ARTICLE pubs.acs.org/Langmuir
DNA Melting Temperature Assay for Assessing the Stability of DNA Polyplexes Intended for Nonviral Gene Delivery Anja Schallon,† Christopher V. Synatschke,‡ Dmitry V. Pergushov,§ Valerie Jer^ome,† Axel H. E. M€uller,‡ and Ruth Freitag*,† †
Chair for Process Biotechnology and ‡Chair for Macromolecular Chemistry II, University of Bayreuth, Germany Department of Polymer Science, School of Chemistry, Moscow State University, Moscow, Russia
§
bS Supporting Information ABSTRACT: Many synthetic polycations have the ability to form complexes with the polyanion DNA, yet only a few, most notably poly(ethylene imine) (PEI), are efficient gene-delivery vehicles. Although a common explanation of this observation relies on the buffering capacity of the polycation, the intracellular stability of the complex may also play a role and should not be neglected. Assays typically used to follow complex formation, however, often do not provide the required information on stability. In this article, we propose the change in the DNA melting temperature observable after complex formation to be a significant indicator of complex stability. For a given DNA/ polycation ratio, changes in the melting temperature are shown to depend on the polycation chemistry but not on the DNA topology or the polycation architecture. Effects of changes in the DNA/polycation ratio as well as the effect of polycation quaternization can be interpreted using the melting temperature assay. Finally, the assay was used to follow the displacement of DNA from the complexes by poly(methacrylic acid) or short single-stranded DNA sequences as competing polyanions.
’ INTRODUCTION The intentional genetic modification of mammalian cells, socalled transfection, is the basis for DNA medicine but also the initial step in the creation of recombinant cell lines (e.g., for the production of protein-based biopharmaceuticals).1 Viruses achieve this goal quite efficiently and hence are the preferred vectors in DNA medicine, but they have limitations with regard to the amount of DNA that can be delivered together with concerns regarding their immunogenicity, pathogenicity, and oncogenicity. Nonviral vectors such as semisynthetic cationic polymers and lipids in general have a better safety profile; however, they are orders of magnitude less effective in transfection.2,3 In addition, polycationic substances are known to exert a certain cytotoxic effect (e.g., refs 4 and 5), presumably via an interaction with anionic cell components such as membrane lipids, an effect that is typically reduced in the case of the corresponding DNA complexes.5,6 Improving nonviral vectors has been an active research area for more than a decade. Progress is slow because of challenges in the area of synthetic polymer chemistry on the one hand and the current difficulties in understanding the intracellular events that are decisive in successful transfection on the other hand. One nonviral transfection agent, discovered in the early 1990s, continues to be the “gold standard” in the field, namely, poly(ethylene imine) (PEI),7 a polycation r 2011 American Chemical Society
known to show an order of magnitude better transfection efficiency than that of most other polycations. Polycationic transfection agents are assumed to facilitate the entry of DNA into the cells by first complexing the polyanionic DNA via electrostatic interaction.8 In particular, when according to common practice an excess of transfection agent is used, the ensuing complexes carry a positive net charge and consequently associate easily with the negatively charged cell surface. The entry of the complexes into the cells is assumed to occur via endocytosis.9 Escape from the endosomes is then necessary to avoid the digestion of the DNA, and here PEI may have an advantage over most other investigated vectors.10 PEI can act as a buffering agent, thereby counteracting the acidification of the endosomes, simultaneously causing a massive water influx and finally the bursting of the vesicle and the release of the complexes and the DNA (proton sponge effect).11,12 Although this hypothesis is convincing and has been the basis of a number of attempts to create improved polycationic vectors, it does not agree with all experimental results. The pH development in endosomal vesicles has been shown to differ from that necessary for the proton sponge effect.13 Other polymers with a similar buffering capability to that Received: May 13, 2011 Revised: July 18, 2011 Published: July 19, 2011 12042
dx.doi.org/10.1021/la201803c | Langmuir 2011, 27, 12042–12051
Langmuir of PEI have failed as transfection agents in the past,14,15 but some polycations, most notably poly(2-(dimethylamino)ethyl methacrylate) (PDMAEMA), a weakly buffering polycation that contains only tertiary amino groups, nevertheless show quite acceptable transfection efficiencies (e.g., ref 16). Moreover, in a recent study, we showed that independent of the frequency of successful genetic modification, complexes made from a variety of different polycations entered the cell and were transported to the nucleus with similar rates and efficiencies.5 This indicates that cellular uptake (and/or endosomal escape) is not the main bottleneck in nonviral transfection. However, we have never been able to detect the polycation within the nucleus of any of the successfully transfected cells. Therefore, we hypothesize that the complex stability and the judicious release of DNA from the complexes in close proximity to the nucleus may present a more important bottleneck for transfection than endosomal escape. In this context, the intracellular stability of the polycationDNA complexes presumably plays a critical role and should be analyzed. The intracellular release of pDNA from the complexes formed in vitro has to involve components from the cytosol. The time and place for this release is crucial to efficient transfection. Complexes that are too stable may not release the DNA at all, whereas complexes that are too weak may release the DNA too early (i.e., before it has reached the nucleus). In both cases, the delivery will fail.17 Most established assays for DNApolycation complex formation, such as the gel retardation assay or the ethidium bromide assay, are useful mainly in determining the needed N/P (polymer-N/DNA-P) ratio for complete complex formation but do not give enough information on complex stability.18 Moreover, the observed effect may also depend on the DNA topology, twist, or purity.19 In this article, we propose that the analysis of the melting temperature of the complexed DNA could be a sensitive indicator of complex stability. As a proof of principle, this method is used to investigate the complexing capability of polycations with different structures and chemistries. In addition, DNA release from the complexes by competing polyanions is evaluated. In these measurements, PEI/DNA-complexes showed peculiarities that could possibly be related to the superior performance of PEI as a gene delivery agent. We conclude that the use of melting temperature changes (ΔT assay) as an indicator of polyplex stability may offer new in vitro screening possibilities for the directed design of improved nonviral transfection agents prior to testing them in cellular systems.
’ MATERIALS AND METHODS Materials. Plastic materials and chemicals including culture media components were obtained from established suppliers and used as received. High-quality water was produced by a Millipore unit. Branched PEI (b-PEI, 25 kDa) was from Sigma-Aldrich, linear PEI (l-PEI, 25 kDa) was from Polysciences, Inc., and linear poly(methacrylic acid) (PMAA, degree of polymerization = 727, PDI = 1.1, Mn = 67.2 kDa) was from Polymer Standards Service. Polymers were prepared as 500 μM stock solutions in Millipore water. Methyliodide (MeI, Fluka, purum, >99.0%), dioxane (p.a. quality, Fisher Scientific), and ethanol (p.a. quality, VWR International) were used as received. For dialysis, a membrane (3500 MWCO, Spectra/Por) of regenerated cellulose was used. PDMAEMA Synthesis and Polycation Quaternization. Starshaped (st-PDMAEMA, 47.5 kDa), linear (l-PDMAEMA, 17 kDa), and highly branched (hb-PDMAEMA, 11.6 kDa) PDMAEMA were
ARTICLE
synthesized via atom-transfer radical polymerization (ATRP) according to previously reported procedures.20,21 For the quaternization of PDMAEMA, 0.52 g (8.81 106 mol of polymer, 3.31 103 mol of nitrogen) of the star-shaped polymer was dissolved in 200 mL of dioxane. Then, 0.470 g (3.31 103 mol, equimolar to the amount of nitrogen) of MeI was added to the solution, and the flask was sealed with a rubber septum. The mixture was stirred at ambient temperature for 2 days before being dialyzed exhaustively against Millipore water. The final product was obtained as a white powder after freeze drying. A 100% quaternization efficiency was determined by 1H NMR spectroscopy.21 PEI quaternization was achieved by dissolving 0.45 g (1.80 105 mol of polymer, 1.04 102 mol of nitrogen) of PEI in 50 mL of ethanol, followed by the addition of 2.963 g (2.09 102 mol, 2-fold excess compared to nitrogen) of MeI and subsequent sealing of the flask with a rubber septum. The solution was stirred at ambient temperature and turned cloudy within a few minutes. When the reaction was stopped after 2 days of continuous stirring, slightly yellow flakes had precipitated in the solution, which immediately dissolved upon the addition of a small amount of water. The clear solution was dialyzed against Millipore water for several days. After the solution was freeze dried, a slightly yellow powder was obtained. On the basis of the C/N ratio obtained from elemental analysis, the quaternization efficiency was approximately 41%. Plasmids. Plasmids pH2B-EGFP (5.1 kb, 53.8% GC22), encoding for nuclei-localized EGFP (enhanced green fluorescent protein) and pTriEx4 (5.2 kb, 45.1% GC, Novagen), were used in the experiments. The plasmids were amplified in E. coli DH5R in TB medium (Roth) to sufficient quantities using standard molecular biology techniques. The plasmid DNAs (pDNAs) were purified using the endofree plasmid kit from Qiagen according to the manufacturer’s instruction. pDNAs were diluted in PCR water (Sigma-Aldrich). The concentration and quality were determined by the A260/280 ratio and by electrophoresis in 0.8% agarose gels buffered in TBE buffer (90 mM Tris base, 90 mM boric acid, 2 mM EDTA, pH 8). Linearization of the plasmid was performed with BamHI (Fermentas) according to the manufacturer’s instruction, and linearized pDNA was thereafter purified by isopropanol precipitation. Preparation of Polycation/DNA Complexes. Our research was driven by the need to understand better the formation and stability of DNA polyplexes intended for nonviral gene delivery. As a consequence, our experiments focused on standard conditions, reagents, and protocols for transfection purposes, and the preparation of polycation/ DNA complexes followed established procedures in molecular biology. In particular, pDNA (2 mg/mL stock solution Milli-Q water) was diluted in 150 mM NaCl solution to an end concentration of 15 μg/ mL (ethidium bromide assay), 25 μg/mL (gel retardation assay, ΔT assay), or 1 mg/mL (μ-DSC). A pH of 6.5 was determined for the pDNA solution. When staining agents were to be used, these were incubated with pDNA before complexation. Polycations were dissolved at a concentration of 50 μM (PEI: concentration of nitrogen = 29 mM, pH 9; st-PDMAEMA: concentration of nitrogen = 15 mM, pH 8) in Milli-Q water and used without further purification. For complex formation, enough of the polycation stock solution to reach the desired N/P (polymer nitrogen/DNA phosphate) ratio was added in a single drop to the pDNA solution and immediately vortex mixed at full speed, followed by incubation at room temperature for 30 min. Afterward, complexes were used without further purification. The indicated N/P ratio represents the molar ratio of nitrogen in the polymers to phosphorus in the pDNA. Ethidium Bromide Assay. pDNA complexation with cationic polyelectrolytes was followed by quenching of the ethidium bromide (EtBr, Roth) fluorescence as described previously.23 Briefly, 15 μg/mL pDNA pre-equilibrated with EtBr (0.4 μg/mL) was complexed with increasing amounts of cationic polymer in a 150 mM aqueous NaCl solution in black 96-well plates (Nunc). The samples were allowed to stabilize for 30 min before the fluorescence was measured using a 12043
dx.doi.org/10.1021/la201803c |Langmuir 2011, 27, 12042–12051
Langmuir
ARTICLE
fluorescence microplate reader (Genios Pro, Tecan) with appropriate filters (Ex 544 (20)/Em 590 (25)). A sample containing only pDNA and EtBr was used to calibrate the machine to 100% fluorescence against a background of 0.4 μg/mL of EtBr in 150 mM aqueous NaCl solution. The percentage of dye displaced upon polycation binding was calculated using the following equation relative displacement ¼ 1
Fobs F0 FDNA F0
where Fobs, F0, and FDNA are the fluorescence intensities of a given sample, the dye in buffer alone, and the dye complexed to pDNA alone. Gel Retardation Assay. The gel retardation assay was carried out in 0.8% (w/v) agarose gels and TBE buffer. The running time was 90 min, and the applied voltage was 80 V. For each polymer/DNA system, complexes were prepared of 5 μg pDNA (25 μg/mL) at the indicated N/P ratios in a 150 mM aqueous NaCl solution and left for 30 min at room temperature for complex formation prior to analysis. After electrophoresis, gels were stained with EtBr (10 μg/mL) and the position of DNA-containing bands was visualized under UV light (254 nm). Turbidity Measurements. For the turbidity measurements, an automatic titrator (Titrando 809, Metrohm) equipped with a turbidity probe (λ0 = 523 nm, Spectrosense, Metrohm) and a temperature sensor (Pt 1000, Metrohm) was used. Stock solutions of the polyelectrolytes were prepared in Milli-Q water. The concentration of both polyion solutions was set to 2 mM with respect to the repeating units of each polymer (base-molar concentration). In this manner, the concentration did not depend on the molecular weight of the polymers. The molar polymer concentration of the solutions was as follows: PMAA = 2.56 μM, PDMAEMA = 6.62 μM, and PEI = 3.45 μM. Samples for the turbidity measurements were prepared by adding the respective polycation solution to the polyanion solution in a homemade glass vessel equipped with a magnetic stirring bar. A polycation/polyanion basemolar ratio of 5 (20 mL/4 mL) was used, yielding slightly turbid solutions that were then titrated with a 4 M aqueous NaCl solution, which was added every 8 s in 50 μL steps. A data point for the turbidity was recorded every 2 s until a clear solution with no significant further change in the transmission was obtained. The amount of NaCl necessary to obtain optically transparent systems, which is a consequence of the dissociation of interpolymer salt bonds between oppositely charged polymeric counterparts, was calculated from the intersection of the linear increase in transmission before and the steady state after clarification of the system. Differential Scanning Microcalorimetry (μ-DSC). The calorimetric measurements were performed with a Setaram μDSC III using closed “batch” cells at a scanning rate of 0.5 °C/min in the temperature range from 20 to 80 °C. Complexes were prepared in a 150 mM aqueous NaCl solution with a concentration of 1 mg/mL DNA at an N/P ratio of 5 for both PEI and PDMAEMA. The 150 mM NaCl solution was used as a reference. ΔT Assay (DNA Melting Temperature Change Assay). For the determination of melting temperatures, the pDNA (25 μg/mL, free or complexed) and the SYBR green I dye (0.083X, Agilent Technologies Stratagene product division) were mixed in a 150 mM aqueous NaCl solution and incubated for 10 min. After the addition of the cationic polymers at a suitable N/P ratio, the solution was vortex mixed for 10 s and incubated for 30 min at room temperature in the dark. The mixture was then placed in RealTime tubes (Agilent Technologies) in a RealTime PCR machine (MX3005P, Agilent Technologies) equipped with an SYBR green filter (BP 504 ( 12 nm), and the temperature was increased from 25 to 95 °C at a rate of 1 °C/min. The melting temperature of the DNA was defined as the inflection point of the sigmoid curve when plotting the relative SYBR green fluorescence versus temperature, with the relative fluorescence given by the ratio of the
Figure 1. Gel retardation assay of pH2B-EGFP (0.5 μg, 25 μg/mL) complexed in 150 mM NaCl with b-PEI and st-PDMAEMA at the indicated N/P ratios. Gel electrophoresis was performed in 1 TBE buffer with 0.8% agarose gels. Noncomplexed pDNA shows typical bands of supercoiled (sc) and open circle (oc) pDNA. The size of the DNA ladder is denoted as base pairs (bp). fluorescence at a given temperature relative to the fluorescence at the starting temperature (T = 25 °C). Displacement of pDNA by Competing Polyanions. To investigate the possible displacement of pDNA from the complexes by another polyanion, poly(methacrylic acid) (PMAA) was added after the complexes had been formed as described above. To decrease the influence of the pH, PBS (137 mM NaCl, 1.47 mM KH2PO4, 2.68 mM KCl, 8.1 mM Na2HPO4, pH 7.4) was used to dilute the PMAA to 50 μM. In addition to PMAA, pDNA displacement was also attempted by short single-stranded (20-mer) oligonucleotides (ssDNA, 50 -GCCTACATACCTCGCTCTGC-30 , MWG Eurofins) and yeast tRNA (80 bp, Sigma), which were both diluted in Milli-Q water and otherwise processed using the same protocol as for PMAA. Statistical Analysis. Group data are reported as the mean ( SD. For transfection results, the Student’s t test was used to determine whether data groups differed significantly from each other. The statistical significance was defined as having P values
