Research Article For reprint orders, please contact: [email protected]
Efficient targeted gene delivery by a novel PAMAM/DNA dendriplex coated with hyaluronic acid
Aim: To design and develop a novel target-specific DNA-delivery system using hyaluronic acid (HA) –polyamidoamine (PAMAM) conjugates (P–HA). Materials & methods: The coupling of HA to the PAMAM dendrimer was analyzed by 1H-NMR and elemental analysis (CHN). Their properties were characterized in terms of size and zeta-potential and evaluated for in vitro and in vivo transfection efficiency. Results: The designed covalent HA-dendriplexes enhanced gene transfection of pCMV-Luc reporter gene in overexpressing CD44-receptor cancer cells. They were also more efficient in transfecting MDA-MB231 cells than conventional PEI-polyplexes. The cytotoxicity of the covalent HA-dendriplexes was lower than when using conventional polyethylenimine-polyplexes. In vivo studies showed that these targeted complexes were also efficient for delivering pCMVLuc in different organs of healthy mice, as well as in tumors of C57BL/6 animals. Conclusions: The HA-dendriplexes developed in this work may offer an advantageous alternative to conventional cationic polymer-based formulations for DNA delivery into cancer cells in an efficient and safe manner.
Koldo Urbiola1, Carmen Sanmartín2, Laura BlancoFernández1 & Conchita Tros de Ilarduya1 Department of Pharmacy & Pharmaceutical Technology, School of Pharmacy, University of Navarra, C/Irunlarrea 1, 31080 Pamplona, Spain 2 Department of Organic & Pharmaceutical Chemistry, University of Navarra, Spain *Author for correspondence: Tel.: +34 948 425 600 (ext. 80 6375) ctros@ unav.es 1
Original submitted 13 February 2013; Revised submitted 24 February 2014 Keywords: gene therapy • gene transfer • hyaluronic acid/hyaluronan • nanoparticle • nanotechnology • polyamidoamine dendrimer
Gene therapy focuses on the delivery of therapeutic genes to cells and promises considerable advances in the treatment of several important diseases. However, there is an important need to improve the activity of gene delivery aimed vectors. Because viral vectors are related with toxicity or excessive immune responses [1] , in recent years polymer nanocarriers that can be used as nonviral gene vectors have aroused high interest in DNA delivery. In this respect, cationic, polymer-based, non-viral delivery systems provide several advantages over viral vectors, including simplicity of production and the ability to package plasmids of any size; however, the lack of biodegradability and targeted delivery and the relatively high cytotoxicity of some of them, such as polyethylenimine (PEI), remain of concern. Polyamidoamine (PAMAM) dendrimers, described by Tomalia et al. [2] are positive-
10.2217/NNM.14.45 © 2014 Future Medicine Ltd
charged dendrimers. PAMAM dendrimers have attracted great interest because of their low polydispersity index, controlled molecular weight and versatility [3] and have been successfully developed as gene and drug carriers and imaging agents [4–6] . They also have a large tolerable dose range between transfection and cytotoxicity. Early studies by Szoka et al. [7] showed that dendrimers above generation 4 (G4) can efficiently transfect different tumor cell lines in vitro. Cationic PAMAM dendrimers with amine terminal groups are known as good condensing agents of DNA or RNA by electrostatic interactions, leading to the formation of more compact particles termed dendriplexes. These structures are able to protect the plasmid DNA against degradation and facilitate the endosome escape by proton sponge effect [7,8] .
Nanomedicine (Epub ahead of print)
part of
ISSN 1743-5889
Research Article Urbiola, Sanmartín, Blanco-Fernández & Tros de Ilarduya For reprint orders, please contact: [email protected] Hyaluronic acid (HA) is a ubiquitous high molecular weight glycosaminoglycan and the principal compound of the extracellular matrix. It is biodegradable, has low toxicity and is a biocompatible polyanion formed by the repetition of the dimer d-glucuronic acid and N-acetyl-d-glucosamine. The ability of HA to interact electrostatically, shield surface charge, act as a targeting agent, and improve biodistribution or the toxicity profile have focused interest on it. It is one of the few polyanions that can coat DNA/polycation complexes without disrupting their structures. On the other hand, it has been approved for injection by the US FDA. A receptor for HA, CD44, is known to be overexpressed on various tumor cell surfaces [9] and in many solid tumor cells including lung, breast, colorectal, gastric, pancreatic, renal, hepatocellular and cervical cancer, as well as melanoma [10] . It has been also implicated in metastatic spread in various cancer types [11] . The degree of intracellular uptake of HA-complexes depends on the receptor density of the cells. Higher CD44 receptor density correlates with higher uptake of HA-complexes. However, the normal cells with low CD44 receptors will not be affected by HA-complexes [12] . This property makes this system tumor-selective. The carboxyl groups of HA are known to be the recognition sites for HA receptors and hyaluronidase [13] and they are essential for transcriptional activity [14] . In this respect, the cellular uptake of hyaluronan-coated DNA/polyplexes via receptor-mediated endocytosis is considered the most favorable mechanism for therapeutic applications. Since HA presents interesting features for improving transfection efficacy with nonviral systems, its use has been explored in the last years for the efficient delivery of pDNA and RNA. It could work both as a protecting coat against blood components and as a ligand for targeted cells. Some studies reported that addition of hyaluronate improved the uptake of genes complexed to different polycationic carriers via the HA-receptors on the surfaces of tumor cells [15–17] . This fact has been principally explored for polyethylenimine (PEI)-based polyplexes [18–24] and it has been also used for targeting dioleoyl phosphatidylethanolamine (DOPE)-based lipoplexes [25] . Nevertheless, addition of HA to PAMAM based-formulations has not been widely explored. Han et al. [26] described an alteration in the intracellular distribution of doxorubicin by the siRNA/PAMAM/HA complex and Chen et al. [27] described the effect of the mixing orders of siRNA/hyperbranched poly(amido amine)/HA on transfection activity. In our group we have previously evaluated the use of different nontargeted PAMAM dendriplexes for gene delivery [28,29] . Now, the aim of this work is to design
10.2217/NNM.14.45
Nanomedicine (Epub ahead of print)
and evaluate a novel polymeric gene delivery vector designed with the objectives of reducing the toxicity of PAMAM, and increasing specific interactions with tumor cells. In this way, we have developed a nanocarrier containing PAMAM–HA and DNA to achieve an enhanced therapeutic effect. Material & methods Materials
High-molecular-weight hyaluronic acid (HA; sodium salt, 1600 kDa), the cationic polyamidoamine dendrimer (PAMAM-G5, 28,825 kDa, 128 primary amines) and branched polyethylenimine (bPEI, 25 kDa) were provided by Sigma-Aldrich (Madrid, Spain). The plasmid pCMVLuc-VR1216 (BioServe Technologies, MD, USA) encoding luciferase gene was amplified in Escherichia coli, isolated and purified using an EndoFree® QUIAGEN Plasmid Giga Kit (QUIAGEN, Germany) and used in the transfection studies. Plasmid purity was confirmed by agarose gel electrophoresis followed by ethidium bromide staining. The pCpG-hCMV-SCEP-eFluc plasmid (kindly provided by Manfred Ogris) encodes an optimized firefly luciferase cDNA under the CMV-EF1alpha hybrid promoter (SCEP) with hCMV enhancer. HEPES glucose buffer (HEPES 10 mM, 10% glucose, pH 7.4; BHG) was prepared from D(+)glucose and N-(2-hydroxyethyl) piperazine-N′-[2-ethanesulfonic acid] (Sigma-Aldrich, Madrid, Spain). Alamar Blue Dye was purchased from Accumed Technologies (OH, USA). Cell culture
B16F10 (murine melanoma), MDA-MB231 (human breast adenocarcinoma), MCF-7 (human breast adenocarcinoma) and Neuro2A (murine neuroblastoma) cell lines were obtained from the American Type Culture Collection (MD, USA). Cell lines were maintained at 37ºC under 5% CO2 in Dulbecco’s Modified Eagle Medium (DMEM) GlutaMAX™, supplemented with 10% (v/v) heat-inactivated fetal bovine serum and penicillin (100 U/ml) and streptomycin (100 μg/ml, Gibco BRL Life Technologies, Barcelona, Spain). Cells were trypsinized twice a week. Synthesis, purification & characterization of PAMAM–HA conjugates
PAMAM–HA (P–HA) conjugates were prepared as follows: first, a solution of HA in distilled water was oxidized by the addition of potassium peryodate (1:2 molar ratio) until the hydroxyl is oxidized to carbonyl groups at room temperature for 48 h. After that, ethylene glycol (10%) was added to stop the reaction
future science group
Efficient targeted gene delivery by a novel PAMAM/DNA dendriplex coated with hyaluronic acid
Research Article
For reprint orders, please contact: [email protected] and the total volume was dialyzed (membrane pore size 12,000 Da) at 4°C with NaCl 0.1 M for 24 h and distilled water for 24 h more before lyophilization. Second, methyl alcohol was removed by rotary evaporation of the commercial PAMAM solution. Then, distilled water was added to the film in order to get an aqueous solution of PAMAM and oxidized HA was added with mechanic stirring for 48 h. After this time, sodium borohydride was added in order to reduce the unreacted HA (1:1 molar ratio PAMAM:sodium borohydride) and finish the amination reaction. Before lyophilization, the final product was dialized for 24 h against NaCl 0.1 M and for 24 h with distilled water. Finally, it was purified by centrifugation-based ultrafiltration with an Amicon (100,000 MWCO; Millipore, MA, USA) in order to remove the unreacted PAMAM and to avoid the formation of large aggregates. The new covalent compound was characterized by 1H NMR. The spectra were recorded on a Bruker 400 Ultrashield™ spectrometer (Rheinstetten, Germany) using tetramethylsilane (TMS) as the internal standard. Proton chemical shifts are reported in ppm relative to internal TMS (0.0 ppm). All NMR spectra were acquired at ambient temperature using D2O as solvent. Elemental microanalyses were carried out on vacuum-dried samples using a LECO CHN-900 elemental analyzer with carbon, hydrogen and nitrogen determinator. Preparation of HA–PAMAM–DNA dendriplexes
Electrostatically assembled nanoparticles were formed by adding, in this order: BHG, pDNA, the desired amount of HA (1, 5 or 10 μg HA/μg DNA) and PAMAM G5. The P–HA covalent dendriplexes were formed by mixing equal volumes of PAMAMG5 solution and pDNA. When P–HA was added, PAMAM-G5 was partially substituted by increasing amounts of P–HA; the degree of replacement is presented as a percentage (0, 16, 25, 50 and 100%) of P–HA. After the addition of the components, nanoparticles were incubated for 15 min at room temperature. Complexes were formulated at N/P 6 ratio (positive charges of polymer/negative charges of pDNA). Previous studies by our group indicated that at this ratio, DNA was completely condensed by PAMAM in order to obtain small and stable nanoparticles [28] .
were diluted in buffer HEPES–glucose (BHG) and measured at least three times. Gel retention studies
For gel retention studies, naked DNA, plain, electrostatic dendriplexes (1, 5 or 10 μg HA/μg DNA) and complexes with HA covalently attached (16, 25, 50 or 100% of P-HA) were prepared at a final concentration of 50 μg/ml as previously described and electrophoresed through a 0.8% agarose gel using Tris–Borate–EDTA buffer (100 mM Tris, 90 mM boric acid, 1 mM EDTA, pH 8.4) with ethidium bromide. The gel was run for 30 min at 100 mV and visualized under UV illumination using a camera Doc 2000 (BioRad, CA, USA). DNase I protection assay
Naked DNA (5 μg), plain, electrostatic dendriplexes containing 1, 5 or 10 μg HA/μg DNA and complexes with HA covalently attached (16, 25, 50 or 100% of P–HA) were prepared as previously described at a final concentration of 50 μg/ml in 100 μl of total volume. DNAse I (Invitrogen; 1 U/μg DNA) was added to each sample and the mixtures were incubated at 37°C for 30 min. 6 μl of EDTA (0.25 M) were added to inactivate the DNase I. Dendriplexes were disassembled by adding 5 μl of SDS (15%) and were incubated for 5 min at room temperature. Then, the samples were electrophoresed 45 min at 80 mV. The integrity of the plasmid in each formulation was compared with untreated DNA as control. CD44 expression
The expression of CD44 receptors was assured by flow cytometry in different cell lines. B16F10, MDAMB231, MCF-7 and Neuro2A cells were trypsynized and washed twice with phosphate-buffered saline (PBS). Then, 500,000 cells were incubated with anti-human/mouse CD44 FITC (eBiosciences, CA, USA) following manufacturer instructions in PBS for 30 min at 4°C in darkness. After that, cells were washed three times and fluorescence was measured by flow cytometry (FACSCalibur™, BD Biosciences, CA, USA). At least 10,000 events were collected for each measurement. Cells without staining were introduced as control. In vitro transfection activity
Particle size & surface charge determinations
Particle size was determined by dynamic light scattering and the overall charge by zeta-potential measurements using a particle analyzer (Zetasizer Nano, Malvern Instruments, Malvern, UK). Final pDNA concentration was 10 μg/ml. Samples of complexes
future science group
For each transfection, 100,000 cells were seeded in 1 ml of medium in 48-well culture plates and incubated for 24 h at 37ºC in 5% CO2. The primary growth medium was removed and replaced with 0.3 ml of medium containing 10% FBS and 0.2 ml of complexes formulated in BHG containing 1 μg of
10.2217/NNM.14.45
Research Article Urbiola, Sanmartín, Blanco-Fernández & Tros de Ilarduya For reprint orders, please contact: [email protected] pDNA per well. After 4 h, complexes were removed and replaced with fresh culture medium. A total of 48 h later, cells were washed with PBS and lyzed using 100 μl of Reporter Lysis Buffer (Promega, WI, USA) at room temperature for 10 min, followed by two freeze–thaw cycles. The cell lysate was centrifuged for 2 min at 12,000 × g to pellet debris. A 20 μl sample of the supernatant was assayed for total luciferase activity using the luciferase assay reagent (Promega), according to the manufacturer’s protocol. A luminometer (Sirius-2, Berthold Detection Systems, Innogenetics, Diagnóstica y Terapéutica, Barcelona, Spain) was used to measure luciferase activity. The Bio-Rad Dc Protein Assay (Bio-Rad Laboratories, USA), using fetal bovine serum (FBS) albumin as the standard, was used for quantifying protein content. Data were expressed as nanograms of luciferase (based on a standard curve for luciferase activity) per milligram of protein. Toxicity studies
Cell viability was quantified by Alamar Blue assay. 100,000 cells per well were seeded and grown overnight in 48-well culture plates. Cells were transfected as described previously. After 4 h, complexes were removed and substituted by new fresh media. A total of 48 h later, media was removed and 1 ml of 10% (v/v) Alamar Blue dye in medium supplemented with 10% FBS was added to each well. After 2.5 h of incubation at 37ºC, 200 μl of the supernatant was assayed by measuring absorbance at 570 and 600 nm. Cell viability was calculated according to the formula: % viability = (A570 -A600 ) of treated cells x100/(A570 -A600 ) of control cells. Data were acquired in a Power Wave XS spectrophotometer with KC Junior (Biotek, VT, USA) software. In vivo studies
8–10 weeks female Balb-C mice were purchased from Harlan Iberica (Barcelona, Spain). Animals were handled with the approval of the Committee on Animal Research at the University of Navarra, Pamplona, Spain (075/11). Mice were housed under standard conditions with ad libitum access to water and standard diet. Mice were distributed randomly in groups of eight and a unique dose of 60 μg of pCMVLuc was administered intravenously in 200 μl via the tail vein. At 24 h after injection, mice were sacrified and the organs were harvested and washed with cold saline. Afterwards organs were homogeneized with lysis buffer (Promega) using an homogeneizer (Mini-BeadBeater, BioSpec Products Inc., OK, USA) and centrifuged at 2000 rpm for 10 min at 4ºC. As described before, 20 μl of the supernatant was analyzed for luciferase activity. For tumoral studies B16F10 (2 × 105 cells)
10.2217/NNM.14.45
Nanomedicine (Epub ahead of print)
in 50 μl of PBS were inoculated subcutaneously into the right flank of C57BL/6 mice. Animals were intratumorally or intravenously injected with dendriplexes containing 60 μg of pCpG-hCMV-SCEP-eFluc plasmid when tumor sizes reached 5–6 mM in diameter (7–10 days post-inoculation of cells). A total of 24 h after injection mice were sacrified and tumors were harvested for luciferase expression quantification as described previously. Statistical analysis
Results are reported as the mean values ± standard deviation. Statistical analysis was performed with SPSS 15.0 (SPSS, IL, USA). The different formulations were compared with ANOVA (Tukey post hoc adjust). Differences were considered statistically significant at p < 0.05. The different transfection activities in vitro were compared with ANOVA (Tukey post hoc adjust). In vivo results were compared with Mann–Whitney U test. Differences were considered statistically significant at p < 0.05. Results Synthesis & characterization of PAMAM–HA conjugates
P-HA conjugate was synthetized via amine bond formation, by amination between the amino groups of PAMAM dendrimer G5 (128 primary amino groups, MW: 28,826 Da) and the carboxyl groups of HA (Figure 1) . H-NMR characterization
1
The structural characterization of products including the new conjugate P-HA was confirmed by 1H-NMR in D2O to observe functional groups (Figure 2) . In Figure 2A , peaks belonging to PAMAM G5 at 2.35 ppm (NCH2 CH2CO), 2.54 (CONHCH2CH2N), 2.64 (NCH2CH2CO), 3.1 (CONHCH2CH2NH2) and 3.25 (NCH2CH2NHCO) are observed. In Figure 2B a signal at 1.9 ppm is detected, corresponding to the methyl peak of acetomide group (H of –NCOCH3). Proton peaks of C2-C6 at (3.3–3.8) ppm and C1 at (4.4–4.6) ppm were also observed. The 1H-NMR spectrum of the physical mixture of PAMAM and HA (Figure 2C) displays similar data to those observed for the separate compounds. In the spectrum of the new covalent compound (P-HA) (Figure 2D), the disappearance of some peaks and appearance of new signals is detected, thus confirming that HA was successfully conjugated to G5 PAMAM dendrimer. In this respect, signals at 2.64 and (3.1–3.2) ppm have disappeared and new peaks at (2.35–2.41), 2.95 and 3.35 ppm have appeared, indicating the formation of a new compound. A displace-
future science group
Efficient targeted gene delivery by a novel PAMAM/DNA dendriplex coated with hyaluronic acid
Research Article
For reprint orders, please contact: [email protected] NH2
H 2N
HN
NH O
O N
n = 64 COOH
CH2OH OH
O
O O
CH2OH
KIO4
OH
O
O O
O NH
COO
OH OH
COCH3
Hyaluronic acid
PAMAM G5
COOH
CH2OH OH
PAMAM G5
O
O O
NaBH4 O
O NH
O
O
NH
NH
COCH3
PAMAM G5
Oxidized hyaluronic acid
NH
H3COC PAMAM G5
PAMAM–hyaluronic acid conjugate
Figure 1. Scheme of the synthesis of the covalent compound PAMAM–hyaluronic acid.
ment of the peaks compared with the signals in the spectra of HA and PAMAM separately is observed. The weak signal at 1.9 ppm corresponding to the HA can be attributed to the low proportion of the ligand in the conjugate and to the introduction of HA to the PAMAM. In order to confirm that the grafted material has been included and to support the interpretation of 1D spectra, the 2D homonuclear correlation spectrum (1H-1H-COSY) was acquired for a period of 72 h (Figure 2E) . The proton peaks of methylene groups in the new product appeared at 2.35 ppm corresponding to the methylene group N-CH2-CH2-CO that is coupled with the methylene connected to nitrogen (2.72 ppm). The triplet in 2.39 ppm is coupled with the signal at 3.18 ppm corresponding to the methylene group of PAMAM described in 1D spectra. Interestingly, the NMR spectrum of grafted material revealed a rather small homonuclear coupling to long distance among protons of the hyaluronic moiety peaks of C2–C6 with one of the methylene group of the alkyl chains of the PAMAM. It is possible to observe two double-doblets at 3.44–3.66 ppm. These changes in the position and multiplicity of the signals related to physical mixture confirmed the formation of the proposed structure. The signal above 4.5 ppm present in the five spectra of Figure 2 corresponds to D2O. Elemental analysis
Elemental analysis was carried out with PAMAM-G5, HA and the covalent compound P-HA. The results corresponding to the new compound compared with
future science group
HA and PAMAM indicated the presence of PAMAM in the derivative due to the increase in the proportion of carbon and nitrogen until similar values to PAMAMG5 (Table 1) . The reproducibility of the percentage of the elements C, H and N, revealed the presence of a new compound different from the simple physical mixture. Particle size & zeta-potential of PAMAM–HA dendriplexes
Complexes containing different amount of HA covalent or electrostatically bounded were analyzed in terms of particle size and surface charge (zeta-potential) in the presence of plasmid DNA (pCMVLuc; 10 μg/ml) (Table 2) . In the HA-electrostatically bounded complexes, differences were observed in size and surface charge by adding the ligand. The size of the complexes showed approximately a twofold increase when 5 or 10 μg of HA/μg of pDNA were added. Zeta-potential of plain-dendriplexes (nontargeted) was positive (30 mV), but the addition of increasing amounts of HA (5 or 10 μg of HA/μg of pDNA) led to a decrease in the values to -30 mV, resulting in a shielding of the positive surface charge, probably due to the fact that the ligand, HA, is negatively charged. In the new covalent conjugate, size and zeta-potential were studied as a function of the percentage of PAMAM-G5 substituted by P-HA to achieve a N/P ratio of 6 (Table 2) . No significant differences in size were observed by adding increasing amounts of the ligand HA. In all cases, particle size was close to
10.2217/NNM.14.45
Research Article Urbiola, Sanmartín, Blanco-Fernández & Tros de Ilarduya For reprint orders, please contact: [email protected] A
5.5
B
5.0
4.5
4.0
3.5 ppm
3.0
2.5
2.0
5.5
C
5.5
5.0
4.5
4.0
3.5 ppm
3.0
2.5
2.0
5.0
4.5
4.0
3.5 ppm
3.0
2.5
2.0
D
5.0
4.5
4.0
3.5 ppm
3.0
2.5
2.0
5.5
5
4.5
4 3.5 3 2.5 ppm (t2)
2
E
5.0
4.5
4.0
3.5 3.0 ppm (t2)
2.5
2.0
Figure 2. Representative1H-NMR spectra of (A) PAMAM, (B) hyaluronic acid, (C) physical mixture, (D) covalent compound P-HA and (E) the 2D homonuclear correlation spectrum (1H-1H-COSY) of the covalent product.
10.2217/NNM.14.45
Nanomedicine (Epub ahead of print)
future science group
Efficient targeted gene delivery by a novel PAMAM/DNA dendriplex coated with hyaluronic acid
Research Article
For reprint orders, please contact: [email protected] Table 1. Elemental CHN analysis of PAMAM-G5, hyaluronic acid and PAMAM–hyaluronic acid. Compound
C (%)
H (%)
N (%)
PAMAM-G5
52.6 ± 0.1
8.8 ± 0.3
25.0 ± 0.5
Hyaluronic acid
35.6 ± 0.2
5.6 ± 0.1
2.5 ± 0.3
PAMAM–hyaluronic acid
43.6 ± 0.4
7.9 ± 0.0
19.5 ± 0.2
Data are expressed as the mean ± standard deviation of four determinations.
100 nm. Zeta-potential measurements did not show either important modifications, despite the presence of the negatively charged targeting molecule HA, showing positive values in all cases. Comparing both formulations, electrostatically assembled complexes showed bigger particle sizes and lower zeta-potential values in the presence of 5 and 10 μg HA/μg DNA than the covalent P-HA dendriplexes. Complexes were stable and the polydispersity index was kept around or under 0.3 in all cases. Measurements were carried out in triplicate immediately after preparation of the complexes. Formation of PAMAM–HA dendriplexes
Analysis of the formation of complexes was performed by examining the retardation in the migration of the plasmid DNA during agarose gel electrophoresis. Figure 3 shows the electrophoretic immobilization of DNA. The migration of naked plasmid DNA (lane 1) lead to two bands corresponding to open circular and supercoiled morphology. The other lanes (2–9) showed the retention of DNA in the electrostatic and covalent complexes containing different amounts of HA. Stability of dendriplexes against DNase I digestion
To study whether P-HA dendriplexes can protect DNA against enzymatic degradation, naked DNA and P-HA complexes were incubated with DNase I. Figure 4 shows that naked plasmid DNA (lane 2) was completely digested within 30 min of incubation with
the enzyme, whereas DNA in P-HA dendriplexes (lanes 3–10) was protected in electrostatic or covalent complexes. In the case of covalent P-HA dendriplexes with 100% of P-HA substituted, a slight smear of degradation is detected (lane 10). CD44 receptor expression
To test the feasibility of HA-receptor-mediated targeting by P–HA dendriplexes, cell surface densities of CD44, a well-characterized HA receptor, was measured in B16F10, MDA-MB231, MCF-7 and Neuro2A cells, by using an anti CD44 antibody by flow cytometry. These data will allow us also to correlate the transfection efficiency results with the level of expression of the CD44 receptor in the cells. As shown in Figure 5, B16F10 and MDA-MB231 cells showed higher binding to FITC-CD44 antibody than the isotype control, indicating the presence of CD44 receptors in these cells. However, the binding of FITC-CD44 antibody to the MCF-7 cells was much lower, suggesting relatively lower expression of CD44 receptors. Among these cell lines, Neuro2A showed the lowest peak shift, indicating that 0% of cells expressed CD44. The CD44-positive proportion of B16F10, MDA-MB231 and MCF-7 cells was 80–90, 75–90 and 10–50%, respectively. In vitro transfection activity in B16F10 & MDA-MB231 cells
Plain (nontargeted) and targeted HA-dendriplexes containing 1 μg of DNA were prepared by covalent or
Table 2. Size and zeta-potential of electrostatic and covalent dendriplexes formulated at ratio N/P 6 as a function of the amount of hyaluronic acid and the percentage of PAMAM-G5 substituted by PAMAM–hyaluronic acid. Electrostatic
Covalent
μg HA/μg DNA
Size (nm)
PDI
ZP (mV)
% P–HA
Size (nm)
PDI
ZP (mV)
0
87.9 ± 1.3
0.24
30.6 ± 0.8
0
99.0 ± 0.9
0.37
28.7 ± 0.5
1
103.0 ± 1.4
0.13
20.6 ± 0.5
16
101.6 ± 1.0
0.34
21.0 ± 0.8
5
194.7 ± 2.2
0.20
-28.5 ± 1.2
25
120.2 ± 2.1
0.32
25.2 ± 0.1
10
171.7 ± 1.6
0.17
-30.0 ± 1.1
50
111.9 ± 1.3
0.33
22.7 ± 3.6
100
101.0 ± 1.0
0.37
23.7 ± 1.6
Data are expressed as the mean ± standard deviation of three measurements. P–HA: PAMAM–hyaluronic acid; PDI: Polydispersion index; ZP: Zeta-potential.
future science group
10.2217/NNM.14.45
Research Article Urbiola, Sanmartín, Blanco-Fernández & Tros de Ilarduya For reprint orders, please contact: [email protected]
1
2
3
4
5
6
7
8
measurable luciferase expression was detected with naked DNA either in B16F10 or in MDA-MB231 cells. These results correlate with the CD44 receptor expression observed in Figure 5.
9
Specificity of targeting to the HA receptor
Figure 3. Retardation assay of plain (nontargeted) nanoparticles (lane 2), electrostatic dendriplexes containing 1, 5 or 10 μg HA/μg DNA (lanes 3–5, respectively) and covalent complexes with 16, 25, 50 and 100% of P-HA (lanes 6–9, respectively). Naked DNA was included as control (lane 1).
electrostatic binding of the ligand and evaluated for their capacity to transfect B16F10 and MDA-MB231 cells (Figure 6) . The reason for using PAMAM– DNA–HA electrostatically assembled nanoparticles as control is because one of the objectives of this work was to compare the transfection activity of electrostatically bound HA nanoparticles with the particles prepared by covalent binding of the ligand, in order to know if this led to more efficient targeted complexes. In B16F10 cells (HA+ receptor) an increase in the transfection is observed in the presence of the ligand HA when the new covalent dendriplexes are added. Compared with nontargeted complexes an 1.5 (p = 0.05) and 1.6 (p < 0.01)-fold increase is detected by adding 50 or 100% P–HA, respectively. The same behavior was observed in MDA-MB231 cells, where a 2.5 (p < 0.001), 2.7 (p < 0.001) and 2.5 (p < 0.001)fold increase is observed by adding 16, 50 or 100% P–HA, respectively. In this case, transfection levels are higher than by adding control PEI-polyplexes to the cells. On the other hand, the level of transfection in MDA-MB231 cells resulted to be much higher than in B16F10 cells. Electrostatic dendriplexes did not show any improvement in gene expression when containing the ligand HA in both cell lines. No DNase-1 1
2
3
4
5
6
7
8
9
10
Figure 4. Protection assay. Untreated control DNA (lane 1) and DNase I treated: naked DNA (lane 2), plain (nontargeted) dendriplexes (lane 3), electrostatic dendriplexes containing 1, 5 or 10 μg HA/μg DNA (lanes 4–6, respectively) and covalent complexes with 16, 25, 50 and 100% of P-HA (lanes 7–10, respectively).
10.2217/NNM.14.45
Nanomedicine (Epub ahead of print)
In order to study if the uptake of HA-dendriplexes was mediated via interaction with the HA receptor (CD44), plain (nontargeted) and HA-dendriplexes were transfected into the low-expressing or deficient HA-receptor cell lines MCF-7 and Neuro2A (Figure 6) . No increase in transfection activity by adding the targeted complexes was observed in these cells, either with electrostatic or covalent HA-dendriplexes. Electrostatic complexes showed lower values of gene expression compared with the covalent ones. On the other hand, a competition assay by adding an excess of free HA (1 mg/ml) to MDA-MB231 cells, showed a decrease in transfection efficiency of HAdendriplexes (Figure 7) . These results confirm the receptor-mediated endocytosis mechanism. Cytotoxicity of PAMAM–HA dendriplexes
Cell viability following transfection was determined by the Alamar blue assay as described in the section titled ‘Material & methods’. A viability higher than 80% is observed in all transfected wells, independently of the cell line used and the amount of the ligand (Figure 6) . The relative cytotoxicity of the complexes was also determined by the total amount of extractable cellular proteins in the cell lysate per well, and results confirmed the results with the Alamar blue assay (data not shown). It is important to note that in all cases the viability with the P–HA dendriplexes was higher than with the control PEI-polyplexes. In vivo transfection activity
In vivo luciferase activity in Balb-C healthy mice were carried out 24 h after the administration of the dendriplexes. Plain dendriplexes led to low luciferase activity values in liver and heart, whereas the lung showed the highest activity (Figure 8A) . In the case of HA- dendriplexes the highest gene expression is obtained in the liver. 50% of P-HA containing dendriplexes produced a statistically significant 30- and 19-fold increase in the heart (p < 0.05) and the liver (p
Efficient targeted gene delivery by a novel PAMAM/DNA dendriplex coated with hyaluronic acid
Aim: To design and develop a novel target-specific DNA-delivery system using hyaluronic acid (HA) –polyamidoamine (PAMAM) conjugates (P–HA). Materials & methods: The coupling of HA to the PAMAM dendrimer was analyzed by 1H-NMR and elemental analysis (CHN). Their properties were characterized in terms of size and zeta-potential and evaluated for in vitro and in vivo transfection efficiency. Results: The designed covalent HA-dendriplexes enhanced gene transfection of pCMV-Luc reporter gene in overexpressing CD44-receptor cancer cells. They were also more efficient in transfecting MDA-MB231 cells than conventional PEI-polyplexes. The cytotoxicity of the covalent HA-dendriplexes was lower than when using conventional polyethylenimine-polyplexes. In vivo studies showed that these targeted complexes were also efficient for delivering pCMVLuc in different organs of healthy mice, as well as in tumors of C57BL/6 animals. Conclusions: The HA-dendriplexes developed in this work may offer an advantageous alternative to conventional cationic polymer-based formulations for DNA delivery into cancer cells in an efficient and safe manner.
Koldo Urbiola1, Carmen Sanmartín2, Laura BlancoFernández1 & Conchita Tros de Ilarduya1 Department of Pharmacy & Pharmaceutical Technology, School of Pharmacy, University of Navarra, C/Irunlarrea 1, 31080 Pamplona, Spain 2 Department of Organic & Pharmaceutical Chemistry, University of Navarra, Spain *Author for correspondence: Tel.: +34 948 425 600 (ext. 80 6375) ctros@ unav.es 1
Original submitted 13 February 2013; Revised submitted 24 February 2014 Keywords: gene therapy • gene transfer • hyaluronic acid/hyaluronan • nanoparticle • nanotechnology • polyamidoamine dendrimer
Gene therapy focuses on the delivery of therapeutic genes to cells and promises considerable advances in the treatment of several important diseases. However, there is an important need to improve the activity of gene delivery aimed vectors. Because viral vectors are related with toxicity or excessive immune responses [1] , in recent years polymer nanocarriers that can be used as nonviral gene vectors have aroused high interest in DNA delivery. In this respect, cationic, polymer-based, non-viral delivery systems provide several advantages over viral vectors, including simplicity of production and the ability to package plasmids of any size; however, the lack of biodegradability and targeted delivery and the relatively high cytotoxicity of some of them, such as polyethylenimine (PEI), remain of concern. Polyamidoamine (PAMAM) dendrimers, described by Tomalia et al. [2] are positive-
10.2217/NNM.14.45 © 2014 Future Medicine Ltd
charged dendrimers. PAMAM dendrimers have attracted great interest because of their low polydispersity index, controlled molecular weight and versatility [3] and have been successfully developed as gene and drug carriers and imaging agents [4–6] . They also have a large tolerable dose range between transfection and cytotoxicity. Early studies by Szoka et al. [7] showed that dendrimers above generation 4 (G4) can efficiently transfect different tumor cell lines in vitro. Cationic PAMAM dendrimers with amine terminal groups are known as good condensing agents of DNA or RNA by electrostatic interactions, leading to the formation of more compact particles termed dendriplexes. These structures are able to protect the plasmid DNA against degradation and facilitate the endosome escape by proton sponge effect [7,8] .
Nanomedicine (Epub ahead of print)
part of
ISSN 1743-5889
Research Article Urbiola, Sanmartín, Blanco-Fernández & Tros de Ilarduya For reprint orders, please contact: [email protected] Hyaluronic acid (HA) is a ubiquitous high molecular weight glycosaminoglycan and the principal compound of the extracellular matrix. It is biodegradable, has low toxicity and is a biocompatible polyanion formed by the repetition of the dimer d-glucuronic acid and N-acetyl-d-glucosamine. The ability of HA to interact electrostatically, shield surface charge, act as a targeting agent, and improve biodistribution or the toxicity profile have focused interest on it. It is one of the few polyanions that can coat DNA/polycation complexes without disrupting their structures. On the other hand, it has been approved for injection by the US FDA. A receptor for HA, CD44, is known to be overexpressed on various tumor cell surfaces [9] and in many solid tumor cells including lung, breast, colorectal, gastric, pancreatic, renal, hepatocellular and cervical cancer, as well as melanoma [10] . It has been also implicated in metastatic spread in various cancer types [11] . The degree of intracellular uptake of HA-complexes depends on the receptor density of the cells. Higher CD44 receptor density correlates with higher uptake of HA-complexes. However, the normal cells with low CD44 receptors will not be affected by HA-complexes [12] . This property makes this system tumor-selective. The carboxyl groups of HA are known to be the recognition sites for HA receptors and hyaluronidase [13] and they are essential for transcriptional activity [14] . In this respect, the cellular uptake of hyaluronan-coated DNA/polyplexes via receptor-mediated endocytosis is considered the most favorable mechanism for therapeutic applications. Since HA presents interesting features for improving transfection efficacy with nonviral systems, its use has been explored in the last years for the efficient delivery of pDNA and RNA. It could work both as a protecting coat against blood components and as a ligand for targeted cells. Some studies reported that addition of hyaluronate improved the uptake of genes complexed to different polycationic carriers via the HA-receptors on the surfaces of tumor cells [15–17] . This fact has been principally explored for polyethylenimine (PEI)-based polyplexes [18–24] and it has been also used for targeting dioleoyl phosphatidylethanolamine (DOPE)-based lipoplexes [25] . Nevertheless, addition of HA to PAMAM based-formulations has not been widely explored. Han et al. [26] described an alteration in the intracellular distribution of doxorubicin by the siRNA/PAMAM/HA complex and Chen et al. [27] described the effect of the mixing orders of siRNA/hyperbranched poly(amido amine)/HA on transfection activity. In our group we have previously evaluated the use of different nontargeted PAMAM dendriplexes for gene delivery [28,29] . Now, the aim of this work is to design
10.2217/NNM.14.45
Nanomedicine (Epub ahead of print)
and evaluate a novel polymeric gene delivery vector designed with the objectives of reducing the toxicity of PAMAM, and increasing specific interactions with tumor cells. In this way, we have developed a nanocarrier containing PAMAM–HA and DNA to achieve an enhanced therapeutic effect. Material & methods Materials
High-molecular-weight hyaluronic acid (HA; sodium salt, 1600 kDa), the cationic polyamidoamine dendrimer (PAMAM-G5, 28,825 kDa, 128 primary amines) and branched polyethylenimine (bPEI, 25 kDa) were provided by Sigma-Aldrich (Madrid, Spain). The plasmid pCMVLuc-VR1216 (BioServe Technologies, MD, USA) encoding luciferase gene was amplified in Escherichia coli, isolated and purified using an EndoFree® QUIAGEN Plasmid Giga Kit (QUIAGEN, Germany) and used in the transfection studies. Plasmid purity was confirmed by agarose gel electrophoresis followed by ethidium bromide staining. The pCpG-hCMV-SCEP-eFluc plasmid (kindly provided by Manfred Ogris) encodes an optimized firefly luciferase cDNA under the CMV-EF1alpha hybrid promoter (SCEP) with hCMV enhancer. HEPES glucose buffer (HEPES 10 mM, 10% glucose, pH 7.4; BHG) was prepared from D(+)glucose and N-(2-hydroxyethyl) piperazine-N′-[2-ethanesulfonic acid] (Sigma-Aldrich, Madrid, Spain). Alamar Blue Dye was purchased from Accumed Technologies (OH, USA). Cell culture
B16F10 (murine melanoma), MDA-MB231 (human breast adenocarcinoma), MCF-7 (human breast adenocarcinoma) and Neuro2A (murine neuroblastoma) cell lines were obtained from the American Type Culture Collection (MD, USA). Cell lines were maintained at 37ºC under 5% CO2 in Dulbecco’s Modified Eagle Medium (DMEM) GlutaMAX™, supplemented with 10% (v/v) heat-inactivated fetal bovine serum and penicillin (100 U/ml) and streptomycin (100 μg/ml, Gibco BRL Life Technologies, Barcelona, Spain). Cells were trypsinized twice a week. Synthesis, purification & characterization of PAMAM–HA conjugates
PAMAM–HA (P–HA) conjugates were prepared as follows: first, a solution of HA in distilled water was oxidized by the addition of potassium peryodate (1:2 molar ratio) until the hydroxyl is oxidized to carbonyl groups at room temperature for 48 h. After that, ethylene glycol (10%) was added to stop the reaction
future science group
Efficient targeted gene delivery by a novel PAMAM/DNA dendriplex coated with hyaluronic acid
Research Article
For reprint orders, please contact: [email protected] and the total volume was dialyzed (membrane pore size 12,000 Da) at 4°C with NaCl 0.1 M for 24 h and distilled water for 24 h more before lyophilization. Second, methyl alcohol was removed by rotary evaporation of the commercial PAMAM solution. Then, distilled water was added to the film in order to get an aqueous solution of PAMAM and oxidized HA was added with mechanic stirring for 48 h. After this time, sodium borohydride was added in order to reduce the unreacted HA (1:1 molar ratio PAMAM:sodium borohydride) and finish the amination reaction. Before lyophilization, the final product was dialized for 24 h against NaCl 0.1 M and for 24 h with distilled water. Finally, it was purified by centrifugation-based ultrafiltration with an Amicon (100,000 MWCO; Millipore, MA, USA) in order to remove the unreacted PAMAM and to avoid the formation of large aggregates. The new covalent compound was characterized by 1H NMR. The spectra were recorded on a Bruker 400 Ultrashield™ spectrometer (Rheinstetten, Germany) using tetramethylsilane (TMS) as the internal standard. Proton chemical shifts are reported in ppm relative to internal TMS (0.0 ppm). All NMR spectra were acquired at ambient temperature using D2O as solvent. Elemental microanalyses were carried out on vacuum-dried samples using a LECO CHN-900 elemental analyzer with carbon, hydrogen and nitrogen determinator. Preparation of HA–PAMAM–DNA dendriplexes
Electrostatically assembled nanoparticles were formed by adding, in this order: BHG, pDNA, the desired amount of HA (1, 5 or 10 μg HA/μg DNA) and PAMAM G5. The P–HA covalent dendriplexes were formed by mixing equal volumes of PAMAMG5 solution and pDNA. When P–HA was added, PAMAM-G5 was partially substituted by increasing amounts of P–HA; the degree of replacement is presented as a percentage (0, 16, 25, 50 and 100%) of P–HA. After the addition of the components, nanoparticles were incubated for 15 min at room temperature. Complexes were formulated at N/P 6 ratio (positive charges of polymer/negative charges of pDNA). Previous studies by our group indicated that at this ratio, DNA was completely condensed by PAMAM in order to obtain small and stable nanoparticles [28] .
were diluted in buffer HEPES–glucose (BHG) and measured at least three times. Gel retention studies
For gel retention studies, naked DNA, plain, electrostatic dendriplexes (1, 5 or 10 μg HA/μg DNA) and complexes with HA covalently attached (16, 25, 50 or 100% of P-HA) were prepared at a final concentration of 50 μg/ml as previously described and electrophoresed through a 0.8% agarose gel using Tris–Borate–EDTA buffer (100 mM Tris, 90 mM boric acid, 1 mM EDTA, pH 8.4) with ethidium bromide. The gel was run for 30 min at 100 mV and visualized under UV illumination using a camera Doc 2000 (BioRad, CA, USA). DNase I protection assay
Naked DNA (5 μg), plain, electrostatic dendriplexes containing 1, 5 or 10 μg HA/μg DNA and complexes with HA covalently attached (16, 25, 50 or 100% of P–HA) were prepared as previously described at a final concentration of 50 μg/ml in 100 μl of total volume. DNAse I (Invitrogen; 1 U/μg DNA) was added to each sample and the mixtures were incubated at 37°C for 30 min. 6 μl of EDTA (0.25 M) were added to inactivate the DNase I. Dendriplexes were disassembled by adding 5 μl of SDS (15%) and were incubated for 5 min at room temperature. Then, the samples were electrophoresed 45 min at 80 mV. The integrity of the plasmid in each formulation was compared with untreated DNA as control. CD44 expression
The expression of CD44 receptors was assured by flow cytometry in different cell lines. B16F10, MDAMB231, MCF-7 and Neuro2A cells were trypsynized and washed twice with phosphate-buffered saline (PBS). Then, 500,000 cells were incubated with anti-human/mouse CD44 FITC (eBiosciences, CA, USA) following manufacturer instructions in PBS for 30 min at 4°C in darkness. After that, cells were washed three times and fluorescence was measured by flow cytometry (FACSCalibur™, BD Biosciences, CA, USA). At least 10,000 events were collected for each measurement. Cells without staining were introduced as control. In vitro transfection activity
Particle size & surface charge determinations
Particle size was determined by dynamic light scattering and the overall charge by zeta-potential measurements using a particle analyzer (Zetasizer Nano, Malvern Instruments, Malvern, UK). Final pDNA concentration was 10 μg/ml. Samples of complexes
future science group
For each transfection, 100,000 cells were seeded in 1 ml of medium in 48-well culture plates and incubated for 24 h at 37ºC in 5% CO2. The primary growth medium was removed and replaced with 0.3 ml of medium containing 10% FBS and 0.2 ml of complexes formulated in BHG containing 1 μg of
10.2217/NNM.14.45
Research Article Urbiola, Sanmartín, Blanco-Fernández & Tros de Ilarduya For reprint orders, please contact: [email protected] pDNA per well. After 4 h, complexes were removed and replaced with fresh culture medium. A total of 48 h later, cells were washed with PBS and lyzed using 100 μl of Reporter Lysis Buffer (Promega, WI, USA) at room temperature for 10 min, followed by two freeze–thaw cycles. The cell lysate was centrifuged for 2 min at 12,000 × g to pellet debris. A 20 μl sample of the supernatant was assayed for total luciferase activity using the luciferase assay reagent (Promega), according to the manufacturer’s protocol. A luminometer (Sirius-2, Berthold Detection Systems, Innogenetics, Diagnóstica y Terapéutica, Barcelona, Spain) was used to measure luciferase activity. The Bio-Rad Dc Protein Assay (Bio-Rad Laboratories, USA), using fetal bovine serum (FBS) albumin as the standard, was used for quantifying protein content. Data were expressed as nanograms of luciferase (based on a standard curve for luciferase activity) per milligram of protein. Toxicity studies
Cell viability was quantified by Alamar Blue assay. 100,000 cells per well were seeded and grown overnight in 48-well culture plates. Cells were transfected as described previously. After 4 h, complexes were removed and substituted by new fresh media. A total of 48 h later, media was removed and 1 ml of 10% (v/v) Alamar Blue dye in medium supplemented with 10% FBS was added to each well. After 2.5 h of incubation at 37ºC, 200 μl of the supernatant was assayed by measuring absorbance at 570 and 600 nm. Cell viability was calculated according to the formula: % viability = (A570 -A600 ) of treated cells x100/(A570 -A600 ) of control cells. Data were acquired in a Power Wave XS spectrophotometer with KC Junior (Biotek, VT, USA) software. In vivo studies
8–10 weeks female Balb-C mice were purchased from Harlan Iberica (Barcelona, Spain). Animals were handled with the approval of the Committee on Animal Research at the University of Navarra, Pamplona, Spain (075/11). Mice were housed under standard conditions with ad libitum access to water and standard diet. Mice were distributed randomly in groups of eight and a unique dose of 60 μg of pCMVLuc was administered intravenously in 200 μl via the tail vein. At 24 h after injection, mice were sacrified and the organs were harvested and washed with cold saline. Afterwards organs were homogeneized with lysis buffer (Promega) using an homogeneizer (Mini-BeadBeater, BioSpec Products Inc., OK, USA) and centrifuged at 2000 rpm for 10 min at 4ºC. As described before, 20 μl of the supernatant was analyzed for luciferase activity. For tumoral studies B16F10 (2 × 105 cells)
10.2217/NNM.14.45
Nanomedicine (Epub ahead of print)
in 50 μl of PBS were inoculated subcutaneously into the right flank of C57BL/6 mice. Animals were intratumorally or intravenously injected with dendriplexes containing 60 μg of pCpG-hCMV-SCEP-eFluc plasmid when tumor sizes reached 5–6 mM in diameter (7–10 days post-inoculation of cells). A total of 24 h after injection mice were sacrified and tumors were harvested for luciferase expression quantification as described previously. Statistical analysis
Results are reported as the mean values ± standard deviation. Statistical analysis was performed with SPSS 15.0 (SPSS, IL, USA). The different formulations were compared with ANOVA (Tukey post hoc adjust). Differences were considered statistically significant at p < 0.05. The different transfection activities in vitro were compared with ANOVA (Tukey post hoc adjust). In vivo results were compared with Mann–Whitney U test. Differences were considered statistically significant at p < 0.05. Results Synthesis & characterization of PAMAM–HA conjugates
P-HA conjugate was synthetized via amine bond formation, by amination between the amino groups of PAMAM dendrimer G5 (128 primary amino groups, MW: 28,826 Da) and the carboxyl groups of HA (Figure 1) . H-NMR characterization
1
The structural characterization of products including the new conjugate P-HA was confirmed by 1H-NMR in D2O to observe functional groups (Figure 2) . In Figure 2A , peaks belonging to PAMAM G5 at 2.35 ppm (NCH2 CH2CO), 2.54 (CONHCH2CH2N), 2.64 (NCH2CH2CO), 3.1 (CONHCH2CH2NH2) and 3.25 (NCH2CH2NHCO) are observed. In Figure 2B a signal at 1.9 ppm is detected, corresponding to the methyl peak of acetomide group (H of –NCOCH3). Proton peaks of C2-C6 at (3.3–3.8) ppm and C1 at (4.4–4.6) ppm were also observed. The 1H-NMR spectrum of the physical mixture of PAMAM and HA (Figure 2C) displays similar data to those observed for the separate compounds. In the spectrum of the new covalent compound (P-HA) (Figure 2D), the disappearance of some peaks and appearance of new signals is detected, thus confirming that HA was successfully conjugated to G5 PAMAM dendrimer. In this respect, signals at 2.64 and (3.1–3.2) ppm have disappeared and new peaks at (2.35–2.41), 2.95 and 3.35 ppm have appeared, indicating the formation of a new compound. A displace-
future science group
Efficient targeted gene delivery by a novel PAMAM/DNA dendriplex coated with hyaluronic acid
Research Article
For reprint orders, please contact: [email protected] NH2
H 2N
HN
NH O
O N
n = 64 COOH
CH2OH OH
O
O O
CH2OH
KIO4
OH
O
O O
O NH
COO
OH OH
COCH3
Hyaluronic acid
PAMAM G5
COOH
CH2OH OH
PAMAM G5
O
O O
NaBH4 O
O NH
O
O
NH
NH
COCH3
PAMAM G5
Oxidized hyaluronic acid
NH
H3COC PAMAM G5
PAMAM–hyaluronic acid conjugate
Figure 1. Scheme of the synthesis of the covalent compound PAMAM–hyaluronic acid.
ment of the peaks compared with the signals in the spectra of HA and PAMAM separately is observed. The weak signal at 1.9 ppm corresponding to the HA can be attributed to the low proportion of the ligand in the conjugate and to the introduction of HA to the PAMAM. In order to confirm that the grafted material has been included and to support the interpretation of 1D spectra, the 2D homonuclear correlation spectrum (1H-1H-COSY) was acquired for a period of 72 h (Figure 2E) . The proton peaks of methylene groups in the new product appeared at 2.35 ppm corresponding to the methylene group N-CH2-CH2-CO that is coupled with the methylene connected to nitrogen (2.72 ppm). The triplet in 2.39 ppm is coupled with the signal at 3.18 ppm corresponding to the methylene group of PAMAM described in 1D spectra. Interestingly, the NMR spectrum of grafted material revealed a rather small homonuclear coupling to long distance among protons of the hyaluronic moiety peaks of C2–C6 with one of the methylene group of the alkyl chains of the PAMAM. It is possible to observe two double-doblets at 3.44–3.66 ppm. These changes in the position and multiplicity of the signals related to physical mixture confirmed the formation of the proposed structure. The signal above 4.5 ppm present in the five spectra of Figure 2 corresponds to D2O. Elemental analysis
Elemental analysis was carried out with PAMAM-G5, HA and the covalent compound P-HA. The results corresponding to the new compound compared with
future science group
HA and PAMAM indicated the presence of PAMAM in the derivative due to the increase in the proportion of carbon and nitrogen until similar values to PAMAMG5 (Table 1) . The reproducibility of the percentage of the elements C, H and N, revealed the presence of a new compound different from the simple physical mixture. Particle size & zeta-potential of PAMAM–HA dendriplexes
Complexes containing different amount of HA covalent or electrostatically bounded were analyzed in terms of particle size and surface charge (zeta-potential) in the presence of plasmid DNA (pCMVLuc; 10 μg/ml) (Table 2) . In the HA-electrostatically bounded complexes, differences were observed in size and surface charge by adding the ligand. The size of the complexes showed approximately a twofold increase when 5 or 10 μg of HA/μg of pDNA were added. Zeta-potential of plain-dendriplexes (nontargeted) was positive (30 mV), but the addition of increasing amounts of HA (5 or 10 μg of HA/μg of pDNA) led to a decrease in the values to -30 mV, resulting in a shielding of the positive surface charge, probably due to the fact that the ligand, HA, is negatively charged. In the new covalent conjugate, size and zeta-potential were studied as a function of the percentage of PAMAM-G5 substituted by P-HA to achieve a N/P ratio of 6 (Table 2) . No significant differences in size were observed by adding increasing amounts of the ligand HA. In all cases, particle size was close to
10.2217/NNM.14.45
Research Article Urbiola, Sanmartín, Blanco-Fernández & Tros de Ilarduya For reprint orders, please contact: [email protected] A
5.5
B
5.0
4.5
4.0
3.5 ppm
3.0
2.5
2.0
5.5
C
5.5
5.0
4.5
4.0
3.5 ppm
3.0
2.5
2.0
5.0
4.5
4.0
3.5 ppm
3.0
2.5
2.0
D
5.0
4.5
4.0
3.5 ppm
3.0
2.5
2.0
5.5
5
4.5
4 3.5 3 2.5 ppm (t2)
2
E
5.0
4.5
4.0
3.5 3.0 ppm (t2)
2.5
2.0
Figure 2. Representative1H-NMR spectra of (A) PAMAM, (B) hyaluronic acid, (C) physical mixture, (D) covalent compound P-HA and (E) the 2D homonuclear correlation spectrum (1H-1H-COSY) of the covalent product.
10.2217/NNM.14.45
Nanomedicine (Epub ahead of print)
future science group
Efficient targeted gene delivery by a novel PAMAM/DNA dendriplex coated with hyaluronic acid
Research Article
For reprint orders, please contact: [email protected] Table 1. Elemental CHN analysis of PAMAM-G5, hyaluronic acid and PAMAM–hyaluronic acid. Compound
C (%)
H (%)
N (%)
PAMAM-G5
52.6 ± 0.1
8.8 ± 0.3
25.0 ± 0.5
Hyaluronic acid
35.6 ± 0.2
5.6 ± 0.1
2.5 ± 0.3
PAMAM–hyaluronic acid
43.6 ± 0.4
7.9 ± 0.0
19.5 ± 0.2
Data are expressed as the mean ± standard deviation of four determinations.
100 nm. Zeta-potential measurements did not show either important modifications, despite the presence of the negatively charged targeting molecule HA, showing positive values in all cases. Comparing both formulations, electrostatically assembled complexes showed bigger particle sizes and lower zeta-potential values in the presence of 5 and 10 μg HA/μg DNA than the covalent P-HA dendriplexes. Complexes were stable and the polydispersity index was kept around or under 0.3 in all cases. Measurements were carried out in triplicate immediately after preparation of the complexes. Formation of PAMAM–HA dendriplexes
Analysis of the formation of complexes was performed by examining the retardation in the migration of the plasmid DNA during agarose gel electrophoresis. Figure 3 shows the electrophoretic immobilization of DNA. The migration of naked plasmid DNA (lane 1) lead to two bands corresponding to open circular and supercoiled morphology. The other lanes (2–9) showed the retention of DNA in the electrostatic and covalent complexes containing different amounts of HA. Stability of dendriplexes against DNase I digestion
To study whether P-HA dendriplexes can protect DNA against enzymatic degradation, naked DNA and P-HA complexes were incubated with DNase I. Figure 4 shows that naked plasmid DNA (lane 2) was completely digested within 30 min of incubation with
the enzyme, whereas DNA in P-HA dendriplexes (lanes 3–10) was protected in electrostatic or covalent complexes. In the case of covalent P-HA dendriplexes with 100% of P-HA substituted, a slight smear of degradation is detected (lane 10). CD44 receptor expression
To test the feasibility of HA-receptor-mediated targeting by P–HA dendriplexes, cell surface densities of CD44, a well-characterized HA receptor, was measured in B16F10, MDA-MB231, MCF-7 and Neuro2A cells, by using an anti CD44 antibody by flow cytometry. These data will allow us also to correlate the transfection efficiency results with the level of expression of the CD44 receptor in the cells. As shown in Figure 5, B16F10 and MDA-MB231 cells showed higher binding to FITC-CD44 antibody than the isotype control, indicating the presence of CD44 receptors in these cells. However, the binding of FITC-CD44 antibody to the MCF-7 cells was much lower, suggesting relatively lower expression of CD44 receptors. Among these cell lines, Neuro2A showed the lowest peak shift, indicating that 0% of cells expressed CD44. The CD44-positive proportion of B16F10, MDA-MB231 and MCF-7 cells was 80–90, 75–90 and 10–50%, respectively. In vitro transfection activity in B16F10 & MDA-MB231 cells
Plain (nontargeted) and targeted HA-dendriplexes containing 1 μg of DNA were prepared by covalent or
Table 2. Size and zeta-potential of electrostatic and covalent dendriplexes formulated at ratio N/P 6 as a function of the amount of hyaluronic acid and the percentage of PAMAM-G5 substituted by PAMAM–hyaluronic acid. Electrostatic
Covalent
μg HA/μg DNA
Size (nm)
PDI
ZP (mV)
% P–HA
Size (nm)
PDI
ZP (mV)
0
87.9 ± 1.3
0.24
30.6 ± 0.8
0
99.0 ± 0.9
0.37
28.7 ± 0.5
1
103.0 ± 1.4
0.13
20.6 ± 0.5
16
101.6 ± 1.0
0.34
21.0 ± 0.8
5
194.7 ± 2.2
0.20
-28.5 ± 1.2
25
120.2 ± 2.1
0.32
25.2 ± 0.1
10
171.7 ± 1.6
0.17
-30.0 ± 1.1
50
111.9 ± 1.3
0.33
22.7 ± 3.6
100
101.0 ± 1.0
0.37
23.7 ± 1.6
Data are expressed as the mean ± standard deviation of three measurements. P–HA: PAMAM–hyaluronic acid; PDI: Polydispersion index; ZP: Zeta-potential.
future science group
10.2217/NNM.14.45
Research Article Urbiola, Sanmartín, Blanco-Fernández & Tros de Ilarduya For reprint orders, please contact: [email protected]
1
2
3
4
5
6
7
8
measurable luciferase expression was detected with naked DNA either in B16F10 or in MDA-MB231 cells. These results correlate with the CD44 receptor expression observed in Figure 5.
9
Specificity of targeting to the HA receptor
Figure 3. Retardation assay of plain (nontargeted) nanoparticles (lane 2), electrostatic dendriplexes containing 1, 5 or 10 μg HA/μg DNA (lanes 3–5, respectively) and covalent complexes with 16, 25, 50 and 100% of P-HA (lanes 6–9, respectively). Naked DNA was included as control (lane 1).
electrostatic binding of the ligand and evaluated for their capacity to transfect B16F10 and MDA-MB231 cells (Figure 6) . The reason for using PAMAM– DNA–HA electrostatically assembled nanoparticles as control is because one of the objectives of this work was to compare the transfection activity of electrostatically bound HA nanoparticles with the particles prepared by covalent binding of the ligand, in order to know if this led to more efficient targeted complexes. In B16F10 cells (HA+ receptor) an increase in the transfection is observed in the presence of the ligand HA when the new covalent dendriplexes are added. Compared with nontargeted complexes an 1.5 (p = 0.05) and 1.6 (p < 0.01)-fold increase is detected by adding 50 or 100% P–HA, respectively. The same behavior was observed in MDA-MB231 cells, where a 2.5 (p < 0.001), 2.7 (p < 0.001) and 2.5 (p < 0.001)fold increase is observed by adding 16, 50 or 100% P–HA, respectively. In this case, transfection levels are higher than by adding control PEI-polyplexes to the cells. On the other hand, the level of transfection in MDA-MB231 cells resulted to be much higher than in B16F10 cells. Electrostatic dendriplexes did not show any improvement in gene expression when containing the ligand HA in both cell lines. No DNase-1 1
2
3
4
5
6
7
8
9
10
Figure 4. Protection assay. Untreated control DNA (lane 1) and DNase I treated: naked DNA (lane 2), plain (nontargeted) dendriplexes (lane 3), electrostatic dendriplexes containing 1, 5 or 10 μg HA/μg DNA (lanes 4–6, respectively) and covalent complexes with 16, 25, 50 and 100% of P-HA (lanes 7–10, respectively).
10.2217/NNM.14.45
Nanomedicine (Epub ahead of print)
In order to study if the uptake of HA-dendriplexes was mediated via interaction with the HA receptor (CD44), plain (nontargeted) and HA-dendriplexes were transfected into the low-expressing or deficient HA-receptor cell lines MCF-7 and Neuro2A (Figure 6) . No increase in transfection activity by adding the targeted complexes was observed in these cells, either with electrostatic or covalent HA-dendriplexes. Electrostatic complexes showed lower values of gene expression compared with the covalent ones. On the other hand, a competition assay by adding an excess of free HA (1 mg/ml) to MDA-MB231 cells, showed a decrease in transfection efficiency of HAdendriplexes (Figure 7) . These results confirm the receptor-mediated endocytosis mechanism. Cytotoxicity of PAMAM–HA dendriplexes
Cell viability following transfection was determined by the Alamar blue assay as described in the section titled ‘Material & methods’. A viability higher than 80% is observed in all transfected wells, independently of the cell line used and the amount of the ligand (Figure 6) . The relative cytotoxicity of the complexes was also determined by the total amount of extractable cellular proteins in the cell lysate per well, and results confirmed the results with the Alamar blue assay (data not shown). It is important to note that in all cases the viability with the P–HA dendriplexes was higher than with the control PEI-polyplexes. In vivo transfection activity
In vivo luciferase activity in Balb-C healthy mice were carried out 24 h after the administration of the dendriplexes. Plain dendriplexes led to low luciferase activity values in liver and heart, whereas the lung showed the highest activity (Figure 8A) . In the case of HA- dendriplexes the highest gene expression is obtained in the liver. 50% of P-HA containing dendriplexes produced a statistically significant 30- and 19-fold increase in the heart (p < 0.05) and the liver (p
