Magnetic nanoparticles as a new approach to improve the efficacy of gene therapy against differentiated human uterine fibroid cells and tumor-initiating stem cells

Abstract

Uterine fibroid(s) (UF/UFs) are benign tumors commonly found in women of reproductive age. The long-term outcomes of myomectomies are often hampered by high rates of recurrence (up to 60%).

Objective

To study whether efficient transduction and subsequent elimination of fibroid tumor initiating stem cells during debulking of tumor cells will aid in completely eradicating the tumor as well as decreasing the likelihood of recurrence.

Design

We have developed a localized non-surgical adenovirus-based alternative for the treatment of UFs. Combining viral based gene delivery with nanotechnology provides an opportunity to develop more efficient targeted viral gene therapy. Magnetic nanoparticles (MNPs) complexed to adenovirus, in the presence of an external magnetic field, accelerate adenovirus transduction.

Setting

Research laboratory located in Georgia Regents University, an academic research institution.

Patients N/A Interventions

MNPs complexed to adenovirus (AD GFP) or (AD LacZ) were used to transfect differentiated human fibroid cells in vitro. Main Outcome Measures rate of transduction and tumor growth inhibition.

Results

We observed a significant increase in transduction efficiency among differentiated human fibroid cells at 2 different multiplicities of infection (MOI); 1 and 10 respectively, with MNPs as compared to adenovirus-alone. Human fibroid stem cells transfected with AD-LacZ expressed β-Galactosidaze at (MOI) of 1, 10, and 50 at percentages of 19%, 62%, and 90%, respectively, which were significantly enhanced with MNPs.

Conclusion

When applied with adenovirus herpes simplex thymidine kinase, magnetofection significantly suppressed proliferation and induced apoptosis in both cell types. Through the use of magnetofection, we will prove that a lower viral dose will effectively increase the overall safety profile of suicide gene therapy against fibroid tumors.

Introduction

Uterine fibroids (UFs), also known as uterine leiomyomas are benign neoplasms of the myometrium and represent the most common solid tumor in reproductive- aged women (1, 2). These tumors occur in 77% of women overall with clinical manifestation in 25% of those affected by age 45 (2–5). Although benign, they commonly cause severe symptoms such as heavy, irregular and prolonged menstrual bleeding, and anemia. Other common symptoms include pelvic discomfort, bowel and bladder dysfunction caused by pressure due to anatomical placement, and/or positioning of the fibroids. UFs have also been associated with subfertility and recurrent spontaneous abortion (6–10).These clinical complications seriously impact women’s health and quality of life. UFs are the most common indication for the more than 600, 000 hysterectomies performed in the US annually. Hysterectomy is an invasive major surgery. is oftentimes associated with significant morbidity, possible mortality, and imposes a huge economic impact on the US healthcare delivery system.(10, 11)

For women with symptomatic UFs, that desire future fertility, only limited conservative methods of treatment are available to manage fibroids without compromising subsequent chances of achieving a healthy pregnancy. Due to various factors, more women are delaying childbearing, which has led to an increase in the number of nulligravida patients with symptomatic UFs(5). Despite the burden of suffering, many women affected are averse to surgery and actively seek fertility-preserving alternatives.(12–14) We have previously shown that intratumoral gene therapy, a localized method of UF treatment, has the ability (15–19), to ablate UFs without interfering with ovulation, uterine blood supply or systemic ovarian function. (14, 18) The use of adenovirus complexes for the treatment of UFs serves as a novel and minimally invasive therapeutic option, for this growing group of patients (14).

Adenoviruses are among the most robust gene delivery tools and offer immense promise in the field of gene therapy. Our group has a proven track-record in the utilization of these unique vectors for the development of a localized non-surgical alternative for the treatment of UF tumors. (16) (15–18) This technique allows for the successful ablation of UFs without interfering with ovulation, uterine blood supply, or systemic ovarian function oftentimes associated with other UF treatment modalities.

Though the efficacy and safety profile of replication-incompetent adenoviruses is outstanding, their clinical use with systemic or even localized delivery is hampered by adverse reactions, including thrombocytopenia (20), acquired immune responses mediated by cytotoxic lymphocytes against viral and/or transgene products,(21, 22) and in some cases, may lead to the potentially life-threatening systemic cytokine syndrome (23–25). The latter acute toxic effects are due to activation of the innate immune system and exhibit a steep dependence on vector dose i.e., decreased viral load, and equates to a lessened likelihood of severe immune reaction. Its occurrence varies substantially among subjects, and the likelihood of eliciting such an immune response cannot be readily predicted. Our research group, has developed a method which targets therapeutic adenoviruses towards fibroid lesions and minimizes any potential delivery beyond the tumor lesion. In our approach, we have genetically modified the adenovirus with a targeting short peptide composed of 3 amino acids (Glycine, Arginine, and Aspartic Acid) collectively referred to as the RGD peptide motif. The RGD peptide motif is expressed on the virus capsid to utilize different internalization pathways other than the well-known Coxacie-adenovirus receptor (CAR), commonly expressed on many normal cells.(17) The advantage of the CAR independent RGD pathway is its utilization of the integrin internalization pathway which is highly expressed on fibroid tumor cells as compared to surrounding normal myometrium(17, 26, 27).

We have found that the integration of gene therapy and nanotechnology serves as yet another approach that can be used in minimizing the required dosage of tumor-targeted adenovirus while sufficiently increasing the efficiency of transduction. MNPs conjugated to adenoviral vectors, in the presence of an external magnetic field, have been shown to greatly enhance targeted gene transfer into tumor cells (28). These magnetic nanoparticles accelerate transduction kinetics, a technique referred to as magnetofection (29). The magnetofection method was developed to overcome biological barriers against the delivery of efficient gene therapy through the use of nucleic acids or viral vectors associated with MNPs (29, 30).

The principle of magnetofection is to associate transfection reagents or viruses with specific magnetic nanoparticles; thereby, forming molecular complexes. Resulting molecular complexes are then concentrated and transported into cells supported by an appropriate magnetic field (29, 30). Through the exertion of a magnetic force upon gene vectors, we were able to rapidly increase the concentration of the applied vector dose on cells, so that 100% of the cells come in contact with a high vector dose, thereby promoting cellular uptake.

This approach has not yet been evaluated against human UF tumor cells. Our aim is three-fold: 1) to enhance the efficiency of transduction while maintaining or minimizing viral dose, 2) to enable targeting to fibroid tumor tissue and avoid surrounding healthy myometrium, and 3) to validate that our approach can transduce and eliminate fibroid stem cell populations. The latter would be a novel paradigm-shifting improvement in UF therapeutics. Eliminating tumor-forming fibroid stem cells, would likely prevent tumor recurrence, a major challenge in the field of UFs, and likely prevent the development of new fibroid lesions. Currently available treatment modalities are not capable of affecting fibroid stem cells. Utilization of magnetofection in fibroid gene therapy is novel and innovative. It represents the natural evolution and progression of therapeutic options available. Our group aims to pursue the design and development of cutting-edge approaches which are localized, effective, safe, and fertility-preserving therapies in the treatment of uterine fibroids. In this work, we demonstrate that magnetofection does indeed enhance adenoviral gene therapy, increase lethality against human fibroid cells, and for the first time, demonstrates that such an approach is efficient in eliminating human fibroid tumor initiating stem cells.

Materials and Methods

Cell culture

All experiments were done in accordance with biosafety guidelines of Georgia Regents University and conducted after protocol approval by Georgia Regents University Institutional Review Board. Human immortalized fibroid cells and uterine smooth muscle cells were in-kind gifts provided by Darlene Dixon, PhD (National Institute of Environmental Health Sciences, Research Triangle Park, NC, US). Cells were cultured and maintained in smooth muscle cell basal media (SmBM, Lonza Walkersville, MD, US) containing 10% fetal bovine serum (FBS, Lonza Walkersville, MD, US), 0.1% insulin, 0.2% human fibroblast growth factor–basic (hFGF-B), 0.1% gentamycin sulfate, amphotericin-B (GA-1000), and 0.1% human epidermal growth factor (h-EGF) (Lonza, Walkersville, MD, US). Human fibroid stem cells were isolated and characterized in our laboratory as previously described (31, 32). For cell proliferation assays, cells were grown in a 24-well culture plate and transduced with adenoviral vectors (0–100 PFU/cell) followed by ganciclovir (GCV) therapy (Sigma Co, St. Louis, MO, US) at a concentration of 10 µg/ml for 3–7 days. Cell growth was measured with the MTT kit (Sigma Co, St. Louis, MO, US) per the manufacturer's instructions. AD-GFP vector was used in our initial proof of concept experiments. We then used adenovirus vector encoding HSV-1TK gene under transcriptional control of the Rous sarcoma virus (AD-HSV1TK), as described by Chen et. al. (33). An adenovirus vector expressing a marker gene coding for bacterial β-galactosidase (AD-Lac Z), was used as a negative control. Both viruses were in-kind gifts from Dr. Savio Woo (Mount Sinai School of Medicine, New York, NY, US). The AD-RGD-TK vector, a modified adenovirus vector contains the herpes simplex virus thymidine kinase gene. This complex was prepared in Dr. David Curiel’s lab (Washington University, St. Louis, MO). A genetic incorporation of the RGD-4C motif into the H1 loop of the fiber was used to enhance the infectivity of the vector by permitting binding to the αvβ3 and αvβ5 integrins as previously described.(34). Magnetic nanoparticles were purchased from OZ Biosciences (Parc Scientifique de Luminy, France).

• Preparation of adenovirus−magnetic nanoparticles (MNPs) complexes for magnetofection

Magnetic nanoparticles adenovirus complexes were prepared by mixing 75 µL of Adenomag magnetic nanoparticle suspension with 15 µl of 6.5× 1011 VP of AD-RGD-TK, equal to 6.5 × 109 PFU/ml diluted in 500 µL of serum free culture media, a lower ratio than recommended by the manufacturer of Adenomag (OZ Biosciences, Parc Scientifique de Luminy, Marcella, France). After a 20-minute incubation period at room temperature (RT), different dilutions of the complexes were performed to cover the required multiplicities of infection (MOIs), so that 50 µL of the complexes were added to the cells. AD-GFP, AD-RGD Luciferase, and AD-RGD-TK were utilized in preparation of the complex for proof of infectivity to wild type adenovirus as well as genetically modified AD-RGD variant, and cytotoxicity of AD-RGD-TK adenovirus complexes. Magnetic complexes were applied to cells in a drop-wise manner. Cells were then incubated for 30 min on a super magnetic plate (OZ Biosciences, France) at 37°C, 5% CO2. The magnetic plate was then removed and cells were cultured under standard cell culture conditions until the end of the experiments.

• MNPs characterization

The morphological examination of MNP was carried out through the use of a transmission electron microscope (TEM) (JEOL JEM 1230 Transmission Electron Microscope) with 2% phosphotungstic acid negative staining. The mean size and zeta potential of the nanoparticles were determined by nanoparticle tracking analysis with ZetaView PMX 110 (Particle Metrix, Meerbusch, Germany) and corresponding software ZetaView 8.02.28. Samples were diluted 1 to 10 in 1× PBS and sonicated prior to ZetaView measurement. Zeta potential was measured using 0.05 × PBS instead of 1× PBS for conductivity at approximately 500 µS/cm. Per OZ Bioscience Company, nanoparticles have a hydrodynamic radius of 200–300 nm, with 10–15 nm, metal cores with positive zeta potential. We also characterized adenovirus size by electron microscope before and after conjugation with the MNP through incubation with the viral particles for 20 minutes at RT. All measurements were taken at pH 7.4, 25° C.

• Comparing Adeno-GFP (AD-GFP) regular transduction versus magnetofection

Upon completion of both regular transduction and magnetofection protocols, human fibroid tumor cells and human fibroid stem cells with AD-GFP were washed, and fixed with 4% paraformaldehyde (PFA). GFP expression was detected under an inverted fluorescence microscope equipped with a digital camera (Axiovert, Zeiss Carl Zeiss Microscopy Ltd, Cambridge, MA, US). Images were taken with a 20× Plan-Neofluar dry lens. The same exposure for luminosity and contrast were applied to each slide. Images presented, represent three independent experiments.

• Comparative analysis of AD-RGD luciferase regular transduction versus magnetofection AD-RGD luciferase transfection and magnetofection were performed on human fibroid tumor cells and human fibroid stem cells. Luciferase bioluminescence assays were conducted for group comparisons.

  • Boosting suicidal gene therapy of human fibroid tumor cells and human fibroid stem cells by MNPs and applied to a magnetic field
    Human fibroid tumor cells and human fibroid stem cells were seeded in flat-bottom 24-well plates, at 40,000 cells/well density for human fibroid tumor cells and 2,000 cells/well density for human fibroid stem cells. Cells were then incubated under standard cell culture conditions. After 24h incubation, media in each well was replaced with 600 µL of fresh cell culture media containing 10% FBS; 50 µL of MNPs complexed AD-RGD-TK, AD- Lac Z (negative control) or WT-AD (positive control) virus alone, and prepared as previously described. Cells with MNPs complexed AD-RGD-TK were placed on a magnetic plate in the CO2 incubator for 30 minutes, removed and maintained under regular cell culture conditions. After 24h of infection, media was replenished with fresh cell culture media, and cells were further cultivated until evaluation. All samples were measured in triplicate. Cell survival was assessed by a luciferase assay in stable cells expressing luciferase or by a MTT-based respiration activity assay.
  • MTT Assay: Cells were Infected with Adenovirus RGD thymidine kinase (AD-RGD-TK), adenovirus B-galactosidase (AD- Lac Z), considered the wild type adenovirus (WT-AD) (negative control). Untreated cells were washed with PBS and incubated for 3h in 100 µL of 1 mg/mL MTT solution which was prepared in PBS with 5 mg/ mL glucose. Afterwards, 100 µL of solubilization solution with Dimethyl Sulphoxide (DMSO Sigma, St. Louis.MO, US) was added to dissolve formazan. Optical density was measured at 590 nm and cell viability was expressed as respiration activity normalized to the reference data for untreated cells.
  • Luciferase Assay: Cells were washed with PBS and lysed with 100 µL of lysis buffer (0.1% Triton X-100 in 250 mM Tris; pH 7.8) per well. After incubation for 15–20 min at RT, 50 µL of cell lysate was drawn from each well and transferred into a 96-well black flat-bottom plate. Finally, 100 µL of luciferase buffer was added. Cells were then washed with PBS, and 100 µL of 35 µM D-Luciferin in PBS was added.

Detection of the bystander effect in human fibroid stem cells

Tumor cells transduced with an adenovirus vector expressing the herpes simplex virus thymidine kinase (HSV-TK) gene are rendered sensitive to the anti-herpetic drug, ganciclovir. The bystander effect refers to the observation that not all cells need to be transduced, in order to be killed. This is attributed to the capacity to transfer small cytotoxic molecules from transfected to non-transfected cells due to direct contact and the presence of gap junctions, thus conferring the cytotoxicity to them. We wanted to check for this bystander effect in human fibroid stem cells so that we could account for its action in deducing the viral load required for their complete eradication. Human fibroid stem cells were grown in 100 mm tissue culture plates until 70–80% confluence was achieved and then transfected with the AD-RGD TK as previously described. The next day, cells (as well as cells from uninfected plates) were trypsinized and counted. Different ratios of uninfected/infected cells were cocultured, for a total of 4,000 cells and plated in 24-well plates. 24h later media was removed and replaced with media containing GCV 10ug/ml. Five days later, viable cell counts were determined using MTT Assay.

Isolation of protein and immunoblots

Proliferating cell nuclear antigen (PCNA), and BCL2-associated X protein (BAX) expression was examined using western blot analysis. Ice-cold radio-immune-precipitation assay (RIPA) buffer supplemented with phosphatase and protease inhibitors (50 mM sodium vanadate, 0.5 mM phenylmethylsulphonyl fluoride, 2 mg/ml aprotinin, and 0.5 mg/ml leupeptin) (Sigma, St. Louis, MO, US) was used for protein extraction and for the elimination of homogenates and cell lysate. Protein concentrations were measured by Bradford protein assay (Bio-Rad, Hercules, CA, US). Total protein samples (40 µg) were filtered via SDS-PAGE (Polyacrylamide Gel Electrophoresis) (10% acrylamide gel) using the Bio-Rad Trans-Blot system (Bio-Rad, Hercules, CA, US) and transferred to membranes. Membranes were blocked with 5% non-fat milk in phosphate-buffered saline (Gibco, Grand Island, NY, US) containing 0.1% Tween 20 (Sigma, St. Louis, MO, US)(PBS-T), incubated for 1h, washed in PBS-T, and hybridized with primary antibodies, specific antibodies for PCNA (BAX Santa Cruz Biotechnology, Inc. Dallas, Texas US). In addition, membranes were incubated with a secondary antibody to bind β-actin at a 1:15,000 dilution (Sigma, US), which served as an internal control. Incubation with secondary antibodies and the detection of the antigen-antibody complex were performed using a Super Signal™ West Dura Extended Duration Substrate (Life Technologies, Grand Island, NY, US). Immunoblots of PCNA (36 kDa), BAX (27 kDa), β-actin (43 kDa) were quantified with a Bio-Rad Image Lab. densitometer (Bio-Rad, Hercules, CA US).

Caspase-3 Assay

Caspase-3 enzyme activity was measured in human fibroid stem cells using the Caspase assay system (Abcam, Cambridge, MA, US) based on Caspase 3 enzyme’s ability to release yellow chromophore p-nitro aniline (pNA) from the colorimetric substrate (Ac-DEVD-pNA) provided in the Caspase assay system. Tissue lysates were centrifuged at 13,000 rpm. Reading was at 405 (excitation)/510(emission) using the 96-plate reader Synergy HT with Gen 5 software (Biotek Instruments, Inc., Winooski, VT, US) according to the manufacturer’s instructions. Caspase-3 activity values were normalized against the total tissue protein content measured

Data analysis

Statistical analysis of the samples were done using a Student’s t-test, where P-values ≤0.05 were considered statistically significant with 95% confidence intervals. All statistical analyses were performed using GraphPad Prism version 6.00 for Mac (GraphPad Software, San Diego, US).

Results

MNPs size and surface charge identification

MNPs size ranged from 115 to 214 nm in size, with a mean of 129±71nm and zeta potential of 18.12 mV by ZetaView PMX 110 (Particle Metrix, Meerbusch, Germany) and corresponding software ZetaView 8.02.28.Transmision electron microscopy (JEOL JEM 1230, Peabody, MA) showed the shape of adenovirus and MNPs. Covalent conjugation with adenovirus and MNPs was confirmed by the close intimate approximation of nanoparticles around the surface of adenovirus (Fig. 1).

Figure 1.

Transmission electron micrographs of Adenovirus AD RGD TK, MNP and MNP AD RGD TK. Size characterization of Adenovirus and magnetic nanoparticles (MNPs) by transmission electron microscopy. All samples were negatively stained with phosphotungstic acid solution and observed under transmission electron microscope at different magnifications: (A) Adenovirus (B) MNPs (C), (D) Adenovirus conjugated to MNPs. Bar size is 50 nanometer in A and D and 200 nm in B and C. The upper panel represents the color-enhanced version of the actual micrographs in the lower panel.

MNPs enhance AD-GFP transduction of human fibroid tumor cells

The purpose of conducting this experiment was to evaluate whether conjugating adenoviral vectors with MNPs, would enhance the ability of adenoviral vectors to transfect human fibroid tumor cells. We compared the transduction efficiency of non-conjugated AD-GFP to that of MNPs conjugated AD-GFP at two different multiplicities of infection (MOI), 5 and 10 PFU/cell (Fig.2A). As shown in Fig.2B, there was a significant increase in the number of transfected cells from 8.06% in unconjugated AD GFP group to 28.56 % ± 8.9% in the MNPs conjugated group (p < 0.005), for MOI 5 and an increase from 25.38% in the unconjugated AD GFP group to 85.65 ± 7.57% in the MNPs conjugated group for MOI 10 (p= 0.005).

Figure 2.

in vitro Ad GFP magnetofection: (incubation of Adenovirus with MNPs on magnet for 20 minutes). Human fibroid tumor cells (1 × 106 in each of 6 wells) incubated for 20 minutes with MNPs conjugated adenovirus or regular virus with exposure to external magnetic wells. (A) MNPs boosted Ad-GFP transfection in human fibroid tumor cells by fluorescent microscopy (B) Quantitative analysis to compare the transfection efficiency of Ad GFP with or without MNPs, we noted a significant increase in the transfection rate after conjugating Ad GFP with MNPs at both MOI 5 and 10 (P<0.005)

MNPs enhance AD-RGD-luciferase transduction into human fibroid tumor cells

A previous report from our group (16) demonstrated that by genetically altering adenovirus to express RGD peptide in the viral capsid, transduction efficiency of the adenovirus was markedly enhanced against human fibroid tumor cells. In this study, we aimed to investigate whether further enhancement of transduction efficiency could be accomplished by means of MNPs-adenoviral conjugations.

As shown in Supplementary Fig. 1, MNPs significantly enhanced transduction efficiency of the modified virus AD-RGD-Luc (as reflected in bioluminescence intensity measured by luciferase assay) from 23.43 % in the unconjugated AD RGD LUC to 46.20 ± 5.7% in the MNPs group at MOI 5 and from 44.67±7.8% to 85.77%± 8.6% at MOI 10 10, respectively when compared to the unconjugated AD-RGD-Luc against human fibroid tumor cells (P = 0.005).

MNPs enhance the ability of AD-RGD-TK to suppress proliferation of human fibroid tumor cells

We recently used AD-RGD-TK followed by ganciclovir treatment to efficiently induce apoptosis and inhibit proliferation of human fibroid cells (16). In order to assess the ability of MNPs to further enhance the anti-fibroid ability of AD-RGD-TK, ascending MOIs of MNPs-conjugated versus unconjugated virus were tested against human fibroid cells in vitro. As shown in Supplementary Fig. 2, there was a significant decrease in the percentage of viable cells from 45% ± 1.34 % to 33% ± 1.38% at MOI 25, from 39± 2.6 % to 25 ± 1.43% at MOI 50 and from 32 ± 2.15 % to 21 ± 0.98% at MOI 75 (p<0.0001).

Magnetically enhanced AD-RGD-TK is superior to unconjugated AD-RGD-TK in inducing apoptosis markers in human fibroid tumor cells

As recently reported, part of the anti-fibroid effect of targeted AD-RGD-TK is via induction of apoptosis and inhibition of proliferation in these tumorigenic cells (16). In this work, we wanted to evaluate if MNPs conjugation can further enhance the anti-fibroid capabilities of targeted adenoviral vectors. MNPs conjugated AD-RGD-TK decreased the expression of proliferation-related proliferating cell nuclear antigen (PCNA) and increased the expression of apoptosis-related BAX in human fibroid tumor cells as detected by western blot as compared to AD-RGD-TK alone. (Supplementary Fig. 3)

Adenovirus vector readily transduces human fibroid stem cells

Some of the challenges faced in the treatment of UFs is its high incidence as well as its prolific rate of recurrence (35, 36). Recurrence has been observed after both medical (e.g., treatment with GnRH analogues) (37, 38) and surgical treatment (e.g., myomectomy and myolysis) (14, 39). This is likely due to the inability of these approaches to arrest growth or eliminate tumor forming fibroid stem cells (40–42). Currently, there is no available fibroid treatment strategy designed to target fibroid tumor-forming stem cells. Our group recently isolated and characterized human fibroid tumor-forming stem cells as well as myometrial stem cells (40). In our present work, we push the envelope further by evaluating the ability of adenovirus AD-LacZ to transduce human fibroid stem cells as a first step in building a strategy towards eliminating fibroid tumor-forming stem cells. We demonstrated that human fibroid stem cells are susceptible to transduction by AD-LacZ serotype 5 at MOI of 10, and 25 with percentages of 62% ±12.3, and 90%±23.6, respectively (Fig. 3). These findings led us to the next question, can herpes simplex thymidine kinase suicide gene therapy be used to eradicate human fibroid tumor forming stem cells?

Figure 3.

Transfection of fibroid stem cells by AD-LacZ X Gal staining of human fibroid stem cells after transfection with Adenovirus with Ad- Lac Z reporter gene at multiplicity of infection (MOI) 10 (A) and 25 (B). X-gal is an analogue of lactose and therefore hydrolyzed by the B-galactosidase enzyme giving intensely blue products (Arrows) DAPI, a fluorescent stain that binds strongly to A-T rich regions in DNA, is used as a general nuclear stain to demonstrate total numbers of C/D. E is a numerical representation of results observed in A and B and calculated as described in materials and methods. Results represent 3 independent experiments.

AD-RGD-TK/GCV effectively reduces proliferation of human fibroid stem cells

In view of the potent efficacy of the AD-RGD TK/GCV system against human differentiated fibroid cells tested in vitro (16, 18), we examined its efficacy against human fibroid stem cell. We detected a significant decrease in the percentage of viable human fibroid stem cells within the AD-RGD TK/GCV-treated group at MOI 50 (P<0.005), and 75 (P< 0.0001), as compared to non357 transduced fibroid stem cells (Supplementary Fig.4)

AD-RGD-TK bystander killing effect is operational in transfected human fibroid stem cells

In human fibroid cell lines, the bystander effect of AD-TK/GCV was robust as we have previously reported (43). This raised the question as to whether this unique feature of TK/GCV suicide gene therapy approach could be operational in fibroid stem cells as well. To further investigate this approach, different ratios of AD-RGD-TK transfected fibroid stem cells were cocultured with untransfected wild type cells (WT) and treated for 5 days with 10 µg/ml GCV. By increasing the percentage of transfected cells (10, 20, 50, 70%) in the cell mixture, we observed a significant decrease in cell viability, respective to WT untransfected cells when transfected cell ratios were between 20 and 70% (*P<0.0001) (Supplementary Fig.5). Our data suggest that human fibroid stem cells exhibit a strong bystander effect, as cell numbers significantly decreased when as little as 20% of the cell mixture was infected, and near maximal cell killing ability occurred when 70% of cells were transfected. This could enhance the effectiveness of our therapeutic modality, especially in large fibroid lesions where infecting every tumor and/or tumor-forming stem cell might not be attainable. These encouraging results motivated us to develop an additional targeting strategy for this robust modified adenoviral vector, one that would aim for complete eradication of tumor initiating stem cells and prevent tumor recurrence.

MNPs enhance transfection of AD-GFP to human fibroid stem cells

After demonstrating the enhancing effect of MNPs on adenovirus transfection of human fibroid tumor cells, we wanted to test the same strategy towards fibroid tumor-forming stem cells. Comparing transduction efficiency of AD-GFP with or without MNPs, we found a significant increase in the percentage of GFP Positive cells by 23.66%±6.4, 25.45 %±7.2 and 29%±7.9 at MOI 5, 10, and 25 respectively in cells transfected with conjugated versus unconjugated virus, (P< 0.005) (Fig.4).

Figure 4.

(A) In vitro Ad GFP magnetofection: (incubation of Ad GFP with MNPs for 20 minutes) of fibroid stem cells (1 × 10 3 / cm2 in each of 6 wells), followed by 20 minute exposure to the magnetic field. (B) We observed that MNPs significantly enhanced transfection efficiency at the 3 different MOIs (P< 0.005).

AD-RGD-TK’s ability to suppress proliferation is enhanced by magnetically enforced transfection in human fibroid stem cells

In order to assess MNP’s ability to enhance adenoviral mediated cell death of human fibroid stem cells, our replication-defective adenovirus AD-RGD-TK/GCV was conjugated to MNPs and tested against human fibroid stem cells. This assessment allowed us to determine the level and functionality of thymidine kinase as well as analyze MNP’s role in this novel cell type. Upon transfection of human fibroid stem cells by AD-RGD-TK system with or without MNPs at MOI 50, we observed a significant reduction in the number of viable fibroid stem cells by 55% ± 12.3 in regular transfection group (P < 0.005) and by 78% ±16.75 in MNPs group (P<0.001) as compared to control untreated cells (considered as 100%). When comparing the two suicidal gene therapy strategies, the MNPs enhanced approach accomplished significantly higher cell killing than that of Ad-RGD-TK alone (P<0.01) (Supplementary Fig.6).

Magnetically enhanced AD-RGD-TK is superior to unconjugated AD-RGD-TK in inducing apoptosis markers in fibroid stem cells

To further evaluate the ability of MNPs to enhance AD-RGD-TK induced apoptosis in human fibroid tumor-forming stem cells, these cells were transfected with MNPs-conjugated AD-RGD-TK compared to unconjugated vectors both at MOI of 25. MNPs conjugated AD-RGD TK decreased the expression of proliferation-related proliferating cell nuclear antigen (PCNA) and increased the expression of apoptosis-related BAX (P<0.05) in human fibroid stem cells (Supplementary Fig.7).

Magnetically enhanced AD-RGD-TK is superior to unconjugated AD- RGD-TK in elevating caspase-3 enzyme activity

Upon in vitro assessment of Caspase-3 enzyme activity for comparing magnetofection versus unconjugated transduction of AD-RGD-TK/GCV, there was an increase in Caspase 3 enzyme activity by 3 and 6 folds in regular transfection and magnetofection groups, respectively. P=0.05 (Supplementary Fig.8).

Discussion

Replication-deficient adenoviral vectors have demonstrated great potential as gene therapy vectors. (44) However, researchers still face challenges associated with tissue specific targeting and vector-mediated immunogenicity. Development of a specific and efficient technique for the in vivo delivery of adenoviral vectors to target tissue with effective minimal dosing is a pivotal step forward towards making the use of adenovirus’ gene therapy a clinically acceptable tool. UFs serve as an attractive target for gene therapy due to several inherent biological features. UFs are localized and well circumscribed in the uterus with fibrous capsules that could conceivably simplify targeting of the viral load to the tumor (45). We recently applied several gene therapy strategies to UFs; including the use of an adenovirus-delivered dominant-negative mutant estrogen receptor under a cytomegalovirus (CMV) promoter (AD5-DNER);(46), an AD-herpes simplex thymidine kinas/ganciclovir (HSV-TK/GCV); and finally, an AD RGD modified virus (17, 47, 48). We have experienced considerable success in applying these approaches to the leiomyoma nude mice model as well as the Eker rat model (49).

To optimize our approach toward human fibroid gene therapy, as well as improve the safety profile of this novel treatment, we have presented the utility of several modified AD vectors in human fibroid tumor cells. Our goal was to identify the best AD vector to enable the targeting of therapeutic genes to human fibroid cells with minimal effect on normal myometrial cells, as well as all extra-uterine tissues and organs. To achieve the levels of enhanced transduction efficiency required in the context of UFs, it was necessary to route the AD5 vectors via CAR-independent pathways as UFs typically exhibit low expression of that receptor. One such modification was the insertion of a short peptide (21 amino acid) composed of Arginine, Glycine and Aspartate (RGD) to the H1 loop of the wild AD5 fiber knob domain to reroute AD5 binding to the cellular membrane integrins.(50) The widespread distribution of RGD sequence and integrin binding sites is an important consideration in gene therapy protocols contemplating systemic (intravenous or intraperitoneal) delivery of such vectors. A unique feature of our work is taking advantage of the well-circumscribed easily accessible nature of UF lesions. As we are designing this approach as a localized gene therapy protocol where the therapeutic vector will be directly delivered by intratumoral injection, we mainly focused on the distinct features between UF lesions and adjacent myometrium. As we have described in our previously published work (Nair et. al., 2013) adding RGD motif to adenoviral fiber increased transduction into human fibroid as compared to normal myometrial cells.

Our overarching goal was to selectively target and ablate fibroid tumor tissue without spreading viral particles to surrounding healthy myometrium or extra uterine tissues/organs. There is a critical need to guide and confine the therapeutic vector towards the fibroid tumor mass. This targeted area serves as the site for the delivery of our therapeutic gene. What is readily possible in traditional cell culturing becomes a real challenge when dealing with an animal model or finally in humans. Directing molecules to the site where they are needed, remains difficult without invasive methods (51). In addition, impaired cell transduction as a result of immune reactions remains a challenge in adenovirus gene therapy, especially in clinical scenarios, where repeated delivery is required (52). Throughout the past decade, several experiments using plasmids and/or viruses have shown that magnetofection offers a number of advantages for in vivo gene targeting (53–59) including: (1) high cellular uptake is reached within minutes, and (2) targeted and confined gene expression is made possible by magnetic focusing of MNPs at the desired site of action in a non-invasive manner.(60) Interestingly, the use of a magnetic field to direct a therapeutic modality is not totally alien to the field of UFs. A similar technique is used in the FDA approved magnetic resonance guided focused ultrasound (MRgFUS) therapy, a procedure that has been used successfully for the ablation of fibroid tumors since 2003 (61). In MRgFUS, high-energy focused ultrasound waves are delivered to tumor tissue under magnetic resonance imaging (MRI) causing thermal coagulation of the targeted tissue (62). We have utilized magnetic resonance, in much of the same fashion, to generate a localized field in which to inject the virus to the tumor tissue and restrict its spread into nearby as well as distant tissues. (60) Infusion of the virus will occur until a substantial viral load is achieved; total intracellular entrapment of the virus is sustained; and, the risk of collateral and/or distant spread is eliminated. The nanoparticle/vector complexes might be delivered precisely to the fibroid lesion by imaging-guided injection techniques under guidance of ultrasound or MRI for example, using currently available technology.(63, 64) Subsequently, the same diagnostic MRI field can be utilized to create a limited transient (minutes) magnetic field to concentrate and trap the adenovirus-conjugated magnetic complexes within the fibroids and to limit any potential spread to nearby or distant organs. Such an approach will maximize the therapeutic cytotoxic effect of these adenoviral vectors within the fibroid tumor lesions while improving safety by minimizing external spread.

Several groups have used Affinity complexion of Adenovirus with MNPs to achieve facilitated CAR-independent cellular uptake and enhanced transgene expression in cells otherwise non-permissive to adenoviral transduction.(65) (66). For example, Fallini et.al established that the magnetofection method for the transfection of primary motor neurons proved to be efficient, robust and non-toxic. Their optimized protocol allowed for efficient transfection of motor neurons during different stages of differentiation, high levels of expression over several days, and the co-expression of at least three proteins in one cell. The group deemed the approach cost-effective, easy to perform, and required minimal use of specialized equipment. (67) In a more recent study, researchers demonstrated that MNPs treatment of EGFR-positive lung cancer cells resulted in abrogation of cell cycle arrest in the G2/M phase and induced DNA damage. Consequences of MNP treatment included induction of autophagy, apoptosis, and DNA damage which resulted in effective tumor growth inhibition both in vitro and in vivo. This study established proof-of-concept, by demonstrating a novel mechanism of MNP action – its ability to produce antitumor activity in lung cancer cells (68).Using nanoparticles to enhance adenovirus-mediated gene therapy is novel to the field of UFs. We needed to first evaluate the utility of such an innovative approach in vitro against representative human fibroid cell models, as described in this work. Our future plans include assess ing the efficacy and safety of such an approach in an appropriate in vivo preclinical model of uterine fibroids. We are currently working on designing magnetic jackets to befit the animals in order to expose the tumor region to a well-determined magnetic field immediately after regional injection of the MNPs conjugated virus.

In our current work, we use MNPs as a platform for Adenovirus mediated gene delivery as an approach to address a number of limitations inherent to this gene delivery vector and have found that it significantly enhances transduction efficiency. Our genetically modified CAR-independent vector AD-RGD-TK combined with magnetically enhanced delivery demonstrated a significantly higher apoptosis induction rate in UF tumor cells as well as human fibroids stem cells as compared to unconjugated vectors.(16) This important enhancement will permit the utilization of lower viral doses to achieve a therapeutically adequate effect and, as a result, a reduction in possible systemic reactions associated with adenovirus and an improved safety profile. Successful dose reduction is of paramount importance as clinical fibroid lesions can reach very large sizes and may require enormous viral loads for total eradication.

Our group has consistently pursued the development of novel therapeutic approaches for the total ablation of fibroid tumors, especially those apt to addressing the major clinical challenge of disease recurrence. To that end, the scope of our current research is focused on further validating nanoparticle-assisted gene therapy against fibroid tumor-initiating stem cells. These cells are believed to play a pivotal role in both the initiation and recurrence of UFs, specifically, after surgical procedures e.g., myomectomy and myolysis.(69)

For the first time, we have tested the susceptibility of human fibroid stem cells to the suicidal gene therapy approach, and demonstrated the effective transfection and obliteration of these cells utilizing that approach. Furthermore, we found that magnetofection enhancement is also operational with significant increases in cell apoptosis and diminution of tumor cell proliferation. This novel strategy of eliminating fibroid tumor-forming stem cells looks very promising in not only tumor eradication, but also, for the prevention of its regrowth/recurrence. The reported novel stem cell transfection data are in agreement with that of breast cancer stem cells targeted by oncolytic adenovirus which succeeded in stopping bone metastasis (70) by eradicating tumor renewing stem cells. A group of researchers developed a doxorubicin-encapsulated nanoparticle surface-decorated with chitosan that can specifically target the CD44 receptors of these tumor initiating stem cells. This nanoparticle system was engineered to increase the cytotoxicity of the doxorubicin six-fold in comparison to the use of free doxorubicin in eliminating CD44+ cancer stem-like cells residing in 3D mammary tumor spheroids (71). Another group tried to target serous cancer stem cells by restoration of apoptosis pathway activity in serous cancer stem cells which dramatically helped sensitize them to platinum therapy resulting in tumor eradication (72).These studies and ours confirm that development of targeted therapies that are toxic to tumor stem cells while sparing normal stem cells could lead to more effective methods for eradicating this ominous population of cells. The MNPs enhanced adenovirus-mediated suicide gene therapy approach, after appropriate preclinical in vivo evaluation, could provide a safe and effective non-surgical approach for the treatment of UFs with durable long-term favorable outcomes, minimal treatment failures and limited or no disease recurrence.