Image-Guided Breast Tumor Therapy Using a siRNA Nanodrug

Abstract

Iron oxide nanoparticles offer a feasible tool for combined imaging and delivery of siRNA to tumors, stimulating active interest in exploring different imaging and delivery platforms suitable for detection by a variety of modalities. In this study we describe the synthesis and testing of a tumor-targeted nanodrug (MN-EPPT-siBIRC5) that is designed to specifically shuttle siRNA to human breast tumors. The nanodrug binds the tumor-specific antigen uMUC-1, which is found on over 90% of human breast adenocarcinomas. MN-EPPT-siBIRC5 consists of superparamagnetic iron oxide nanoparticles (for magnetic resonance imaging), the dye Cy5.5 (for near-infrared optical imaging), peptides (EPPT) that specifically target uMUC-1, and a synthetic siRNA that targets the tumor-specific anti-apoptotic gene BIRC5. Nanodrug uptake by human breast adenocarcinoma cells resulted in a significant downregulation of BIRC5. Following intravenous delivery into subcutaneous mouse models of breast cancer, the nanodrug demonstrated a preferential tumor uptake, which could be visualized by MRI and near-infrared optical imaging. Furthermore, MRI could be employed to quantitatively monitor nanodrug bioavailability in the tumor tissue throughout the course of treatment. Intravenous injection of the agent once a week over two weeks resulted in the induction of considerable levels of necrosis and apoptosis in the tumors translating into a significant decrease in tumor growth rate. Our strategy permits the simultaneous tumorspecific delivery of siRNA to tumors and the imaging of the delivery process. More generally, it illustrates the potential to apply this approach to many human cancer studies, including for basic tumor biology and therapy.

INTRODUCTION

RNA interference (RNAi) holds considerable potential as a molecular therapeutic tool due to its broad applicability and exquisite specificity (1–3). The main hurdle to the application of RNAi in vivo is the difficulty in achieving efficient delivery of the siRNA to the target tissue, due to phenomena such as RNase degradation, interaction with blood components, and inefficient translocation across the cell membrane. To address these obstacles, many approaches have been proposed for in vivo siRNA delivery. They include liposome-mediated delivery of siRNA in stable nucleic acid-lipid particles (SNALP) (4), polymer-based delivery using atelocollegen and chitosan (5, 6), conjugation to cholesterol (7, 8), and complexing with positively charged peptides or proteins (9–11), to name a few. Various nanoparticle carriers shave been proposed for siRNA delivery in vitro (12–14) and in vivo (15, 16), also reviewed in (17).

In order to evaluate the success of siRNA-mediated therapy, it is also very important to monitor its bioavailability following in vivo administration as well as the associated therapeutic effect, since this will help to develop more successful delivery strategies. In this regard, non-invasive imaging plays an important role, as a technology that permits the in vivo monitoring of siRNA delivery.

Magnetic resonance imaging (MRI) represents a suitable modality for this purpose, because it is characterized by a high spatial resolution, tomographic capability, and the potential to provide quantitative information about contrast agent abundance in tissue (18). At present, few reports have described the application of MRI for image-guided siRNA delivery in vitro (19) and in vivo (20–22).

Previously, we have demonstrated the feasibility of MRI-guided siRNA delivery to tumors, using myristoylated polyarginine-conjugated magnetic nanoparticles that accumulate in tumor tissue through the enhanced permeability and retention effect (20). Despite their capability to mediate very efficient silencing in vivo, these nanoparticles exhibited a high degree of nonspecific uptake, especially by the liver, leading to a more rapid degradation and a reduced effective dose of the agent at the tumor site. A more biologically relevant approach towards improving the bioavailability of the siRNA complex would involve the implementation of a tumor-targeted design. This is a necessary step in order to transform our original delivery agent into an effective, clinically relevant nanodrug.

In the present study, we address this issue by constructing a tumor-selective probe (MN-EPPT-siBIRC5) which consists of magnetic nanoparticles (MN for magnetic resonance imaging), labeled with near-infrared dye Cy5.5 (for optical imaging) and conjugated to a peptide (EPPT), which targets the tumor-specific antigen uMUC-1, as well as to siRNA against the anti-apoptic gene birc5, which encodes survivin. We have previously established the in vivo tumor-targeting properties of the MN-EPPT platform in a variety of adenocarcinoma models, including breast cancer (23–25). Our present results demonstrate the application of this platform for the efficient image-guided delivery of siRNA to breast tumors and the mediation of a robust therapeutic effect, illustrating the potential of this agent as a novel cancer nanodrug.

MATERIALS AND METHODS

Nanodrug Synthesis

The EPPT peptide was synthesized by general Fmoc chemistry using 2-6H-Benzotriazole-1-yl-9,1,3,3-tetramethylammonium hexa fluorophosphate (HBTU) and 1-hydroxybenzotriazole (HOBt) activating agents. Amino acids and resin were purchased from EMD Chemicals (Gibbstown, NJ). All other reagents were purchased from Advanced Chemtech (Louisville, KY), Sigma-Aldrich (St. Louis, MO), GE Life Sciences (Piscataway, NJ) and Fisher-Scientific (Pittsburgh, PA) and used without further purifications. The sequence Cys-(PEG)₂-Tyr-Cys(Acm)-Ala-Arg-Glu-Pro-Pro-Thr-Arg-Thr-Phe-Ala-Tyr-Trp-Gly-Lys(FITC)-CONH₂ (EPPT) was synthesized in a 0.1 mmol scale of Rink amide methyl benzyl hydro amine (MBHA) resin. The Lys (Dde) side chain on the resin was selectively cleaved using 2% hydrazine in DMF and coupled with Fluorescein isothiocyanate (FITC). Finally the resin was cleaved by 5 ml of cocktail mixture (81.5% TFA: 5% thioanisole: 5% phenol: 5% water: 2.5% EDT: 1% TIS), precipitated in cold-ether, purified by high-performance liquid chromatography (HPLC) using eluting solvents A and B (A = 0.1% trifluoroacetic acid (TFA) in 95% water + 5% acetonitrile and B = 0.1% TFA in 90% acetonitrile + 10% water) and characterized by Matrix-assisted laser desorption/ionization (MALDI) mass spectrometry; 2827.17 calculated and 2828.60 found.

The parental MN (crosslinked dextran-coated superparamagnetic iron oxide nanoparticles) were synthesized as described in (26).

The subsequent steps are outlined in Figure 1A. The MN-Cy5.5 precursor was synthesized as described in (23, 25). The targeting EPPT peptide was coupled to MN-Cy5.5 for the resultant MN-EPPT precursor probe using a protocol modified from (23, 25). Briefly, the nanoparticles were conjugated to the heterobifunctional cross-linker, N-succinimidyl 3-(2-pyridyldithio) propionate (SPDP, Pierce Biotechnology, Rockford, IL) and the product was purified using a Sephadex PD-10 column (GE Life Sciences, Piscataway, NJ) against a 100 mM sodium phosphate solution at pH 8.1. The peptide, EPPT, was attached to this linker via the sulfhydryl reactive pyridyl disulfide residue in a 100 mM sodium phosphate solution, pH 8.1.

FIGURE 1.

Synthesis and characterization of MN-EPPT-siRNA. (A) Flow-chart of the synthesis. (B) Nanodrug characterization. The nanodrug consisted of dextran-coated magnetic nanoparticles (MN) triple-labeled with Cy5.5 dye, EPPT peptides, and synthetic siRNA duplexes. (C) Gel electrophoresis demonstrating dissociation of siRNAs from the nanoparticles under reducing conditions.

The siRNA oligos were then conjugated to MN-EPPT using a protocol modified from (20, 27). Five-prime-sense thiol-modified birc5-targeting and scrambled siRNA duplexes were designed and synthesized by Dharmacon (Lafayette, CO). For conjugation, the siRNA was de-protected prior to co-incubation with the magnetic nanoparticles, as recommended by the supplier (Dharmacon, Lafayette, CO). Briefly, the duplex was de-protected with a solution of 3% TCEP followed by a 9.5 M ammonium acetate/ethyl alcohol precipitation. Following de-protection and precipitation, the siRNA was dissolved in 50 mM NaCl, 10 mM EDTA at pH 8 and incubated with the magnetic nanoparticles overnight at 4°C. The final nanodrug was purified using a G-50 Sephadex Quick Spin column (Roche Applied Science, Indianapolis, IN).

Nanodrug Characterization

The nanoparticle concentration was determined based on iron concentration, measured spectrophotometrically, as described in (28). The conjugation ratio of Cy5.5 and EPPT-FITC to MN was quantified spectrophotometrically, as described in (16). The number of EPPT-FITC peptides and Cy5.5 molecules per nanoparticle were obtained from absorbance at 494 nm (extinction coefficient, 68,000 M⁻¹ cm⁻¹) and 678 nm (extinction coefficient, 250,000 M⁻¹cm⁻¹), respectively, In order to characterize the conjugation efficiency and dissociation stoichiometry of siRNA from MN, we performed gel electrophoresis of MN-EPPT-siRNA, following incubation in the reducing buffer dithiothreitol (100 mM DTT for 1 hr at 37°C). The siRNA standard and the nanodrug treated with the reducing agent were electrophoresed on a 15% TBE gel (Invitrogen, Carlsbad, CA) at 200V for 45 minutes. After electrophoresis, the gel was stained with 0.5μg/ml ethidium bromide for 30 minutes, and visualized using a Molecular Imager FX scanner (Bio-Rad, Hercules, CA). The images were analyzed using the software Quantity One, version 4.4.0 (Bio-Rad, Hercules, CA).

Cell Line

All of the studies were performed using the BT-20, and/or CAPAN-2 and LS-174T cell lines (human breast, pancreatic and colorectal adenocarcinoma, ATCC #HTB-19, HTB-80, and CL-188, respectively). The cell lines were authenticated based on viability, recovery, growth, morphology, and isoenzymology by the supplier (American Tissue Collection Center, ATCC, Manassas, VA).

Flow cytometry

Nanodrug uptake by BT-20, CAPAN-2, and LS-174T cells was analyzed using flow cytometry. Cells were incubated with MN-EPPT-siBIRC5 for 48 hrs at 37°C, washed, and analyzed in the FL4 channel (Cy5.5, MN) and the FL2 channel (DY547, siRNA). To correlate nanodrug uptake to uMUC-1 expression, after incubation with the nanodrug, the cells were fixed in 2% paraformaldehyde, incubated for 1 hr at 4°C with a uMUC-1 specific monoclonal antibody (FITC-labeled mouse anti-human CD227, BD Biosciences, San Jose, CA) and analyzed in the FL4 channel (Cy5.5, MN) and the FL1 channel (FITC, uMUC-1-specific antibody). Flow cytometry was performed using a FACSCalibur (Becton Dickinson, San Diego, CA, USA) equipped with the Cell Quest software package (Becton Dickinson, San Diego, CA).

MR Imaging of Cell Phantoms

BT-20 cells were treated as described for flow cytometry. The cells were then pelleted in 0.2 ml PCR tubes and imaged. Imaging was performed using a 9.4T Bruker horizontal bore scanner (Billerica, MA) equipped with ParaVision 3.0 software. The imaging protocol consisted of coronal T2 weighted spin echo (SE) pulse sequences with the following parameters: SE TR/TE = 3000/[8, 16, 24, 32, 40, 48, 56, 64]; FOV = 32 ×32 mm; matrix size = 128 × 128 pixels; slice thickness = 0.5 mm; in-plane resolution = 250 × 250 μm². Image reconstruction and analysis were performed using Marevisi 3.5 software (Institute for Biodiagnostics, National Research Council, Canada). T2 maps were constructed according to established protocol by fitting the T2 values for each of the eight echo times (TE) to a standard exponential decay curve.

T2 relaxation times were calculated by manually segmenting out the cell pellet on MR images.

Real-Time Quantitative RT-PCR

To assess birc5 knock-down by the nanodrug, BT-20, CAPAN-2, and LS-174T cells were incubated with MN-EPPT-siBIRC5 or the control MN-EPPT-siSCR (100 μg iron/ml) for 48 hrs at 37°C. Total RNA was extracted using the Rneasy Mini kit, according to the manufacturer’s protocol (Qiagen Inc., Valencia, CA). Relative levels of uMUC-1 mRNA were determined by real-time quantitative RT-PCR (TaqMan protocol). TaqMan analysis was performed using an ABI Prism 7700 sequence detection system (PE Applied Biosystems, Foster City, CA). The PCR primers and TaqMan probe specific for MUC-1 mRNA were designed using Primer express software 1.5. Primer and Probe sequences were as follows:

• Forward primer, 5′-ACAGGTTCTGGTCATGCAAGC-3′ (nucleotides 64-84 in the 5′ non-repetitive region);

• Reverse primer, 5′-CTCACAGCATTCTTCTCAGTAGAGCT-3′ (nucleotides 139-164 in the 5′ non-repetitive region);

• TaqMan Probe, 5′-FAM-TGGAGAAAAGGAGACTTCGGCTACCCAGA-TAMRA-3′ (nucleotides 96-124 in the 5′ non-repetitive region).

Eukaryotic 18S rRNA TaqMan PDAR Endogenous Control reagent mix (PE Applied Biosystems, Foster City, CA) was used to amplify 18S rRNA as an internal control, according to the manufacturer’s protocol.

Tumor Model

Five to six weeks old female nu/nu mice (n = 8; Massachusetts General Hospital Radiation Oncology breeding facilities) were injected subcutaneously with 3×10⁶ BT-20 tumor cells (American Type Culture Collection, Manassas, VA). Animals were used in experiments on days 10–14 after the inoculation, when tumors were ~ 0.5 cm in diameter. All animal experiments were performed in compliance with institutional guidelines and according to the animal protocol approved by the Subcommittee on Research Animals Care (SRAC) at Massachusetts General Hospital.

Animal Treatment

Treatment with MN-EPPT-siBIRC5 or the control MN-EPPT-siSCR probe involved systemic administration through the tail vein at a dose of 10 mg Fe/kg, 400 nmoles siRNA/kg once a week over the course two weeks. Based on our previous biodistribution data (23) the amount of probe/siRNA delivered to the tumor with each injection is about 2 mg Fe/kg, which is equivalent to ~ 8 nmoles siRNA/kg.

In vivo MR Imaging

MRI was performed before and 24-hrs after each nanodrug injection using a 9.4T Bruker horizontal bore scanner (Billerica, MA) equipped with ParaVision 3.0 software. The imaging sequences were the same as for in vitro MRI. Image reconstruction and analysis were performed as described for in vitro MRI.

Tumor volumes and T2 relaxation times were calculated by manually segmenting out the tumor on MR images. Quantitative evaluation of differential tumor growth by MRI was based on multislice T2-weighted images. The volume was estimated by adding up the number of voxels occupied by tumor and multiplying by voxel volume. For T2 map analysis of relaxation times, the terminal slices were not included in order to avoid interference from partial volume effects.

In vivo and ex vivo Optical Imaging

In vivo optical imaging was performed immediately after each MR imaging session. Animals were placed prone into a whole-body animal imaging system (IVIS Spectrum, Caliper Life Sciences, Hopkinton MA), equipped with 10 narrow band excitation filters (30-nm bandwidth) and 18 narrow band emission filters (20-nm bandwidth) that assist in significantly reducing auto-fluorescence by the spectral scanning of filters and the use of spectral un-mixing algorithms (Caliper life science Corporation, Hopkinton, MA). Imaging was performed using a 675-nm excitation and a 720-nm emission filter. The fluorescence imaging settings (exposure time: 0.5 sec., F-stop: 2; Binning: medium) were kept constant for comparative analysis. Gray scale white-light photographs and epifluorescent images were acquired and superimposed. For ex vivo imaging, excised tumors and adjacent muscle tissue were placed in the optical imaging system and imaged as above. The images were reconstructed using the Living image software version 3.1 (Caliper life science Corporation, Hopkinton, MA).

Immunohistochemistry and in situ Apoptosis Detection

To detect the nanodrug in tumor tissues, tumors were embedded in Tissue-Tek O.C.T. Compound (Sakura Fineteck, Japan) and snap-frozen in liquid nitrogen. Tumors were then cut into 7-μm frozen sections, fixed in 2% paraformaldehyde, washed, counterstained with VECTASHIELD Mounting Medium with DAPI (Vector Laboratories, Burlingame, CA) and analyzed by fluorescence microscopy. Microscopy was performed using a Nikon Eclipse 50i fluorescence microscope equipped with an appropriate filter set (Chroma Technology Corporation, Rockingham, VT). Images were acquired using a CCD camera with near infrared sensitivity (SPOT 7.4 Slider RTKE; Diagnostic Instruments, Sterling Heights, MI) and analyzed using SPOT 4.0 advanced version software (Diagnostic Instruments, Sterling Heights, MI). Fluorescence was collected in the green channel for detection of the FITC label on EPPT peptides, blue channel for DAPI and in the NIR channel for detection of the Cy5.5 label on MN nanoparticles.

To evaluate levels of apoptosis in tumor cells, we performed a terminal deoxynucleotidyl transferase –mediated dNTP nick end-labeling (TUNEL) assay (Apoptag Fluorescein In Situ Apoptosis Detection kit, Chemicon International, Temecula, CA) according to the manufacturer’s protocol. The nuclei were counterstained with DAPI and examined in the DAPI (nuclei) and FITC (apoptotic nuclei) channels. For quantitative analysis of apoptotic levels, a total of 100 nuclei over two separate slides were taken into account. To evaluate levels of necrosis, we stained sections with H&E and analyzed them by light microscopy.

Statistical Analysis

All data were represented as means ± SD. Statistical analysis was performed using two-tailed Student’s t test and linear regression where indicated. A p-value ≤ 0.05 was considered statistically significant.

RESULTS

MN-EPPT-siRNA synthesis

The MN-EPPT-siRNA (BIRC5/SCR) probes consisted of dextran-coated magnetic nanoparticles (MN), triple-conjugated to Cy5.5, EPPT, and siRNA (Fig. 1B). A ratio of four Cy5.5 molecules and eight EPPT peptides per nanoparticle was obtained. In order to characterize the conjugation efficiency and dissociation stoichiometry of siRNA from MN, we performed gel electrophoresis of MN-EPPT-siRNA, following incubation in the reducing buffer dithiothreitol (DTT). The resultant siRNA:MN molar ratios reached 7 for MN-EPPT-siBIRC5 and 7.15 for MN-EPPT-siSCR (Fig. 1C).

In vitro uptake and silencing efficacy of MN-EPPT-siBIRC5

To assess the cellular uptake of MN-EPPT-siBIRC5 by human BT-20 breast adenocarcinoma cells, we performed flow cytometry, following incubation with the nanodrug. A representative dot plot is shown in Figure 2A. A considerable 97.9 ± 0.8 % of the cells were labeled with MN-EPPT-siBIRC5. The good cellular co-localization between fluorescence in the FL2 channel (Dy547, siRNA) and the FL4 channel (Cy5.5, MN) indicated stability of the nanodrug upon cellular uptake. In addition, to demonstrate that the uptake is governed by the expression of MUC-1, we co-stained the cells with a MUC-1 specific monoclonal antibody. The cellular co-localization between fluorescence in the FL1 (MUC-1, FITC) and the FL4 channel (Cy5.5, MN) suggested MUC-1 specific uptake (Fig. 2B). Similar results were obtained with human pancreatic and colorectal adenocarcinoma cell lines (Suppl. Fig. 1A and B), illustrating the broad applicability of our approach.

FIGURE 2.

In vitro testing of MN-EPPT-siRNA uptake and silencing efficacy in BT-20 human breast adenocarcinoma cells. (A) Flow cytometry to assess nanodrug uptake. Representative FL2 (Dy547, siRNA) vs. FL4 (Cy5.5, MN) dot plots showing that 97.9 ± 0.8 % of the cells were labeled with the nanodrug. The good cellular co-localization between fluorescence in the two channels indicated stability of the nanodrug. (B) Flow cytometry to assess nanodrug uptake as a function of uMUC-1 positivity. The cellular co-localization between fluorescence in the FL4 (Cy5.5, MN) and FL1 (FITC, uMUC-1-specific antibody) channels suggested that the nanodrug uptake by the cells is representative of uMUC-1 abundance. (C) In vitro T2 weighted MRI. There was a significant shortening of the T2 relaxation times of the cells incubated with MN-EPPT-siBIRC5, relative to PBS-incubated controls, suggesting that the cellular uptake of the nanodrug could be detected by MRI. (D) qRT-PCR of human breast adenocarcinoma cells incubated with MN-EPPT-siBIRC5 or control probes. There was a significant knock-down of birc5 mediated by MN-EPPT-siBIRC5 relative to the MN-EPPT-siSCR control (p = 0.004, n = 4).

Since one of our primary goals is to detect nanodrug bioavailability by MRI, we performed in vitro T2 weighted MR imaging on breast cancer cells, following incubation with MN-EPPT-siBIRC5. Quantitative analysis revealed a significant (p = 0.001, n = 4) shortening of the T2 relaxation times of the cells following incubation with the nanodrug, indicating that the uptake can be visualized and quantified by MRI (Fig. 2C). Finally, to see if MN-EPPT-siBIRC5 can mediate effective silencing of the target gene, we isolated total RNA from the breast adenocarcinoma cell lines 48 hrs after incubation with the nanodrug, and performed quantitative gene expression analysis (qRT-PCR). As seen in figure 2D, treatment with MN-EPPT-siBIRC5 led to a significantly lower expression of birc5 (survivin) relative to the siRNA scrambled, MN-siSCR, probe (p = 0.004, n = 4). Analogous levels of silencing were also seen in human pancreatic and colorectal adenocarcinoma cells (Suppl. Fig. 1C). These findings imply that the MUC-1 tumor-antigen targeted approach to siRNA delivery mediates a robust gene silencing effect.

In vivo imaging of MN-EPPT-siBIRC5 delivery to tumors

In order to monitor the delivery of the nanodrug to tumors, we performed in vivo magnetic resonance imaging (MRI) of mice, implanted subcutaneously with human breast adenocarcinoma tumors. Superparamagnetic iron oxide nanoparticles are characterized by their strong T2 magnetic susceptibility effects. Their presence in tissue is reflected by marked shortening of T2 relaxation times resulting in a loss of signal (darkening) on MR images. In this experiment, tumors appeared characteristically bright on pre-contrast T2 images. Following injection of the nanodrug, there was a notable decrease in T2 relaxation time (signal) associated with the tumors (Fig. 3A), as a result of MN-EPPT-siBIRC5 accumulation. We also monitored tumor T2 relaxation times, as a quantitative measure of nanodrug abundance in tumor tissue during the entire course of therapy. Treatment with the nanodrug involved weekly i.v. injections over the course of two weeks. Imaging was performed before and after each injection for a total of four data points. As shown in Fig. 3B, injection of the nanodrug resulted in a significant decrease of tumor T2 relaxation times (p = 0.02, n = 4). Furthermore, post-injection tumor T2 relaxation times remained significantly lower that the pre-injection values at all time points (pre-injection: 44.2±0.8, post-injection range: 37.5±0.6 − 28.1±4.2, p < 0.02, n = 4), indicating that the selected treatment time-course ensured persistence of the nanodrug in tumor tissue throughout the experiment.

FIGURE 3.

In vivo MR imaging in a subcutaneous breast tumor model. (A) Representative pre- and post-contrast T2-weighted images (top) and color-coded T2 maps (bottom) of tumor-bearing mice injected intravenously with MN-EPPT-siBIRC5 (10 mg iron/kg). The tumors (outlined) were characteristically bright (T2 long) pre-contrast. Twenty-four hours after injection, there was a loss of signal intensity (T2 shortening), associated with the tumors, indicative of nanodrug accumulation. (B) Quantitative analysis of tumor T2 relaxation times. T2 map analysis revealed a marked shortening of tumor T2 relaxation times 24 hrs after nanodrug injection, indicating accumulation of MN-EPPT-siBIRC5.

Independent macroscopic confirmation of nanodrug delivery to the tumors was obtained by near-infrared optical imaging of the same animals immediately after each MR-imaging session. The Cy5 5 near infrared dye attached to MN allowed for the generation of NIRF signal sufficient to detect the tumor-selective delivery of the nanodrug to the tumors (Fig. 4A). These findings were further corroborated by ex vivo imaging of excised tumors. The strong fluorescence associated with the tumors but not with adjacent muscle tissue indicated preferential accumulation of the nanodrug in the tumors (Fig. 4B).

FIGURE 4.

Optical imaging in a subcutaneous breast tumor model. (A) In vivo imaging. The bright near-infrared signal associated with the tumors compared to surrounding tissue, reflected tumor-selective delivery of MN-EPPT-siBIRC5. (B) Ex vivo imaging. There was bright near-infrared fluorescence, associated with the tumors. By contrast, the fluorescence of adjacent muscle tissue was at background levels. (C) Fluorescence microscopy. The co-localization between fluorescence in the green (green, FITC, EPPT) and near-infrared (red, Cy5.5, MN) channels reflected the integrity of the nanodrug after persistence in the circulation. The tissues were counterstained with DAPI (nuclei, blue).

The tumoral uptake observed by MRI and optical imaging was characteristic of both MN-EPPT-siBIRC5 and MN-EPPT-siSCR, as expected, based on the fact that both are targeted to uMUC-1 through the EPPT peptide (Suppl. Fig. 2). Quantitative MRI T2 map analysis revealed that there were no significant differences between the tumor T2 relaxation times of animals injected with MN-EPPT-siBIRC5 or MN-EPPT-siSCR at any treatment time point (p > 0.05, n = 4). This indicated that any phenotypic effects associated with MN-EPPT-siBIRC5 administration are a function of siRNA activity and are not influenced by a potential differential probe bioavailability.

Finally, to analyze the uptake of the nanodrug by tumor cells, following in vivo delivery, we performed fluorescence microscopy of frozen tumor sections. The good co-localization between signal originating from the Cy5.5 label on MN and the FITC label on the EPPT peptide indicated that the nanodrug remained intact after persistence in the circulation (Fig. 4C). Taken together, these findings attested to the favorable bioavailability of MUC-1 targeted MN-siRNAs delivered as part of the MN complex. They demonstrated that the MN-EPPT-siRNA nanodrug successfully targeted tumors and that this process could be detected by in vivo MR and optical imaging.

Therapeutic effects of MN-EPPT-siBIRC5 delivery to tumors

Since birc5 encodes the anti-apopototic Survivin proto-oncogene, silencing of birc5 can mediate a therapeutic effect by inducing necrotic/apoptotic tumor-cell death. We tested the therapeutic potential of MN-EPPT-siBIRC5 by hematoxylin & eosin staining (necrosis) and TUNEL assay (apoptosis) of tumor sections derived from breast-tumor bearing animals treated with the nanodrug. We observed the induction of considerable levels of both necrosis (Fig. 5A) and apoptosis (Fig. 5A and B). There was a 5-fold increase in the fraction of apoptotic nuclei in experimental tumors vs. controls injected with MN-EPPT-siSCR (p = 0.003, Fig. 5B). This translated into a nearly two-fold decrease in tumor growth rate (Fig. 5C). Since there were no differences in uptake of MN-EPPT-siBIRC5 and MN-EPPT-siSCR (Suppl. Fig. 2), we concluded that the therapeutic effect observed here was due to siBIRC5 action. These data indicated that the siRNA oligo remained functional in vivo as part of the MN-EPPT-siRNA complex and illustrated the potential of this technology for tumor therapy.

FIGURE 5.

MN-EPPT-siBIRC5 therapeutic effects. (A) Hematoxylin & eosin (top) and TUNEL (bottom) staining of tumor tissue from mice treated with MN-EPPT-siBIRC5 or MN-EPPT-siSCR. In the H&E sections, there were extensive eosinophilic areas in the experimental but not control tumors, suggesting the induction of necrosis by the siBIRC5-bearing nanodrug. There was a visible increase in the fraction of apoptotic (green) nuclei in the TUNEL-stained tissues from experimental mice relative to controls, indicating the induction of apoptosis. The tissues were counterstained with DAPI (nuclei, blue). (B) Quantitative analysis of the TUNEL experiments. There was a five-fold induction of apoptosis in the tumor tissues of experimental animals relative to controls (p = 0.003). (C) Relative tumor volume measurements of MN-EPPT-siBIRC5 and MN-EPPT-siSCR injected animals over the course of treatment. There was a significant decrease in tumor growth rate in the experimental animals relative to controls, apparent by 8 days after the beginning of treatment (p < 0.01, n = 4).

DISCUSSION

The discovery of the RNAi mechanism of post-transcriptional gene regulation (29) and its relevance to mammalian gene silencing (30) opened up a new possibility for intervention against disease. Therapeutic strategies that rely on RNAi are unique because they would involve the modular design of RNA duplexes capable of mediating powerful phenotypic reprogramming over a broad range of disease applications and with specificity sufficient to correct single-nucleotide mutations. The interest in developing RNAi into a therapeutic modality has engendered a variety of ideas and approaches that have as their final goal the transformation of the siRNA duplex into a molecular drug. Several of these approaches are undergoing clinical trials (31).

Our review of the literature, however, reveals the apparent lack of clinically-relevant image-guided strategies for siRNA-based therapy. While versatile, most nanoparticle delivery systems do not possess imaging capability. We feel that the “image-guided” property is important because imaging could provide key information about the persistence of the agent in the circulation, its relative accumulation in the target tissue, and the associated phenotypic or therapeutic effect.

Here, we illustrate the potential of this approach by designing a tumor-targeted, imaging-capable nanodrug and applying it for MRI-guided breast tumor treatment in a murine xenograft model of the disease. MRI serves the dual purpose of reporting on the relative accumulation of the nanodrug in tumor tissue, as a guide to selecting an optimal treatment time-course, and quantifying the change in tumor volume over the course of treatment, as an indicator of therapeutic response.

In addition to being detectable by MRI, our nanodrug is tumor-targeted. Tumor targeting is deemed necessary to achieve optimized tumoral bioavailability and a minimized non-specific uptake by vital organs. The tumor-selectivity of our nanodrug derives from the tumor-associated uMUC-1 antigen, which is underglycosylated and overexpressed in over 90% of breast cancers and more than 50% of all cancers in humans (32), attesting to the broad applicability of this delivery strategy.

As a therapeutic moiety, we have utilized siRNA to the human birc5 gene. Birc5 encodes survivin, a member of the inhibitor of apoptosis protein (IAP) family, which is highly expressed by most cancers and associated with chemotherapy resistance, increased tumor recurrence, and shorter patient survival, making anti-survivin therapy an attractive cancer treatment strategy (33).

Because of the modular nature of our agent, the described method is not only novel but also highly relevant to a variety of neoplastic and other diseases that can be manipulated at the level of gene expression. Ultimately, we believe that the core elements of the developed technology would be applicable in a clinical setting, since related iron oxides are already in clinical use (34).