Cadmium oxide (CdO, AR), stearic acid (98%), selenium powder, octylamine (OA, 99%), 1-hexadecylamine (HAD, 90%), and diethylzinc (ZnEt2) were obtained from Aladdin Co., Ltd. (Xi'an, China). Trioctylphosphine oxide (TOPO, 98%), trioctylphosphine (TOP, 95%), poly(acrylic acid) (PAA, molecular weight (MW) 1,800), 1-ethyl-3-[3-dimethylaminoporpyl] carbodiimide hydrochloride (EDC, 98.5%), and N-hydroxysuccinimide (NHS, 98%) were obtained from Sigma-Aldrich Co., Ltd. (St. Louis, MO, USA). Bovine serum albumin (BSA, 99.9%) was purchased from MP Biomedicals Company (Santa Ana, CA, USA). Bis(trimethylsilyl) sulfide ((TMS)2S) was purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). Liquid paraffin, chloroform, ethanol, hydrochloric acid (HCl), 2-(4-morpholino)ethanesulfonic acid (MES), N,N-dimethylformamide (DMF), paraformaldehyde, and Tween-20 were purchased from Sinopharm Chemical Regent Co., Ltd. (Shanghai, China).
Synthesis of CdSe and CdSe/ZnS core-shell QDs
Highly luminescent core-shell CdSe/ZnS QDs were prepared in high temperature via the pyrolysis of organometallic reagents in a coordinating solvent [26–28]. We select 200°C with and without HAD for synthesis of green- and red-emitting CdSe QDs. The molar ratio of CdO/Se/stearic acid in liquid paraffin was 1:1:4, and the crude QD products were purified by chloroform and ethanol. For the ZnS shell, equal molar ratios of (TMS)2S and ZnEt2 as precursors of Zn and S, and TOP/TOPO were used, and 90°C was used for shell growth. The final core-shell product was repurified and redispersed into aliquot chloroform for later use. About 10 ml of deionized water was added to the solution to prevent evaporation of chloroform for long-period storage (see Additional file 1 for synthesis details of QDs).
Synthesis and characterization of amphiphilic polymer
The amphiphilic polymer is synthesized as follows: in ambient temperature, 0.2 g of PAA (MW 1,800) was added to a flask containing 10 ml DMF. Under slight stirring for 1 h, 137 μl of OA was added, and the solution was continuously stirred for another 30 min. In an individual vial, 0.47 g EDC was dissolved in 0.5 ml DMF and injected to the reaction solution dropwisely. The reaction solution was mixed vigorously overnight to produce amphiphilic polymers (with 50% of the carboxylic acid functional groups modified with an aliphatic chain). Next, 0.25 M HCl was added drop by drop to the polymer solution under vigorous stirring, resulting in a milky and opaque colloid solution. The colloid was centrifuged at 8,000 rpm/min for 10 min. The supernatant was discarded, and the jelly-like precipitant was washed with 0.25 M HCl twice to remove any by-products and impurities. The final precipitate was collected and freeze dried to remove trace amounts of water, giving a dry, white powder. Fourier transform infrared (FTIR) spectroscopy (Equinox 55, Bruker, Karlsruhe, Germany) was used to verify the formation of amide bond and carboxylic groups.
Preparation and characterization of amphiphilic polymers conjugated with QDs
An aliquot of amphiphilic polymer powder was resuspended in MES buffer (0.1 mol/l, pH 6.0) for later use. As-prepared QDs (200 μl, 0.15 mmol) dissolved in chloroform and amphiphilic polymer solution (2.0 ml, 0.45 mmol) were added to 8 ml of deionized water in an open container. The solution was stirred and sonicated for 30 min until the chloroform evaporated completely in the final products. Afterward, the hydrated colloid (polymer-coated QDs, PQDs) was further purified by size exclusion chromatography (Superdex 75, Pharmacia Biotech, AB, Uppsala, Sweden), yielding a transparent, homogeneous, and strong fluorescent solution.
After purification, the purified solution was then concentrated under reduced pressure using a rotary evaporator at approximately 15°C. For assessment of the size distribution and monodispersity of the PQDs, the primal QDs of CdSe, CdSe/ZnS, and purified PQDs were pipetted onto a carbon transmission electron microscopy (TEM) grid; the solvents were wicked away slowly after 15 min. For the PQDs, the grids were counterstained with a 1% phosphotungstic acid solution (pH adjusted to 6) for 30 s. The staining solution was wicked away similarly. All of the prepared grids were imaged (TEM, JEM-2100 F system, JEOL Ltd., Tokyo, Japan) and compared to determine size distribution of the QDs and the degree of polymer coating. For further size analysis, the as-prepared QDs and PQDs were measured using Zetasizer Nano ZSP (Malvern Instruments, Ltd., Worcestershire, UK). In addition, the optical properties of the prepared CdSe, CdSe/ZnS, and PQDs were measured using UV-visible and fluorescence spectrophotometer (Cary 50 Conc, Varian, Palo Alto, CA, USA; F-4600, Hitachi, Tokyo, Japan). The QD concentration was determined using Beer's law after measuring the absorbance value using spectrophotometry [29, 30].
In order to estimate the surface charge and functional group character, we further characterized the polymer and PQDs by using 1% agarose gel electrophoresis. The agarose gel was prepared using standard techniques, and the prepared polymer and PQDs were added into the loading well. The gel was run in 0.5× TBE buffer (pH 8.0) for 30 min at 100 V and imaged with Tanon 2500 gel imaging system (Tanon, Shanghai, China) under 365-nm exciting light. Afterward, the gel was stained with lead acetate (1%) and potassium chromate (1%) for 5 min, respectively, and imaged in the Tanon 2500 system with white light to visualize the carboxyl group contained in the amphiphilic polymer. The migration rates of polymer and PQDs were compared to validate the success of QDs' surface coating.
Effects of pH and ionic strength on the stability of PQDs
In order to evaluate the effects of a wide pH range and high salt concentration on the colloidal stability of the PQDs, the PQD colloids were dispersed in varied pH buffers, PQDs/buffer = 1:1 (v/v), and pH ranged from 2 to 13 (Additional file 1: details of preparation of a series of buffer solutions). The resulting PL spectra were background-corrected, integrated, and normalized to the intensity of PQDs in pH = 7, set as 100%. The stability effect of ionic strength was carried out as follows: dispersions of PQDs were placed in fluorescence cuvettes (1-cm optical path) containing an equal concentration of PQDs but various concentrations of sodium chloride. The lack of volumes was replenished with deionized water (pH = 7). The PL emission from PQDs without NaCl added was set to 100%. The resulting PL spectra were normalized to the emission form slat-free solution.
Preparation of BRCAA1 antibody- and Her2 antibody-conjugated QD nanoprobes
The BRCAA1 monoclonal antibody was conjugated with red PQDs, whereas humanized Her2 monoclonal antibody was conjugated with green PQDs. The optimum mole ratio of PQDs to antibody is 5:3 . The cross-linking reaction was done by using standard EDC-NHS procedure in ambient temperature and dark place for 2 h with continuous mixing. The mixture was then purified by chromatography (Superdex 75, Pharmacia Biotech, AB, Uppsala, Sweden) to remove the free antibody residues. The resultant BRCAA1 antibody- and Her2 antibody-conjugated PQDs were stored at 4°C for later use.
Afterward, the prepared PQDs and specific monoclonal antibody conjunction were analyzed in 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE, Beyotime, Shanghai, China). The gel was run in a standard SDS buffer for 90 min at 120 V. Firstly, the gel was imaged with UV light to determine PQD position, and then, the gel was stained with Coomassie Brilliant Blue fast staining solution and imaged with white light to determine protein position.
The coupling rate of the PQDs and monoclonal antibody was estimated by a NanoDrop device (Thermo Scientific, Wilmington, DE, USA). Before coupling reaction, we measured the total concentration of monoclonal antibody. After coupling reaction, we estimated the monoclonal antibody concentration in the eluenting phase of chromatography and calculated the coupling rate according to the following equation:
BRCAA1 antibody- and Her2 antibody-conjugated QDs for targeted imaging of MGC803 cells in vitro
The overnight incubated MGC803 and GES-1 cells were fixed with 4% paraformaldehyde for 10 min and permeated with 0.5% (v/v) Tween-20 for 20 min. Then, these cells were blocked for 20 min in PBS containing 1% (w/v) BSA. After being washed with PBS twice, these cells were incubated at 4°C with BRCAA1 antibody- and Her2 antibody-conjugated QDs overnight. Unbound probes were removed by washing three times with PBS. Afterward, these cells were imaged under a fluorescence microscope (TS100, ×400, Nikon Co., Tokyo, Japan) and laser scanning confocal microscope in oil immersion objective (Nikon A1si+, ×1,000).
After attaining the fluorescence images, the gastric cancer cells were dissociated from the glass culture dish and sectioned as routine for TEM imaging.
BRCAA1 antibody- and Her2 antibody-conjugated QDs for targeted imaging of gastric cancer cells in vivo
To quantitatively analyze the fluorescence intensity from PQD-labeled MGC803 cells, macro fluorescence images were acquired using PQD-labeled MGC803 cells which were diluted with PBS to a final concentration from 2 × 102 to 2,048 × 102 cells/200 μl. Afterward, 200 μl of the prepared cell solutions were added to polystyrene TC-treated 96-well microplates (Corning® Life Sciences, Corning, NY, USA, #3603). Fluorescence intensity was measured in a Bruker In-Vivo F PRO system (Bruker Corporation, UK), and the resulting background-corrected data was curve fitted to single exponentials. Signal curve fitting was done using the software Origin (OriginLab, Northampton, MA, USA; http://www.originlab.com/).
All of the following animal studies complied with current ethical considerations: Approval (SYXK-2007-0025) of the Institutional Animal Care and Use Committee of Shanghai JiaoTong University (Shanghai, China) was obtained. Nude mice (male, 18 to 22 g, 4 to 5 weeks old) were obtained from the Shanghai LAC Laboratory Animal Co. Ltd., Chinese Academy of Sciences (Shanghai, China, SCXK2007-0005), and housed in a SPF-grade animal center.
Pathogen-free athymic nude mice were housed in a vivarium accredited by our University. Male athymic nude mice (4 to 6 weeks old) were used to establish subcutaneous gastric cancer models; 1.5 × 106 MGC803 cells suspended in 100 μl DMEM were subcutaneously injected into the left anterior flank area of each mouse. Four weeks later, tumors were allowed to grow to approximately 5 mm in diameter, and the prepared Her2 antibody-conjugated QDs (red, emission peak 657 nm) were injected into the mice via the tail vein for 6 h. Whole-animal imaging and ex vivo organ imaging were performed using the Bruker In-Vivo F PRO system. The excitation and emission filters were set to 410 and 700 nm (band pass, ±15 nm), respectively, and exposure time was set to 3 s. Collected images were analyzed using the imageJ software (NIH ImageJ; http://rsb.info.nih.gov/ij/), which uses spectral unmixing algorithms to separate autofluorescence from quantum dot signals.