Mutagenesis, Vol. 18, No. 3, 259-264,
May 2003
© 2003 UK Environmental Mutagen Society/Oxford University Press
Evaluating the genetic toxicology of DNA-based products using existing genetic toxicology assays
Molecular Toxicology, Biomedical Sciences, Faculty of Medicine, Imperial College School of Science, Technology and Medicine, London SW7 2AZ, UK and 1 Genetic Toxicology, GlaxoSmithKline, Ware, Herts SG12 0DP, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Unlike the development of drugs based on small chemical entities there are no conventional regulatory toxicity studies established for DNA-based products. As the potential for insertional mutagenesis is of particular concern for gene therapy, we have investigated the mutagenicity of model non-viral DNA-based products at the HPRT locus in Chinese hamster V79 cells. Cultures were transfected with a pL3112BSKS plasmid in combination with a number of non-viral transfection facilitators: Effectene, Lipofectamine 2000 and ExGen 500. The plasmid contains a green fluorescent protein gene, which was used as a reporter of transfection efficiency. Flow cytometry was used to analyse large numbers of cells. Small scale transient transfection efficiencies (7–90%) were obtained at low cytotoxicity, however, scaling up the process led to decreased transfection and increased cytotoxicity. Stable transfection (chromosomal) was observed, but only at very low levels (<1.5%). Two of the non-viral delivery facilitators (Effectene and ExGen 500) themselves induced mutation at the HPRT locus, although they were considerably less potent than the positive control ethyl methanesulphonate. In transfection experiments, neither of the non-viral delivery facilitators (Effectene and Lipofectamine 2000) had a mutagenic effect, whereas with ExGen 500 there was evidence of a mutagenic effect, consistent with the mutagenicity observed with the non-viral transfection facilitator alone. Moreover, treatment with the plasmid pL3112BSKS itself was able to induce a 3- to 7-fold increase in the mutation frequency at the HPRT locus. Our studies highlight some of the problems associated with using exisiting genetic toxicology testing procedures for the assessment of DNA-based products used in novel gene therapy approaches. However, given the limited level of sophistication of the current approach, the data suggest that non-viral gene therapy may present a detectable mutation risk and more appropriate testing strategies are needed to evaluate the nature of the risk.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Gene therapy is the term used for a new generation of therapeutics with the ability to correct inherited or acquired diseases at the molecular level of DNA. To date over 400 clinical trials have been approved world wide for gene therapies aimed at the treatment of various human diseases, ranging from monogenic disorders like cystic fibrosis to multifactorial conditions such as cancer and vascular disease (Mhashilkar et al., 2001
). In the UK, the conduct of clinical trials involving gene therapy is monitored by the Gene Therapy Advisory Committee (GTAC), and within research institutions by Local Ethical Review Committees. The approval of gene therapy trials in the USA is dependent on institutional review by at least two Clinical Ethical Review Boards and also by the Recombinant DNA Advisory Committee (RAC) who advise the US Food and Drug Administration (FDA). Despite these safeguards, there is concern that the conduct and reporting of approved gene therapy trials is not always transparent (Mhashilkar et al., 2001; Sneyers et al., 2001) and that information regarding safety and efficacy are not universally available to regulators, clinicians and scientists world wide. Moreover, unlike the development of drugs based on small chemical entities, which adhere to guidelines agreed by the International Conference on the Harmonisation of Technical Requirements for the Registration of Pharmaceuticals for Human Use (ICH), there is as yet no definitive world wide consensus for the pre-clinical safety assessment of gene therapy products.
In the European Community, the European Agency for the Evaluation of Medicinal Products guidance on the quality, pre-clinical and clinical aspects of gene transfer medicinal products (CPMP/BWP/3088/99) was enforced in 2001. In the USA, the FDA recently published a position document reviewing the safety assessment of DNA-based products (www.fda.gov/cber/gdlns/somegene.txt). A number of specific issues were addressed, including: (i) the integration and persistence of inserted DNA in the genome; (ii) the potential for gonadal disposition, the risk of reproductive toxicity and transmission of altered genetic material to subsequent generations; (iii) immunogenic effects such as ectopic protein production and abnormal immune responses to vectors. The latter issue was dramatically brought to public attention by the recent death of a research subject, Jesse Gelsinger, as a direct result of gene transfer in a clinical trial (Nevin, 2000). However, in terms of genetic toxicology, it is the first two issues that are of primary interest. The integration of DNA-based products into the host cell genome could potentially lead to insertional mutagenesis and, therefore, there is a perceived genotoxic risk from gene manipulation. In theory, this could lead to genetic instability, gene mutation, translocation, gene activation and/or gene inactivation and, in turn, somatic or, in the case of germ cells, inherited mutagenesis and/or carcinogenesis. Some of these concerns have recently been illustrated by the work of Miller et al. (2002) with adeno-associated virus (AAV) vectors. They investigated the sites of provirus integration in transduced HeLa cells and reported that four of nine rescued AAV proviruses were integrated within genes, and one within an exon, and that the integrated vectors were associated with chromosomal deletions and other chromosomal rearrangements. A key issue surrounding these results is whether AAV vectors induce chromosomal instability or integrate into existing breaks. In a separate case, Li et al. (2002) recently showed that a retroviral gene marker, dLNGFR, induced leukaemia in secondary mouse recipients caused by insertional mutagenesis resulting in the activation of the Evi1 oncogene and a synergistic effect with the gene marker resulting in cell signal transduction interference. Lastly, the RAC, which advises the FDA, temporarily halted the progress of a gene therapy trial because the viral vector was found to be present in the semen of a patient (Boyce, 2001). Moreover, during the genesis of this manuscript, two cases of leukaemia were reported in patients receiving gene therapy treatment (see
Nature 419, 545, 2002[Medline] and
Nature 421, 305, 2003[Medline]). These reports demonstrate the unpredictable nature of DNA integration and highlight the potential concerns described by the FDA.
In order for the pharmaceutical industry to bring DNA-based products to the market it is likely that some form of non-clinical safety assessment for insertional mutagenesis will be required to support late phase clinical development and drug registration. The aim of the current study was to establish whether existing pre-clinical genotoxicity assays could be used to evaluate the potential of DNA-based therapies for insertional mutagenesis by measuring mutation induction at the HPRT locus in Chinese hamster V79 cells. Although viral-mediated gene therapy is recognized as a much more efficient gene delivery system, it is also potentially more hazardous. Thus in the first instance, a very simple model of gene therapy was employed, i.e. a non-targeted plasmid vector harbouring an enhanced green fluorescent protein (EGFP) reporter gene and a number of non-viral carrier facilitators [transfection facilitators (TF)], namely Effectene (Qiagen), Lipofectamine 2000 (GibCo BRL) and ExGen 500 (MBI Fermentas) were tested. The results of these studies are presented and discussed in terms of current and future strategies to address this issue.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine serum (FBS), trypsin–EDTA, penicillin/streptomycin and OptimemTM were from Invitrogen (Paisley, UK). All other reagents, except where indicated were from Sigma Aldrich (Poole, UK).
Cell line and culture conditions
The Chinese hamster fibroblast V79 cells were cultured in DMEM supplemented with 10% FBS, 2 mM L-glutamine, 100 IU/ml penicillin and 100 µg/ml streptomycin and maintained at 37°C in a humidified atmosphere of 5% CO2/95% air. Geneticin (0.8 mg/ml) was used to select for G418 resistance in V79 cells transfected with pL3112BSKS.
Plasmids and purification procedure
Plasmids pL3112BSKS and pCAGGSCre were kindly provided by Dr Tadashi Kaname (Clinical Sciences, Imperial College, London, UK). Plasmid pL3112BSKS (9.5 kb) contains the EGFP reporter gene attached to a neomycin resistance gene. pCAGGSCre (5.9 kb) does not contain the same gene cassette and was used for sham transfections. Plasmids were propagated in Escherichia coli JM109 cells and purified using the Qiagen Maxi prep kit, according to the manufacturer’s instructions. A bacterial colony from a freshly streaked LB ampicillin (100 µg/ml) plate was inoculated into a 2 ml starter culture and then 200 µl was transferred to an overnight culture containing 100 µg/ml ampicillin. Plasmids were verified by restriction enzyme digestion.
Transfection of pL3112BSKS and pCAGGSCre using non-viral reagents
Exponentially growing V79 cells were transfected in triplicate with each plasmid separately using various TFs: EffecteneTM (Qiagen), Lipofectamine 2000TM (LP2000) (Invitrogen) and ExGen 500 (MBI Fermentas). The concentrations of plasmid and DNA were derived from optimization experiments carried out on 24-well plates and analysed by flow cytometry. Cells were seeded at a cell density of 1.5x106 (3x106 for LP2000) in T75 flasks 24 h prior to transfection.
The transfection procedure used with Effectene was according to the Qiagen EffecteneTM transfection reagent handbook. Briefly, complexes were prepared by mixing 3.47 µg/flask plasmid with appropriate amounts of Effectene to give DNA:Effectene ratios of 1:10, 1:25, 1:50 and 1:100. Each complex solution was incubated for 10 min at room temperature and transferred to the cells in DMEM, to a final volume of 10 ml. The cells were incubated at 37°C, 5% CO2 for 16 h and then the medium was replaced with fresh, complete DMEM medium and incubated for a further 24 h.
Transfection with LP2000 was carried out according to the manufacturer’s guidelines for transfection of adherent cells. Complexes were prepared by adding 16.54 µg plasmid/flask with the appropriate amount of LP2000 to give DNA:LP2000 ratios of 1:1, 1:3, 1:6 and 1:10. The complexes were diluted with 1.67 ml OptimemTM and incubated for 20 min at room temperature before being transferred to the cells with 8.33 ml DMEM complete medium. The flasks were incubated at 37°C, 5% CO2 for 24 h.
Transfection with ExGen 500 was carried out according to the manufacturer’s guidelines. The plasmid DNA (10 µg/flask) was diluted in 150 mM sodium chloride to a final volume of 400 µl, vortexed and centrifuged. ExGen 500 was prepared using 3, 6, 9 and 12 equivalents in a 90 µl volume using 150 mM sodium chloride. ExGen 500 was added to the DNA and vortexed immediately. The complexes were incubated at room temperature for 10 min and then overlaid onto the cells in 5 ml medium. The flasks were incubated at 37°C, 5% CO2 for 24 h.
Transfection efficiency
After transfection, cells were trypsinized and pelleted by centrifugation at 1000 r.p.m. for 5 min. The cells were resuspended in PBS–EDTA (5 mM) to a final concentration of 1x106 cells/ml, and 250 µl was transferred to flow cytometry (FACs) tubes with propidium iodide (1 µg/ml). The samples were examined with a Beckton Dickenson FACSCalibur machine. The flow cytometry parameters were adjusted and set for the sham transfection sample and then kept constant throughout the experiment. The acquired data were gated such that the cells were scattered in the middle of the forward and side scatter plot and the FL-1 gain was adjusted so that the entire control population resided in the first log on the FL-1 histogram. Viable cells, i.e. those not stained with propidium iodide, were selected and ~10 000 cells were analysed per treatment. The histograms for the transfected population of cells were overlaid on the control histogram (i.e. sham transfection) and the difference between the peak areas on the graph was calculated as the percentage transfection efficiency.
For determination of stable transfectants, trypsinized cells were centrifuged at 1000 r.p.m. for 5 min, resuspended in DMEM complete medium and adjusted to a cell density of 1000 and 10 000 cells/well on 6-well plates with 3 ml of medium and G418 (0.8 mg/ml) solution. The plates were incubated at 37°C, 5% CO2 for a period of 3 weeks with the G418 solution being replaced every 3–4 days. After this incubation period stable colonies were counted and verified as expressing GFP using fluorescence microscopy.
Treatment of cells with transfection facilitator and pL3112BSKS alone
In addition to the transfection assays, V79 cells were also assessed for cytotoxicity and genotoxicity after ‘treatment’ with the non-viral TFs and pL3112BSKS alone. The conditions and incubation times used for the TFs were kept the same as for transfection studies except that the pL3112BSKS was omitted. Treatment with plasmid alone involved adding various amounts of the pL3112BSKS plasmid (0.34–34.7 µg/T75 flask) to replicate cultures and incubating the cells at 37°C for 24 h. The pL3112BSKS plasmid concentration range was selected on the basis that the high concentration would provide a 100-fold increase over the concentration of plasmid used in the transfection study with Effectene.
Cell survival (colony-forming ability)
Cells, treated or transfected, were plated at a density of 100 cells/per well in 6-well plates. The plates were incubated at 37°C, 5% CO2 for a period of 7 days. After this incubation period, colonies were stained with methylene blue in 50% methanol and counted.
Mutagenicity at the HPRT locus
Prior to plating 1.5x106 V79 cells into T75 flasks, they were treated with HAT for 3 days to reduce background hrpt mutant cells. Cells treated with facilitator alone or transfected cells were maintained in exponential growth for 7 days to allow phenotypic expression of induced hprt mutations. Following this period, the cells were trypsinized, counted and plated at a density of 5x105 cells/plate (90 mm diameter) with 10 ml medium supplemented with 5 µg/ml 6-TG to select for hprt- mutants. The plates were incubated at 37°C, 5% CO2 for a minimum of 7 days before the resulting colonies were stained with methylene blue in 50% methanol and counted.
The direct acting mutagen and known rodent carcinogen ethyl methanesulphonate (EMS) (350 µg/ml for 4 h or 200 µg/ml for 24 h) was used as positive control for the cytotoxicity and genotoxicity assays.
Statistical analysis
Data sets were compared using Dunnets t-test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Exponentially growing V79 cells were transfected with plasmid pL3112BSKS, in the absence or presence of the TFs Effectene, LP2000 or ExGen 500, and assayed for transient and stable transfection efficiency, cell survival and mutation induction at the HPRT locus.
Transfection efficiency
Initially, studies were performed in 24-well plates to determine the optimum plasmid concentration to use with each non-viral TF. The highest transfection efficiencies for Effectene, LP2000 and ExGen 500 were obtained with 0.8, 1 and 2 µg pL3112BSKS DNA, respectively (Table I). The optimum plasmid and TF concentrations were scaled up proportionally for transfection in T75 flasks.
|
In large scale transfection studies (T75 flasks), FACs analysis of transfected V79 cells showed that a ratio of 25:1 Effectene:pL3112BSKS produced the best transient transfection efficiency for this TF, i.e. 37.5 ± 1.4% (mean ± SD) (Figure 1
). These data were confirmed by fluorescence microscopy (Figure 2). Stable transfection efficiencies were determined following selection for the integrated neo gene with Geneticin and, once more, a ratio of 25:1 Effectene:DNA produced the highest transfection efficiency, although the frequency of stably transfected cells (0.75%) was much lower than the transient transfection rate (Figure 1, inset).
|
|
The highest transient transfection efficiency was obtained using LP2000 (48.5 ± 1.3%). Optimization of transfection was obtained with a ratio of 3:1 LP2000:DNA (Figure 3
), and the stable transfection frequency was 0.5% (Figure 3, inset). The best transient transfection efficiency with ExGen 500 (32 ± 1.4%) was observed using 6 equiv. of ExGen 500 (Figure 4), and the stable transfection efficiency was 1.4 ± 0.2% (Figure 4, inset). Image analysis of cells transfected using LP2000 and ExGen 500 (data not shown) confirmed the FACs results.
|
|
Colony survival after treatment or transfection
Cytotoxicity (as determined by relative colony survival) was only observed following transfection with Effectene:pL3112BSKS at a ratio of 100:1, giving a decrease in cell survival of 50% (Table II
and Figure 5). Treatment with Effectene alone showed dose-dependent cytotoxicity to V79 cells. Interestingly, treatment was significantly more toxic than transfection at the 25:1, 50:1 and 100:1 concentrations, indicating a protective effect of the DNA:TF complex compared to the TF alone. With both transfection and treatment with LP2000 there was significant cytotoxicity at all TF:plasmid ratios tested (Table II), with no significant difference between treatment and transfection. For ExGen 500 there was again dose-dependent toxicity after both treatment and transfection, and at the highest ratio used (i.e. 12 equiv.) only 47 and 43% of cells survived, respectively (Table II). There was no significant difference between treatment and transfection. In experiments in which cells were treated with pL3112BSKS alone, significantly reduced clonogenicity was again observed, being 78.3 ± 6.7% at the highest concentration used. In all experiments, the positive control EMS (either 350 or 250 µg/ml) was found to be cytotoxic, significantly lowering clonogenicity to between 26 and 53% of the vehicle control.
|
|
Mutation frequency of V79 cells at the HPRT locus
There was no evidence of induced mutation at the HPRT locus after transfection with Effectene (Figure 6
). Interestingly, however, treatment with Effectene alone at the highest dose used did significantly (P < 0.05) increase the mutation frequency (Table II
and Figure 6), but only at very cytotoxic concentrations (i.e. only 15% colony survival). In fact, at this dose the treatment was significantly different compared with transfection, demonstrating that it was the Effectene that was inducing the mutagenic effect and that the presence of the plasmid DNA was able to attenuate this mutagenicity (Table II and Figure 6). As expected, the positive control EMS (350 µg/ml) induced a statistically significant (P < 0.001) increase in the hprt mutation frequency (Figure 6 and Table II).
|
There was no evidence of induced mutation at the HPRT locus after transfection or treatment with the other TF, LP2000. The EMS control (200 µg/ml) again induced a significant change in hprt mutation frequency, 372 ± 118 mutants/106 cells (Table II
). Treatment with ExGen 500 induced a small increase in mutation frequency at high doses compared with the negative control. There was a marginally higher effect with treatment compared with transfection, but this difference was only significant at the highest dose of ExGen 500 employed (i.e. 12 equiv.). After treatment of cells with pL3112BSKS alone there was evidence of mutation at the HPRT locus, which followed an inverse dose–response relationship (Table II). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The aim of the present study was to investigate the use of a conventional genotoxic assay (gene mutation at the HPRT locus) to determine insertional mutagensis as a result of cellular transfection with a DNA plasmid. To our knowledge, this is the first report to systematically assess the mutagenic potential of transfection using the TFs Effectene, LP2000 and ExGen 500. The use of the non-viral TFs produced relatively high transient transfection efficiencies in the optimization studies, i.e. 91, 66 and 45% for Effectene, LP2000 and ExGen 500, respectively. However, the transfection efficiency decreased when the experiments were scaled up to permit transfection of large numbers of cells (i.e. 3–10x106 cells) for the hprt assay (and obtain reliable gene mutation frequency and colony survival data). The maximum transfection efficiencies obtained in the hprt assays were 38, 49 and 32% for Effectene, LP2000 and ExGen 500, respectively. The decrease in transfection efficiency was probably due to the increased area of the culture flasks, as the concentration of the reagents was kept the same.
The stable transfection efficiency for each combination of TF and plasmid was determined by plating transiently transfected cells under selection with the antibiotic Geneticin. The results of the stable transfection studies confirmed the optimum ratios of TF and plasmid observed by FACs analysis of the transient transfection experiments. However, as expected, the stable transfection efficiencies were considerably lower (~65-fold lower) than the transient transfection efficiencies. The results showed that 0.75, 0.5 and 1.4% of cells were stably transfected after 14 days selective (G418) culturing, for Effectene, LP2000 and ExGen 500, respectively.
There was some evidence of cytotoxicity after transfection with the non-viral TF, especially with high ratio (100:1) Effectene:pL3112BSKS, where the transfection efficiency was decreased as a result of cytotoxicity. Interestingly, treatment of V79 cells with Effectene alone was significantly more cytotoxic compared with transfection (i.e. Effectene + plasmid). This may have been due to changes in pH caused when the facilitator complexes with the plasmid vector (Zuidam and Barenholz, 1999). In all cases, the effect of treatment with the non-viral TF was more cytotoxic than transfection. Treatment with plasmid pL3112BSKS alone caused negligible cytotoxicity.
Consistent with the published literature, the positive control, EMS, was mutagenic at the HPRT locus in the V79 cells (O’Donovan, 1990; Davies et al., 1993). The treatment of V79 cells with Effectene and LP2000 was negative for gene mutation at the HPRT locus, except at the highest dose of Effectene (which was highly cytotoxic). Treatment with ExGen 500 appeared to induce a concentration-dependent increase in mutation frequency at the HPRT locus. To our knowledge, this is the first report of the potential mutagenicity of this type of reagent.
The effect of treatment with the plasmid pL3112BSKS (carrying the GFP reporter gene) on gene mutation at the HPRT locus was investigated and naked DNA was found to be mutagenic in V79 cells a low concentrations (
3.47 µg DNA). Transfection with the plasmid using ExGen 500 induced a 2- to 7-fold increase in mutation frequency, similar to that seen with treatment of the cells with ExGen 500 alone. In contrast, there was no significant effect of transfection with the plasmid using the TFs Effectene and LP2000. This may be due, in part, to the low stable transfection efficiency (<1.5%) observed with V79 cells. Clearly, the logistics of cell culture are a significant factor in the analysis of HPRT gene mutations in this type of study, particularly as the assay is based on negative selection. The mutation frequency is determined at a single gene locus in the hprt assay and this reduces the sensitivity of the approach, especially as plasmid insertion would be expected to occur randomly within the genome (based on observations in our own laboratory using FISH to detect plasmid intergration; data not shown).
Current EMEA guidance on DNA vaccines (and viral vectors) specifically require that the extent of integration of a new gene transfer product into the host genome is investigated. The site, distribution, extent, clearance and transcription of vector gene expression desired and undesired (where integration is not intended) should be evaluated, e.g. by quantitative PCR, in situ PCR and reverse transcriptase PCR. The results should be discussed in relation to spontaneous mutation rates if possible. Depending on the extent of integration of DNA into the host genome and the clinical indication, studies may be required to investigate the potential for tumour formation or disruption of normal gene expression. Given the limited level of sophistication of current regulatory genotoxicity procedures, our experiments suggest that their use for the evaluation of novel gene therapies based on DNA products will not be without ambiguity. Therefore, a new generation of assay needs to be identified that is based on more sensitive and direct reporting systems with, ideally, positive selection for mutations so that large cell populations can be analysed.
|
Acknowledgments |
|---|
These studies were funded by the BRSRC and GlaxoSmithKline (Ware, UK) through a CASE studentship awarded to C.C.S.
|
Notes |
|---|
2 To whom correspondence should be addressed. Tel: +44 020 7594 3188; Fax: +44 020 7594 3050; Email: n.gooderham{at}ic.ac.uk
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
-
Boyce,N. (2001) Trial halted after gene shows up in semen. Nature, 414, 677.[Medline]
Davies,M.J., Phillips,B.J., Anderson,D. and Rumsby,P.C. (1993) Molecular analysis of mutation at the HPRT locus of Chinese hamster V79 cells induced by ethyl methanesulphonate and mitomycin C. Mutat. Res., 291, 117–124.[Web of Science][Medline]
Li,Z., Dullmann,J., Schiedlmeier,B. et al. (2002) Murine leukemia induced by retroviral gene marking. Science, 296, 497.
McGarrity,L.J., Dobrovolsky,V.M., Morris,S.M. et al. (2001) Detection of mutation in Chinese hamster ovary cells using a fluorescent reporter. Environ. Mol. Mutagen., 37, 51.[CrossRef]
Mhashilkar,A., Chada,S. and Roth,J.A. (2001) Gene therapy: therapeutic approaches and implications. Biotechnol. Adv., 19, 279–297.[CrossRef][Web of Science][Medline]
Miller,D.G., Rutledge,E.A. and Russell,D.W. (2002) Chromosomal effects of adeno-associated virus vector integration. Nature Genet., 30, 147–148.[CrossRef][Web of Science][Medline]
Nevin,N.C. (2000) What has happened to gene therapy? Eur. J. Pediatr., 159, S240–S242.
O’Donovan,M.R. (1990) Mutation assays of ethyl methanesulphonate, benzidine and benzo[a]pyrene using Chinese hamster V79 cells. Mutagenesis, 5, S9–S13.
Sneyers,M., Dumon,J.-C. and Vaerenbergh,B.V. (2001) Gene therapy clinical trials in Belgium. Hum. Gene Ther., 12, 1361–1365.[CrossRef][Web of Science][Medline]
Zuidam,N.J. and Barenholz,Y. (1999) Characterisation of DNA–lipid complexes commonly used for gene delivery. Int. J. Pharm., 183, 43–46.[CrossRef][Web of Science][Medline]
Received on July 15, 2002; accepted on February 3, 2003.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  F. L. Martin and K. T. Semple Environmental health impacts: occurrence, exposure and significance, Lancaster University, UK, 9-10 September 2003 Mutagenesis, September 1, 2004; 19(5): 423 - 429. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||