Comparison of cDNA and genomic forms of tyrosine hydroxylase gene therapy of the brain with Trojan horse liposomes

BACKGROUND: The present study examines whether chromosomal derived forms of therapeutic genes can be delivered to brain following intravenous administration. The brain expression of a rat tyrosine hydroxylase (TH) cDNA is compared to the brain expression of a plasmid DNA encoding the 18 kb rat TH gene. METHODS: TH gene expression is measured in cell culture and in vivo in brain in experimental Parkinson's disease (PD). A total of four eukaryotic expression plasmids encoding rat TH were engineered wherein the size of the TH expression cassette ranged from 1.5 kb, in the case of the cDNA form of the gene, to 17.5 kb, in the case of the largest size genomic construct. The TH expression plasmids were delivered to either cultured cells or to rat brain in vivo with Trojan horse liposomes (THLs), which target the non-viral plasmid DNA to cells via cell membrane receptors. RESULTS: The pattern of TH gene expression in cell culture and in vivo was similar: the cDNA form of the TH gene was fast-acting with short duration of action, and the genomic form of the TH gene was slow-acting with longer duration of action. The most sustained replacement of striatal TH enzyme activity in experimental PD was produced by combination gene therapy where both the cDNA and the genomic forms of the TH gene were administered simultaneously. CONCLUSIONS: Eukaryotic expression plasmids encoding genomic forms of therapeutic genes, as large as 18 kb, can be successfully incorporated in THLs and delivered to brain following intravenous administration.     

Electroporation enhances reporter gene expression following delivery of naked plasmid DNA to the lung

BACKGROUND: Existing methods of non-viral airway gene transfer suffer from low levels of efficiency. Electroporation has been used to enhance gene transfer in a range of tissues. Here we assess the usefulness of electroporation for enhancing gene transfer in the lungs of mice and sheep. METHODS: Naked plasmid DNA (pDNA) expressing either luciferase or green fluorescent protein (GFP) was delivered to mouse lungs by instillation. Following surgical visualisation, the lungs were directly electroporated and the level and duration of luciferase activity was assessed and cell types that were positive for GFP were identified in lung cryosections. Naked pDNA was nebulised to the sheep lung and electrodes attached to the tip of a bronchoscope were used to electroporate airway segment bifurcations, Luciferase activity was assessed in electroporated and control non-electroporated regions, after 24 h. RESULTS: Following delivery of naked pDNA to the mouse lung, electroporation resulted in up to 400-fold higher luciferase activity than naked pDNA alone when luciferase was under the control of a cytomegalovirus (CMV) promoter. Following delivery of a plasmid containing the human polyubiquitin C (UbC) promoter, electroporation resulted in elevated luciferase activity for at least 28 days. Visualisation of GFP indicated that electroporation resulted in increased GFP detection compared with non-electroporated controls. In the sheep lung electroporation of defined sites in the airways resulted in luciferase activity 100-fold greater than naked pDNA alone. CONCLUSIONS: These results indicate that electroporation can be used to enhance gene transfer in the lungs of mice and sheep without compromising the duration of expression.     

Bacterial magnetic particles (BMPs)-PEI as a novel and efficient non-viral gene delivery system

BACKGROUND: Non-viral methods of gene delivery, especially using polyethylenimine (PEI), have been widely used in gene therapy or DNA vaccination. However, the PEI system has its own drawbacks, which limits its applications. METHODS: We have developed a novel non-viral delivery system based on PEI coated on the surface of bacterial magnetic nanoparticles (BMPs). The ability of BMPs-PEI complexes to bind DNA was determined by retardation of plasmid DNA in agarose gel electrophoresis. The transfection efficiency of BMPs-PEI/DNA complexes into eukaryotic cells was determined by flow cytometric analysis. The MTT assay was invited to investigate the cytotoxicity of BMPs-PEI/DNA complexes. The expression efficiency in vivo of BMPs-PEI bound to the plasmid pCMVbeta encoding beta-galactosidase was evaluated intramuscularly inoculated into mice. The immune responses of in vivo delivery of BMPs-PEI bound plasmid pcD-VP1 were determined by MTT assay for T cell proliferation and ELISA for detecting total IgG antibodies. RESULTS: BMPs-PEI complexes could bind DNA and provide protection from DNase degradation. The transfection efficiency of BMPs-PEI/DNA complexes was higher than that in PEI/DNA complexes. Interestingly, in contrast to PEI, the BMPs-PEI complex was less cytotoxic to cells in vitro. We further demonstrated that the BMPs-PEI system can deliver an exogenous gene to animals and allow it to be expressed in vivo. Such expression resulted in higher levels of humoral and cellular immune responses against the target antigen compared to controls. CONCLUSIONS: We have developed a novel BMPs-PEI gene delivery system with a high transfection efficiency and low toxicity, which presents an attractive strategy for gene therapy and DNA vaccination.     

Transient non-viral cutaneous gene delivery in burn wounds

BACKGROUND: Gene transfer to burn wounds could present an alternative to conventional and often insufficient topical and systemic application of therapeutic agents to aid in wound healing. The goals of this study were to assess and optimize the potential of transient non-viral gene delivery to burn wounds. METHODS: HaCaT cells were transfected with luciferase or beta-galactosidase transgene using either pure plasmid DNA (pDNA) or complexed with Lipofectamine 2000, FuGENE6, or DOTAP-Chol. Expression was determined by bioluminescence and fluorescence. Forty male Sprague-Dawley rats received naked pDNA, lipoplexes, or carrier control intradermally into either unburned skin, superficial, partial, or full-thickness scald burn. Animals were sacrificed after 24 h, 48 h, or 7 days, and transgene expression was assessed. RESULTS: Gene transfer to HaCaT cells showed the overall highest expression for DOTAP/Chol (77.85 ng luciferase/mg protein), followed by Lipofectamine 2000 (33.14 ng luciferase/mg protein). pDNA-derived gene transfer to superficial burn wounds showed the highest expression among burn groups (0.77 ng luciferase/mg protein). However, lipoplex-derived gene transfer to superficial burns and unburned skin failed to show higher expression. CONCLUSIONS: Lipofectamine 2000 and DOTAP/Chol lipoplex showed significantly enhanced gene transfer, whereas no transfection was detectable for naked DNA in vitro. In contrast to the in vitro study, naked DNA was the only agent with which gene delivery was successful in experimental burn wounds. These findings highlight the limited predictability of in vitro analysis for gene delivery as a therapeutic approach.     

In vivo electroporation and ubiquitin promoter--a protocol for sustained gene expression in the lung

BACKGROUND: Gene therapy applications require safe and efficient methods for gene transfer. Present methods are restricted by low efficiency and short duration of transgene expression. In vivo electroporation, a physical method of gene transfer, has evolved as an efficient method in recent years. We present a protocol involving electroporation combined with a long-acting promoter system for gene transfer to the lung. METHODS: The study was designed to evaluate electroporation-mediated gene transfer to the lung and to analyze a promoter system that allows prolonged transgene expression. A volume of 250 microl of purified plasmid DNA suspended in water was instilled into the left lung of anesthetized rats, followed by left thoracotomy and electroporation of the exposed left lung. Plasmids pCiKlux and pUblux expressing luciferase under the control of the cytomegalovirus immediate-early promoter/enhancer (CMV-IEPE) or human polyubiquitin c (Ubc) promoter were used. Electroporation conditions were optimized with four pulses (200 V/cm, 20 ms at 1 Hz) using flat plate electrodes. The animals were sacrificed at different time points up to day 40, after gene transfer. Gene expression was detected and quantified by bioluminescent reporter imaging (BLI) and relative light units per milligram of protein (RLU/mg) was measured by luminometer for p.Pyralis luciferase and immunohistochemistry, using an anti-luciferase antibody. RESULTS: Gene expression with the CMV-IEPE promoter was highest 24 h after gene transfer (2932+/-249.4 relative light units (RLU)/mg of total lung protein) and returned to baseline by day 3 (382+/-318 RLU/mg of total lung protein); at day 5 no expression was detected, whereas gene expression under the Ubc promoter was detected up to day 40 (1989+/-710 RLU/mg of total lung protein) with a peak at day 20 (2821+/-2092 RLU/mg of total lung protein). Arterial blood gas (PaO2), histological assessment and cytokine measurements showed no significant toxicity neither at day 1 nor at day 40. CONCLUSIONS: These results provide evidence that in vivo electroporation is a safe and effective tool for non-viral gene delivery to the lungs. If this method is used in combination with a long-acting promoter system, sustained transgene expression can be achieved.     

Structure of the mouse glutamate decarboxylase 65 gene and its promoter: preferential expression of its promoter in the GABAergic neurons of transgenic mice

GABA is synthesized by glutamate decarboxylase (GAD), which has two forms, GAD65 and GAD67. To elucidate the molecular mechanisms of mouse GAD65 (mGAD65) gene expression, we isolated and characterized the mGAD65 gene. The mGAD65 gene was found to be divided into 16 exons and spread over 75 kb. The sequence of the first exon and the 5'-flanking region indicated the presence of potential neuron-specific cis-regulatory elements. We used transgenic mice to examine the expression pattern conferred by a 9.2-kb promoter-proximal DNA fragment of the mGAD65 gene fused to the bacterial lacZ reporter gene. Transgenic mice showed high beta-galactosidase activity specifically in brain and testis. They also showed characteristic patterns of transgene expression in olfactory bulb, cerebellar cortex, and spinal cord, a similar expression pattern to that of endogenous mGAD65. However, no transgene expression was observed in the ventral thalamus or hypothalamus, in which high mGAD65 gene expression levels have been observed. These results suggest that the 9.2-kb DNA fragment of the mGAD65 gene is associated with its tissue-specific expression and its targeted expression in GABAergic neurons of specific brain regions but that additional regulatory elements are necessary to obtain fully correct expression.     

Enteroendocrine cell expression of a cholecystokinin gene construct in transgenic mice and cultured cells

CCK is predominantly expressed in subsets of endocrine cells in the intestine and neurons in the brain. We evaluated the expression of a CCK gene construct in transgenic mice and cultured cells to identify a genomic region that directs correct tissue- and cell-specific expression in enteroendocrine cells. The CCKL1 transgene contained 6.4 kb of mouse Cck fused to lacZ. Expression was evaluated in three transgenic lines (J11, J12, J14) by measurement of beta-galactosidase in tissue homogenates and frozen sections. Correct tissue-specific expression was observed, with beta-galactosidase activity detected in intestine and brain. However, there were differences seen in cell-specific expression in the intestine. Line J14 exhibited expression in CCK-endocrine cells, with expressing cells arising at the normal time during fetal development. However, transgene expression in line J12 intestine was limited to neurons of the enteric nervous system, which reflect an early fetal expression pattern for CCK. Analysis of an additional 15 transgenic founder mice demonstrated intestinal expression in 40% of transgenics, with expressing mice following either an endocrine cell pattern or a neuronal pattern in approximately equal numbers. CCKL1 transfection analysis in cultured cells also demonstrated enteroendocrine cell expression, with 100-fold enhanced activity in the enteroendocrine cell line STC-1 compared with nonendocrine cell lines. The results suggest that the minimal cis-regulatory DNA elements necessary for appropriate CCK expression in enteroendocrine cells reside within the 6.4-kb mouse genomic fragment.     

PhiC31 integrase mediates integration in cultured synovial cells and enhances gene expression in rabbit joints

BACKGROUND: Gene transfer to synovium in joints has been shown to be an effective approach for treating pathologies associated with rheumatoid arthritis (RA) and related joint disorders. However, the efficiency and duration of gene delivery has been limiting for successful gene therapy for arthritis. The transient gene expression that often accompanies non-viral gene delivery can be prolonged by integration of vector DNA into the host genome. We report a novel approach for non-viral gene therapy to joints that utilizes phage phiC31 integrase to bring about unidirectional genomic integration. METHODS: Rabbit and human synovial cells were co-transfected with a plasmid expressing phiC31 integrase and a plasmid containing the transgene and an attB site. Cells were cultured with or without G418 selection and the number of neo-resistant colonies or eGFP cells determined, respectively. Plasmid rescue, PCR query, and DNA sequence analysis were performed to reveal integration sites in the rabbit and human genomes. For in vivo studies, attB-reporter gene plasmids and a plasmid expressing phiC31 integrase were intra-articularly injected into rabbit knees. Joint sections were used for histological analysis of beta-gal expression, and synovial cells were isolated to measure luciferase expression. RESULTS: We demonstrated that co-transfection of a plasmid expressing phiC31 integrase with a plasmid containing the transgene and attB increased the frequency of transgene expression in rabbit synovial fibroblasts and primary human RA synoviocytes. Plasmid rescue and DNA sequence analysis of plasmid-chromosome junctions revealed integration at endogenous pseudo attP sequences in the rabbit genome, and PCR query detected integration at previously characterized integration sites in the human genome. Significantly higher levels of transgene expression were detected in vivo in rabbit knees after intra-articular injection of attB-reporter gene plasmids and a plasmid expressing phiC31 integrase. CONCLUSION: The ability of phiC31 integrase to facilitate genomic integration in synovial cells and increase transgene expression in the rabbit synovium suggests that, in combination with more efficient DNA delivery methods, this integrase system could be beneficial for treatment of rheumatoid arthritis and other joint disorders.     

In vitro and in vivo evaluation of intrinsic immunogenicity of reporter and insulin gene therapy plasmids

BACKGROUND: Plasmid DNA vectors offer the potential of safe gene therapy avoiding viral vector-mediated toxicity and immunogenicity. As plasmid DNA is bacterial in origin, presence of bacterial lipopolysaccharide (LPS) or unmethylated CpG dinucleotides may stimulate host innate immunity. METHODS: Primary cultures of mouse and rat dendritic cells were established and incubated with bacterial lipopolysaccharide; immunostimulatory CpG oligodeoxynucleotide; control GpC oligodeoxynucleotide; and a range of (pVR1012) plasmids encoding transgenes with increasing CpG content (wild-type and mutant human preproinsulin; non-mammalian eukaryotic eGFP reporter gene; and bacterial beta-galactosidase reporter gene). IL-12 secretion was assayed to determine in vitro plasmid immunogenicity. Local inflammatory response following intramuscular injection of these plasmids, with or without a non-ionic carrier SP1017, was characterised in vivo. RESULTS: Dose-responsive LPS and CpG stimulation of IL-12 secretion from dendritic cells was demonstrated. All plasmids induced significant IL-12 secretion in comparison to control unstimulated cells. The beta-galactosidase plasmid had highest CpG content and induced significantly higher IL-12 secretion than constructs containing a eukaryotic transgene. Injection of rat muscle with the beta-galactosidase construct induced greater inflammatory response than human preproinsulin constructs. This was further enhanced by SP1017. At 2 days post-injection, monocyte/macrophage injection site infiltration predominated with CD8-positive lymphocytes predominating at 7 days. There was no evidence of transgene expression in infiltrating immune cells. CONCLUSIONS: Dendritic cell immunostimulation may be employed as an in vitro bioassay of innate immune response to plasmid DNA vectors during evaluation for clinical gene therapy.