Monoclonal anti-human EGFR antibody was obtained from Cetuximab, Merck KGaA, Germany. ¹³¹I was provided by the Beijing Atomic Gaoke Limited by Share Ltd. The nanoparticles were obtained from the Institute of Nanobiotechnology, School of Materials Science and Engineering, Tianjin Key Laboratory of Composites and Functional Materials, Tianjin University.

The amphiphilic BSA-PCL conjugate was synthesized as previously described (16), and the nanosized self-assembly of the amphiphilic BSA-PCL conjugate was obtained via emulsion-solvent evaporation method. Briefly, 4 mg of BSA-PCL conjugate was dissolved in 4 ml of phosphate buffer (PB, 0.1 M, pH 7.4) at room temperature and sonicated in a bath sonicator for 10 min. During the ultrasonic treatment process, 2 ml of dichloromethane was slowly injected into the PB via syringe. Dichloromethane was then evaporated with a vacuum rotary evaporator, and self-assembly of the cetuximab-decorated BSA-PCL conjugate was prepared with the same process. The obtained PB solutions of the self-assemblies of the BSA-PCL and cetuximab-decorated BSA-PCL conjugates were stored at 4°C.

U251 and U87 human glioblastoma cells were purchased from Cell Resource Center, Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences/Peking Union Medical College (Beijing, China). The U251 and U87 cells which overexpressed EGFR were cultured as a monolayer in DMEM media supplemented with 100 U/ml of penicillin, 100 M/ml streptomycin, and 10% FBS in a humidified atmosphere containing 5% CO₂ at 37°C.

We used the well-established direct labeling method for anti-EGFR-nanoparticles EGFR-BSA-PCL and BSA-PCL (18,19). EGFR-BSA-PCL and BSA-PCL were labeled with ¹³¹I (Beijing Atomic Hi-Tech Co., Ltd.) using chloramine-T method. The same amounts of EGFR-BSA-PCL or BSA-PCL were diluted in PB to a total volume of 100 µl (1 mg/ml), and ~37 MBq ¹³¹I was added. Up to 100 µl of chloramine-T (5 mg/ml in PB) was added. After 60 sec of incubation, complete oscillation using an oscillator was simultaneously needed. The reaction was stopped by adding 100 µl of sodium metabisulfite (5 mg/ml in PB), and complete oscillation for 60 sec was also needed. To separate the labeled EGFR-BSA-PCL or BSA-PCL from the low-molecular-weight compounds, centrifugation method was adopted to remove the small molecules. Our group used a centrifuge tube (Amicon^® Pro Purification System, Merck Millipore). The specific radioactivity was approximately 333 and 296 MBq/mg for EGFR-BSA-PCL and BSA-PCL, respectively. The ratio of radioiodine label was also calculated.

As previously mentioned, cells were planted in the confocal small dish (5×10³ cells/dish). Using the above steps, EGFR-BSA-PCL and BSA-PCL (marked with FITC) were added to each dish and incubated another 4 and 12 h, respectively. After incubating the rejected nanoparticles, the cells were washed twice or thrice repeatedly with sterile PBS and fixed with 0.5 ml of paraformaldehyde. Extracellular localization of fluorescent EGFR-labeled nanoparticles was then periodically visualized and recorded using Laser Scanning Confocal Microscope (Leica TCS SP2; Leica, Munich, Germany).

Flow cytometry was used to evaluate the binding of EGFR-BSA-PCL and BSA-PCL to target cells (EGFR⁺: U251 and U87). U251 and U87 cells were seeded into a cell-culture flask (5×10⁶ cells) and incubated for 24 h. Up to 1 ml each of EGFR-BSA-PCL and BSA-PCL (1 mg/ml, marked with FITC) was added, and the cells were incubated for 4 or 12 h. After incubation, the supernatant was discarded, and the plates were washed two to four times in PBS and then used for trypsinization. After centrifugation, 250 µl of the cell suspension was obtained, and another 250 µl of paraformaldehyde was added. The cells were then analyzed by a BD FACSCalibur flow cytometry (BD Biosciences, Franklin Lakes, NJ, USA) using 490 nm laser source.

MTT assays were conducted to evaluate the cytotoxicity of ¹³¹I-labeled EGFR-BSA-PCL and BSA-PCL. U251 and U87 cells were seeded into 96-well microplates with 1×10⁴ cells/well. The ¹³¹I-EGFR-BSA-PCL, ¹³¹I-BSA-PCL, EGFR-BSA-PCL, or BSA-PCL under different concentration gradients were added and then incubated for another 4 h. After removing all the nanoparticles, 20 µl of MTT reagent [3-(4,5 dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide; Sigma] was added (5 mg/ml) into each well. The plates were incubated for 4 h, and the MTT-containing medium was removed and replaced with DMSO. The amount of the blue formazan compound indicated the number of living cells and was determined using a spectrophotometer (492 nm). Finally, after gently shaking the microplates for 10 min to dissolve the formazan, each sample with six replicates (n=6) was analyzed on a BioTek ELX 800 microplate reader (BioTek Instruments, Inc., Winooski, VT, USA) at λ=492 nm.

Briefly, the cells (10⁵/well) were plated in triplicates in 96-well plates. Subsequently, 0.37–3.7 MBq ¹³¹I-EGFR-BSA-PCL and ¹³¹I-BSA-PCL were added into each well for 4 h. The medium was then completely removed, and the cells were quickly washed twice with ice-cold PBS. Up to 200 µl of 10% FBS-DMEM was added into each well. A γ-counter (LKB Gamma 1261; LKB Instruments, Mt. Waverley, Australia) was used to calculate the counts per minute (CPM) after 24 h.

This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and China Regulations For the Administration of Affairs Concerning Experimental Animals. The protocol was approved by the Committee on the Ethics of Animal Experiments of the Tianjin Medical University General Hospital. All surgery was performed under sodium pentobarbital anesthesia, and all efforts were made to minimize suffering.

Four-week-old BALB/c nude female mice were obtained from the PLA Military Academy of Medical Sciences Laboratory Animal Center. The mice were normally bred and maintained under specific-pathogen-free conditions with constant temperature and humidity range. The nude mice (n=27) were inoculated with U87 in the subcutaneous tissue upper back. A total volume of 50 µl was inoculated in the subcutaneous tissue upper back of the mice using a syringe. The U87 cells were slowly injected, and the syringe was kept in for 1–2 min before being retracted to avoid cells from ascending through the injection canal. Based on preliminary studies, the total span of the experiments was set to 21 days to ensure sufficient tumor development. However, the mice developed signs of considerable tumor burden, which is defined as loss of >20% of the mice initial body weight, and the diameter of the tumor deposit was >1 cm. After 21 days, the animals were divided into three experimental groups with three mice each: groups A, B, and C were treated by intratumoral injection with 74 MBq (370 MBq/ml) ¹³¹I-EGFR-BSA-PCL, 74 MBq (370 MBq/ml) ¹³¹I-BSA-PCL, and 74 MBq (370 MBq/ml) ¹³¹I, respectively.

At 4, 24, and 72 h after intratumoral injection, single-photon emission computed tomography (SPECT) (Discovery VH; GE Healthcare, Milwaukee, WI, USA) imaging was previously performed (26). To reduce the exposure of salivary and thyroid glands to unwanted radiation, all the mice were administered with 1% sodium perchlorate solution via their drinking water for 4 h before the experiment.

All experiments were performed in triplicate unless otherwise indicated. Statistical analysis was performed using SPSS software (SPSS 15.0). Results are presented as mean ± SD. Statistical significance was examined using Student's t-test. P<0.05 was considered statistically significant. For different experimental groups, we compared the difference between different drugs at the same medication time and at the same dose of the drug; for the same group comparison, we compared the efficacy of the same drug in different administration times, and compared the efficacy of the same medication time at different doses.

Confocal microscopy was used to analyze the binding and internalization of targeted and non-targeted BSA-PCL. Specific fluorescence staining on U251 and U87 cells was shown upon incubation with EGFR-BSA-PCL and BSA-PCL after 4 or 12 h at 37°C (Fig. 1). The fluorescence intensity of the EGFR-BSA-PCL group was stronger than that of the BSA-PCL group in both U251 and U87 cells regardless of similar conditions at 37°C for 4 or 12 h of incubation.

The binding activity of the EGFR-BSA-PCL and BSA-PCL conjugates was further confirmed by flow cytometry using the U251 and U87 cells that express EGFR. FCM analyses revealed results consistent with those observed in the fluorescence microscopy analyses. Hence, the binding and uptake of EGFR-BSA-PCL were significantly higher than those of BSA-PCL in both the U87 and U251 cells after 4 or 12 h of incubation at 37°C. The uptake of EGFR-BSA-PCL was 70.5 and 59.2% in the U251 and U87 cells, respectively, after 4 h of incubation. The uptake of EGFR-BSA-PCL was 76.0 and 60.0% in the U251 and U87 cells, respectively, after 12 h of incubation (Fig. 2). These findings implied that EGFR-BSA-PCL had special binding efficiency in the U251 and U87 cells which expressed EGFR.

Cellular binding and uptake of the two different ¹³¹I-labeled BSA-PCLs were evaluated by confocal fluorescence microscopy and flow cytometry in the U251 and U87 cell lines. The targeting efficiency of ¹³¹I-EGFR-BSA-PCL was considerably higher than that of ¹³¹I-BSA-PCL, EGFR-BSA-PCL, or naked BSA-PCL in both U251 and U87 cell lines. Analysis of cell viability determined by MTT cytotoxicity experiment; (7,27–29). Cytotoxicity was determined using MTT assay as previously reported. Time- and dose-dependent MTT studies were performed (Fig. 3, Table I). The data showed that 0.925 MBq ¹³¹I-EGFR-BSA-PCL or ¹³¹I-BSA-PCL had the strongest inhibition effect on tumor cells at 24, 48, 72 and 96 h. Experiments also showed that ¹³¹I-EGFR-BSA-PCL had better tumor inhibition effect than ¹³¹I-BSA-PCL with extended time. After incubating the nanoparticles for 4 h, the cell viability of the U251 and U87 cells was determined after 24, 48, 72 and 96 h, respectively. The cells incubated with ¹³¹I or cetuximab alone have no effect on the survival of glioma cells (data not shown).

Confocal microscopy, flow cytometry, and MTT analyses revealed that both ¹³¹I-EGFR-BSA-PCL and ¹³¹I-BSA-PCL could inhibit cell growth. However, the inhibition effect of ¹³¹I-EGFR-BSA-PCL was better than that of ¹³¹I-BSA-PCL in both U251 and U87 cells. Radionuclide uptake results were consistent with the CPM of ¹³¹I-EGFR-BSA-PCL, which was always higher than ¹³¹I-BSA-PCL in both U251 and U87 cells. After 4 h incubation with ¹³¹I-EGFR-BSA-PCL and ¹³¹I-BSA-PCL, the radioiodine uptake was distributed in a parabola fashion in the U251 and U87 cells (Fig. 4). When radioiodine uptake reached the peak, the CPM decreased as the dose of ¹³¹I-labeled nanoparticles increased. In the U251 cells, 0.925 MBq ¹³¹I-EGFR-BSA-PCL and 1.85 MBq ¹³¹I-BSA-PCL had the highest CPM. In the U87 cells, 1.85 MBq ¹³¹I-EGFR-BSA-PCL and 2.775 MBq ¹³¹I-BSA-PCL had the highest CPM.

The biodistribution of ¹³¹I-EGFR-BSA-PCL, ¹³¹I-BSA-PCL, and ¹³¹I in mice bearing U87 tumor xenografts at 4, 24, and 72 h and the ratio of radioactivity counts at different time points are shown in Fig. 5.

SPECT images showed that after the injection of the drug, the development of the tumor site was the most clear (Fig. 6). This finding indicated that the drugs are mainly gathered at the tumor site, and with time, the development of the tumor tissues gradually fade; the other parts of the nude mice develop gradually, indicating that the drug entered through the bloodstream to other parts of the body and the retention time of ¹³¹I-EGFR-BSA-PCL in the tumor tissues was longer than that of ¹³¹I-BSA-PCL and ¹³¹I. The ratio of radioactivity count with tumor tissue and nude body at 4 h after injection was the highest, followed by a decreasing trend. In the ratio of radioactive counts at different time points, ¹³¹I-EGFR-BSA-PCL group was higher than the ¹³¹I-BSA-PCL and ¹³¹I groups, and the decreasing trend was slower, indicating that the residence time of ¹³¹I-EGFR-BSA-PCL in the tumor was the longest.