Gene delivery with bisphosphonate-stabilized calcium phosphate

0 downloads 0 Views 578KB Size Report
Nov 24, 2010 - The polyanionic nature and high molecular weight of free nucleic ... complex with the nucleic acid drug to protect the sensitive genetic material from ... acids and calcium phosphate occurs presumably between the calcium ions and .... equal amount of CaCl2 solution containing DNA was slowly added while.
Contents lists available at ScienceDirect

Journal of Controlled Release j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j c o n r e l

Gene delivery with bisphosphonate-stabilized calcium phosphate nanoparticles Elisabeth V. Giger a, Josep Puigmartí-Luis b, Rahel Schlatter a, Bastien Castagner a, Petra S. Dittrich b, Jean-Christophe Leroux a,⁎ a b

Institute of Pharmaceutical Sciences, Department of Chemistry and Applied Biosciences, ETH Zürich, Wolfgang-Pauli-Str. 10, 8093 Zürich, Switzerland Laboratory of Organic Chemistry, Department of Chemistry and Applied Biosciences, ETH Zürich, Wolfgang-Pauli-Str. 10, 8093 Zürich, Switzerland

a r t i c l e

i n f o

Article history: Received 26 August 2010 Accepted 8 November 2010 Available online 24 November 2010 Keywords: Calcium phosphate Nanoparticles Bisphosphonates Gene delivery

a b s t r a c t Nucleic acid drugs are promising new therapeutics, due to their possible applications in a wide variety of diseases and their strong targeting potential and associated lower off-target effects compared to conventional pharmaceuticals. However, their poor intracellular bioavailability and rapid degradation hinder their development as drugs. Therefore, efficient delivery is a major challenge. Various systems have been developed to overcome this problem. The entrapment of genetic material into nanoparticles constitutes a promising approach to increase the in vitro and in vivo transfection activity. Calcium phosphate-DNA co-precipitates have been used for gene delivery for more than 35 years and have the advantage of being nontoxic, easy to produce, and having the ability to complex nucleic acids leading to efficient transfection. Conventional synthetic methods yield particles that are only stable for a short period of time. Herein is proposed a versatile, surfactant-free method to stabilize calcium phosphate-DNA nanoparticles based on the use of poly(ethylene glycol)-functionalized bisphosphonate. The strength of the interaction between the bisphosphonate and the calcium phosphate enabled the formation of stable, but bioresorbable particles of around 200 nm, which exhibited physical stability over several days. Additionally, the nanoparticles revealed good and sustained ability to transfect cells while displaying low toxicity. © 2010 Elsevier B.V. All rights reserved.

1. Introduction The delivery of exogenous genes offers a tremendous but still underexploited therapeutic potential. While promising clinical trials have begun in the 1990s but, there are still no approved gene therapy systems, with the exception of a sole virus-based product sold only in China (Gendicine®). The major reasons for the clinical failure of gene delivery systems are the lack of sustained gene expression and the challenges associated with the development of safe and efficient delivery systems. The polyanionic nature and high molecular weight of free nucleic acids prevent them from efficiently crossing the negatively-charged plasma membrane. Inside the cell, further hurdles such as escape from the endosome and transport to the nucleus, need to be overcome [1,2]. For successful transfection, a delivery system should form a stable complex with the nucleic acid drug to protect the sensitive genetic material from premature degradation. Additionally, the delivery system must be innocuous [2,3]. Different approaches for gene administration have been explored including viral and non-viral (e.g., polyplexes and lipoplexes) carriers and physical methods (e.g., gene gun and electro-

⁎ Corresponding author. Wolfgang-Pauli-Str. 10, HCI H 301, 8093 Zürich, Switzerland. Tel.: + 41 44 633 73 10; fax: + 41 44 633 13 14. E-mail address: [email protected] (J.-C. Leroux). 0168-3659/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jconrel.2010.11.012

poration). Non-viral vehicles have been extensively investigated to improve the pharmacokinetic properties and cellular transport of nucleic acid drugs [4–8]. Depending on the material used, they can circumvent some of the problems associated with viral systems such as the strong immune responses occurring in some patients. The most common non-viral gene delivery systems consists of nanoparticles (NP) that are lipid or polymer-based [9,10]. However, non-viral transfection is less efficient than its viral counterpart, especially under in vivo conditions. Moreover, owing to their polycationic nature, almost all synthetic gene delivery carriers are associated with a significant toxicity, which limits their use in the clinic [11]. Calcium phosphate-DNA co-precipitation is an in vitro transfection method that has been employed for more than 30 years in a wide variety of cell lines [12]. Calcium phosphate is bioresorbable, biocompatible, and has an adsorptive capacity for nucleic acids. Furthermore, the process of preparation is straightforward and the production costs are low. The interaction between nucleic acids and calcium phosphate occurs presumably between the calcium ions and the negatively-charged phosphate groups in the backbone of the nucleic acids [13]. However, calcium phosphate-DNA particles tend to grow quickly, which causes reproducibility problems in vitro. Factors like time of incubation, pH, temperature, or the presence of serum strongly influence the activity of calcium phosphate particles [14,15]. The exact mechanism of transfection using calcium phosphate particles is still not clear. The particles have been shown to enter

GENE DELIVERY

Journal of Controlled Release 150 (2011) 87–93

GENE DELIVERY

88

E.V. Giger et al. / Journal of Controlled Release 150 (2011) 87–93

the cells via endocytosis, and it has been suggested that the Ca2+ ions can themselves act as a membranolytic agents. Nevertheless, the precise release mechanism from the endosome and the nuclear delivery is still not understood [16–19]. Given the intrinsic advantage of calcium phosphate as a condensing material for DNA in comparison to complex and often toxic cationic materials (or other inorganic compounds), it is currently being considered for in vivo gene therapy and other biomedical applications [20–24]. Unfortunately, the large size and instability of the calcium phosphate-DNA particles preclude their parenteral injection for systemic delivery. Several approaches to stabilize calcium phosphate-DNA particles have been investigated [25–31]. For example, addition of increasing concentrations of magnesium to calcium has been shown to slow down particle growth [27]. In addition, Sokolova et al. [28] have coated calcium phosphate NP with DNA, have added an additional layer of calcium phosphate to protect the DNA, and have stabilized them further with another DNA layer to prevent agglomeration. Such a method inhibits the growth of the particles and protects the genetic material, but is relatively complex and may not be suitable for systemic gene delivery owing to the lack of stealth properties. Intravenously injected NP have to be modified to circumvent their rapid uptake by cells of the mononuclear phagocyte system. Poly(ethylene glycol) (PEG) is the most widely used moiety for surface modification of NP. Block copolymers such as PEG-blockpoly(aspartic acid) or PEG-block-poly(methacrylic acid) have been described to stabilize calcium phosphate NP, and provide a biocompatible PEG coating. However the physical stability of such coatings in physiologically relevant media has to be established [25,26,29]. Microemulsion methods have also been described, but they require the use of potentially toxic surfactants to stabilize the NP [32,33]. Recently, Huang and co-workers [30] have presented an elegant but relatively complex strategy which involved the use of organic solvents to prepare stable calcium phosphate NP. The latter were coated with a liposome shell that was further modified with a PEG layer conjugated to a targeting ligand. Bisphosphonates (bp) are used to treat bone diseases such as osteoporosis due to their strong affinity for hydroxyapatite, which is a naturally occurring crystalline form of calcium phosphate and a component of bone tissue. The two phosphonate groups and the socalled “bone hook” hydroxyl group all contribute to the tight interaction with calcium [34,35]. Therefore, the binding of bp to calcium phosphate is stronger than that of surfactants and has been shown to inhibit the aggregation of calcium phosphate NP in vitro [36,37]. Furthermore, treatment with bp is safe and well tolerated [38]. Herein is developed an original solvent-free and surfactant-free procedure, which relies on PEG-bp or other bp derivatives to produce stable calcium phosphate-DNA NP (Fig. 1). The amount of bp derivative used for stabilizing the NP is in the μM range. Because of the nontoxic nature of bp and calcium phosphate and the extreme

simplicity of the preparation process, such a system may represent an attractive and safe alternative for gene transfection. 2. Materials and methods 2.1. Materials 1-hydroxy-PEG750-1,1-bisphosphonic acid (PEG-bp) and 4-hydroxy-N,N,N-trimethyl-4,4-diphosphonobutan-1-ammonium (Amm-bp) were from Surfactis Technologies SAS (Angers, France) (supplementary Scheme 1). HeLa cells were purchased from ATCC (Manassas, VA). Agarose, SYBR® green, Dulbecco's Modified Eagle Medium (DMEM) with GlutaMax™, penicillin/streptomycin, fetal bovine serum (FBS), phosphate buffered saline (PBS), Trypsin-EDTA, and Lipofectamine™ were bought from Invitrogen (Carlsbad, CA). CellTiter® 96 Aqueous Non-Radioactive Cell Proliferation Assay was from Promega (Dübendorf, Switzerland). Label IT® Fluorescein Plasmid Delivery Control was purchased from Mirus (Madison, WI). Green fluorescent protein (GFP)-DNA was generated by cloning the green fluorescent protein into the VR1012 vector that carries a kanamycin-resistance gene (Courtesy of E. Walter), inserted into Top10 Escherichia coli (Invitrogen) and purified using the HiSpeed® Plasmid Maxi Kit (Qiagen, Hilden, Germany). Glacial acetic acid and ethylenediaminetetraacetic acid (EDTA) was from Hänseler (Herisau, Switzerland) and Acros Organics (Morris Plains), respectively. All other chemicals were purchased from Sigma (St. Louis, MO). 2.2. Preparation of calcium phosphate (nano)particles GFP-DNA was diluted in water to a final concentration of 6.66 μg/ml, and mixed at a volume ratio of 10:1 with 2.5 M CaCl2, 4-(2-hydroxyethyl)1-piperazineethanesulfonic acid (HEPES) buffer 2× (280 mM NaCl, 1.5 mM Na2HPO4, 50 mM HEPES, and pH 7.23) in a three-neck flask. An equal amount of CaCl2 solution containing DNA was slowly added while air was introduced into the HEPES buffer. Immediately after the addition of the CaCl2 +DNA solution, the same volume of bp derivatives (final concentrations 6, 10 and 25 μM) was mixed with the calcium phosphateDNA co-precipitate. The procedure for the calcium phosphate control was identical except that no bp derivatives were added to stabilize the particles. The final concentration of GFP-DNA was 10 μg/ml. 2.3. Preparation of calcium phosphate particles on microchip To prepare NP on a microchip, a polydimethylsiloxane chip with 4 channels was used. The CaCl2 solution containing DNA and the HEPES buffer were injected via a syringe pump system into the two middle channels of the microchip. Then, a solution containing 20 μM PEG-bp was used as auxiliary streams (on the side of the chip) in order to stabilize the produced calcium phosphate NP. 2.4. Particle size determination and zeta potential Particle size and zeta potential were determined by dynamic light scattering (DLS) and Doppler laser anemometry, respectively, using a DelsaNano C Particle Analyzer (Beckman Coulter, Krefeld, Germany). The CONTIN method included in the software of the instrument was used to calculate the hydrodynamic diameter of the NP. 2.5. X-ray diffraction (XRD)

Fig. 1. Schematic representation of the preparation of calcium phosphate NP. The fluorescence microscopy picture shows HeLa cells transfected with PEG-NP (10 μM).

XRD measurements of calcium phosphate particles (CP-control) and nanoparticles coated with PEG-bp (PEG-NP) were performed on an X'Pert PRO-MPD (PANalytical B.V., Almelo, the Netherlands) with Cu Kα radiation operated at 39 kV and 45 mA.

2.6. Scanning electron microscopy (SEM) Calcium phosphate (nano)particles were prepared as described above. A droplet of the samples was deposited onto a silicium platelet, which was taped on a grid, and water and salt was removed with blotting paper. SEM was performed on a FEI Quanta 200 FEG microscope (FEI, Eindhoven, The Netherlands). The voltage of the incident electron beam was at 10 kV. 2.7. Determination of the entrapment efficiency To determine the extent of DNA incorporation into the NP, solutions with empty and DNA-loaded NP were prepared as described above. The solutions were centrifuged at 4 °C for 15 min at 15,000 × g. DLS showed that none of the samples scattered light, confirming that all of the NP were in the pellet. The amount of unbound DNA in the supernatant was then determined by measuring its optical density (OD) at 260 nm. NP not containing DNA were used as blank. Entrapment efficiency (EE) was determined using the following equation: EE = (pDNA0 − pDNAS) / pDNA0 × 100, where pDNA0 and pDNAS are the OD of the initial DNA solution and supernatant, respectively. 2.8. Determination of phosphate content The amount of phosphate incorporated in the particles was calculated by assaying the free phosphate concentration after filtration of the particles. Briefly, NP were ultrafiltered on a 100-kDa centrifugal filter unit Amicon Ultra-4 (Millipore AG, Zug, Switzerland) for 1 min at 3000 ×g, and the phosphate content in the filtrate was determined using a colorimetric inorganic phosphate assay (with malachite green) as described elsewhere [39]. 2.9. Serum stability experiments The stability of the encapsulated GFP-DNA was assessed by incubating the particles with complete medium containing 10% FBS (1:4, v/v) for 2.5 h at 37 °C. The (nano)particles were loaded together with 10 μl gel loading solution and 0.6 μl SYBR green on a 0.7% agarose gel. The gel was run in TAE buffer (40 mM Tris base, 1 mM EDTA, 20 mM glacial acetic acid, pH 8.5) for 40 min at 90 V, and the bands of GFP-DNA were visualized by UV transilluminator (Gel Doc™ XR, Bio-Rad, Hercules, CA). As a control, (nano)particles were incubated for the same time period at room temperature.

89

stop trypsinization. Then, the cells were collected and washed with PBS/ 2% FBS. Data for 10,000 cells were collected on a BD FACSCanto™ Flow Cytometer (BD Biosciences, San Jose, California) and data was analyzed using the FlowJo software (Tree Star, Inc., Ashland, OR). Transfection efficiency was defined as percentage of GFP-positive cells. 2.12. Uptake studies in HeLa cells NP were incubated with HeLa cells as described for the transfection experiments, except that the incubation time was limited to 5 h. To track the NP, the latter were prepared using fluorescently-labeled Label IT® Fluorescein Plasmid Delivery Control instead of GFP-DNA. The concentration of DNA used was the same as for the NP containing GFP-DNA. The cells were analyzed by flow cytometry. Using the FlowJo software, uptake was determined as percentage of FITC-positive cells and normalized mean fluorescence values were calculated. 2.13. Cytotoxicity experiments In vitro cytotoxicity was determined by the MTS assay [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4sulfophenyl)-2 H-tetrazolium, inner salt] (CellTiter® 96 Aqueous NonRadioactive Cell Proliferation Assay, Promega). HeLa cells were seeded in a 96-well plate at a density of 4000 cells/well in 100 μl complete medium and cultured for 24 h at 37 °C in a humidified atmosphere containing 5% CO2. The medium was then replaced by 100 μl/well fresh medium and different amounts of (nano)particles and Lipofectamine™ as a control were added to the cells. After a 48 h incubation period, they were rinsed twice with PBS and incubated with 100 μl medium without phenol red plus 20 μl CellTiter 96® Aqueous One Reagent containing the tetrazolium compound MTS for about 2 h. MTS was thereby bioreduced by cells into a formazan product. The absorbance of formazan was measured at 490 nm with a plate reader (Infinite M-200, Tecan, Männedorf, Switzerland). Cell viability was calculated according to the following equation: cell viability [%] = (OD490 sample / OD490 control) × 100, where OD490 sample represents the optical density of the wells treated with calcium phosphate NP (modified and unmodified) and OD490 control represents the wells treated with growth medium only. 2.14. Statistical analysis Results for multiple groups were evaluated by ANOVA and the Holm–Sidak method for multiple comparisons. P b 0.05 was considered significant.

2.10. Cell culture 3. Results and discussion HeLa cells were grown in complete medium (DMEM with GlutaMAX supplemented with 10% FBS and 1% antibiotics (penicillin/streptomycin). Subcultivation was done twice weekly. For transfection and uptake experiments cells were grown to about 80% confluence before harvesting. 2.11. In vitro transfection experiments in HeLa cells The transfection efficiency of bp-coated calcium phosphate NP was assessed in HeLa cells, and compared to that of uncoated particles and Lipofectamine™ lipoplexes. The cells were seeded at a density of 50,000 cells/well into 24-well tissue-treated plates 24 h before transfection. The medium was replaced by medium containing 10 or 50% FBS an hour before transfection. The calcium phosphate NP, control particles, and lipoplexes were added to the cells at a final DNA concentration of 0.5 μg/well, and the incubated for 48 h. Cells were monitored by fluorescence microscopy (Axiovert 35, Zeiss, Jena, Germany) and pictures of cells were taken by confocal laser scanning microscopy using a Leica SP2-AOBS (Leica, Wetzlar, Germany). The cells were then trypsinized with trypsin 0.05% for 5 min, followed by the addition of complete medium to

The method of preparation of the calcium phosphate NP was remarkably straightforward. CaCl2 (containing plasmid DNA) was mixed with HEPES buffer containing sodium phosphate with agitation by a stream of air [40]. The rapidly growing calcium phosphate NP were stabilized by immediate addition of an aqueous solution of PEG-bp. The prepared NP were not further purified since the free ions and DNA do not influence the uptake and transfection process. If needed, the particles could eventually be purified by size exclusion chromatography. The sizes of NP prepared with different concentrations of bp (6, 10, and 25 μM) were determined by DLS (Fig. 2A). NP without DNA had a mean diameter and polydispersity index (PDI) of 130–180 nm and 0.1–0.2, respectively. The size of the PEG-bp-coated NP remained relatively constant over 72 h, confirming the physical stability of the system. The particles covered with another bp derivative, Amm-NP, were slightly larger than the PEG-bp-coated ones (PEG-NP) (160 vs. 140 nm) and were less stable over time. The concentrations of the bp derivatives tested (6, 10, and 25 μM) did not have an influence on the initial particle size. Lower bp concentrations

GENE DELIVERY

E.V. Giger et al. / Journal of Controlled Release 150 (2011) 87–93

GENE DELIVERY

90

E.V. Giger et al. / Journal of Controlled Release 150 (2011) 87–93

led to large aggregates because particle growth was not efficiently inhibited. This is presumably due to insufficient coverage of the NP surface. To determine the influence of phosphate concentration on NP size, particles were prepared in the presence of 1.5, 3 and 6 mM phosphate. Calcium ions are present in excess in the process of calcium phosphate precipitation and therefore only the phosphate concentration in the solution can be varied. Increasing phosphate concentration led to larger particles (supplementary Figure S1), presumably because the precipitation process is faster and larger particles are already present when the bisphosphonate derivative is added. Additionally, a delay of 10 and 30 s between calcium and phosphate mixing and the addition of PEG-bp were introduced. Particle size increased with increasing incubation time suggesting that the bp interrupts the rapid growth of the particles or prevents their aggregation (as seen in Fig. 3). The zeta potential of the particles was determined by laser Doppler anemometry. The PEG-bp decorated NP not containing DNA had slightly negative zeta potentials (−5 to −8 mV depending on the concentration of the bp derivative), while the NP coated with Ammbp possessed a positive zeta potential (11 to 18 mV) (Table 1). XRD measurement of CP-control and PEG-NP revealed that the particles were partly crystalline, with peaks characteristic of hydroxyapatite. The highest intensity peak was at around 31.8° 2Θ. Other peaks were observed at 26.1°, 39.9°, 45.6°, and several smaller peaks were between 47 and 56°. These values correspond to those of hydroxyapatite (JCPDS card no. 74-0566), as reported by other groups when preparing calcium phosphate particles in a similar way [27,41]. CP-control and PEG-NP were further characterized by SEM. It can be seen that the calcium phosphate co-precipitates aggregated into

Fig. 2. Sizes and PDI (♦) of DNA free (A) and GFP-DNA NP (B). Data presented as mean± SD (n = 3).

large assemblies (Fig. 3A), while PEG-NP were more dispersed (Fig. 3B). The SEM images confirmed the sizes of the PEG-NP measured by DLS. Alternatively, NP could be prepared using a 4-channel microfluidic platform. Calcium chloride and HEPES buffer containing phosphate were injected via a syringe pump system into the microchip from the two middle channels. A solution containing PEG-bp was used as auxiliary streams in order to stabilize the calcium phosphate NP. This reaction yields a final suspension that contained NP of ~ 200 nm with a PDI of less than 0.3. These NP were stable for the time tested (96 h) (supplementary Figure S2). NP prepared in the presence of GFP-DNA (6.6 μg/ml) and PEG-bp (10 and 25 μM) were slightly larger than those prepared in the absence of DNA (~220 nm), and were stable for 48 h (Fig. 2B). The PDI of the DNA-containing NP was around 0.2, similar to the DNA-free NP. At a final bp concentration of 6 μM, the DNA-containing NP were stable for less than 24 h. DNA-loaded PEG-NP had a more negative potential (−16 to −21 mV) than the empty NP as would be expected from the incorporation of the negatively-charged nucleic acid (Table 1). Lower bp concentrations resulted in more negative zeta potentials, presumably because of the lower charge shielding of the entrapped DNA. The entrapment efficiency of DNA into the NP was determined by spectrophotometry after separating the unentrapped DNA by centrifugation. Entrapment varied around 70% for bp concentrations comprised between 6 and 25 μM (supplementary Table S1). The DNA loading level was then determined by calculating the amount of NP using the phosphate assay (see Materials and methods section) and the theoretical calcium content for such particles [27]. At a DNA incorporation efficiency of 70%, it was found that 4.7 μg/ml of DNA was entrapped into 68 μg/ml of calcium phosphate, corresponding to a loading of 6.9% (w/w). The serum stability of DNA incorporated in the different NP was tested by incubating them at 37 °C with complete cell medium containing 10% FBS. The DNA embedded in NP and calcium phosphate co-precipitate was still visible by gel electrophoresis after 2.5 h of incubation, while

Fig. 3. Scanning electron micrographs of CP-control particles (A) and PEG-NP (B).

91

Table 1 Zeta potential of NP prepared with different concentrations of bp derivatives. Mean ± SD (n = 3). bp concentration [μM]

6

10

25

PEG-NP [mV] Amm-NP [mV] PEG-NP + DNA [mV] Amm-NP + DNA [mV]

−5.2 ± 1.5 12.0 ± 1.2 −21.0 ± 1.1 −19.1 ± 4.0

−7.5 ± 1.0 11.1 ± 0.7 −17.3 ± 2.2 −17.6 ± 2.0

−7.8 ± 1.2 18.3 ± 6.0 −16.5 ± 2.4 −19.0 ± 5.1

free DNA showed more degradation (supplementary Figure S3), indicating that NP protected DNA, at least partially, from the serum nucleases. To study the cellular uptake of the NP, HeLa cells were incubated for 5 h with NP loaded with fluorescently-labeled DNA (supplementary Figure S4). Longer incubation times were not investigated to avoid bias which could occur as a result of the degradation of the fluorescent DNA. The mean fluorescence intensity and the percentage of cells taking up the NP were determined by flow cytometry. After 5 h of incubation, Amm-NP, CP-control, and lipoplexes showed similarly high normalized mean fluorescence intensity, while the intensity of NP stabilized with 10 μM PEG-bp was about 4-fold lower. This implies that the cells take up fewer NP per cell (supplementary Figures S4B and S5). Hardly any fluorescence could be detected when free labeled DNA was given to the cells. Interestingly, the proportion of cells taking up NP was relatively high for all formulations, although lower for PEG-NP (Figure S3A). It is known that cellular uptake of NP decorated with PEG is less efficient [42]. However, cells did internalize some PEG-NP as shown by the high fluorescent-positive proportion of cells (63%). In this case, the uptake of PEG-NP remained possible because the PEG chain length was relatively short (750 g/mol) [43]. In terms of in vivo applications, a higher PEG chain length would be desirable to ensure long circulation times to the NP. However, it may also further decrease the cell internalization. In such a case the use of active ligand-mediated targeting should be envisaged [44,45]. Transfection experiments in complete medium containing 10% (Fig. 4A and supplementary Figure S6) or 50% (supplementary Figure S7) FBS were performed with NP loaded with a GFP-plasmid. HeLa cells were incubated for 48 h with the NP and the transfection efficiency assessed by flow cytometry. The transfection efficiency of free DNA was below 1% and therefore negligible. Calcium phosphate NP prepared in the presence of 6 and 10 μM of PEG-bp or Amm-bp exhibited transfection efficiency of around 60%, similar to that achieved with the lipoplex formulation Lipofectamine™ (Figs. 1 and 4A). At 25 μM bp, transfection efficiency was lower (20–30%) compared to NP decorated with less bp. At this concentration both bp derivatives seemed to interfere with the transfection process. Although this finding cannot be rationalized at this stage, it could be related to an incomplete release of the intact DNA. In the presence of 50% FBS the transfection efficiency was generally maintained (supplementary Figure S7), though decreased significantly for the PEG-NP prepared with 25 μM bp. To confirm that the transfection was mediated by discrete bp-coated NP, CP-controls were supplemented with free PEG-bp (6, 10, and 25 μM) and the mixtures incubated with the cells (supplementary Figure S7, right side). In this case, the transfection efficiencies at all PEG-bp concentrations were around 20%, and thus significantly lower than PEG-NP prepared in the presence of 6 and 10 μM PEG-bp. Transfection efficiency was comparable to the levels of the CP-control containing no bp, indicating that the observed transfection was not mediated via unbound bp. The effect of the physical stability of the NP on the transfection efficiency was also investigated. Calcium phosphate NP were prepared as described above and incubated at room temperature for 48 h prior to assessing their transfection activity (Fig. 4B). The unstable CP-control could only transfect 3% of the cells 48 h after preparation, indicating that most of the precipitate was too large to enter the cells. This result

Fig. 4. Transfection efficiencies as determined by flow cytometry 48 h after adding the formulated GFP-DNA to HeLa cells. The particles were added to the cells immediately (A) or 48 h after their preparation (B). Data presented as mean ± SD (n = 3–5). *The difference in transfection efficiency of Lipofectamine and PEG-NP compared to CP-control, Amm-NP and free DNA is significant (p ≤ 0.05).

correlates with a previous study where it was found that 20 min postpreparation, calcium phosphate-DNA transfection efficiency was 3–5% [14]. PEG-NP showed the same high transfection efficiency (65%) as when they were used immediately after preparation, confirming their physical stability for at least 48 h. Amm-NP, which were less stable than PEG-NP transfected the cells with a lower efficiency (8%). Physical stability is an important advantage for in vivo use as the NP do not have to be used immediately after preparation. Finally, the toxicity of the different formulations was tested on HeLa cells (Fig. 5). Interestingly, PEG- and Amm-NP displayed an overall lower cytotoxicity than CP-control and the Lipofectamine™ lipoplex. At low concentrations, similar to those used for the transfection experiments (first data points in Fig. 5), cell viability of the bp-NP remained high (90–95%). This result was comparable to that of CP-control but substantially higher than that observed with Lipofectamine™ (40%). Furthermore the free bp derivatives had no influence on cell viability at all tested concentrations (16, 160, and 1600 μM), which were up to ~1000 times higher than that used for the transfection studies (supplementary Figure S8). These findings further confirm that the bp-NP are safe, at least in vitro. 4. Conclusion In conclusion, this work shows that stable calcium phosphate-DNA NP could be produced by using bp derivatives. These NP could be manufactured either manually or via microfluidic methods. More

GENE DELIVERY

E.V. Giger et al. / Journal of Controlled Release 150 (2011) 87–93

GENE DELIVERY

92

E.V. Giger et al. / Journal of Controlled Release 150 (2011) 87–93

Fig. 5. Cytotoxicity of PEG-NP (●), Amm-NP (○), CP-control (▼), and Lipofectamine (△) as a function of DNA concentration. Data presented as mean ± SD (n = 4). The concentration used for the transfection experiments was 0.9 μg/ml.

importantly, the bp-NP were stable over time and could transfect efficiently cells in vitro using physiologically relevant media. As opposed to conventional calcium phosphate-DNA precipitate, the transfection efficiency was preserved upon short term storage and the NP were found to be substantially less cytotoxic than the lipoplexes. Further experiments will aim at investigating the impact of PEG length on PEG-NP activity as well as the addition of a targeting ligand to trigger receptor-mediated endocytosis of the NP. This work has implication beyond the field of nucleic acid delivery and could also prove useful for in vivo imaging approaches involving calcium phosphate NP. Acknowledgments Dr. K. Kunze is acknowledged for his support with the SEM experiments. The authors thank Mr. Michael Schinhammer for help with XRD measurements and Ms. Lorine Brülisauer for her help with the confocal laser scanning microscopy experiments. Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10.1016/j.jconrel.2010.11.012. References [1] J. Alper, Drug delivery: breaching the membrane, Science 296 (2002) 838–839. [2] D. Luo, W.M. Saltzman, Synthetic DNA delivery systems, Nat. Biotechnol. 18 (2000) 33–37. [3] D.J. Glover, H.J. Lipps, D.A. Jans, Towards safe, non-viral therapeutic gene expression in humans, Nat. Rev. Genet. 6 (2005) 299–310. [4] P. Xu, G.K. Quick, Y. Yeo, Gene delivery through the use of a hyaluronateassociated intracellularly degradable crosslinked polyethyleneimine, Biomaterials 30 (2009) 5834–5843. [5] J.J. Green, B.Y. Zhou, M.M. Mitalipova, C. Beard, R. Langer, R. Jaenisch, D.G. Anderson, Nanoparticles for gene transfer to human embryonic stem cell colonies, Nano Lett. 8 (2008) 3126–3130. [6] X. Zhang, K.K. Sharma, M. Boeglin, J. Ogier, D. Mainard, J.-C. Voegel, Y. Mély, N. Benkirane-Jessel, Transfection ability and intracellular DNA pathway of nanostructured gene-delivery systems, Nano Lett. 8 (2008) 2432–2436. [7] M. Elsabahy, N. Wazen, N. Bayó-Puxan, G. Deleavey, M. Servant, M.J. Damha, J.-C. Leroux, Delivery of nucleic acids through the controlled disassembly of multifunctional nanocomplexes, Adv. Funct. Mater. 19 (2009) 3862–3867. [8] M.-A. Yessine, M.-H. Dufresne, C. Meier, H.-U. Petereit, J.-C. Leroux, Protonactuated membrane-destabilizing polyion complex micelles, Bioconjug. Chem. 18 (2007) 1010–1014. [9] S.D. Li, L. Huang, Gene therapy progress and prospects: non-viral gene therapy by systemic delivery. Gene Ther. 13 (2006) 1313-1319. [10] S.D. Patil, D.G. Rhodes, D.J. Burgess, DNA-based therapeutics and DNA delivery systems: a comprehensive review, AAPS J. 7 (2005) E61–77.

[11] M. Ogris, S. Brunner, S. Schüller, R. Kircheis, E. Wagner, PEGylated DNA/transferrinPEI complexes: reduced interaction with blood components, extended circulation in blood and potential for systemic gene delivery, Gene Ther. 6 (1999) 595–605. [12] F.L. Graham, A.J. Van der Erb, A new technique for the assay of infectivity of human adenovirus 5 DNA, Virology 52 (1973) 456–467. [13] M. Okazaki, Y. Yoshida, S. Yamaguchi, M. Kaneno, J.C. Elliott, Affinity binding phenomena of DNA onto apatite crystals, Biomaterials 22 (2001) 2459–2464. [14] M. Jordan, A. Schallhorn, F.M. Wurm, Transfecting mammalian cells: optimization of critical parameters affecting calcium-phosphate precipitate formation, Nucleic Acids Res. 24 (1996) 596–601. [15] J.V. O'Mahoney, T.E. Adams, Optimization of experimental variables influencing reporter gene expression in heptoma cells following calcium phosphate transfection, DNA Cell Biol. 13 (1994) 1227–1232. [16] A. Loyter, G. Scangos, D. Juricek, D. Keene, F.H. Ruddle, Mechanisms of DNA entry into mammalian cells: II. Phagocytosis of calcium phosphate DNA co-precipitate visualized by electron microscopy, Exp. Cell Res. 139 (1982) 223–234. [17] E. Orrantia, P.L. Chang, Intracellular distribution of DNA internalized through calcium phosphate precipitation, Exp. Cell Res. 190 (1990) 170–174. [18] V. Sokolova, A. Kovtun, R. Heumann, M. Epple, Tracking the pathway of calcium phosphate/DNA nanoparticles during cell transfection by incorporation of redfluorescing tetramethylrhodamine isothiocyanate-bovine serum albumin into these nanoparticles, J. Biol. Inorg. Chem. 12 (2007) 174–179. [19] S. Zaitsev, I. Buchwalow, A. Haberland, S. Tkachuk, I. Zaitseva, H. Haller, M. Böttger, Histone H1-mediated transfection: role of calcium in the cellular uptake and intracellular fate of H1-DNA complexes, Acta Histochem. 104 (2002) 85–92. [20] J. Schwiertz, A. Wiehe, S. Gräfe, B. Gitter, M. Epple, Calcium phosphate nanoparticles as efficient carriers for photodynamic therapy against cells and bacteria, Biomaterials 30 (2009) 3324–3331. [21] E.İ. Altinoğlu, T.J. Russin, J.M. Kaiser, B.M. Barth, P.C. Eklund, M. Kester, J.H. Adair, Near-infrared emitting fluorophore-doped calcium phosphate nanoparticles for in vivo imaging of human breast cancer, ACS Nano 2 (2008) 2075–2084. [22] M. Kester, Y. Heakal, T. Fox, A. Sharma, G.P. Robertson, T.T. Morgan, E.İ. Altinoğlu, A. Tabaković, M.R. Parette, S.M. Rouse, V. Ruiz-Velasco, J.H. Adair, Calcium phosphate nanocomposite particles for in vitro imaging and encapsulated chemotherapeutic drug delivery to cancer cells, Nano Lett. 8 (2008) 4116–4121. [23] T.T. Morgan, H.S. Muddana, E.İ. Altinoğlu, S.M. Rouse, A. Tabaković, T. Tabouillot, T. J. Russin, S.S. Shanmugavelandy, P.J. Butler, P.C. Eklund, J.K. Yun, M. Kester, J.H. Adair, Encapsulation of organic molecules in calcium phosphate nanocomposite particles for intracellular imaging and drug delivery, Nano Lett. 8 (2008) 4108–4115. [24] M. Epple, K. Ganesan, R. Heumann, J. Klesing, A. Kovtun, S. Neumann, V. Sokolova, Application of calcium phosphate nanoparticles in biomedicine. J. Mater. Chem. 20 (2010) 18-23. [25] Y. Kakizawa, S. Furukawa, K. Kataoka, Block copolymer-coated calcium phosphate nanoparticles sensing intracellular environment for oligodeoxynucleotide and siRNA delivery, J. Control. Release 97 (2004) 345–356. [26] Y. Kakizawa, S. Furukawa, A. Ishii, K. Kataoka, Organic–inorganic hybridnanocarrier of siRNA constructing through the self-assembly of calcium phosphate and PEG-based block aniomer, J. Control. Release 111 (2006) 368–370. [27] E.H. Chowdhury, M. Kunou, M. Nagaoka, A.K. Kundu, T. Hoshiba, T. Akaike, Highefficiency gene delivery for expression in mammalian cells by nanoprecipitates of Ca–Mg phosphate, Gene 341 (2004) 77–82. [28] V.V. Sokolova, I. Radtke, R. Heumann, M. Epple, Effective transfection of cells with multishell calcium phosphate-DNA nanoparticles, Biomaterials 27 (2006) 3147–3153. [29] Y. Kakizawa, K. Kataoka, Block copolymer self-assembly into monodispersive nanoparticles with hybrid core of antisense DNA and calcium phosphate, Langmuir 18 (2002) 4539–4543. [30] J. Li, Y.-C. Chen, Y.-C. Tseng, S. Mozumdar, L. Huang, Biodegradable calcium phosphate nanoparticle with lipid coating for systemic siRNA delivery. J. Control. Release 142 (2010) 416-421. [31] M. Zhang, K. Kataoka, Nano-structured composites based on calcium phosphate for cellular delivery of therapeutic and diagnostic agents, Nano Today 4 (2009) 508–517. [32] S. Bisht, G. Bhakta, S. Mitra, A. Maitra, pDNA loaded calcium phosphate nanoparticles: highly efficient non-viral vector for gene delivery, Int. J. Pharm. 288 (2005) 157–168. [33] I. Roy, S. Mitra, A. Maitra, S. Mozumdar, Calcium phosphate nanoparticles as novel non-viral vectors for targeted gene delivery, Int. J. Pharm. 250 (2003) 25–33. [34] H. Fleisch, Bisphosphonates: mechanisms of action, Endocr. Rev. 19 (1998) 80–100. [35] R.G.G. Russell, M.J. Rogers, Bisphosphonates: from the laboratory to the clinic and back again, Bone 25 (1999) 97–106. [36] H. Fleisch, R.G.G. Russell, D.F. Marion, Diphosphonates inhibit hydroxyapatite dissolution in vitro and bone resorption in tissue culture and in vivo, Science 165 (1969) 1262–1264. [37] N.M. Hansen, R. Felix, S. Bisaz, H. Fleisch, Aggregation of hydroxyapatite crystals, Biochim. Biophys. Acta 451 (1976) 549–559. [38] P.D. Papapetrou, Bisphosphonate-associated adverse events, Hormones 8 (2009) 96–110. [39] S.G. Carter, D.W. Karl, Inorganic phosphate assay with malachite green: an improvement and evaluation, J. Biochem. Biophys. Methods 7 (1982) 7–13. [40] R.E. Kingston, C.A. Chen, H. Okayama, in: J.S. Bonifacino, M. Dasso, J.B. Harford, J. Lippincott-Schwartz, K.M. Yamada (Eds.), Curr Protoc Cell Biol., vol. 20, 2003. [41] D. Olton, J. Li, M.E. Wilson, T. Rogers, J. Close, L. Huang, P.N. Kumta, C. Sfeir, Nanostructured calcium phosphates (NanoCaPs) for non-viral gene delivery: influence of the synthesis parameters on transfection efficiency, Biomaterials 28 (2007) 1267–1279.

[42] R. Gref, Y. Minamitake, M.T. Peracchia, V. Trubetskoy, V. Torchilin, R. Langer, Biodegradable long-circulating polymeric nanospheres, Science 263 (1994) 1600–1603. [43] T. Yuda, K. Maruyama, M. Iwatsuru, Prolongation of liposome circulation time by various derivatives of polyethyleneglycols, Biol. Pharm. Bull. 19 (1996) 1347–1351.

93

[44] T.M. Allen, Long-circulating (sterically stabilized) liposomes for targeted drug delivery, Trends Pharmacol. Sci. 15 (1994) 215–220. [45] M.C. Woodle, Sterically stabilized liposome therapeutics, Adv. Drug Deliv. Rev. 16 (1995) 249–265.

GENE DELIVERY

E.V. Giger et al. / Journal of Controlled Release 150 (2011) 87–93