This article is mentioned in:

Introduction

Gastric cancer is the fourth most common malignancy and accounts for >740,000 cancer-associated mortalities/year globally (1,2). Despite great improvements made in therapeutic methods in recent years, the prognosis is still unsatisfactory (3). Novel combinations of conventional chemotherapies, including the SPIRITS trial (S1 plus cisplatin vs. S1) demonstrated improved overall survival in patients treated with S1 plus cisplatin (13.0 months) compared with those treated with S1 alone (11.0 months) (4). Additionally, other combinations of cytotoxic agents, including docetaxel, irinotecan, capecitabine and oxaliplatin, have also been reported to prolong survival. However, therapeutic efficacy is still limited by two major factors: Drug resistance and side effects (5–7).

Human epithelial growth factor receptor 2 (HER2) is over-expressed in a significant proportion of gastric cancers (6–23%) (8). It is associated with tumor invasion, metastasis, chemoresistance and poor prognosis (9). Trastuzumab, as a recombinant humanized monoclonal antibody that targets the extracellular domain IV of HER2, is one of the most promising targets in human malignancy in recent years (10). In trial investigating the use of trastuzumab for gastric cancer, the addition of trastuzumab (Herceptin®) to chemotherapy significantly improved overall survival (13.8 months, 95% CI 12–16) compared with chemotherapy alone (11.1 months, 95% CI 10–13) in patients with HER2-positive gastric cancer (11), which indicated that combined chemotherapy with trastuzumab may be a novel treatment for patients with HER2-positive advanced gastric cancer. However, the majority of patients with gastric cancer still develop acquired resistance to trastuzumab (12). To achieve improved benefits for HER2-targeted therapy, the development of novel drug delivery systems that may decrease the dosage and periods of molecular targeted therapy are urgently required.

Nanoparticle (NP)-based therapeutics offer an innovative method to overcome the limitations of current agents (13). NPs possess unique properties that enable them to be used as imaging probes, which may be traced via magnetic resonance imaging (MRI) and therapeutic agents at the same time (14); however, they can be further loaded to deliver specific drugs (15–16) or target specific molecules (17,18). Superparamagnetic iron oxide (SPIO), known to be a highly efficient T2 contrast agent for MRI, is an ideal small molecular probe for medical use (19). Although SPIO exploit an enhanced permeability and retention effect (EPR) for tumor uptake, EPR is still inefficient with relatively low concentrations of NPs reaching tumors (20). Vascular heterogeneity commonly exists in large tumors, particularly metastases, and leads to unpredictable rates of NP extravasation as well as decreased perfusion and overall uptake (21,22). Therefore, the development of active targeting NPs is required. However, reports on SPIO with a combined feature of targeted intracellular drug release and imaging function are rare.

In the present study, a dual-functioning NPs conjugate, Au-Fe3O4, for HER2 targeted oxaliplatin delivery and intracellular drug release triggered via pH, was developed. In order to illustrate its targeting and therapeutic potential, cell culture and animal experiments of oxaliplatin-Au-Fe3O4-Herceptin NPs against human gastric cancer cell line SGC-7901 were conducted. It was demonstrated in the present study that selective targeting of HER2-positive gastric cancer cells using oxaliplatin-Au-Fe3O4-Herceptin NPs may increase the efficacy and decrease the side effects of oxaliplatin chemotherapy.

Materials and methods

Preparation of Au-Fe3O4 NPs modified with poly(ethylene glycol) (PEG)

For typical synthesis of 8 nm Au NPs, a precursor solution containing hexane (10 ml), oleylamine (10 ml) and 0.1 g of HAuCl4.4H2O (Aladdin Shanghai Biochemical Technology Co., Ltd., Shanghai, China) was prepared and magnetically stirred at 15°C under low flow N2 (15°C, 15 MPar). Following 10 min, a premixed solution (at 15°C) containing tetrabutylammonium bromide (0.5 mmol), hexane (1 mmol) and oleylamine (1 mmol) was injected into the precursor solution; the solution changed color to deep purple within 5 sec. The mixture was incubated at 15°C for 1 h prior to the precipitation by absolute ethyl alcohol. The precipitated Au NPs were collected by centrifugation (11,688 × g, 8 min), dissolved in absolute ethyl alcohol and detected via transmission electron microscope (TEM; JEOL 1230; JEOL, Ltd., Tokyo, Japan) at 80–200 kV.

Au NPs (0.48 mmol), octadecene (20 mmol), oleic acid (8 mmol) and oleylamine (8 mmol) were then mixed prior to adding iron acetylacetonate (6 mmol) and dodecanediol (20 mmol) into the solution at 180°C for 2 h followed by 300°C for 1 h. Following cooling to room temperature, Au-Fe3O4 NPs were collected by centrifugation (16,501 × g, 10 min) and detected via TEM at 80–200 kV.

Surface modification of Au-Fe3O4 NPs

For Au-Fe3O4 modification, 50 ml α,ω-Bis{2-[(3-carboxy-1-oxopropyl)amino]ethyl}PEG (Mr=2,000; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), 1.0 mg of N-hydroxysuccinimide (NHS; Sigma-Aldrich; Merck KGaA), 1.25 g of dicyclohexylcarbodiimide (Aladdin Shanghai Biochemical Technology Co., Ltd., Shanghai, China) and 1.5 g of dopamine hydrochloride were dissolved in a mixture of CHCl3 (20 mmol), dimethylformamide (8 mmol) and anhydrous Na2CO3 (6 mmol). The solution was stirred at 37°C for 2 h, Au-Fe3O4 NPs (5 mg) were added, and the resulting solution was stirred overnight at 37°C under a N2 flow. The modified NPs were precipitated by adding cyclohexane (25 mmol), collected by centrifugation at 16,501 × g, dried under N2 flow at 40°C (23). Surfactants and other salts were removed via dialysis (molecular mass cut off, 14,000 kDa; Spectrum Laboratories, Inc., Rancho Dominguez, CA, USA) for 24 h in PBS or water. The final iron concentration of the particles was determined by inductively coupled plasma mass spectrometry (ICP-MS; Element XR, Thermo Fisher Scientific, Inc., Waltham, MA, USA), which is a type of mass spectrometry capable of detecting metals and several non-metals at concentrations as low as one part in 1015 (part/quadrillion) on non-interfered low-background isotopes. The ICP multi-element standard solution IV (1,000 mg/kg, Merck KGaA, Darmstadt, Germany) was used as background equivalent concentration in the ICP-MS measurement. The nitric acid (HNO3) and hydrochloric acid (HCl) used were ultra-pure 100 grade (Merck KGaA). High purified water was obtained by a Milli-Q system (Merck KGaA). The operating conditions of ICP-MS are listed in Table I.

Table I. Operating conditions of ICP-MS.

Table I.

Operating conditions of ICP-MS.

Parameter	Setting
Radio frequency power	1300 W
Plasma gas flow rate	16 l/min
Auxiliary gas flow rate	0.85 l/min
Carrier gas flow rate	1.12 l/min
Nebulizer	Micromist nebulizer with uptake rate of 200 l/min
Spray chamber	Isomist spray chamber
Sampling cone	1.0 mm
Skimmer cone	0.7 mm
Sampling depth	2.6 mm
Replicate of mesurement	5 times
Measured element and resolution mode (m/Δm)	57Fe (4000), 195Pt (4000)

Conjugation of Herceptin and oxaliplatin to Au-Fe3O4 NPs

To conjugate the anti-HER2 antibody Herceptin, Au-Fe3O4 NPs in methyl ester sulfonate (Aladdin Shanghai Biochemical Technology Co., Ltd.) were mixed with 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC; Sigma-Aldrich; Merck KGaA) for 15 min at 4°C. Sulfo-NHS (Sigma-Aldrich; Merck KGaA) was added into the solution, which was then subjected to PD-10 column (GE Healthcare Life Sciences, Uppsala, Sweden) filtering to remove excessive EDC and sulfo-NHS, according to the manufacturer's protocol. Herceptin (Genentech Inc., San Francisco, CA, USA) was added into the conjugate following the alteration of pH to 7.4. The antibody-conjugated NPs were separated from unbound Herceptin and Au-Fe3O4 NPs using 300 K ultra-filtration (Optima MAX-TL, Beckman Coulter, Inc., Brea, CA, USA). The peptide bond of Au-Fe3O4-Herceptin was recorded by infrared (IR) spectroscopy on a Nicolet™ iS10 spectrometer (Thermo Fisher Scientific, Inc.) in the range between 2,500–400 cm−1 and a resolution of 2 cm−1. The Herceptin mass spectrum was detected by matrix-assisted laser-desorption ionization/time-of-flight (TOF)/TOF using Ultraflex III TOF/TOF MS (Bruker-Michrom, Inc., Auburn, CA, USA). MALDI-TOF MS measurements were taken using reflectron positive-ion mode. Acceleration was performed at 25 kV. Laser power was set as high as possible allowing baseline separation of isotopic peaks. Sample spectra were acquired by summing 25,000 laser shots at a frequency of 2,000 Hz, using a window from m/z 2,000 to 20,000 (Table II). To confirm the structures of the peaks, tandem mass spectrometry (MALDI-TOF/TOF MS/MS) was performed using laser-induced disassociation. Data acquisition was performed with the FlexImaging 3.0 software (Bruker-Michrom, Inc.).

Table II. Operating conditions of MALDI-TOF MS.

Table II.

Operating conditions of MALDI-TOF MS.

Parameters	Settings
Mode	Reflectron positive-ion mode
Laser	200 Hz smartbeam-I laser
Delay	80 ns
Ion source	1 voltage, 25 kV; 2 voltage, 23.4 kV
Lens voltage	6 kV
Mass range	2,000–20,000 kDa
Software	FlexImaging 3.0 software (Bruker-Michrom, Inc.)

For oxaliplatin-binding ligand synthesis, ethyl bromoacetate (8 mmol) and KI (3 mmol) were added to a solution of cystamine dihydrochloride (1.125 g, 5 mmol) in 100 ml acetone and 10 ml Et3N, stirred for 6 h at room temperature and the insoluble solid was removed via filtration. The filtrated crude products was dried and purified on a Biotage flash chromatography (Isolera™, Biotage Inc. Uppsala, Sweden) with SiliaFLASH™ cartridges (SiliCycle Inc. Quebec City, Canada) according to the manufacturer's protocol (24). An appropriate solvent mixture (petrol/EtOAc, 3:1) was used as eluent, obtaining a yield of ~80% (Table III). The carboxylic groups were deprotected in methanol solution (0.496 g, 1 mmol), 5 ml of 1 M NaOH aqueous solution was added, stirred for 30 min at room temperature and a small amount of water was added if any precipitate remained. Following 24 h of stirring, 20 ml distilled water was added, the solution was acidified to pH 3.0 using 1 M HCl and the resulting precipitate was collected by centrifugation (16,501 × g) at 37°C and washed with EtOH/H2O (1:1). The detection of ligands of antibody-coupled Au-Fe3O4 NPs was preceded by nuclear magnetic resonance (NMR; Bruker Advance 2B/400 MHz; Bruker Corporation).

Table III. Chromatographic conditions.

Table III.

Chromatographic conditions.

Parameters	Settings
Column	SiliaFLASH™
Sample volume	8 ml
Band length	6 mm
Application rate	10 s/µl
Scanner	Camag TLC scanner III
Scanning speed	20 mm/s
Data resolution	100 µm/step
Measurement mode	Absorption
Flow rate	10 ml/min
Pressure	8 psi

The antibody-coupled Au-Fe3O4 NPs (1 mg) were mixed with oxaliplatin binding ligand solution (described previously) for 6 h at room temperature, uncoupled ligands were removed using a PD-10 column. Oxaliplatin (20 mg/ml; Eloxatin; Sanofi S.A., Paris, France) was added to the NP solution, stirred overnight at room temperature in the dark and unconjugated oxaliplatin was removed via low speed centrifugation (12,000 × g) followed by purification on a PD-10 column. The amount of oxaliplatin bound to the NPs was determined via ICP-MS under the same experimental conditions as previously stated (Element XR, Thermo Fisher Scientific, Inc.).

Oxaliplatin release assay

Oxaliplatin-Au-Fe3O4-Herceptin NPs (50 µg/2 ml) were placed into Slide-A-Lyzer™ Dialysis Cassettes (molecular mass cut off, 2,000 kDa; Thermo Fisher Scientific, Inc.), which was in 30 ml PBS at room temperature. At the time intervals of 1, 2, 4, 8 and 19 h, 1 ml PBS was sampled. The platinum concentration was determined via Element XR ICP-MS (Thermo Fisher Scientific, Inc.) additional details can be found in Table I.

Cell lines and animals

Stomach adenocarcinoma cell line SGC-7901 was purchased from American Type Culture Collection (Manassas, VA, USA). SGC-7901 cells were cultured in RPMI-1640 medium (Hangzhou Gino Biomedical Technology Co., Ltd., Hangzhou, China) supplemented with 10% fetal bovine serum (FBS; Hangzhou Sijiqing Biological Engineering Materials, Co., Ltd., Hangzhou, China) and 1% penicillin/streptomycin (Genom Biologic). Cells were incubated in a standard humidified incubator in 5% CO2 at 37°C and passaged every 3–5 days using trypsin-EDTA (Genom Biologic). A total of 50, 4–6-week-old female BALB/c nu/nu mice (18–22 g, Shanghai Institutes for Biological Sciences, Shanghai, China) used in the present study were housed 5 to a cage under pathogen-free conditions and maintained on a daily 12/12 h light/dark cycle at Zhejiang University Laboratory Animal Center (Hangzhou, China), a fully accredited Association for Accreditation of Laboratory Animal Care facility. Animals were housed in a sterile environment, fed sterilized food and water ad libitum. All animal studies were carried out in strict accordance with the guidelines for the welfare and use of animals in cancer research of the Committee of the National Cancer Research Institute. The Zhejiang University Animal Research Committee approved the present study protocol. All surgery was performed under sodium pentobarbital anesthesia (40 mg/kg), and all efforts were made to minimize suffering (25). Animal health and behavior was monitored every 12 h, and were euthanized when the mice display early markers associated with mortality or a poor prognosis, including abdominal distension, emaciation or cachexy.

Cell viability assay (cytotoxicity assay)

The cytotoxicity of NPs on SGC-7901 was determined using MTS assays (26). Briefly, SGC-7901 cells (4×103 cells/well) were seeded into 96-well plates and incubated overnight at 37°C. Au-Fe3O4 NPs with different Fe concentrations (0, 50, 100, 150, 200 and 250 mg/l) were added to the designed wells and incubated at 37°C for 24, 48 and 72 h. Following incubation at 37°C with MTS solution (CellTiter96®AQueous One Solution Cell Proliferation Assay, Promega Corporation, Madison, WI, USA) for 2 h, the absorbance was measured using a TU-1901 Ultraviolet-Visible Spectrophotometer (Beijing Purkinje General Instrument, Co, Ltd., Beijing, China) at 570 nm. The same method was used to measure the cytotoxicity of Au-Fe3O4-Herceptin and oxaliplatin-Au-Fe3O4-Herceptin at Fe concentrations of 0, 30, 50, 80 and 100 mg/l.

Cellular internalization of NPs

Oxaliplatin-Au-Fe3O4-Herceptin NPs with 0.01 mg/ml Fe were added to SGC-7901 cells until they were 70% confluent, cultured with RPMI-1640 medium (10% FBS, 1% penicillin) and incubated for 2 h at 37°C. Cells (1×105/ml) were plated on sterile glass slides were first fixed with 2.5% glutaraldehyde (Aladdin Shanghai Biochemical Technology Co., Ltd.) in phosphate buffer (pH 7.0, Aladdin Shanghai Biochemical Technology Co., Ltd.) for 4 h and then with 1% OsO4 (Aladdin Shanghai Biochemical Technology Co., Ltd.) in phosphate buffer (pH 7.0) for 1 h at room temperature. The specimens were placed in a 1:1 mixture of absolute acetone and Spurr's resin (AGAR Scientific, Ltd., Stansted, UK) for 1 h at room temperature, then placed in capsules contained embedding medium (Sigma-Aldrich; Merck KGaA) and heated at 70°C for 9 h. The specimen sections of cells were stained with uranyl acetate and alkaline lead citrate for 15 min consecutively and observed under TEM of Model H-7650 (Hitachi, Ltd., Tokyo, Japan) at 80 kV.

Xenograft tumor model and in vivo imaging by MRI

A total of 20 4–6-week-old female BALB/c nu/nu mice weighing 18–22 g were used for tumor implantation. Sub-confluent SGC-7901 cells were harvested and re-suspended in PBS to a final concentration of 1.0×108/ml. Tumor cell suspension was subcutaneously injected into the flank of each mouse. After 4 weeks, when the subcutaneous mass reached a diameter of 1.0 cm, prior to the development of signs of distress, oxaliplatin-Au-Fe3O4-Herceptin NPs complex was injected into the mouse model via the caudal vein at an equivalent iron concentration (0.5 mg/kg). Following 24 h, relaxation time measurements (T1 and T2) of an aqueous solution of oxaliplatin-Au-Fe3O4-Herceptin and NPs complex in vivo were acquired using a clinical magnetic resonance scanner (Signa EXCITE 3.0T HD; GE Healthcare Life Sciences) without contrast. An 11-cm circular coil was used for all the MRI studies. With regards to the T2-weighted sequence, a repetition time of 750 msec and an echo time of 50–300 msec were employed. To alleviate animal suffering, all mice were euthanized by the end of MRI studies via carbon dioxide asphyxiation followed by cervical dislocation, as the majority of mice become moribund within days of this time point. Subcutaneous tumor masses were harvested and immediately immersed in 10% neutral buffered formalin (pH 6.8–7.2, Thermo Fisher Scientific, Inc.) for 2 h at room temperature and then paraffin embedded and sliced (4 µm).. Iron staining was performed using an iron stain kit (Sigma-Aldrich; Merck KGaA) according to the manufacturer's protocol. Images (magnification, ×200) were evaluated with a light microscope (Leica DMi1, Leica Microsystems GmbH, Wetzlar, Germany).

Statistical analysis

Statistical analysis was performed with Student's t-test and one-way analysis of variance using SPSS 10.0 software (SPSS, Inc., Chicago, IL, USA). Bonferroni correction was applied for multiply comparisons dividing the significance level by the number of tested variables. All experiments were performed at least in triplicates and are expressed as the mean ± standard deviation (SD). P<0.05 was considered to indicate a statistically significant difference.

Results and Discussion

Synthesis and characterization of Au-Fe3O4 NPs

The 8–20 nm Au-Fe3O4 NPs were synthesized by decomposing iron pentacarbonyl on the surfaces of Au NPs in the presence of oleic acid and oleylamine. TEM was used to characterize the synthesis of the dumbbell-like Au-Fe3O4 NPs. Au NPs were observed as uniform spheres with an average size of 8 nm, based on the size (diameter)-distribution of ~100 NPs. The Au NPs within the Au-Fe3O4 NP complexes appeared to be black under TEM images (Fig. 1A) due to the heavy atom effect (27).

Figure 1. Synthesis and characterization of oxaliplatin-Au-Fe3O4-Herceptin NPs. (A) Transmission electron microscope images of dumbbell-shaped structure of 8–20 nm Au-Fe3O4 NPs, formed of 8 nm Au NPs (black) and 20 nm Fe3O4 NPs (grey). (B) Mass spectrometry analysis of Au-Fe3O4-Herceptin NPs. A specific peak for human epithelial growth factor receptor antibody, Herceptin (Mr=148), at m/z ~148,000 was detected by matrix-assisted laser-desorption ionization-TOF-TOF. (C) Infrared spectrum of Au-Fe3O4-Herceptin NPs. Specific double peaks of carbonyl bands at 1580 and 1650 cm−1 (arrow) representing the amide bonds linking the polyethylene glycol to the silane. (D) NMR of the oxaliplatin-binding ligand (D2O, 300 MHz) revealed peaks at δ 3.51 (8.00 H, s, 3-H), 2.97–3.02 (3.80 H, t, 2-H) and 2.72–2.76 (3.72 H, t, 1-H). (E) The NMR analog of oxaliplatin-binding ligand. (F) The released oxaliplatin level from pH=6.0 group was significantly increased compared with pH=8.0 group. (G) The released oxaliplatin level at pH=7.3 after 4 h of incubation was significantly compared with the free oxaliplatin group. *P<0.01.*P<0.01 vs. pH=8.0. NMR, Nuclear Magnetic Resonance; NPs, nanoparticles; TOF, time of flight; Pt, oxiplatin.

Xu et al (28) reported the structure of Au-Fe3O4 NP in 2009. Compared with conventional single-component Au or Fe3O4 NPs, Au-Fe3O4 NPs possess unique advantages. First, the structure contains magnetic (Fe3O4) and optically active plasmonic (Au) units and can therefore be stably detected optically and magnetically, compared with the fast signal loss observed in common fluorescent labeling. Secondly, the presence of Fe3O4 and Au surfaces facilitate the attachment of different chemical functional groups, enabling a variety of target-specific imaging and delivery applications. Thirdly, the size of NPs can be controlled to optimize their magnetic and optical properties, and as small particles are only capable of accommodating a few DNA strands, proteins, antibodies or therapeutic molecules, kinetic cell targeting and drug release assays can be performed (23).

Due to the high surface area-to-volume ratio, NPs have a tendency to aggregate and absorb plasma proteins upon intravenous injection, leading to rapid clearance by the reticuloendothelial system (RES) (29). The size of the Au-Fe3O4 NPs in the present study is suitable for in vivo applications, as it has been demonstrated that NPs between 10–150 nm in diameter can effectively escape rapid clearance by the RES. NPs are commonly protected with a polymer coating to improve their dispersity and stability (30). In the present study, PEG was used to align the oil phase of the NPs to the water phase, dopamine was reacted with the oleic acid on the surface of the Fe3O4 NPs, then the amino group of dopamine was reacted with the carboxyl group of PEG under catalysis of the EDC-sulfo-NHS system. This ensured that PEG was only connected to the Fe3O4 NPs. The ζ-potential of the dumbbell-shaped Au-Fe3O4 NPs prior to surface modification was 7.1±1.2 mV. Following being loaded with PEG, the dumbbell-like Au-Fe3O4 NPs possessed a ζ-potential of 13.8±1.6 mV. It is notable that the surface charge density of NPs is an important parameter that determines their colloidal stability and cellular interactions (31).

Fe3O4 surface functionalization and multifunctional probe loading

Herceptin is a recombinant, DNA-derived, humanized monoclonal antibody glycoprotein that selectively targets the extracellular domain of HER2 (32). Herceptin contains 1,328 amino acids and has a Mr of ~148 (33). Herceptin has been approved for the clinical treatment of HER2-overexpressing breast cancer, metastatic gastric cancer and gastro-esophageal junction adenocarcinoma (34,35). In the present study, the EDC-sulfo-NHS system was used to activate the carboxyl groups of Au-Fe3O4 NPs modified with PEG, as aforementioned. Removal of excess EDC and sulfo-NHS prior to addition of Herceptin was used to avoid self-crosslinking of the antibody. Following activation using the EDC-sulfo-NHS system, Herceptin was conjugated to the modified Au-Fe3O4 NPs through PEG (Mr=2,000) and dopamine via a condensation reaction with the formation of a peptide bond. Mass spectrometry analysis of oxaliplatin-Au-Fe3O4-Herceptin NPs revealed a specific peak for Herceptin at m/z ~148,000 (Fig. 1B), which was not detected in Au-Fe3O4 NPs. IR spectrum analysis of oxaliplatin-Au-Fe3O4-Herceptin NPs revealed the presence of a peptide bond, indicated by specific double split peaks between 1,580–1,650 cm−1 (Fig. 1C).

Oxaliplatin has long been used as a chemotherapeutic agent (36); however, it does not specifically target tumor cells and can be taken-up by any rapidly growing cells, leading to toxic side effects (37). The present study proposed that the conjugation of oxaliplatin onto Au-Fe3O4-Herceptin NPs, which can actively accumulate in HER2 positive tumor tissues, may greatly decrease the side effects of oxaliplatin. The Au particle of the Au-Fe3O4-Herceptin NPs was coated with thiol (HS)-PEG-NH2 with the thiol for oxaliplatin binding. NMR results of the ligand (D2O, 300 MHz) revealed three peaks at δ 3.51 (8.00 H, s, 3-H), 2.97–3.02 (3.80 H, t, 2-H) and 2.72–2.76 (3.72 H, t, 1-H), as presented in Fig. 1D. The NMR of the analog is depicted in Fig. 1E. Under certain conditions, the disulfide bond of the ligand was broken. One sulfur atom of the ligand was connected to an Au NP and another to oxaliplatin. In the present study, according to the weight percentage of Pt/Au (13.2%), giving a Pt/Au atom ratio of 13:100, ~2,084 Pt units were bound to each Au NP.

The results of drug release profiles in vitro at different pH values indicated that oxaliplatin-Au-Fe3O4-Herceptin NPs released 13, 21 and 32% of oxaliplatin in the first h and a total of 25, 43 and 58% by 4 h at pH 8.0, 7.0 and 6.0, respectively (P=0.003) (Fig. 1F). It appears that oxaliplatin became more prone to detach from the conjugated NPs in lower pH conditions, and that Pt release is pH dependent. When human serum has a pH of ~7.35, only ~25% of oxaliplatin will be released from the conjugated NPs by 4 h of incubation (28), which delayed the oxaliplatin release by ~4 times compared with free oxaliplatin (80% release, Fig. 1G). However, following uptake by target cells, particularly in the endosomal/lysosomal compartments, which have a pH of ~5.0, >60% of oxaliplatin was released from oxaliplatin-Au-Fe3O4-Herceptin NPs (28). This pH dependent release character could reduce the systemic toxicity effect to a great extent and is ideal for the selective targeting therapy in vivo.

Cellular uptake and cytotoxicity of oxaliplatin-Au-Fe3O4-Herceptin NPs in SGC-7901 cells in vitro

The SGC-7901 cell line is known to overexpress HER2 and is an ideal cell line to assess the efficacy of HER2 targeted therapy (38,39). Oxaliplatin-Au-Fe3O4-Herceptin NPs (0.01 mg/ml Fe) were incubated with SGC-7901 cells for 2 h. TEM image analysis revealed the presence of NPs in the endosome/lysosome, which indicated the successful uptake of NPs through endocytosis (Fig. 2A).

Figure 2. Cellular internalization and cytotoxicity of NPs. (A) Transmission electron microscope image of the oxaliplatin-Au-Fe3O4-Herceptin NPs in SGC-7901 cells via endocytosis (arrow) following 2 h incubation. (B) The viability of SGC-7901 cells was not significantly affected until treated with Au-Fe3O4 with a Fe concentration >100 µg/ml. (C) The viability of SGC-7901 cells was significantly decreased following incubation with oxaliplatin-Au-Fe3O4-Herceptin, compared with control. *P<0.01 vs. 0 µg/ml Fe. NPs, nanoparticles.

The cytotoxicity of NPs depends on their concentration and the surfactants used (28). In the present study, the water-soluble Au-Fe3O4 NPs were tested over a concentration range of 0–250 mg Fe/l. The MTS assay (Fig. 2B) revealed that the Au-Fe3O4 NPs did not induce appreciable cell viability (~100%) for ≤100 µg/ml Fe; however, cell viability decreased in a concentration-dependent manner at a Fe concentrations >100 µg/ml, with only half of the cells viable at a Fe concentration of 150 µg/ml (P=0.005). Therefore, within the safe zone (0–100 µg/ml), Au-Fe3O4 did not inhibit cell growth and the cytotoxicity of the NPs itself to the tumor cells may be negligible.

The cytotoxicity assays were further performed to assess the effect of oxaliplatin-Au-Fe3O4-Herceptin NPs on the viability of HER2-positive SGC-7901 cells in vitro (Fig. 2C). The control experiments demonstrated that the cytotoxicity of oxaliplatin-Au-Fe3O4-Herceptin NPs was markedly increased compared with Au-Fe3O4 NPs (P=0.005). The oxaliplatin-Au-Fe3O4-Herceptin NPs had a half-maximal inhibitory concentration toward SGC-7901 cells of 75 µg/ml (Fe concentration). The toxicity was considered to be as a result of two sources, one from Herceptin, the other from oxaliplatin released by hydrolysis.

A variety of specific antigens, including HER2, are frequently overexpressed on the surface of tumor cells. These antigens provide specific targets, which could be selectively bound by monoclonal antibodies. Therefore, when linked to a monoclonal antibody, NP conjugates may enable target-specific delivery via high affinity antibody-antigen interactions and receptor-mediated endocytosis (40–42). A number of cationic NPs have been indicated to enter cells by transiently generating holes in the cell membrane (43). However, most of the uptake of NPs into mammalian cells and macrophages usually occurs via endocytosis (44,45). In the present study, TEM indicated that the platinum-tethered NPs synthesized in the present study were internalized by SGC-7901 cells via endocytosis. Therefore, conjugation of Herceptin to Au-Fe3O4 NPs is an effective way for targeted internalization.

Imaging of the selective uptake and accumulation of oxaliplatin-Au-Fe3O4-Herceptin NPs in HER2-expressing gastric cancer cells in vivo

The specific targeting efficiency of oxaliplatin-Au-Fe3O4-Herceptin NPs with various iron concentrations towards HER2-positive tumor cells was investigated in vivo using a xenograft model of SGC-7901 and T2-weighted MRI without contrast. As depicted in Fig. 3A, the T2 relaxation time decreased with the increase of the iron concentration in T2-weighted MRI (Fe concentration of 0.0, 1.5, 3.0, 6.0, 12.0, 25.0, 50.0, 100.0, 200.0, 400.0 µg/ml).

Figure 3. Biodistribution of NPs. (A) T2-weighted MRI of xenograft tumor mouse model receiving oxaliplatin-Au-Fe3O4-Herceptin NPs. Accumulation of the NPs (low signal) was detected inside the percutaneous tumor mass (left arrow), compared with the group treated with Au-Fe3O4 NPs alone (right arrow) The color spectrum indicated the association between NPs concentration and the color displayed in T2 MRI. (B) Intracellular iron NPs accumulation around the tumor vessel (arrow) were marked blue using Prussian Blue staining (magnification, ×200). NPs, nanoparticles; MRI, magnetic resonance imaging.

For the purpose of acquiring adequate NPs accumulation at site and a safe dosage for the mice, in the present study, a dose of 0.35 mg/kg Fe was selected for the in vivo experiments based on the in vitro cytotoxicity assays. A total of 24 h following caudal vein injection, the mice were sodium pentobarbital anesthetized and received MRI examination. In T2-weighted imaging, the signal intensity of subcutaneous mass was significantly decreased in the oxaliplatin-Au-Fe3O4-Herceptin NPs group, compared with the group treated with Au-Fe3O4 NPs alone, which demonstrated no detectable intensity alteration. It indicated that the HER2-labeled NPs exhibit increased specificity in their attachment to SGC-7901 cells.

Following the sacrifice, subcutaneous tumor masses were paraffin embedded and stained with Prussian blue. Fe4[Fe(CN)6]3 sediment (blue) was then observed around the tumor feeding vascular wall, which indicated oxaliplatin-Au-Fe3O4-Herceptin NPs aggregation. These results demonstrated that the oxaliplatin-Au-Fe3O4-Herceptin NPs could specifically accumulate in the xenograft tumor mass with HER2 expression via blood flow. The mice were re-imaged by MRI two days following the first scan and no marked signal loss was observed. The study for long term tracking of the NPs in vivo remains underway.

Conclusion

In the present study, a dual-functioning NPs conjugate Au-Fe3O4 for HER2 targeted oxaliplatin delivery and intracellular drug release triggered via pH was developed. In order to illustrate its targeting and therapeutic potential, cell culture and animal experiments of oxaliplatin-Au-Fe3O4-Herceptin NPs against human gastric cancer cell line SGC-7901 were conducted. It was demonstrated that selective targeting of HER2-positive gastric cancer cells using oxaliplatin-Au-Fe3O4-Herceptin NPs may increase the efficacy and decrease the side effects of oxaliplatin chemotherapy. Oxaliplatin-Au-Fe3O4-Herceptin NPs with high stability in aqueous solution, HER2 targeting, pH dependent drug release and MRI detectable capability were successfully developed in the present study. Evaluation of oxaliplatin-Au-Fe3O4-Herceptin NPs in vivo by HER2 positive gastric cancer cell line SGC-7901 revealed that NPs demonstrated selective uptake and accumulated in HER2-expressing SGC-7901 xenograft tumor mass, which makes them a promising multifunctional platform for simultaneous magnetic traceable and HER2 targeted chemotherapy for gastric cancer.

Acknowledgements

The authors would like to thank Professor Shouheng Sun of Brown University (Providence, RI, USA) for assisting with the synthesis of Au-Fe3O4.

Funding

Zhejiang medical health science and technology plan (grant nos. 2013KYA164 and 2015KYB171), Zhejiang Provincial Natural Science Foundation (grant nos. LY15E010005, LQ15H190001 and LY18H180005).

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Authors' contributions

DL managed the experiment design, data collection, animal experiment, MRI analysis and manuscript preparation; XWL and CLC assisted the preparation of Au-Fe3O4 NPs and the surface modification. CL, CBZ and WDZ managed the cell culture and assisted the establishment of xenograft tumor model and in vivo imaging. JGZ assisted the animal experiment and MRI analysis. JF and KC in charge of the preparation of Au-Fe3O4 NPs, surface modification and ligation of HER2 and oxaliplatin. LC managed the experiment design and manuscript preparation. All authors have approved this manuscript.

Ethics approval and consent to participate

All animal studies were carried out in strict accordance with the guidelines for the welfare and use of animals in cancer research of the Committee of the National Cancer Research Institute. The Zhejiang University Animal Research Committee approved the present study protocol.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References