enhance or improve upon cationic liposome-mediated transfection by precondensation of nucleic acids with a number of different possible polycationic agents leading to some improved transfection efficiencies.17181920212223 Various polycations have been identified which are efficient at enhancing cationic liposome-mediated transfection. Of these, poly-L-lysine (pLL) and protamine sulfate (PA) have proven the most effective to date, enabling increases in transfection of sometimes over 30-fold compared with cationic liposome-mediated transfection, in a variety of non-neuronal cell lines.1721 Fritz et al22 have described increases in cationic liposome-mediated transfection using a recombinant human H1 histone protein incorporating a nuclear localisation signal (NLS-H1). In addition, the nonhistone chromosomal high mobility group 1,2 (HMG-1,2) proteins have been used with some effect as part of HVJ liposome systems.2024

Protamine is a naturally occurring polycation found in the head of spermatozoa. The role of protamine is to condense DNA in sperm and aid in its transfer to the egg nucleus. This nuclear localising property of protamine makes it particularly attractive for transfection applications. Also, protamine sulfate is a smaller defined peptide system (4–4.25 kDa), as compared with pLL that embraces a range of average chain lengths and molecular weights (typically 18–19.2 kDa). Smaller polycationic agents such as protamine would be expected to show less likelihood of immunogenic responses in the target tissue and nucleic acid condensation should be easier to control. For this reason, we have been exploring the possibility that other small cationic peptides may be enhancers of cationic liposome-mediated transfection. To this end we have begun examining peptides and proteins closely associated with virus DNA. Here we report the results of studies carried out with two cationic peptides, μ (mu) and Vp1. Vp1 is a 19 amino acid residue cationic peptide and the only structural peptide of polyomavirus to exhibit DNA binding properties.2526 Mu is a 19 amino acid residue polycationic peptide known to be associated with the core complex of adenovirus (Table 1).2728 Vp1 but not mu contains an embedded classical nuclear localisation signal (NLS) similar to that found in HMG-1,2 and NLS-H1.2226 We found that mu showed promising transfection enhancing properties even with differentiated, post-mitotic neuronal cells indicating the possible usefulness of this peptide for future in vivo applications.

Table 1 1 mu and Vp1 protein sequences

Results

DNA binding analysis

The capacity of mu to bind plasmid DNA (pDNA) was studied by gel retardation assay. This was compared directly with the capacity of Vp1. The pDNA used throughout was pCMVβ that expresses a β-galactosidase reporter gene. Varying amounts of both peptides were incubated with pCMVβ at room temperature for approximately 20 min and then analysed by agarose gel electrophoresis. Without the addition of peptide, supercoiled and relaxed circular pCMVβ migrated in the expected manner (Figure 1b, c; lane 8). Beginning at a pCMVβ:mu ratio of 1:0.25 (w/w) the migration of pCMVβ was slightly affected, but at a ratio 1:0.5 (w/w) the migration of pCMVβ was substantially retarded. At ratios above 1:0.5 pCMVβ was unable to migrate into the agarose gel at all and the ability of ethidium bromide to interchelate pCMVβ was significantly reduced. By contrast, Vp1 was unable to retard or otherwise affect pCMVβ up to a pCMVβ:Vp1 ratio of 1:32 (w/w) (Figure 1).

Figure 1
figure1

Mu peptide is more efficient at binding pDNA than Vp1. (a) pCMVβ (1 μg) was incubated with 0 μg (lane 2), 5 μg (lane 3), 10 μg (lane 4), 15 μg (lane 5), 20 μg (lane 6), 25 μg (lane 7) and 30 μg (lane 8) of BSA for 20 min at room temperature in HBS. Samples were then analysed on a 1% agarose gel for altered mobility. (b) pCMVβ (1 μg) was incubated with 0.25 μg (lane 2), 0.5 μg (lane 3), 1 μg (lane 4), 2 μg (lane 4), 4 μg (lane 6), 6 μg (lane 7) and 0 μg (lane 8) mu peptide as in a. (c) pCMVβ (1 μg) was incubated with 2 μg (lane 2), 4 μg (lane 3), 6 μg (lane 4), 8 μg (lane 5), 16 μg (lane 6), 32 μg (lane 7) and 0 μg (lane 8) Vp1. In all gels lane 1 corresponds to a 1 kb DNA marker (BRL).

Transfection of undifferentiated ND7 cell line

Previously, cationic liposomes formulated from 3β-[N-(N′,N′-dimethylaminoethane)-carbamoyl]cholesterol (DC-Chol) and DOPE were shown to be able to transfect ND7 cells efficiently, a neuronally derived cell line.29 Transfection was optimal (>40% of cells) using lipoplex mixtures prepared from pDNA (1 μg) complexed with DC-Chol/DOPE (3 μg).29 Typically, maximal levels of transgene expression were obtained between 48–60 h after transfection during which time significant numbers of cell divisions had almost certainly taken place. In the studies described here, transfection of ND7 cells with lipoplex mixtures (pCMVβ:(DC-Chol/DOPE) ratio 1:3, 1:4, or 1:6 (w/w)) was compared with transfection involving lipopolyplex mixtures prepared by combining various quantities of mu with pCMVβ (1 μg) and DC-Chol/DOPE cationic liposomes (Figure 2). Enzyme assays were performed 20–24 h after transfection for convenience (data trends were found to be unaffected by this shorter period of post-transfection cell growth). Results from the gel retardation assay suggested that at a pCMVβ:mu ratio of 1:0.5 (w/w), sufficient mu to neutralise and sequester all pCMVβ was present (Figure 1). However, lipopolyplex mixtures prepared with this pCMVβ:mu ratio were no more effective at mediating transfection than corresponding DC-Chol/DOPE lipoplex mixtures prepared with the same amount of pCMVβ (results not shown). Therefore, we prepared a series of lipopolyplex mixtures with 0.6, 6, 12 or 21 μg of mu so as to investigate the consequences of using larger quantities of mu.

Figure 2
figure2

β-Galactosidase activity in undifferentiated ND7 cells transfected with pCMVβ:mu:(DC-Chol/DOPE) lipopolyplex mixtures and pCMVβ:(DC-Chol/DOPE) lipoplex mixtures. Lipopolyplex mixtures were formed by incubating pCMVβ with mu before the addition of DC-Chol/DOPE cationic liposomes. In each case, pCMVβ (1 μg) was precomplexed with 0.6, 6, 12, or 21 μg of mu following which 3, 4 or 6 μg of DC-Chol/DOPE cationic liposomes were added. ND7 cells were seeded at a density of 4 × 104 cells per well in 24-well culture dishes and grown for 24 h. Before transfection, cells were washed in serum-free media. ND7 cells were exposed to transfection mixtures for 2 h then maintained at 37°C, 5% CO2 for another 20–24 h before being harvested and processed for β-galactosidase enzyme assay. Numbers represent means ± s.d., n = 3.

Several lipopolyplex mixtures were found to perform better than DC-Chol/DOPE lipoplex mixtures. The most effective was pCMVβ:mu:(DC-Chol/DOPE) 1:12:6 (w/w/w). This combination was five-fold more effective at mediating transfection than the best DC-Chol/DOPE lipoplex mixture (pCMVβ:(DC-Chol/DOPE) 1:3 (w/w)) and 11-fold better than pCMVβ alone (Figure 2). By contrast, pCMVβ:mu polyplex mixtures failed to transfect cells more effectively than pCMVβ alone. Gratifyingly, there was no obvious cell loss detected with any of the lipopolyplex mixtures used. Also, the concentration of protein in the cellular lysates used for β-galactosidase reporter gene assays did not significantly vary with nontransfected cells (data not shown). ND7 transfection experiments performed with Vp1-containing lipopolyplex mixtures failed to provide any improvement over lipoplex transfection levels. Finally, parallel experiments were carried out with COS7 cells. Using this cell line, pCMVβ:mu:(DC-Chol/DOPE) 1:12:6 (w/w/w) lipopolyplex mixtures were found to be the most effective at mediating transfection once again, being four-fold more effective than the best DC-Chol/DOPE lipoplex mixtures (pCMVβ:(DC-Chol/DOPE) 1:6 (w/w)) (data not shown). The β-galactosidase reporter gene assay provides a measure of the overall level of β-galactosidase produced, but gives no information regarding the number of cells transfected. For this reason, we also performed cell counts on transfected ND7 cells. Cells were transfected with pCMVβ:mu:(DC-Chol/DOPE) 1:12:6 (w/w/w) lipopolyplex mixtures and a six-fold increase in the number of β-galactosidase positive cells was observed compared with transfection mediated by pCMVβ:(DC-Chol/DOPE) 1:3 (w/w) lipoplex mixtures (Figures 3 and 5.

Figure 3
figure3

Mu enhances cationic liposome mediated transfection efficiency in the neuronal cell line ND7. ND7 neurons were plated in 24-well culture dishes at a density of 4 × 104 cells per well and allowed to grow for 24 h. The undifferentiated ND7 neurons were then transfected with either pCMVβ alone (a), pCMVβ:(DC-Chol/DOPE) lipoplex mixtures, 1:3 (w/w) (b) or with pCMVβ:mu:(DC-Chol/DOPE) lipopolyplex mixtures 1:12:6 (w/w/w) (c). Before transfection, cells were washed in serum-free media. Cells were exposed to transfection mixtures for 2 h then maintained at 37°C, 5% CO2 for another 48 h before being fixed and processed for histochemical detection of X-gal.

A direct comparison was made between lipopolyplex mixtures formulated with mu and mixtures formulated with poly-L-lysine (pLL) and protamine (PA) sulfate. Both pLL and protamine sulfate have been shown to enhance liposome-mediated transfection in other cell types,171819 but to the best of our knowledge neither have been utilised for the transfection of neuronal cells. Lipopolyplexes of pLL or PA were prepared with a variety of component ratios based upon reported results with other cultured cell lines.21 In the event, both pLL and PA containing lipopolyplex mixtures were also found to outperform DC-Chol/DOPE lipoplex mixtures, the most effective being pCMVβ:pLL:(DC-Chol/DOPE) 1:2:6 (w/w/w) and pCMVβ:PA:(DC-Chol/DOPE) 1:2:6 (w/w/w). Within experimental error, both of these particular lipopolyplex mixtures were found to perform identically to pCMVβ:mu:(DC-Chol/DOPE) 1:12:6 (w/w/w) lipopolyplex mixtures (Figure 4).

Figure 4
figure4

A direct comparison was made between the ability of lipopolyplex mixtures formulated with mu and those formulated with poly-L-lysine and protamine sulfate to enhance liposome-mediated transfections in ND7 cells. Those ratios found to be optimal in ND7 cells are compared here. Varying amounts of polycation were mixed with 1 μg pCMVβ before complexing with 6 μg DC-Chol/DOPE liposomes. Transfection was performed in 24-well culture plates at a cell density of 4 × 104 cells/ml and 24 h later cells were processed for β-galactosidase histochemical assay. Transfection efficiency is represented as the number of positive (blue) cells divided by the total number. Each bar represents the mean ± s.d. (n = 3).

Transfection in differentiated ND7s

The ability of mu-containing lipopolyplex mixtures to transfect differentiated ND7 cells was investigated using pCMVβ:mu:(DC-Chol/DOPE) 1:12:6 (w/w/w) lipopolyplex mixtures, the same as those shown to give optimal transfection in undifferentiated ND7 cells. The ND7 cell line is derived from a fusion of primary rat dorsal root ganglia (DRG) neurons and the mouse neuroblastoma N18Tg2.30 ND7 cells can be differentiated in a variety of ways including the withdrawal of serum, cAMP administration or exposure to a combination of reduced serum, cAMP and nerve growth factor (NGF). Differentiation of ND7 cells leads to the expression of cellular properties associated with their parental nociceptive sensory neurons including a significant reduction in cell division and the onset of neurite outgrowth. ND7 cells were transfected with lipopolyplex and lipoplex mixtures 24 h after the onset of differentiation. Cells differentiated by all three methods described above were prepared for transfection experiments. Irrespective of the manner in which differentiation was induced, mu-containing lipopolyplex mixtures outperformed lipoplex mixtures by between six- and 10-fold depending upon the manner of differentiation of the ND7 cells (Figure 5)). Most crucially, in two out of the three cases, the percentage of differentiated ND7 cells transfected by pCMVβ:mu:(DC-Chol/DOPE) 1:12:6 (w/w/w) lipopolyplex mixtures was equal to if not better than the percentage of undifferentiated ND7 cells transfected with the same mixture (Figure 5).

Figure 5
figure5

Transfection efficiency estimated in undifferentiated and differentiated ND7 cells following administration of pCMVβ:mu:(DC-Chol/DOPE) lipopolyplex mixtures 1:12:6 (w/w/w) or pCMVβ:(DC-Chol/DOPE) 1:3 (w/w) lipoplex mixtures. ND7 cells were plated in 24-well culture dishes at a density of 4 × 104 cells per well and allowed to grow for 24 h. These undifferentiated cells (+ serum) were then either transfected directly, or differentiated by a further 24 h growth in one of three different differentiation media. The differentiation media used were serum-free (− serum), normal growth media with 1 mM cAMP (cAMP), or reduced serum (0.5%) containing 1 mM cAMP and 50 ng/ml nerve growth factor (NGF). Before transfection, cells were washed in serum-free media. Cells were exposed to transfection mixtures for 2 h then maintained at 37°C, 5% CO2 for another 48 h before being fixed and processed for X-gal histochemistry and the percentage of transfected cells determined.

Discussion

We have previously shown that DC-Chol/DOPE cationic liposomes are efficient at transfecting the neuronally derived ND7 cell line.29 DC-Chol/DOPE cationic liposomes have been used successfully outside the CNS in a variety of tissues to mediate gene delivery and have undergone clinical trials for gene therapy treatments of cystic fibrosis.3132 Also, DC-Chol/DOPE liposomes have been shown not to exhibit cytotoxic side-effects.3334 For these reasons we have been studying ways to improve upon the efficacy of DC-Chol/DOPE cationic liposome-mediated gene delivery for use with neural cells. Since their inception, there has been a continuous drive to develop more and more effective cationic liposome systems,57 but the vast majority of improvements have been sought by developing new cytofectins such as the successful series of DC-Chol analogues that we recently described.35 One alternative approach has been to prepare combination delivery systems from cationic liposomes and cationic polymers, using the polymer as an enhancing agent for transfection.1718192021 In our case, we decided to draw inspiration from the adenovirus in order to try to enhance cationic liposome-mediated gene delivery. In particular, we were drawn to the powerful DNA-condensing peptide, known as μ (mu), identified in the mature adenovirus core.2728 This 19 amino residue peptide results from the cleavage of a 79-residue precursor by adenovirus-encoded proteinase. Mu contains no known nuclear localisation signals but has been reported to have powerful DNA-condensing properties.27

The condensing power of mu was amply illustrated by our own gel retardation assay data using our test plasmid pCMVβ (Figure 1). While a small shift in pCMVβ (1 μg) mobility was observed with a pCMVβ:mu ratio of 1:0.25 (w/w), retardation was essentially complete at a pCMVβ:mu ratio of 1:0.5 (w/w) consistent with almost complete neutralisation of plasmid charge by associated mu (Figure 1). A simple theoretical calculation is consistent with this observation. A pCMVβ:mu ratio of 1:0.5 (w/w) corresponds to a pCMVβ:mu mole ratio of approximately 1:1000 (assuming an average nucleotide molecular weight of 329). Each pCMVβ contains 7000 bp that yield a total of 14000 negative charges per plasmid molecule from all the available phosphodiester links. By contrast, each mu has 12 basic amino acid residues and a free N-terminal α-amino functional group that could potentially yield 13 distinct positive charges per peptide molecule. Assuming both the number of potential charges and the pCMVβ:mu mole ratio to be correct, there does indeed appear to be sufficient mu associating with each plasmid to neutralise essentially all of the plasmid negative charges. The Vp1 peptide is the same length as mu and has five basic and one acidic amino acid residues together with a free N-terminal α-amino functional group that could potentially yield five distinct positive charges per peptide molecule. However, in spite of being able to present potentially almost half the number of positive charges as mu, Vp1 was found to be a vastly inferior agent for the neutralisation and condensation of pCMVβ plasmid in comparison to mu (Figure 1). This suggests that the nature and order of amino acid residues in a putative nucleic acid binding peptide can have a significant impact upon the charge neutralising and condensing properties of that peptide.

The ability of mu-containing lipopolyplex mixtures to transfect undifferentiated ND7 cells is revealing (Figures 2, 3 and 4). The lipopolyplex mixture optimal for transfection, pCMVβ:mu:(DC-Chol/DOPE) 1:12:6 (w/w/w), was found to contain particles of approximately 500–1000 nm in size by photon correlation spectroscopy measurements (unimodal analysis). These particles were certainly positively charged given the excess of mu and cationic liposomes used to prepare them. The pCMVβ:mu:(DC-Chol/DOPE) 1:12:6 (w/w/w) lipoplex mixture was reproducibly five-fold more effective at mediating ND7 cell transfection in vitro than the best DC-Chol/DOPE lipoplex mixture (pCMVβ:(DC-Chol/DOPE) 1:3 (w/w)), and similar results were also obtained with COS7 cells. Such enhancements are better than we had observed previously in transfection experiments in vitro using our second generation DC-Chol analogue/DOPE cationic liposome systems.35 Moreover, pCMVβ:mu:(DC-Chol/DOPE) 1:12:6 (w/w/w) lipopolyplex mixtures proved to be as equally effective at transfecting ND7 cells as lipopolyplex mixtures formulated using poly-L-lysine (pLL) or protamine sulfate (PA) in place of mu. This suggests that mu offers a very real alternative to both as enhancing agents for cationic liposome-mediated transfection and being a very well defined synthetic system could even surpass both given additional experiments.

Lipopolyplex mixtures formulated with Vp1 peptide were singularly unsuccessful in enhancing transfection. This is perhaps unsurprising in view of the weak plasmid charge neutralising and condensing properties of the Vp1 peptide. However, we had hoped for better in view of the classical nuclear localisation signal (NLS) embedded in the sequence (Table 1). Recent evidence has suggested that nuclear import of nucleic acids may be inefficient during cationic liposome-mediated transfection.10113637 For this reason there have been a number of attempts, including the work described here, to improve transfection efficiency by attempting to enhance nuclear uptake of nucleic acids using polycationic polymers harbouring peptide sequences with known nuclear localising capabilities.172022 However, our results appear to suggest that the presence of the NLS sequences does not promote transfection without adequate nucleic acid charge neutralisation and condensation as well. Indeed the latter seems the more important. This is born out by the experience of Fritz et al22 who observed no difference in transfection efficiencies between recombinant human histone (H1) and a modified version (NLS-H1) containing the SV40 large T antigen nuclear localizing sequence. However, others have suggested that the presence of an NLS does improve the nuclear accumulation of pDNA albeit via specific intracellular pathways.3839 As a specialised adenovirus core peptide, mu may have some influence over these pathways, but that remains to be established.

However, perhaps the most significant observations described here relate to the ability of mu-containing lipopolyplex mixtures to transfect differentiated ND7 cells (Figure 5). Differentiated ND7 cells possess a similar phenotype to their parental peripheral sensory neurons, including the induction of neurite outgrowth, a significant reduction in overall proliferation and a corresponding reduction in transfectability as a result.3035 Hence in effect, differentiated ND7 cells provide a simple ex vivo-like post-mitotic CNS cell system. Successful transfection of these cells would bode well for transfection of primary neurons and CNS cells in vivo. Gratifyingly, lipopolyplex mixtures comprised of pCMVβ:mu:(DC-Chol/DOPE) 1:12:6 (w/w/w) were able to transfect differentiated ND7 cells successfully with levels of transfection equivalent to those seen with undifferentiated cells, except in the instance where differentiation was caused by serum withdrawal (− serum) (Figure 5). Moreover, lipopolyplex transfection levels were consistently between six- and 10-fold higher than those seen with transfection by pCMVβ:(DC-Chol/DOPE) 1:3 (w/w) lipoplex mixtures. These results offer a tantalising glimpse that mu may indeed have some influence over the process of nuclear entry of nucleic acids. However, even if that were not the case, mu appears to have some attractive properties that place it alongside pLL and protamine sulfate as a useful enhancer of cationic liposome-mediated gene delivery.

Materials and methods

Cationic peptides and polypeptides

pLL (15000–30000 kDa) and protamine sulfate were purchased from Sigma (Poole, UK). Peptides Vp1 and Mu were synthesised on a Shimadzu PSSM-8 solid phase peptide synthesiser using a five-fold excess of (9-fluorenyl) methoxycarbonyl (Fmoc)-protected L-amino acids (Novabiochem, Nottingham, UK) and the FastMoc reagents 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetra-methylu- ronium hexafluorophosphate/hydroxybenzotriazole (HBTU/ HOBt) (Advanced Chemtech Europe, Cambridge, UK) as the amide coupling agent. After resin cleavage and deprotection, desalting was performed by gel filtration using a column of P2 Biogel (2 × 28 cm; BioRad, Herts, UK) attached to an FPLC system (Amersham Pharmacia Biotech UK, Bucks, UK) with 0.1% aqueous TFA as eluant at a flow rate of 0.5–0.75 ml/min. Final preparative reverse-phase purification was achieved with a Vydac column (C18, 5 μm, 2 × 25 cm; Hichrom, Berks, UK) attached to a Gilson HPLC system (Anachem, Bedfordshire, UK). Peptides were eluted at 5 ml/min by means of a linear gradient of acetonitrile in 0.1% aqueous TFA and elution monitored at 220–230 nm.

The Vp1 peptide was prepared using a preloaded L-Pro-2-chlorotrityl super acid labile resin (Novabiochem) (100 mg, 1.05 mmol/g, 0.1 mmol). Extended coupling times were used to incorporate all amino acid residues from the sixth (Lys) through to the N-terminal residue. After automated N-terminal Fmoc deprotection with piperidine (20%, v/v) in dimethyl-formamide, the resin was isolated, washed with dimethylformamide (10 ml) and methanol (15 ml), and then dried in vacuo. Crude peptide was cleaved from the resin using ice cooled TFA (8 ml), containing phenol (7%, w/v), ethanedithiol (2%, v/v), thioanisole (4%, v/v) and water (4%, v/v) (known as mixture A), and then precipitated with ice cold methyl-tert-butylether (MTBE) (30 ml). The subsequent pellet was then desalted and the crude peptide mixture purified by reverse phase HPLC. After elution, fractions containing the desired peptide (eluting with acetonitrile 68.5% v/v) were combined and lyophilized to give the peptide as a white powder. Overall yield: 32 mg (15 μmol, 15%); MS (MALDI-TOF) C85H151N26O26S3: [M + H]+ calculated 2049.5, found 2050.2. The sequence was confirmed by amino acid composition and sequence analysis. Homogeneity was judged >95% by HPLC analysis.

The mu peptide was prepared using Gly-Wang resin (Novabiochem) (40 mg, 0.67 mmol/g, 0.03 mmol). Normal coupling times were used throughout. After automated N-terminal Fmoc deprotection as above, the resin was isolated and washed with dichloromethane (20 ml) and methanol (20 ml) after which the resin was dried in vacuo. Crude peptide was cleaved from the resin using mixture A (8 ml) and precipitated with MTBE (30 ml), all as above. Finally, the crude peptide mixture was desalted and purified by reverse phase HPLC. After elution, fractions containing the desired peptide (eluting with acetonitrile 17.2%) were combined and lyophilized to give the peptide as a white powder. Overall yield: 65 mg (26 μmol, 80%); MS (ES) C95H170N52O21S2: [M + H]+ calculated 2440.7, found 2440.6. Homogeneity was judged >95% by HPLC analysis.

DNA binding analysis

The purified peptides were reconstituted in sterile distilled H2O at 3 mg/ml. Peptide and pCMVβ (1 μg) were combined in 20 μl HEPES buffered saline (HBS) (137 mM NaCl, 5 mM KCl, 0.75 mM Na2HPO4, 19 mM HEPES, pH 7.4) for 20 min at room temperature. Peptide:pCMVβ polyplexes were subsequently analysed by agarose gel electrophoresis (1%). Control incubations for general macromolecular pDNA interactions were performed with varying amounts of molecular biology grade purified bovine serum albumin (Sigma).

Cell cultures

ND7 cells are derived from the fusion of a neuroblastoma (N18Tg2) with neonatal rat sensory neurons and are a well-characterised cell line.30 The cell line was maintained in normal growth media (NGM) (Leibovitz's L-15 media (BRL, Paisley, UK) enriched with 10% fetal bovine serum (BRL), 4 g/l glucose, 4 g/l sodium bicarbonate (BRL), 100 IU/ml penicillin/streptomycin (BRL) at 37°C and 5% CO2. The cells were plated on to 24-well plates (Life Technologies, Paisley, UK) at a density (4 × 104 cells per well) that produced 70% confluence after 24 h. These cells were then used directly for transfection or differentiated as described below.

Differentiation of ND7 cells was carried out using three methods previously described.3040 Cells were seeded in NGM at a density of 4 × 104 cells per well in a 24-well culture dish (Nunc). After 24 h, the media was replaced with either: (a) serum-free differentiation media (− serum) (50% Hams F12, 50% DMEM, 5 μg/ml transferrin, 250 ng/ml insulin, 0.3 μM sodium selenite); or (b) NGM supplemented with 1 mM adenosine 3′,5′-cyclic monophosphate (cAMP; Sigma) (cAMP); or (c) low serum nerve growth factor media (NGF) (L-15 supplemented with 2 mM glutamine, 4 g/l glucose, 4 g/l sodium bicarbonate, 100 μg/ml penicillin, 100 g/ml streptomycin, 0.5% fetal bovine serum, 1 mM cAMP, 50 ng/ml nerve growth factor (Alomone Laboratories, Jerusalem, Israel). Differentiated ND7 cells were grown in appropriate media for 24 h at 37°C, 5% CO2 before transfection.

COS-7 cells (derived from Green Monkey kidney) were grown in RPMI 1640 media (BRL) supplemented with 10% fetal bovine serum (BRL) and 100 μg/ml penicillin/streptomycin (BRL).

Plasmid constructs

All experiments were performed using the plasmid pCMVβ (Clontech, Palo Alto, CA, USA) harboring the full-length sequence for E. coli β-galactosidase downstream of the human cytomegalovirus immediate–early promoter/enhancer (Clontech, Hampshire, UK). Stocks of plasmid DNA (1 mg/ml) were prepared using standard molecular cloning techniques and purified using the Qiagen Endotoxin-free plasmid purification system (Qiagen, Dorking, UK).

Liposomes

DC-Chol/DOPE cationic liposomes were prepared as previously described.2936 Briefly, 6 μmol of DC-Chol and 4 μmol of DOPE (supplied at 10 mg/ml in CHCl3) were added to freshly distilled CH2Cl2 (5 ml) under nitrogen. 20 mM HEPES buffer, pH 7.8 (5 ml) was added to the mixture and the whole sonicated for 3 min. Organic solvents were removed under reduced pressure and the resulting liposome suspension was then sonicated for a further 3 min. These liposome stock preparations (1.2 mg/ml) were stored at 4°C before use.

Transfection protocol

Since initial experiments determined that the presence of fetal bovine serum inhibited transfection of ND7 cells serum-free differentiation media was used for all transfection experiments. Appropriate aliquots of pCMVβ (1 μl) and cationic liposomes (2.5–5 μl) were placed at the bottom of a 7 ml sterile Bijou container (Bibby Sterilin, Staffordshire, UK), but not in contact with each other. DNA and liposomes were combined by the addition of 400 μl serum-free differentiation media and gentle shaking. Lipoplex mixtures were then incubated at room temperature for 20 to 30 min before being applied to cells. Lipopolyplex mixtures were generated in the following manner. Peptide (2–7 μl) was placed in the bottom of sterile polystyrene containers alongside, but not in contact with pCMVβ (1 μl) and mixed by the introduction of 400 μl serum-free NGM media. These polyplex mixtures were then incubated at room temperature for 10 min after which time DC-Chol/DOPE cationic lipsomes (2.5–5 μl) were added. Resulting lipopolyplex mixtures were then further incubated at room temperature for 20 min before being applied to cells.

Before transfection, all cells were washed in serum-free media and then exposed to transfection mixtures for 2 h at 37°C, 5% CO2. Following this, transfection mixtures were replaced with complete media and incubation continued for 20–24 or 48 h, as appropriate. Thereafter, cells were harvested for β-galactosidase enzyme assays (Promega, Southampton, UK) or fixed and processed for X-gal histochemistry as described.29 Cell counts were made under × 40 magnification using a Nikon Diaphot inverted microscope. Each transfection experiment was repeated at least three times and at least three separate counts were made for each well.

References

  1. 1

    Wood MJA et al. Inflammatory effects of gene-transfer into the CNS with defective HSV-1 vectors Gene Therapy 1994 1: 283–291

  2. 2

    Byrnes AP et al. Adenovirus gene-transfer causes inflammation in the brain Neuroscience 1995 66: 1015–1024

  3. 3

    Naldini L et al. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector Science 1996 272: 263–267

  4. 4

    Miller AD . Cationic liposomes for gene delivery Angewandte Chemie-International Edition 1998 37: 1769–1785

  5. 5

    Lee RJ, Huang L . Lipidic vector systems for gene transfer Crit Rev Ther Drug Carr Syst 1997 14: 173–206

  6. 6

    Gao X, Huang L . Cationic liposome-mediated gene-transfer Gene Therapy 1995 2: 710–722

  7. 7

    Felgner PL et al. Lipofection – a highly efficient, lipid-mediated DNA-transfection procedure Proc Natl Acad Sci USA 1987 84: 7413–7417

  8. 8

    Farhood H, Serbina N, Huang L . The role of dioleoyl phosphatidylethanolamine in cationic liposome mediated gene transfer Biochim Biophys Acta 1995 1235: 289–295

  9. 9

    Caplen NJ et al. In vitro liposome-mediated DNA transfection of epithelial cell lines using the cationic liposome DC-Chol/DOPE Gene Therapy 1995 2: 603–613

  10. 10

    Labat-Moleur F et al. An electron microscopy study into the mechanism of gene transfer with lipopolyamines Gene Therapy 1996 3: 1010–1017

  11. 11

    Zabner J et al. Cellular and molecular barriers to gene transfer by a cationic lipid J Biol Chem 1995 270: 18997–19007

  12. 12

    Felgner JH et al. Enhanced gene delivery and mechanism studies with a novel series of cationic lipid formulations J Biol Chem 1994 269: 2550–2561

  13. 13

    Sahenk Z et al. Gene delivery to spinal motor neurons Brain Res 1993 606: 126–129

  14. 14

    Iwamoto Y et al. Liposome-mediated Bdnf Cdna transfer in intact and injured rat brain Neuroreport 1996 7: 609–612

  15. 15

    Roessler BJ, Davidson BL . Direct plasmid-mediated transfection of adult murine brain cells in vivo using cationic liposomes Neurosci Lett 1994 167: 5–10

  16. 16

    Zhou X, Huang L . DNA transfection mediated by cationic liposomes containing lipopolylysine: characterization and mechanism of action Biochim Biophys Acta 1994 1189: 195–203

  17. 17

    Sorgi FL, Bhattacharya S, Huang L . Protamine sulfate enhances lipid-mediated gene transfer Gene Therapy 1997 4: 961–968

  18. 18

    Li S, Huang L . Protamine sulfate provides enhanced and reproducible intravenous gene transfer by cationic liposome/DNA complex J Liposome Res 1997 7: 207–219

  19. 19

    Vitiello L et al. Condensation of plasmid DNA with polylysine improves liposome-mediated gene transfer into established and primary muscle cells Gene Therapy 1996 3: 396–404

  20. 20

    Namiki Y, Takahashi T, Ohno T . Gene transduction for disseminated intraperitoneal tumor using cationic liposomes containing non-histone chromatin proteins: cationic liposomal gene therapy of carcinomatosa Gene Therapy 1998 5: 240–246

  21. 21

    Gao X, Huang L . Potentiation of cationic liposome-mediated gene delivery by polycations Biochemistry 1996 35: 9286–9286

  22. 22

    Fritz JD et al. Gene transfer into mammalian cells using histone-condensed plasmid DNA Hum Gene Ther 1996 7: 1395–1404

  23. 23

    Hagstrom JE et al. Complexes of non-cationic liposomes and histone H1 mediate efficient transfection of DNA without encapsulation Biochim Biophys Acta 1996 1284: 47–55

  24. 24

    Isaka Y et al. The HVJ liposome method Exp Nephrol 1998 6: 144–147

  25. 25

    Gillock ET et al. Polyomavirus major capsid protein VP1 is capable of packaging cellular DNA when expressed in the baculovirus system J Virol 1997 71: 2857–2865

  26. 26

    Chang D, Cai X, Consigli RA . Characterization of the DNA binding properties of polyomavirus capsid proteins J Virol 1993 67: 6327–6331

  27. 27

    Anderson CW, Young ME, Flint SJ . Characterization of the adenovirus 2 virion protein, mu Virology 1989 172: 506–512

  28. 28

    Hosokawa K, Sung MT . Isolation and characterization of an extremely basic protein from adenovirus type 5 J Virol 1976 17: 924–934

  29. 29

    McQuillin A et al. Optimization of liposome mediated transfection of a neuronal cell line Neuroreport 1997 8: 1481–1484

  30. 30

    Wood JN et al. Novel cell-lines display properties of nociceptive sensory neurons Proc Roy Soc Lond Ser B Biol Sci 1990 241: 187–194

  31. 31

    Caplen NJ et al. Liposome-mediated Cftr gene-transfer to the nasal epithelium of patients with cystic-fibrosis Nature Med 1995 1: 39–46

  32. 32

    Nabel G, Chang A, Nabel E . Clinical protocol: immunotherapy of malignancy by in vivo gene transfer into tumors Hum Gene Ther 1992 3: 399–410

  33. 33

    Nabel GJ et al. Direct gene-transfer with DNA liposome complexes in melanoma – expession, biological activity, and lack of toxicity in humans Proc Natl Acad Sci USA 1993 90: 11307–11311

  34. 34

    Stewart MJ et al. Gene-transfer in vivo with DNA liposome complexes – safety and acute toxicity in mice Hum Gene Ther 1992 3: 267–275

  35. 35

    Cooper RG et al. Polyamine analogues of 3 β-[N-(N′,N′-dimethylaminoethane)carbomoyl]cholesterol (DC-Chol) as agents for gene delivery Chemistry 1998 4: 137–151

  36. 36

    Thierry AR et al. Characterization of liposome-mediated gene delivery: expression, stability and pharmacokinetics of plasmid DNA Gene Therapy 1997 4: 226–237

  37. 37

    Coonrod A, Li FQ, Horwitz M . On the mechanism of DNA transfection: efficient gene transfer without viruses Gene Therapy 1997 4: 1313–1321

  38. 38

    Sebestyen MG et al. DNA vector chemistry: the covalent attachment of signal peptides to plasmid DNA Nat Biotechnol 1998 16: 80–85

  39. 39

    Hagstrom JE et al. Nuclear import of DNA in digitonin-permeabilized cells J Cell Sci 1997 110: 2323–2331

  40. 40

    Budhrammahadeo V, Lillycrop KA, Latchman DS . The levels of the antagonistic Pou family transcription factors Brn-3a and Brn-3b in neuronal cells are regulated in opposite directions by serum growth factors Neurosci Lett 1995 185: 48–51

Download references

Acknowledgements

This work was supported by a grant from the Motor Neurone Disease Association. SIS thanks the BBSRC and Genzyme for a CASE studentship. We thank the Mitsubishi-Tokyo Pharmaceutical company for supporting the Imperial College Genetic Therapies Centre and the West Riding Medical Research Trust for their support.

Author information

Correspondence to K D Murray.

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • gene therapy
  • peptides
  • CNS
  • post-mitotic

Further reading