Poor prognosis in mammary carcinoma is associated with a certain expression profile of a defined set of genes including osteopontin and bone sialoprotein. Efficient and specific delivery of antisenses (AS) and a protection of the sequences from degradation are the crucial conditions for AS therapeutic efficiency. We hypothesized that effective and safe AS delivery direceted against these genes could be achieved by polymeric nanoparticles (NP) fabricated from a biocompatible polymer. Due to their nano-size range and small negative charge, AS-NP can overcome the absorption barrier offering increased resistance to nuclease degradation, sustained duration of AS administration, and consequently, prolonged antisense action. The ASs designed against OPN and BSP-II were successfully encapsulated in NP composed of the biodegradable and biocompatible polylactide-co-glycolide polymer (PLGA), exhibiting sustained release and stability of the ASs. The therapeutic efficacy of the AS-NP delivery system was examined in vitro, and in a breast cancer bone metastasis animal model of MDA-MB-231 human breast cancer cells in nude rats. Treatment with OPN-AS or BSP-AS loaded NP in comparison with osmotic mini-pumps (locoregional injection and SC implants, respectively) resulted in a significant decrease in both, tumor bone metastasis incidence and in the size of the lesions in rats with metastases. Despite its smaller dose, AS-NP exhibited a better therapeutic efficacy than osmotic mini-pumps in terms of lesion ratio at later time periods (8–12 weeks). It may be concluded that AS delivery by NP is a promising therapeutic modality providing stability of the encapsulated AS and a sustained release.
The most prevalent solid tumors, such as breast, lung and prostate cancers, metastasize into the skeleton and cause either osteolytic (destructive) or osteoblastic lesions.1 Both types are often accompanied by bone pain and increased bone fragility and thus are reason for extended suffering.2, 3 The initial steps include invasion of tumor cells into normal tissue, traversion of small blood vessel walls and thus access to the circulation.4 Cancer cells that survive these initial steps can enter the sinusoids of the bone marrow and migrate across the sinusoidal wall to the endosteal bone surface.1 For homing into the bone compartment tumor cells have to develop specific functions that differ from those in the normal tissue and they presumably express several proteins that assist in this process.3 Recent experiments involving gene expressing profiling of patients with mammary carcinoma have shown that a poor prognosis (short interval to metastasis and death) is associated with a certain expression profile of a defined set of genes.5–9 Osteopontin and the product of a closely related small integrin-binding ligand (SIBLING, N-linked glycoprotein) gene, bone sialoprotein-II (BSP), have both been characterized as promising targets for a therapy by antisense oligonucleotides directed against the RNA of these proteins, thus preventing or reducing lytic skeletal metastasis.10–12
Efficient and specific delivery of antisenses (ASs) and a protection of the sequences from degradation are the crucial conditions for AS therapeutic efficiency. Due to their large molecular size and high negative charge density, the resulting low cellular permeability of AS is a major problem encountered with their therapeutic use.13–15 To improve cellular delivery of AS several methods, employed in DNA delivery, have been developed including viral vectors, liposomes, and other delivery systems.14, 16–18
We hypothesized that effective and safe AS delivery could be achieved by polymeric nanoparticles (NP) fabricated from the biocompatible and biodegradable PLGA. PLGA copolymers are among the few synthetic biocompatible biodegradable polymers approved for human gene therapy use.19 Due to their nano-size range and small negative charge, polymeric nanospheres containing AS can overcome the absorption barrier of the cell membrane by penetrating inside the cell via endocytosis.20, 21 Moreover, the controlled release delivery mode offers increased resistance to nuclease degradation, sustained duration of AS administration, and consequently, prolonged antisense action. To the best of our knowledge this work is the first report on the in vivo administration of OPN and BSP-II AS sequences via PLGA NP.
In this work we examined the characteristics and therapeutic efficacy of a controlled release NP delivery system for AS sequences, designed against OPN and BSP-II. The AS NP efficiency in the inhibition of metastatic bone lysis was evaluated in the rat animal model of mammary carcinoma in comparison to osmotic mini-pump implants.
Material and Methods
PLGA, with a lactide-glycolide ratio of 50:50, inherent viscosity of 0.79 dl/g, and MW of 75,200 was purchased from Birmingham Polymers (Birmingham, AL). OPN AS sequence, labeled with fluoroscein (FITC) was purchased from Alpha DNA (Montreal, Quebec, Canada). Rhodamine-labeled PLGA was synthesized as reported previously.21 Polyvinyl alcohol (PVA) with an average MW range of 15,000–30,000 D, was obtained from Sigma (St Louis, MO). All other chemicals were obtained commercially at the highest analytical grade available. All buffers and solutions were filter sterilized.
Design of the antisense sequences
The suitability of 10 different AS per gene was predicted by using the HUSAR program “Mfold,” which takes RNA folding into account. Application of this program onto the cDNA sequence of OPN (Gene bank acc. no. gi:3360431) and BSPII (Gene bank acc. no. gi:11435526) resulted in the recognition of RNA stretches that probably contain bulges or loops, which are preferentially single stranded and thus allow access of DNA antisense structures. AS of 20 base pairs lengths were selected against these single-stranded regions and synthesized with a phosphorothioate backbone, to increase stability against degrading enzymes. The phosphorothioate derivates have been characterized as described previously11 and the most effective and specific sequences were chosen for modulation of OPN and BSP-II in this study. For OPN this was a stretch corresponding to bp. 1343-1362 (5′-CTAACTTAAAAAACAAAAGA-3′), and for BSP-II a stretch corresponding to bp. 546–565 (5′-GCTTTCTTCGTTTTCA TTTC-3′). A nonsense oligonucleotide (NS), derived from HBV genome, a 20 bp long phosphorothioate oligomer, was used as control (5′-GCGAGGGAGTTCTTCTTCTA-3′).
Preparation and characterization of antisense nanoparticles (AS NP)
AS NP preparation
A double emulsion system and the solvent evaporation technique were used to incorporate OPN and BSP-II AS sequences in PLGA.21 Briefly, 300 μl of TE buffer (10 mM Tris and 1 mM EDTA) containing 1,200 μg of AS (initial loading of 1.7 w/w %) were emulsified in 3% PLGA/chloroform solution (2.3 ml) and sonicated over an ice-bath using a microtip probe sonicator (Microson XL; Misonix, Farmingdale, NY) at 11 W for 30 sec. The resulting primary emulsion was added drop wise to a 2% polyvinyl alcohol (PVA) solution (in 25 ml TE) containing 7.5 mM calcium chloride, and after sonicating this mixture for 1 min a double emulsion was formed. Following overnight chloroform evaporation at 4°C, the resulting particles were collected by ultracentrifugation at 80,000g (TST28 rotor, Beckman's polyallomer centrifuge tubes, 25 × 89 mm), washed three times with sterile double distilled water, resuspended in sterile 2% mannitol water solution, and lyophilized. Dry NP were stored in a vacuum desiccator at 4°C. Several types of NP have been prepared for the various studies, empty NP (blank, serving as control), NP loaded with NS (control), NP containing OSP or BSP AS (AS-NP); and fluorescent NP including, NP loaded with FITC-AS (FITC-NP), fluorescent empty NP (PLGA-rhodamine, Rhodamine-NP), and double labeled NP (AS-FITC and PLGA-rhodamine).
AS NP physicochemical characterization
AS loading: The amount of AS entrapped in the NP was analyzed by dissolving the NP (10 mg) in chloroform (2 ml), and extracting the AS from the polymer solution by repetitive addition of TE buffer (4 times with 0.5 ml). The AS content was determined using the fluorescence Oligreen assay kit (Molecular Probes, Eugene, OR) with excitation at 480 nm and emission at 520 nm, and by UV spectroscopy at λ = 260 nm.
Size, morphology and surface charge: NP size and morphology were evaluated by dynamic light scattering, ALV (NIBS/HPPS GmbH, Langen, Germany) and transmission electron microscope (TEM CN12, Philips, Eindhoven, The Netherlands). Average size and size distribution of empty NP and NP loaded with AS sequences were measured before and after lyophilization. For ALV size measurement, after lyophilization, 1mg of dry NP was suspended in 1 ml of sterile double distilled water. Before the lyophilization, 30 μl of the NP suspension was added to a 1ml of sterile double distilled water. A Zeta-sizer (Malvern Instruments Ltd, UK) was used for evaluation of NP charge. The lyophilized empty and AS loaded particles were suspended in PBS to reach the attenuation index 7-9, and the zeta potential (mV ± SD) was measured.
In vitro release: Following the lyophilization, AS NP (5–20 mg) were suspended in TE buffer (1 ml) and incubated at 37°C on a shaker at 100 r.p.m. (Lab Line Instruments, Melrose Park, IL). At different time points, the buffer was separated from the NP by centrifugation and analyzed for the amount of released AS using the fluorescence Oligreen assay kit. Each time, the NP were resuspended in 1 ml of fresh medium.
AS NP stability: In order to examine the stability of the formulation stored in a vacuum desiccator at 4°C, size and morphology of lyophilized NP were evaluated at different time points over a period of 3 years using ALV and TEM. Structural integrity of the AS sequences, either extracted, or released from the NP in TE buffer at 37°C, was evaluated by gel electrophoresis, using urea-PAGE.
The human breast adenocarcinoma cell line MDA-MB-231 (American Type Culture Collection, HTB-26) was grown as a monolayer in RPMI 1640 medium with 10% (fetal bovine serum) FBS and 2 mM L-glutamine as recommended. The human cancer cell line HeLa (American Type Culture Collection) was grown as a monolayer in DMEM medium with 10% FBS and 2 mM L-glutamine as recommended (all materials of Biological Industries, Beit Haemek, Israel).
NP effect on cells' viability
HeLa cells (1.5 × 104/chamber) were seeded in a Lab-Tek chambered cover glass system and incubated for 24 hr. The cells were treated with empty NP (10 mg/ml) and were incubated for 24 or 48 hr. At each time point, the cells were washed with PBS (×3) and harvested. The total cells' number and amount of living and dead cells were counted, using a hemocytometer. Trypan blue was used for staining dead cells. Nontreated cells were used as a control group. The percentage of living and dead cells was calculated. Statistical differences between NP treated and non-treated groups were tested at each time point by the one-tailed unpaired Student's test. Differences were considered statistically significant at p < 0.05.
In vitro visualization of cellular uptake
MDA-MB-231 and Hela cells (1.5 × 104/chamber) were seeded in a Lab-Tek chambered cover glass system (Nalge Nunc International Corp.) and incubated for 24 hr. Fluorescent NP loaded with FITC-AS or naked FITC-AS sequences were added to the cells and incubated for 4, 24 or 48 hr. Cells were washed with PBS (×3), fixed with absolute methanol for 6 minutes at −20°C, washed (×3) with PBS, and were mounted with fluorescence microscopy mounting media (Sigma). AS NP uptake in comparison to naked AS was observed and recorded by means of confocal microscopy (Zeiss LSM 410, Germany). Nontreated cells and cells incubated with empty NP were determined as background. Confocal cross-sections of cells taken at 24 and 48 hr after treatment were used to verify intracellular (cytoplasmic) localization of the NP and the AS.
Quantification of AS NP cellular uptake by fluorescence activated cell sorting (FACS) analysis
HeLa cells (1.5 × 104/chamber) were seeded in a Lab-Tek chambered cover glass system and incubated for 24 hr. Fluorescent empty NP (PLGA-rhodamine, Rhodamine-NP), nonlabeled NP loaded with fluorescent OPN AS (FITC-AS), double stained NP (AS-FITC and PLGA-rhodamine), and naked fluorescent AS were added to the cells. Nontreated cells and nonlabeled empty NP were used as control groups. The cells were incubated for 4, 8, 24 or 48 hr and at each time point, the cells were washed with PBS (×3), and harvested. The cells were washed (×1,500 RPM, 5 minutes) in FACS medium (PBS, 1% BSA, 0.02% sodium azide) and were suspended in 1 ml FACS medium for flow cytometry analysis. A total of 10,000 cells were counted in each measurement, with two different fluorescent filters, FITC (FL-1) and rhodamine (FL-2).
Number of stained cells
The number of stained cells provides data on the uptake efficiency of naked AS, NP or AS encapsulated in NP. The method takes into account each cell that has successfully internalized labeled material (turned to be fluorescent). The total cell population, presented in a dot-blot chart (Fig. 4a), is divided into four subpopulations: nonstained, stained separately with FITC or rhodamine or double stained. The number of stained cells was calculated from the FACS dot-blot data obtained at 4, 8, 24 and 48 hr after the treatment (n = 4, represented as % of total cell count ±SD). Statistical differences between groups were tested using 2-way ANOVA analysis followed by the Dunnett test. Differences were termed statistically significant at p < 0.05.
Extent and intensity of NP endocytosis
The fluorescent intensity evaluation takes into account the number of successfully internalized cells and also the intensity of fluorescent signals in a single cell. This method provides data on NP and AS uptake as well as fate inside the cells. The fluorescent intensity changes over time due to AS release from the NP, degradation or elimination from the cell.
Fluorescence intensity in cells was evaluated from the FACS histogram data at 4, 8, 24 and 48 hr after the treatment (n = 4, arbitrary units representing the number of stained cells and the intensity of the fluorescent signal in each cell). Statistical differences between groups were tested using the 2-way ANOVA analysis followed by the Dunnett test. Differences were considered statistically significant at p < 0.05.
Inhibition of metastatic bone lysis by OPN and BSP-II AS
Nude rats (RNU strain) were obtained from Harlan (Harlan, Borchen, Germany) at an age of 6–8 weeks. They were housed under specific pathogen-free conditions in a mini-barrier system of the central animal facility, conforming to the standards for care and use of laboratory animals of the DKFZ and NIH. Autoclaved feed and water was given ad libitum to the animals that were maintained under controlled conditions (21 ± 2°C room temperature, 60% humidity and 12 hr light–dark rhythm).
In vivo metastasis model
Subconfluent MDA-MB-231 tumor cells were harvested using 2 mM EDTA in PBS− (phosphate-buffered saline without Ca++ and Mg++) and 0.25% trypsin (Sigma, Taufkirchen, Germany). Cells were counted in a Neubauer's chamber and suspended in RPMI-1640 (5 × 105 cells in 1 ml). For tumor cell implantation, male nude rats were anaesthetized with a mixture of nitrous oxide (1 l/min), oxygen (0.5 l/min) and isoflurane (1–1.5 vol %), and the arterial branches of the right hind leg were dissected.22 A needle was inserted into the superficial epigastric artery (SEA), which is branching off the femoral artery, and 105 MDA- MB-231 cells suspended in 0.2 ml RPMI-1640 were injected. By using vessel clips the tumor cells were directed to the descending genicular and popliteal arteries, which are supplying the knee joint and the muscles of the leg. Resulting bone metastases were observed exclusively in the femur, tibia and fibula of the respective hind leg. The mean lesion size of bone metastasis was followed by X-ray examinations once weekly for 12 weeks. Treatment of these lesions started at the time of tumor cell transplantation by either subcutaneous implants of osmotic mini-pumps (ALZET Osmotic Pumps, Cupertino, CA), or by injecting locoregionally NP loaded with AS. The pumps were filled with 20 mg, respectively of a NS or an AS (OPN or BSP-II). The dose delivered by these pumps was extended to 4 weeks by replacing the pump once after 2 weeks in accordance with their exhaustion rate (6–7mg/kg daily, a total dose of 40 mg/rat over 4 weeks' treatment).
The NP (100 mg in 500 μl PBS, total dose of 600 μg of NS or AS) were administered immediately after tumor inoculation into the same vessel as the cells (superficial epigastric artery), two thirds of the dose was injected into the vessel and one third (using the same needle) into the muscle surrounding the knee joint (inorder to avoid clogging of the artery). Treatment results were observed in groups of 4–5 rats at each arm of the studies, for up to 12 weeks. The results were compared with rats treated by mini-pumps (positive control group) or NS-NP (control group). The lesion size was measured from X-ray pictures using image analysis software, and the mean lesion size and lesion ratio (mean lesion size in treated group in percent of control) were calculated. The mean lesion size was presented in relative units (RU), corresponding to the image analysis of the pixel's count.
Results were expressed as mean ± SD. Statistical differences between groups in the metastatic lesion size were assessed by the one way ANOVA test. When required, this test was followed by the Tukey multiple comparisons post hoc test (p < 0.05 was termed significant). Differences between NP, osmotic mini-pumps and the control groups in metastasis incidence rate were termed significant by the χ2 test at p < 0.05.
Loading, size and morphology
Relatively high encapsulation efficiency was achieved (52 ± 11%, determined as amount entrapped out of the available amount in the formulation, Table 1). The addition of calcium chloride to the exterior phase increased the encapsulation to 52 ± 11% in comparison with only 17% loading efficiency in its absence (data not shown). Average NP size and morphology were not affected by the addition of Ca++ to the formulation. AS loading in the polymeric NP amounted to 4.8 ± 0.68 μg OPN AS, 4.56 ± 1.07 μg BSP-II AS and 6.91 ± 2.82 μg NS per mg polymer. The NP had a spherical morphology as shown by the TEM micrographs (Fig. 1). The mean size of the NP was 234 ± 72 and 236 ± 73 nm for OPN and BSP-II NP, respectively. AS-NP were found to be smaller than empty NP, 235 ± 72 nm and 291 ± 85 nm, respectively. Lyophilization of the NP suspension resulted in a significant increase of the NP size, 293 ± 85 and 438 ± 124 nm, before and after lyophilization, respectively. The introduction of mannitol as a cryoprotectant reduced the final particle size by 100 nm, 324 ± 104 and 438 ± 124 nm, with and without mannitol, respectively. The Zeta-potential of the AS NP was found similar to empty NP. Both had a small negative charge in PBS, −3.1 ± 0.68 mV and −2.6 ± 0.06 mV, AS loaded and empty NP, respectively. The distribution of AS in the NP was visualized using fluorescent AS sequence, labeled with FITC. A fluorescent signal in over 90% of the particles was observed.
Table 1. The physicochemical properties of AS NP examined (Mean ± SD, each batch was measured in triplicate)
NP stability and release
Lyophilized NP maintained their size and morphology for at least 3 years after preparation when stored at 4°C in a vacuum desiccator. Utilization of mannitol as cryoprotectant conveyed a stable and easily resuspendable formulation, whereas NP prepared without mannitol tended to aggregate. Gel electrophoresis studies demonstrated that the ASs encapsulated in NP remained intact, and structural integrity of the AS was maintained (data not shown).
The sustained release of the AS from the NP is shown in Figure 1. About 80% of the loaded AS was released over a 1-month period, with a burst of release (13%) after 1 day.
NP effect on cell viability
First, the possible toxic influence of the NP on HeLa cells' viability was examined. The average total cell counts were 270,000 ± 4,714 cells/ml and 278,333 ± 30,641 cells/ml, 24 and 48 hr after treatment, respectively. The cells' number (living and dead) was similar in treated and nontreated groups 24 hr after treatment (living cells, 95.0% and 96.3%; and dead cells, 5.0% and 5.2% in NP treated and nontreated groups, respectively), as well as after 48 hr (living cells, 94.8% and 95.6%; dead cells, 5.2% and 4.4% in NP treated and nontreated groups, respectively).
Visualization of AS NP uptake
The cellular uptake of fluorescent AS sequences encapsulated in rhodamine-labeled PLGA NP by HeLa and MDA-MB-231 tumor cells was examined. Blank (nonlabeled) empty NP exhibited very low fluorescence, similar to that of nontreated cells, and was determined as background for further analysis. Empty fluorescent NP were shown to internalize and accumulate in the cytoplasm of both HeLa and MDA-MB-231 cancer cells, reaching maximal uptake within the first 24 hr (figure not shown). No significant auto-fluorescence was observed in HeLa cells chosen for further uptake evaluation experiments.
Similarly to the empty NP, AS loaded NP were internalized by the cells and accumulated in the cellular cytoplasm. High NP internalization was achieved by 24 hr (Fig. 2). In contrast, naked AS uptake was decreased overtime (Fig. 2). Confocal cross-section images of HeLa cells 24 and 48 hr after treatment with AS NP (Fig. 2, bottom) verified cell internalization and cytoplasm localization of the NP (green color).
Quantification of AS NP cellular uptake
Number of stained cells: Untreated and treated with blank (nonfluorescent empty) NP cells showed very low fluorescence at all time points, localized under the threshold (101) as seen in the dot-plot (Fig. 3), which was determined as background. The number of cells, internalized with fluorescent AS or NP, was calculated from dot-plot data and is presented in Figure 4. Cells treated with empty NP exhibited a relatively high uptake of 13.6 ± 0.01% already after 4 hr, followed by a gradual increase and thus reaching maximum uptake after 24 hr (44.02 ± 0.68%), and a slight decrease after 48 hr (39.65 ± 0.02%, Fig. 4a). Treatment with naked AS resulted in a significant uptake (17.18 ± 0.83% of the cells) already 4 hr after treatment. The naked AS internalization rate increased to its maximum level after 24 to 48 hr (35.22 ± 19.19% and 30.40 ± 0.11%, respectively, Fig. 4b). In spite of a relatively rapid NP internalization, fluorescent signal from FITC-AS encapsulated in NP was detected in less than 10% of the cells after 8 hr. In contrast, the number of cells loaded with AS was markedly increased after 24 hr, reaching 58.81 ± 25.75% (Fig. 4b).
Extent and intensity of NP endocytosis: The representative histogram charts (Fig. 5) demonstrate the continued uptake over time of AS NP. After 24 hr, a significant increase of NP uptake was exhibited in cells treated with empty rhodamine-NP. AS NP internalization (encapsulated FITC AS) was also observed after 4 hr, but to a lesser extent than naked AS (Fig. 5, top). In contrast, the fluorescence of FITC-AS in NP was markedly increased after 24 hr, and was much higher than that of naked AS (Fig. 5, bottom). Fluorescence intensity, the total cumulative signals from the cells, calculated from the data of the histogram plots, is presented in Figure 6a and 6b. Internalization of naked AS was significant already after 4 hr (41.48 ± 0.03). The AS accumulated inside the cells, reaching 63.22 ± 0.01 and 74.45 ± 0.11, after 24 and 48 hr, respectively (Fig. 6b). In contrast, empty NP uptake was slower, 21.12 ± 0.01 after 4 hr, increasing gradually to 46.43 ± 0.01 after 24 hr (Fig. 6a). The uptake of rhodamine-NP was decreased after 48 hr. Following treatment with AS encapsulated in nonfluorescent NP, relatively low levels of AS were detected inside the cells after 4 and 8 hr, increasing significantly after 24 hr (150.25 ± 0.87, Fig. 6b).
The inhibition of metastatic bone lysis by OPN and BSP-II AS-NP was evaluated in a rat mammary carcinoma metastasis model in comparison to the osmotic mini-pumps delivery system (Table 2, and Figs. 7 and 8). Most animals treated with nonspecific NS delivered by NP (4 of 5 rats) or implanted with NS osmotic mini-pumps (3 of 3 rats) developed bone metastasis (Table 2) and multiple lesions in the femur, tibia and fibula (Fig. 8). The metastatic lesions were detectable for the first time after 4 weeks (mean lesion size of the control group, 16 ± 14 RU, doubled in size until week 8 (38 ± 36) and continued to grow in all rats till the end of the experiment (week 12).
Table 2. Inhibition of metastatic bone lysis by OPN and BSP-II antisenses delivered by osmotic mini-pumps and nanoparticles
Treatment with OPN AS and BSP-II AS loaded NP resulted in a significant decrease in tumor bone metastasis incidence, as assessed by the reduced appearance of osteolytic lesions (Table 2) and in a reduced size of the lesions in metastasis-positive rats (Fig. 7). Two of the 4 rats developed lytic metastasis following the treatment with OPN NP, and 3 of 5 with BSPII NP, but the lesions in the AS NP treatment groups were distinctly smaller in size than those in the control group of NS-NP (Table 2 and Fig. 8b). Both AS sequences successfully prevented metastasis incidence and even caused tumor regression in most animals; 3 of 4 and 3 of 5 animals were found free of visible metastasis at the end the observation period (week 12) following the treatment with OPN and BSP-II NP, respectively (Table 2). Minute osteolytic lesions were found at 6 and 8 weeks and disappeared 10 weeks after the treatment both with OPN and BSP-II AS, delivered by osmotic mini-pumps (Fig. 8a). BSP-II AS had a similar efficiency in inhibition of metastasis in both delivery systems (Table 2 and Fig. 7), but was found less effective than OPN in bone lysis reduction, when delivered by NP (Fig. 8). In comparison with the osmotic mini-pumps delivery system, treatment with NP loaded with AS-OPN was found somewhat more effective in inhibiting the metastatic process (incidence and mean size) and bone lysis. Three of 4 rats, treated with OPN mini-pumps, developed osteolytic lesions with a mean lesion size of 112 ± 182 (Table 2). In contrast, only 1 rat had a small size lesion (4.5 ± 9), which was not detected in theX-ray scans (Table 2 and Fig. 8b) of the group treated with OPN NP.
AS oligonucleotides represent potential therapeutic agents for the inhibition of cancer cell proliferation, migration, tissue homing and progression of the metastasis. One of the major challenges in gene therapy is efficient cellular entry, which could be achieved by NP.21, 23–26 The PLGA-based AS formulation examined exhibited a relatively high encapsulation efficiency of >50% (Table 2). Ca2+ was used as a counter ion for AS in order to reduce its solubility in water. The introduction of calcium appears essential for efficient AS entrapment as was demonstrated for pDNA formulations. It is suggested that calcium cations reduced the AS sequences' negative charge and minimized AS escape to the exterior phase.20
NP size is a significant determinant of the formulation safety and efficacy. Large particles are prone to thrombosis, and are rapidly cleared from the circulation by the mononuclear phagocytic system (MPS).27 On the other hand, particles smaller than 100 nm are capable of penetrating various tissues and escaping the MPS.27, 28 Particles larger than 100 nm are most likely to be taken up by tumor tissue due to enhanced permeability and retention effect.27, 29 The spherical NP produced, in average sizes smaller than 300 nm, and with a narrow size distribution (Table 2, and Fig. 1) seem suitable for a systemic treatment.
Encapsulation of fully phosphorothioated (PT) AS sequences used in this study resulted in smaller NP size in comparison to empty NP (Table 1), and to partially PT AS.21 This is probably due to the relatively larger negative charge of the fully PT AS, contributing to stronger interaction with the calcium cations, resulting in a more compact sequence and a denser NP matrix. In addition, sonication, utilized in the present study, decreasing the double emulsion droplets size,21, 30 resulting consequently in smaller particles as well as a narrow size distribution could be achieved. In contrast to larger DNA sequences, short sonication time (1 minute) is safe for AS 18–20 bases long,31 as is also evidenced from the stability studies. An additional important parameter in the formulation was the utilization of mannitol as a cryoprotectant.32–34 Mannitol inclusion in the formulation prevented aggregation and resulted in a reduced particle size after lyophilization (Table 1). AS NP prepared using the above technique remained stable for a relatively long period of time (over 3 years), and retained both, the NP and AS integrity as well as NP morphology and size. Body distribution, tissue affinity, cellular uptake and activity of formulation are influenced by its surface charge.35, 36 It was shown that the negative charge of nucleic acids, including AS oligonucleotides, can significantly decrease cellular uptake of these molecules, consequently affecting their therapeutic efficiency.37, 38 Due to AS encapsulation in the NP, the large negative surface charge of the AS was significantly decreased; indicating that most of the sequences were distributed inside the NP matrix, rather than on the NP surface.
Controlled release polymeric systems offer the advantage of sustained AS activity and therapeutic effect for prolonged time periods, which can be critical in anticancer and antimetastatic therapy. A sustained release profile of the AS NP formulation in physiologic medium was exhibited for a period of over 1 month (Fig. 1). The sustained release pattern of the AS encapsulated in the NP was confirmed in the cellular uptake experiments. Despite the efficient NP uptake and high AS loading in the NP, relatively high levels of FITC AS were observed inside the cells only after 24 hr (Figs. 4–6), suggesting that the AS FITC signal was shielded by the encapsulating polymer in the NP. Upon release of the AS from the NP the signal was markedly increased correlating with the release profile of the encapsulated AS from the NP (Fig. 1).
Maximal NP uptake was achieved after 24 hr with a slight decrease after 48 hr (Figs. 4 and 6). Since no effect of NP on tumor cell viability was found, the decrease in cellular uptake observed after 48 hr is probably due to cell saturation. Although naked AS was internalized in the cells to a similar extent as AS-NP within the first 8 hr, its uptake efficiency was markedly lower compared to the encapsulated AS at later time points (Figs. 4b and 6b). The limited uptake of naked AS is probably due to its large negative surface charge.37, 38 Moreover, rapid enzymatic degradation and transport out of the cells of unprotected sequences were also shown to decrease intracellular AS accumulation.13, 14, 37 Following the encapsulation in the NP, the AS surface charge was almost eliminated, promoting cellular uptake. Moreover, the NP formulation of controlled AS release conferred protection of the AS sequences from degradation, thus increasing their uptake, and maintaining higher AS amounts in the cell cytoplasm during longer periods of time.
Expression of many genes correlates with metastasis, and some have been shown to play a causal role in this process. OPN and BSP-II AS were shown by our group to effectively knockdown the expression of the respective proteins, and inhibit the colony as well as metastasis formation of pre-exposed MDA-MB-231 human mammary carcinoma cells.11, 12 A mitogenic and antiapoptotic function has been described for OPN39, 40; reduction of such activity was consistent with reduced colony formation in our previous studies,11, 12 and hence could be useful for anticancer treatment in general. Thus, these proteins were considered to be potential targets for AS treatment for the inhibition of bone metastasis formation.
Treatment with OPN-AS or BSP-AS loaded NP resulted in a significant decrease in both, tumor bone metastasis incidence and in the size of the lesions in rats with metastases. Although it was shown that OPN AS is more potent than BSP-II in inhibiting tumor cell proliferation, colony formation and metastatic bone lysis,11, 12 no significant difference between the two was observed probably due to the low number of animals used which precludes the detection of subtle differences. AS-NP effectively inhibited the formation of osteolytic lesions in most animals, and significantly reduced the lesion size in the metastasis positive rats (Table 2 and Figs. 7 and 8). There was no significant difference in efficiency between the 2 delivery systems during the first 4 weeks of treatment. However, the AS-NP possessed a better therapeutic efficacy than the osmotic mini-pumps in terms of lesion ratio at later time periods (8–12 weeks). This could be due to an increased uptake of AS-NP as compared with naked AS. Nevertheless, the total dose administered by the osmotic mini-pumps was much higher, 40 mg vs. 0.6 mg AS from NP. Thus, it is suggested that the NP delivery system enables long therapeutic effect, despite its lower dose, most probably due to protecting the AS from degradation accompanied with a protracted release rate.
It may be concluded that AS delivery by NP is a promising therapeutic modality providing stability of the encapsulated AS and a sustained release. These drugs differ from small-molecule pharmaceuticals in that instead of binding to a pocket of the protein to block certain activities, they prevent the protein from being expressed altogether. This minimizes side effects, and increases the effectiveness of the drug. Further research should focus on targetable NP,38, 41, 42 which may enable systemic administration rather than the regional administration mode utilized in this study.