Cholangiocarcinoma therapy with nanoparticles that combine downregulation of MicroRNA-210 with inhibition of cancer cell invasiveness.

Cholangiocarcinoma (CCA) is the second most common primary liver malignancy with extremely poor therapeutic outcome due to high drug resistance, widespread metastasis and lack of effective treatment options. CCA progression and metastasis are regulated by multiple biological factors including multiple miRNAs and chemokine receptor CXCR4. The goal of this study was to test if nanotherapeutic blockade of CXCR4 by polymeric CXCR4 antagonist (PCX) combined with inhibition of hypoxia-inducible miR-210 cooperatively enhances therapeutic efficacy in CCA through reducing invasiveness, inducing cell killing, and reversing drug resistance.
Methods: We first tested the activity of PCX to inhibit migration of CCA cells. We then prepared PCX/anti-miRNA nanoparticles and analyzed their miRNA delivery efficacy and anticancer activity in vitro. Finally, in vivo biodistribution assay and anticancer activity study were performed in CCA tumor-bearing mice.
Results: Our results show that PCX had a broad inhibitory effect on cell migration, effectively delivered anti-miR-210, and downregulated miR-210 expression in CCA cells. Combination PCX/anti-miR-210 nanoparticles showed cytotoxic activity towards CCA cells and reduced the number of cancer stem-like cells. The nanoparticles reversed hypoxia-induced drug resistance and sensitized CCA cells to standard gemcitabine and cisplatin combination treatment. Systemic intravenous treatment with the nanoparticles in a CCA xenograft model resulted in prominent combined antitumor activity.
Conclusion: Our findings support PCX-based nanoparticles as a promising delivery platform of therapeutic miRNA in combination CCA therapies.

Introduction

Cholangiocarcinoma (CCA) is an aggressive cancer of the biliary duct system. Intrahepatic CCA arises from the epithelial cells of the intrahepatic bile duct and is the second most common primary liver cancer 1. CCA is a deadly disease with dismal prognosis as evident from its low 10% five-year survival rate. Surgery is the only curative option for CCA. However, fewer than 30% of CCA patients are eligible for surgical resection at diagnosis 2. The worldwide incidence and mortality of CCA have increased over the past several decades. Advanced CCA is currently treated with a combination of gemcitabine (GEM) and cisplatin (CDDP). Unfortunately, the therapeutic outcome of systemic GEM/CDDP treatment is poor due to drug resistance 3, 4. The high mortality of CCA is also attributed to early invasion and widespread metastasis 5, 6. No effective treatment options exist for metastatic CCA. These therapeutic challenges highlight the urgent need to develop new therapeutics to reverse drug resistance and inhibit metastasis in CCA.

The C-X-C receptor type 4 (CXCR4) and its ligand (CXCL12) are two key factors in tumor growth, metastasis, angiogenesis, and cancer cell-microenvironment interactions, which make them potential targets for CCA therapy 7, 8. CXCR4 overexpression has been found in more than 20 major human cancer types, including CCA. CXCR4 was upregulated in ~65% of all CCA samples but was undetectable in normal biliary epithelium 9, 10. The upregulation of CXCR4 is highly dependent on multiple transcription factors, growth factors, and hypoxia-inducible factors. The hypoxic tumor microenvironment can induce CXCR4 based on activation of the hypoxia-inducible factor-1α (HIF1α) and transcript stabilization 11. The binding of CXCL12 to CXCR4 activates intracellular signaling to promote migration and invasion of cancer cells. CXCR4 then facilitates metastasis of the primary tumor cells to those sites where CXCL12 is highly expressed 12, 13. CXCL12 is preferentially expressed in the liver, lymph nodes, lung, and bone marrow, which overlaps the metastatic profile of CCA 14. Inhibition of the signaling pathway with CXCR4 antagonists decreases the motility and invasion of CCA cells 15-17. Inhibition of CXCR4 can also reverse drug resistance and improve chemotherapy 18, 19, as documented by the chemosensitizing effect of a CXCR4 antagonist AMD3100 to GEM in human CCA cells 20. Several studies reported that CXCR4 inhibition reduced the stemness of cancer cells to overcome drug resistance 21, 22. Current evidence strongly supports the potential of CXCR4 inhibition to reverse drug resistance and inhibit metastasis in CCA.

MicroRNAs (miRNAs) are small noncoding RNAs that post-transcriptionally regulate gene expression 23. In cancer, miRNAs can act as tumor suppressors or oncogenes and play an important role in tumorigenesis, tumor growth, angiogenesis, and metastasis 24-26. Therapies that target miRNAs exhibit great promise for cancer because single miRNA typically targets multiple genes simultaneously, enabling regulation of multiple signaling pathways involved in various cancers 27. A growing number of studies confirm the important role of miR-25, miR-34a, and let-7c in CCA formation and progression 28-31. Recent evidence already suggests that miRNA therapy can inhibit CCA growth, metastasis, and improve survival in animal models 32-34. Hypoxia in CCA induces the upregulation of miR-210, which helps the cancer cells to adapt to the hypoxic microenvironment through multiple biological pathways 35-37. Oncogenic activity of miR-210 is responsible for cell proliferation, apoptosis, metastasis, DNA repair, cell metabolism, and antitumor drug resistance 38-41. Hence, inhibition of miR-210 provides a promising target for the treatment of CCA.

Despite the great potential of miRNAs in metastatic cancer, their clinical translation has been limited by a lack of efficient and safe systemic delivery systems 42-45. Nanosized polyelectrolyte complexes of miRNA with polycations (polyplexes) have been among the most investigated delivery platforms, which effectively promote miRNA delivery by protecting them from degradation and overcoming multiple biological barriers 46, 47. Traditional polycations used in the preparation of polyplexes are pharmacologically inert with no inherent anticancer activity. Recently, alternative strategies from our lab and others have suggested development of pharmacologically active polymers as a novel approach for combination therapies of cancer 48-51. These delivery systems present several advantages, including simple formulation and high content of active agents.

We have reported a series of polymeric CXCR4 antagonists (PCX) capable of delivering miRNA, siRNA, and DNA 52-56. PCX simultaneously inhibit cancer cell invasion by CXCR4 axis blockade and deliver nucleic acids into various cancer cells for improved anticancer effect. PCX were synthesized from AMD3100 and modified with cholesterol to improve the stability of their nanoparticles for better systemic delivery 57, 58. Due to the key roles of CXCR4 and miR-210 in CCA, we hypothesized that combining inhibition of CXCR4 by PCX and inhibition of hypoxia-inducible miR-210 would cooperatively enhance the therapy of CCA through reducing invasiveness, inducing cell killing and overcoming drug resistance (Scheme ). To test this hypothesis, we used PCX modified with cholesterol to prepare PCX/anti-miR-210 nanoparticles and evaluated their delivery activity and anticancer efficacy in CCA both in vitro and in vivo.

Methods

Materials

Cholesterol-modified polymeric CXCR4 inhibitors PCX (M = 16.7 kDa, M/M = 1.9, cholesterol wt% = 16.8%) were synthesized and characterized as previously described 58. Succinimidyl ester of Alexa Fluor® 647 carboxylic acid was from Life Technologies (Eugene, OR). AlexaFluor 647-labeled PCX polymers (AF647-PCX) were produced according to the manufacturer's instructions and purified by dialysis to remove unreacted dye. AMD3100 was from Biochempartner (Shanghai, China). Dulbecco's modified Eagle medium (DMEM), Dulbecco's phosphate buffered saline (PBS), and fetal bovine serum (FBS) were from Thermo Scientific (Waltham, MA). Hsa-miR-210-3p-Hairpin Inhibitor (anti-miR-210, mature miRNA sequence: 5′-CUGUGCGUGUGACAGCGGCUGA-3′), negative control miR-NC inhibitor (anti-miR-NC, mature miRNA sequence: 5′-UCACAACCUCCUAGAAAGAGUAGA-3′), and carboxyfluorescein (FAM) labeled FAM-anti-miRNA were purchased from Dharmacon (Lafayette, CO). Cell culture inserts (for 24-well plates, 8.0 μm pores, Translucent PET Membrane, cat# 353097) were purchased from BD Biosciences (Billerica, MA). Real-time (RT)-PCR primers were purchased from Invitrogen (Carlsbad, CA). All other reagents were from Fisher Scientific and used as received unless otherwise noted.

Cell culture

Human malignant cholangiocarcinoma Mz-ChA-1 cell line was kindly provided by Dr. Gregory Gores, Mayo Clinic, Rochester, MN. Mz-ChA-1 cells were grown in high-glucose DMEM supplemented with 10% FBS, penicillin (100 U/mL), streptomycin (100 μg/mL), G418 (50 μg/mL), and insulin (0.5 μg/mL) at 37 °C with 5% CO2 in a humidified chamber. To induce hypoxia, cells were incubated in an atmosphere of 2% O2, 5% CO2, and 93% N2 at 37 °C.

Surface expression of CXCR4

Mz-ChA-1 cells were detached with enzyme-free Cell Dissociation Buffer (Thermo Scientific) and suspended in a staining buffer. Cells were stained live with allophycocyanin (APC)-conjugated anti- CXCR4 antibody (Abcam, USA) for 1 h at 4 °C. Isotype-matched negative control was used in the panel of mAb to assess background fluorescence intensity. Samples were analyzed on a BD FACS Calibur flow cytometer (BD Bioscience, Bedford, MA). The results were processed using FlowJo software (Tree Star Inc., Ashland, OR).

Transwell migration

Mz-ChA-1 cells (2 × 105) were seeded into 6-well plates and cultured in complete DMEM medium. The cultured cells were subsequently treated with AMD3100 (300 nM) or PCX (3 μg/mL). After 48 h of incubation, the cells were trypsinized and suspended in medium without serum. Subsequently, 3 × 104 cells were seeded in the top chambers in 300 μL of serum-free medium and 500 μL of complete medium containing 10% FBS was added to the lower transwell chambers. After 12 h, the nonmigrated cells in the top chamber were removed with a cotton swab. The migrated cells were fixed in 100% methanol and stained with 0.2% Crystal Violet solution for 10 min at room temperature. The images were taken by EVOS xl microscope. Three 20× visual fields were randomly selected for each insert, and each group was conducted in triplicate.

Preparation and characterization of nanoparticles

The ability of PCX to condense anti-miRNA was determined by electrophoresis in a 2% agarose gel containing 0.5 μg/mL ethidium bromide (EtBr). PCX/anti-miRNA nanoparticles were prepared by adding a predetermined volume of PCX to an anti-miRNA solution (20 μM in 10 mM HEPES pH 7.4) to achieve the desired w/w ratio and vigorously vortexed for 10 s. Nanoparticles were then incubated at room temperature for 20 min before further use. Nanoparticles formed at various polycation-to-anti-miRNA weight ratios were loaded (20 μL of the sample containing 0.5 μg of microRNA) and run for 15 min at 100 V in 0.5 × Tris/Borate/EDTA buffer. The gels were visualized under UV illumination with a KODAK Gel Logic 100 imaging system. Hydrodynamic diameter and zeta potential of the nanoparticles were determined by dynamic light scattering (DLS) using a ZEN3600 Zetasizer Nano-ZS (Malvern Instruments Ltd., Massachusetts, USA). Morphology was observed under transmission electron microscopy (TEM, Tecnai G2 Spirit, FEI Company, USA) using NanoVan® negative staining (Nanoprobes, USA). Anti-miRNA release from nanoparticles was analyzed by heparin displacement assay. The nanoparticles (w/w = 2) were incubated with increasing concentrations of heparin for 30 min at room temperature. The samples (20 μL of the sample containing 0.5 μg of anti-miRNA) were then analyzed by agarose gel electrophoresis. For the serum stability test, free anti-miRNA solution and nanoparticles (w/w = 2) solution were incubated with FBS (50% v/v) at 37 °C, respectively. Each sample was collected at a specified time interval (0.5, 1, 2, 4, 8, and 24 h) and analyzed by gel electrophoresis. Prior to gel electrophoresis, nanoparticles solution was pre-treated with heparin to replace anti-miRNA from the nanoparticles.

Cellular uptake and intracellular trafficking of nanoparticles

Flow cytometry analysis was performed to study the cellular uptake of the nanoparticles. Mz-ChA-1 cells (5 × 104) were seeded in 24-well plates. After 24 h growth, the cells were incubated for 4 h with nanoparticles prepared with AF647-PCX and FAM-anti-miRNA (w/w =2, 100 nM FAM-anti-miRNA). The cells were then trypsinized, washed with cold PBS, and subjected to analysis using a BD FACS Calibur flow cytometer (BD Bioscience, Bedford, MA). The results were processed using FlowJo software. Intracellular localization of AF647-PCX/FAM-anti-miRNA was also observed by confocal laser scanning microscopy. Cells were cultured on a 20 mm glass-bottom cell culture dish (Nest) using 1 × 105 cells per dish. After 24 h, the medium was replaced with fresh medium and AF647-PCX/FAM-anti-miRNA nanoparticles were added (100 nM FAM-anti-miRNA). After incubation for 4 h, the cells were washed twice with PBS, stained with Hoechst 33258 for 10 min. The cells were visualized using a LSM 800 Laser Scanning Microscope (Zeiss, Jena, Germany). Moreover, to follow the endosomal release of nanoparticles, AF647-PCX/anti-miRNA nanoparticles were incubated with cells for 4 h. Then, the cells were washed twice with PBS and stained with LysoTracker Red DND-99 (Life Technology, USA) for 30 min. Finally, the cells were rinsed three times with PBS and visualized with a confocal microscope.

Quantitative real-time polymerase chain reaction (qRT-PCR)

Cells were seeded in 12-well plates at 1 × 105 cells per well 24 h prior to treatment. Then, the cells were incubated with the nanoparticles containing anti-miRNA (100 nM) in 1 mL of medium for 48 h. The expression levels of miR-210 were evaluated by TaqMan qRT-PCR. The mirVana miRNA Isolation Kit (Ambion, USA) was used for total RNA extraction from cultured cells. 10 ng of total RNA was converted into cDNA using specific primers for miR-210 (or the internal control Z30 (Applied Biosystems, Foster City, CA)) and the TaqMan microRNA reverse transcription kit (Applied Biosystems). qRT-PCR was performed using TaqMan Universal Master Mix II, No AmpErase UNG (2×) and specific primers for miR-210 or Z30 (Applied Biosystems, Foster City, CA) on a Rotor-Gene Q instrument (QIAGEN) according to the manufacturer's instructions. MiRNA expression levels were expressed relative to the internal control according to the comparative threshold cycle (Ct) method.

Apoptosis

Apoptosis was analyzed using the Annexin V-FITC/PI Apoptosis Detection Kit (BioLegend, USA). Briefly, cells were seeded in 12-well plates at a density of 1 × 105 cells per well 24 h prior to treatment. Then, the cells were incubated with the nanoparticles containing anti-miRNA (100 nM) in 1 mL of medium for 48 h. The Annexin V-FITC apoptosis detection was performed using flow cytometry in accordance with the manufacturer's protocol, and the data were processed using FlowJo. Moreover, apoptosis percentage was also quantified by nuclear morphology and visualized by treatment with the fluorescent DNA-binding dye, DAPI (4′, 6-diamidine-2′-phenylindole dihydrochloride). Briefly, cells were stained with 2 μg/mL of DAPI for 30 min at 37 °C. Apoptotic nuclei (condensed, fragmented) were counted and presented as a percent of total nuclei. At least 100 cells were counted per well and experiments were performed in triplicate. Caspase 3/7 activity was measured by enzymatic fluorophore release (Apo-One) according to the manufacturer's protocol (Promega).

Cytotoxicity

Toxicity of the nanoparticles carrying anti-miRNA was evaluated by Cell Titer Blue assay. Briefly, cells were plated in 96-well microplates at a density of 5000 cells/well. After 24 h, cells were incubated with the nanoparticles containing miRNA (100 nM) for another 48 h prior to measuring cell viability. The medium was then removed and replaced with a mixture of 100 μL serum-free media and 20 μL of CellTiter-Blue reagent (CellTiter-Blue Cell Viability Assay, Promega). After a 2 h incubation, fluorescence (560/590 nm) was measured on a SpectraMax iD3 Multi-Mode Microplate Reader (Molecular Devices, CA). The relative cell viability (%) was calculated as [I]treated/[I]untreated × 100%.

Colony formation

Cells were transfected with nanoparticles containing miRNA (100 nM) for 4 h. Then, transfected cells were reseeded in 6-well plates (200 cells per plate). Cells were allowed to grow for 14 days. Cells were fixed in 100% methanol and stained with 0.2% Crystal Violet solution for 10 min. Finally, plates were washed with distilled water and photographed. The number of colonies in each well was quantified.

Aldefluor activity

Cells were transfected with nanoparticles containing miRNA (100 nM) for 4 h and then incubated for 48 h in fresh medium. Then, Aldefluor kit (Stem Cell Technologies, Durham, NC, USA) was used to analyze ALDH enzymatic activity according to the manufacturer's instructions. Cells were suspended in Aldefluor assay buffer containing ALDH substrate (BODIPY-aminoacetaldehyde) and incubated for 30 min at 37 °C. As the negative control, cells were treated with 50 mM diethylaminobenzaldehyde (DEAB) inhibitor before adding ALDH substrate. The cells were subjected to analysis using a BD FACSCalibur flow cytometer (BD Bioscience, Bedford, MA).

Tumor spheroid formation

Cells were transfected with nanoparticles containing miRNA (100 nM) for 4 h and then cells were dissociated and plated in ultralow attachment 6-well plates (Corning, New York) at a density of 5000 viable cells per well. Cells were cultured in serum-free DMEM/F12 medium containing 5 μg/mL insulin (Sigma-Aldrich, St. Louis, MO, USA), 0.4% bovine serum albumin (Sigma-Aldrich, St. Louis, MO, USA), 10 ng/mL basic fibroblast growth factor (Invitrogen, Waltham, MA, USA), and 20 ng/mL human recombinant epidermal growth factor (Invitrogen, Waltham, MA, USA) and observed after 14 days.

Combinational cell killing assay

First, cell viability assay was performed to study the activity of nanoparticles to sensitize cells to chemotherapy. Cells were seeded in 96-well microplates at a density of 5000 cells/well. After 24 h, cells were incubated with indicated concentrations of gemcitabine/cisplatin (GEM/CDDP, w/w=10), PCX/anti-miR-NC (100 nM) plus GEM/CDDP and PCX/anti-miR-210 (100 nM) plus GEM/CDDP, respectively. After incubation for another 48 h, CellTiter-Blue assay was performed to measure cell viability. The IC50 were calculated in GraphPad Prism using a built-in dose-response analysis as the total drugs (gemcitabine plus cisplatin) concentration that achieves 50% growth inhibition relative to untreated cells (n = 3). Moreover, synergy calculations were performed to study the combinational effects between PCX/anti-miR-210 and GEM/CDDP. Briefly, cells were seeded in 96-well microplates at a density of 5000 cells/well. After 24 h, cells were incubated with indicated concentrations of gemcitabine/cisplatin (GEM/CDDP, w/w=10), PCX/anti-miR-210 and PCX/anti-miR-210 plus GEM/CDDP, respectively. After incubation for another 48 h, CellTiter-Blue assay was performed to measure cell viability. The data from cell viability assays were analyzed using Combenefit software (Cancer Research UK Cambridge Institute). The "Mapped Surface" views (Bliss model) were selected as graphical outputs for the synergy distribution.

Xenograft tumor model

All animal experiments followed a protocol approved by the University of Nebraska Medical Center Institutional Animal Care and Use Committee. Animals were placed in a facility accredited by the Association for Assessment and Accreditation of Laboratory Animal Care upon arrival. Male athymic nu/nu mice (6 weeks) were purchased from Charles River Laboratories. The xenograft tumor model was generated by subcutaneous injection of 5 × 106 Mz-ChA-1 cells in 1:1 mixture of PBS/Matrigel (100 µL) into the flank region of the mouse.

Biodistribution

Xenograft tumor mice (~300 mm3) were injected with AF647-PCX/FAM-miRNA nanoparticles (w/w=2, 2.4 mg/kg AF647-PCX, 1.2 mg/kg FAM-miRNA) through tail vein. The mice were sacrificed 24 h post administration, and the tumors and major organs were harvested and subjected to ex vivo fluorescence imaging using Xenogen IVIS 200 (Ex = 640 nm, Em = 680 nm). The fluorescence from each organ was analyzed by the instrument software. Then, the isolated tumors were embedded in an OCT compound, cut into 10 μm sections, stained with DAPI, and imaged with a confocal microscope.

In vivo therapeutic effect

When the tumor sizes reached about 150-200 mm3 at day 10 post-tumor inoculation, mice were randomly assigned to 5 groups (n = 5) and injected with PBS, GEM/CDDP, PCX/anti-miR-NC, PCX/anti-miR-210 and PCX/anti-miR-210 + GEM/CDDP, respectively. PCX/anti-miRNA (w/w = 2, 2.4 mg/kg PCX, 1.2 mg/kg anti-miRNA) were injected at days 11, 13, 15 and 17 through tail vein. GEM/CDDP (15 mg/kg GEM, 1.5 mg/kg CDDP) were given at days 12 and 16 by intraperitoneal injection. Tumor volume was monitored by measuring the perpendicular size of the tumors using digital calipers. The estimated volume was calculated according to the following formula: tumor volume (mm3) = 0.5 × length × width2. Body weight of the mice was also recorded. On days 22, mice were sacrificed, and all tumor tissues and major organs were harvested, fixed in 4% paraformaldehyde, sectioned, and stained with H&E. The apoptosis of tumor cells was determined using Caspase 3 immunohistochemical staining according to the manufacturer's instructions. Blinded histological analysis of the tissues was conducted by a trained pathologist at the UNMC core facility.

Statistical analysis

Data are presented as the mean ± SD. Statistical significance was determined using ANOVA followed by Bonferroni post hoc correction with p < 0.05 as the minimal level of significance.

Results and Discussion

Inhibition of cancer cell migration

We analyzed the expression of CXCR4 in human CCA cells Mz-ChA-1 in normoxic (20% oxygen) and hypoxic (2% oxygen) conditions. The expression of CXCR4 mRNA was quantified with qRT-PCR. Hypoxia increased the CXCR4 expression 5.8-fold when compared with normoxic cells (Figure ). Surface expression of the CXCR4 receptor was then analyzed by flow cytometry (Figure ). The percentage of CXCR4-positive cells increased from 31% under normoxia to 91% in hypoxia. These results confirmed that hypoxia induces CXCR4 expression in CCA cells. High CXCR4 expression typically leads to increased cell motility, migration and invasion, which leads to enhanced metastasis 59.

A migration assay was performed to test if hypoxia enhanced cell migration and if the migration could be inhibited by CXCR4 antagonism of PCX. To avoid the effect of PCX cytotoxicity on migration activity, we carefully selected a safe concentration of PCX (Figure ). PCX at the used concentration of 3 µg/mL showed no cytotoxicity in either normoxia or hypoxia. AMD3100 is a highly specific and potent (nM) small molecule inhibitor of CXCR4. Here, 300 nM of AMD3100 was selected as an effective and safe positive control. To confirm that PCX inhibits CXCR4 in CCA cells, we evaluated the downstream extracellular-signal-regulated kinase (Erk) signaling pathway by Western blot analysis. Erk is a main downstream signaling pathway of CXCR4/CXCL12 axis, which regulates cell survival, migration and invasion 8. As shown in Figure , AMD3100 inhibited SDF-1-induced phosphorylation of Erk in Mz-Cha-1 cells. PCX showed similar activity to inhibit pErk as AMD3100, which confirmed PCX as a CXCR4 inhibitor. Then, the anti-migration activity of PCX was evaluated and compared with AMD3100. As shown in Figure , hypoxia increased the FBS-induced migration of CCA cells ~3-fold when compared with normoxic conditions. AMD3100 effectively inhibited migration under both normoxia (57% inhibition) and hypoxia (63% inhibition). Interestingly, PCX showed better anti-migration activity and almost completely inhibited migration of CCA cells in both conditions (95% inhibition in normoxia, 97% inhibition in hypoxia). To understand why PCX presented such a superior activity, we hypothesized that PCX affected additional signaling pathways involved in cell motility. The bioactive phospholipid lysophosphatidic acid (LPA) and its G-protein-coupled receptors play an important role in cancer migration, invasion and metastasis through regulating GTPases Ras, Rho, and Rac pathways. These LPA-regulated GTPases are normally overexpressed in human CCA and are involved in the invasion and metastasis process. Inhibition of LPA signaling represents an effective approach to inhibit CCA cells migration 60-64. Here, we performed a migration assay using LPA as the chemoattractive agent. As shown in Figure , LPA treatment induced migration of CCA cells. Hypoxia also increased LPA-induced migration when compared with normoxic conditions. As expected, AMD3100 failed to inhibit migration under both normoxia and hypoxia. However, PCX effectively inhibited LPA-induced migration, with 59% inhibition observed in normoxia and 65% inhibition found in hypoxia. AMD3100 reduced CCA cells invasiveness through CXCR4 pathway. However, PCX inhibited invasiveness not only through CXCR4 pathway but also LPA signaling. These findings support the utility of PCX as a potential antimetastatic therapy due to more efficient prevention of CCA migration through multiple pathways than conventional CXCR4 inhibitor.

Besides inhibiting CXCR4, PCX are polycations that form polyplexes with nucleic acids. The ability of PCX to form nanoparticles with miRNA was first evaluated using agarose gel electrophoresis. Nanoparticles were prepared by adding PCX to anti-miRNA solution at increasing polycation/anti-miRNA ratios (Figure ). PCX fully condensed the miRNA at and above PCX/miRNA (w/w) ratios of 2. Partial condensation was observed at lower w/w ratios (0.5-1) as indicated by the smear of the RNA in the gel.

Hydrodynamic size and ζ-potential of PCX/anti-miRNA nanoparticles (w/w = 2) were measured by dynamic light scattering. As shown in Figure , nanoparticles exhibited sizes of 67.5 ± 0.3 nm with low polydispersity index (0.10 ± 0.05). The nanoparticles presented a positive surface charge with ζ potential of 21.5 ± 1.3 mV (Figure ). The shape and morphology of the nanoparticles were analyzed by transmission electron microscopy (TEM) (Figure ) and they were found to be uniform particles with a mostly spherical morphology.

Release of the miRNA from the particles was analyzed by heparin displacement assay. The nanoparticles completely released miRNA above 160 μg/mL heparin (Figure ). Poor stability against degradation by serum nucleases hinders the in vivo application of miRNA. We studied the serum stability of PCX/anti-miRNA nanoparticles in 50% FBS at 37 °C using gel electrophoresis (Figure ). Naked anti-miRNA was rapidly degraded within 2 h of serum incubation. The PCX/anti-miR-210 nanoparticles protected the anti-miRNA for at least 8 h. Intact anti-miRNA was observed even after serum incubation for 24 h. This result confirmed the improved serum stability provided by the PCX particles.

In vitro delivery of anti-miRNA

After preparation and characterization of the nanoparticles, we examined their delivery to CCA cells. We prepared PCX nanoparticles using fluorescently labeled PCX (AF647-PCX) and anti-miRNA (FAM-anti-miRNA). Mz-ChA-1 cells were incubated in normoxic conditions with the AF647-PCX/FAM-anti-miRNA (w/w = 2) for 4 h and observed under a confocal microscope. As shown in Figure , both AF647-PCX (red) and FAM-anti-miRNA (green) were observed in the Mz-ChA-1 cells after 4 h of incubation, indicating effective uptake and internalization of the nanoparticles. Partial colocalization of the PCX and miRNA fluorescence (yellow) suggested that at least part of the particles remained assembled in the cells. Disassembly of a significant fraction of the nanoparticles was evident from the separation of the PCX and miRNA fluorescence signals. The disassembly of the nanoparticles is important for delivery of bioavailable miRNA inhibitors. Cell uptake was also quantified by flow cytometry (Figure ). We found that ~96% CCA cells have internalized both PCX and anti-miRNA. We also quantified the cellular uptake of nanoparticles in hypoxia and found no difference in the uptake percentage or mean fluorescence intensity when compared with the above results in normoxia (Figure ). The lack of difference despite a large increase in CXCR4 expression in hypoxic cells suggests that CXCR4 was not directly involved in the cell uptake of the particles. This conclusion was further supported by a competition assay in which pretreatment of cells with AMD3100 failed to inhibit the internalization of the nanoparticles (Figure ).

Endosomal escape is an important step in delivery of cytoplasmically active nucleic acids like miRNA. Intracellular distribution and trafficking of the PCX nanoparticles were studied using confocal microscopy. Mz-ChA-1 cells were incubated with PCX/FAM-anti-miRNA nanoparticles for 4 h and lysosomes were stained with LysoTracker Red (Figure ). The fluorescence of the FAM-miRNA (green) was localized predominantly in the cytoplasm with only a limited extent of co-localization with the LysoTracker signal (red). This result confirmed endosomal escape of the nanoparticles.

After reaching cytoplasm, free anti-miRNA can inhibit targeted mature miRNA and result in therapeutic effect. Transfection activity was evaluated using nanoparticles prepared with anti-miR-210 and its negative control (anti-miR-NC). CCA cells were treated with PCX/ miRNA under normoxia or hypoxia and miR-210 expression was measured using qRT-PCR (Figure ). As expected, hypoxia induced a 6-fold upregulation of miR-210 expression. Incubation with PCX/anti-miR-210 significantly downregulated the miR-210 levels under both normoxia (~57% decrease) and hypoxia (~85% decrease). PCX/anti-miR-NC had no effect on miR-210 expression. These results confirmed effective delivery of functional anti-miR-210 into CCA cells and inhibition of the targeted miRNA by the PCX nanoparticle.

Anticancer activity in vitro

The CCA cell killing activity of PCX/anti-miR-210 was first studied by determining apoptosis in the treated cells using Annexin V assay (Figure ). PCX/anti-miR-210 treatment increased the number of apoptotic cells (~32%) and necrotic cells (∼13%) compared with the control groups. This pro-apoptotic effect of the nanoparticles was further corroborated by staining with DAPI and evaluation based on nuclear morphology. Upon treatment with PCX/anti-miR-210, Mz-ChA-1 cells demonstrated at least a 26% increase in apoptosis compared to cells treated with PCX/anti-miR-NC (Figure and Figure ). The PCX/anti-miR-210 nanoparticles also increased caspase 3/7 activity (∼3.2-fold) in the CCA cells (Figure ), validating that the nanoparticles induced caspase activation, apoptotic nuclear morphology, and externalization of phosphatidylserine. Next, we studied the contribution of PCX/anti-miR-210 nanoparticles to the overall cell killing activity in Mz-ChA-1 cells using cell viability assay (Figure ). PCX/anti-miR-210 exhibited significantly higher cell killing activity (∼28%) than the untreated and miR-NC controls. Colony formation assay was also performed to study the effect of the nanoparticles on the tumorigenic potential of CCA cells (Figure ). Treatment with PCX/anti-miR-210 reduced colony formation by 92% compared to PBS. As expected, CXCR4 inhibition by PCX alone had no direct cell killing effect in the CCA cells. Overall, the combined results from the anticancer activity assays confirmed the promising activity of PCX/anti-miR-210 nanoparticles in CCA cells due to their effect on apoptosis and colony formation.

Cancer stem cells play important roles in growth, recurrence, and metastasis. CXCR4/CXCL12 axis is involved in modulating the cancer stem cell niche. Inhibition of CXCR4/CXCL12 can decrease cancer stem cells through reducing phosphorylation of ERK and STAT3 65, 66. Knockdown of miR-210 previously reduced stemness of cancer cells by rescuing the expression of Myc antagonist protein 67. The combined inhibition of CXCR4 and miR-210 was thus expected to cooperatively reduce stemness of the CCA cells. Aldehyde dehydrogenase (ALDH) is a reliable marker for cancer stem cells that exhibit high ALDH enzymatic activity. The Mz-ChA-1 cells were treated with the particles, incubated with Aldefluor fluorescent reagent, and ALDH activity was analyzed by flow cytometry (Figure ). In the untreated cells, the ALDH-positive subpopulation of CCA cells was 4.6%. Treatment with PCX/anti-miR-NC reduced the ALDH+ population to 1.4%. Further decrease to 0.8% was observed in cells treated with the combined PCX/anti-miR-210 nanoparticles. Tumor spheroid formation assay was then used to validate the effect of the nanoparticles on cancer stem cells. Compared with the untreated (PBS) control group with large and abundant tumor spheroids, PCX/anti-miR-NC reduced the formation of large spheres. PCX/anti-miR-210 treatment further reduced the activity of cancer cells to form tumor spheroids as indicated by the formation of small and fragmentary spheroids (Figure ). Based on the combined evidence from the ALDH and tumor spheroid assays, we confirmed that both CXCR4 antagonism by PCX and miR-210 inhibition by anti-miR-210 contributed to the reduction of stemness in CCA cancer cells.

Chemosensitizing effect of the nanoparticles

Chemotherapy using GEM and CDDP has limited therapeutic effect in CCA patients due to drug resistance. Cancer stem cells have been shown to be involved in drug resistance through multiple cellular and molecular mechanisms. Decrease of the cancer stem cell population thus emerged as a promising approach to overcome therapy resistance and improve efficacy of cancer therapy 68. After confirming the ability of the nanoparticles to decrease stemness of CCA cells, we aimed to study whether PCX/anti-miRNA nanoparticles could reverse drug resistance and sensitize CCA cancer cells to the chemotherapy. Mz-ChA-1 cells were treated with GEM/CDDP or GEM/CDDP+PCX/anti-miRNA under both normoxia and hypoxia. Then, cell viability was measured and half-maximal inhibition concentrations (IC50) of the GEM/CDDP combination were calculated (Figure ). Treatment with PCX/anti-miR-NC (anti-miRNA 100 nM) in normoxia had no direct cell killing effect (Figure ) but the CXCR4 inhibition by the particles sensitized the cells to the effects of GEM/CDDP as indicated by the decrease of IC50 from 1.48 µg/mL to 0.49 µg/mL (Figure ). Combining the CXCR4 inhibition with miR-210 inhibition further enhanced the chemosensitizing effect of the particles (IC50 ~0.14 µg/mL). Incubation of the CCA cells in hypoxic conditions greatly increased resistance to GEM/CDDP treatment as shown by the more than 3-fold increase of IC50 to >5 µg/mL. Both PCX/anti-miR-NC and PCX/anti-miR-210 nanoparticles sensitized the cancer cells to GEM/CDDP therapy in hypoxic conditions.

The chemosensitizing effect of the PCX/anti-miR-210 nanoparticles was further investigated using synergy analysis. Cell viability was measured in CCA cells treated with different doses and ratios of PCX/anti-miR-210 and GEM/CDDP. Combenefit software was then used to calculate synergy scores for each combination. A positive score (0-100) indicates synergy, a score of 0 is additive, and a negative score indicates antagonism. Higher score indicates stronger synergistic effect. The mapped surface of synergy/antagonism with the Bliss model is shown in Figure . In normoxia, PCX/anti-miR-210 and GEM/CDDP showed synergistic cell killing effect with a score of ~30. Much stronger synergy of PCX/anti-miR-210 and GEM/CDDP was achieved under hypoxia with a score of ~60. These results confirmed that PCX/anti-miR-210 sensitized the CCA cells to chemotherapy in a synergistic manner and reversed hypoxia-induced drug resistance.

Following the promising in vitro findings, we explored the delivery efficacy of the nanoparticles to tumors in vivo. Fluorescently labeled nanoparticles (AF647-PCX/FAM-anti-miRNA) were administered intravenously into mice bearing xenograft Mz-ChA-1 tumors. Following animal sacrifice 24 h post injection, we measured ex vivo fluorescence (AF647) in the excised tissues to analyze the biodistribution (Figure ). We observed significant tumor accumulation of the nanoparticles. As expected, the nanoparticles were also found in the liver, spleen and lung. To evaluate the ability of the nanoparticles to deliver both PCX and anti-miRNA into the tumors, frozen tumor sections were observed under a confocal microscope (Figure ). Both AF647-PCX (red) and FAM-anti-miRNA (green) fluorescence were clearly present in the tumor. The high colocalization (yellow) of AF647-PCX and FAM-anti-miRNA also indicated good stability of the nanoparticles during systemic delivery.

Therapeutic efficacy of the nanoparticles was tested in a xenograft CCA model in nude mice. We confirmed the presence of the hypoxic tumor microenvironment needed for the mechanism of action of our PCX/anti-miR-210 nanoparticles using pimonidazole staining in frozen tumor sections (Figure ). Systemic intravenous treatment with PCX/anti-miR-210 nanoparticles effectively inhibited tumor growth (51%) when compared with PBS (Figure ). Treatment with control PCX/anti-miR-NC nanoparticles or with GEM/CDDP chemotherapy showed negligible effect on tumor growth. These results suggested that CXCR4 inhibition alone has no effect on the growth of CCA tumors. In contrast, combined downregulation of miR-210 and inhibition of CXCR4 by PCX/anti-miR-210 nanoparticles cooperatively achieved antitumor activity. As expected, the CCA tumor model was resistant to GEM/CDDP chemotherapy. We used a suboptimal dose of GEM/CDDP (GEM 15 mg/kg, CDDP 1.5 mg/kg) to better demonstrate the benefits of the nanoparticles. When we treated the animals with the combination of PCX/anti-miR-210 and GEM/CDDP, we observed greatly enhanced tumor growth inhibition (~79%), indicating PCX/anti-miR-210 nanoparticles sensitized the tumors to chemotherapy. At the end of the study, tumors were photographed (Figure ) and weighed (Figure ). The mean tumor weight in the PCX/anti-miR-210 group (0.34 g) was significantly lower than the weight in the PCX/anti-miR-NC group (0.68 g) and the untreated group (~0.81 g). Combination PCX/anti-miR-210 with GEM/CDDP resulted in the lowest tumor weight (~0.15 g). These findings confirmed the strong antitumor and chemosensitizing effect of PCX/anti-miR-210 nanoparticles in CCA.

As shown by the unchanged body weight during the treatment, PCX/anti-miR-210 had no apparent signs of gross toxicity (Figure ). H&E staining of major organs demonstrated no obvious tissue damage when compared to the PBS group (Figure ). Due to the low dose of the chemotherapy drugs in this study, we observed no reduction in body weight or presence of tissue injury caused by GEM/CDDP. Immunohistochemical staining of caspase 3 in the tumors was used to validate the proapoptotic activity of the nanoparticles (Figure ). When compared with the PBS group (2% caspase-3-positive cells), the PCX/anti-miR-210 treatment increased the fraction of apoptotic caspase-3-positive cells to ~10%. Combination of PCX/anti-miR-210 and GEM/CDDP showed even higher proapoptotic activity, with 19% of caspase-3-positive cells. Analysis of the H&E tumor sections also suggested considerably enhanced necrosis in the PCX/anti-miR-210 and GEM/CDDP combination group when compared with the control groups.

Conclusion

In this study, we developed an innovative combination nanoparticle treatment approach through simultaneous inhibition of CXCR4 and miR-210. The results showed that CXCR4-inhibiting polycation PCX could efficiently block the hypoxia-induced migration of CCA cells apparently through CXCR4 and LPA pathways. PCX/anti-miRNA nanoparticles delivered functional anti-miRNA to CCA cells and downregulated miR-210 expression, which resulted in significant cell killing through induction of apoptosis. PCX/anti-miR-210 nanoparticles sensitized CCA cells to GEM/CDDP chemotherapy by reducing stemness and synergistically reversed hypoxia-induced drug resistance. The combination nanoparticles achieved effective systemic delivery to CCA xenograft tumors and enhanced antitumor therapy in vivo through direct tumor growth inhibition and chemotherapy sensitization. The nanoparticles represent a promising dual-function delivery platform for miRNA delivery and provide safe and effective nanomedicines for systemic CCA therapy. Future studies will focus on the evaluation of the nanoparticles in orthotopic and metastatic CCA animal models to facilitate their clinical translation. We will further optimize the formulation of the nanoparticles to improve their systemic delivery to tumors. For example, the high positive charge of the nanoparticles is likely to reduce their circulation time due to increased binding affinity with polyanion components in the blood and opsonization. Optimized PEGylation or zwitterionic modifications of PCX is thus expected to improve their in vivo pharmacokinetic performance.

Supplementary figures.

Click here for additional data file.