Radiolabeled novel mAb 4G1 for immunoSPECT imaging of EGFRvIII expression in preclinical glioblastoma xenografts

Epidermal growth factor receptor mutant III (EGFRvIII) is exclusively expressed in tumors, such as glioblastoma, breast cancer and hepatocellular carcinoma, but never in normal organs. Increasing evidence suggests that EGFRvIII has clinical significance in glioblastoma prognosis due to its enhanced tumorigenicity and chemo/radio resistance, thus the development of an imaging approach to early detect EGFRvIII expression with high specificity is urgently needed. To illustrate this point, we developed a novel anti-EGFRvIII monoclonal antibody 4G1 through mouse immunization, cell fusion and hybridoma screening and then confirmed its specificity and affinity by a serial of assays. Following biodistribution and small animal single-photon emission computed tomography (SPECT/CT) imaging of 125I-4G1 in EGFRvIII positive/negative tumor-bearing mice were performed and evaluated to verify the tumor accumulation of this radiotracer. The biodistribution indicated that 125I-4G1 showed prominent tumor accumulation at 24 h post-injection, which reached maximums of 11.20 ± 0.75% ID/g and 13.98 ± 0.57% ID/g in F98npEGFRvIII and U87vIII xenografts, respectively. In contrast, 125I-4G1 had lower tumor accumulation in F98npEGFR and U87MG xenografts. Small animal SPECT/CT imaging revealed that 125I-4G1 had a higher tumor uptake in EGFRvIII-positive tumors than that in EGFRvIII-negative tumors. This study demonstrates that radiolabeled 4G1 can serve as a valid probe for the imaging of EGFRvIII expression, and would be valuable into the clinical translation for the diagnosis, prognosis, guiding therapy, and therapeutic efficacy evaluation of tumors.


INTRODUCTION
The epidermal growth factor receptor (EGFR), a member of the ErbB family of receptor tyrosine kinases, is a 170 kDa transmembrane glycoprotein that is overexpressed in a variety of solid tumors [1][2][3]. Multiple ligands, including EGF, transforming growth factor-a, amphiregulin, betacellulin, and epiregulin, activate EGFR. Upon binding of the ligand to the EGFR extracellular domain, downstream pathways and several downstream signaling cascades are activated to induce the transcription of genes involved in cell proliferation, tumor invasion, and metastasis [4]. A variety of EGFRtargeting chemotherapeutic drugs, such as tyrosine kinase inhibitors (TKIs), and monoclonal antibodies (mAbs) have been developed to treat cancer patients by blocking EGFR signal transduction [5,6]. However, a large number of studies show that only a minority of patients benefit from these drugs, and the limited efficacy of small molecule TKIs (gefitinib and erlotinib) and mAbs (cetuximab,
Among EGFR mutations, EGFR variant III (EGFRvIII) has attracted increasing attention. EGFRvIII harbors an in-frame deletion of exons 2-7 that encode amino acid residues 6-273. This deletion produces a truncated 150 kDa protein that shows weak but constitutive oncogenic activation in a ligandindependent manner [10][11][12]. Glioblastoma multiforme, the most common and malignant type of primary brain tumors, is often associated with EGFRvIII expression [13,14]. Such expression can be regarded as a marker of poor prognosis because of its potential to confer enhanced tumorigenicity to gliomas [15][16][17]. Recent studies have demonstrated that EGFRvIII is expressed in glioma stem cells (GSCs) and may be a biomarker of GSCs [18]. Moreover, increasing studies have confirmed that EGFRvIII-positive cells are more radioresistant and chemoresistant to EGFR inhibitors because of the constitutively active receptor tyrosine kinase in EGFRvIII [19][20][21]. Thus, it is necessary to examine EGFRvIII expression before and after radio/chemotherapy for patient prognosis and prediction of drug resistance to administer the appropriate individualized therapy. Thus far, several EGFRvIII detection methods have been developed, including reverse transcription-polymerase chain reaction (RT-PCR), real-time RT-PCR, tumor immunohistochemistry, and blood tests. However, these detection methods usually result in inconsistent results due to the limitations of their methodology [22][23][24]. More importantly, these methods are not applicable to noninvasive in vivo detection or real-time monitoring of EGFRvIII expression.
In recent years, molecular imaging has emerged as a novel and rapidly growing multidisciplinary research field with the combination of molecular biology and in vivo imaging [25]. Molecular imaging not only enables noninvasive in vivo imaging, which reflects biological processes at cellular and sub-cellular levels, but also allows real-time monitoring of multiple molecular events and drug effects at molecular and cellular levels. Therefore, molecular imaging has been widely applied to assess disease progression at the molecular pathologic level for early diagnosis of cancer as well as neurological and cardiovascular diseases. Hence, the development of a molecular imaging probe to detect EGFRvIII expression before radiotherapy or chemotherapy would enable more accurate patient prognosis and prediction of drug sensitivity.
In this study, we developed a nuclear molecular imaging probe by labeling a novel anti-EGFRvIII mAb, 4G1, with a radioisotope and evaluated its potential to detect EGFRvIII expression in glioblastoma xenograft models by single-photon emission computed tomography (SPECT) imaging.

Production and characterization of novel mAb against EGFRvIII
After fusion of SP2/0 myeloma cells and spleen cells from immunized BALB/c mice, 157 positive hybridoma clones were obtained after initial ELISA screening. Among them, four hybridoma clones with the highest titer (4G1, 1F1, 7C7 and 4D3) were selected for further expansion after repeated screening. Finally, 4G1 was selected for further study because it had the highest titer, which immunoglobulin subtype was IgG2a.

Affinity and specificity of 4G1
Several experiments were performed to evaluate the affinity and specificity of 4G1. As shown in Figure 1A, the IC 50 value of 125 I-4G1 was 1.83 ± 0.03 nmol/L. To determine the Kd of 125 I-4G1 and number of binding sites per F98 npEGFRvIII cell (Bmax), we performed a saturation binding assay. The Kd value was 4.83 ± 0.12 nmol/L, and the Bmax was approximately 1.21 ± 0.61 × 10 6 sites/cell ( Figure 1B).
The binding assay results showed that 125 I-4G1 exclusively bound to the EGFRvIII protein expressed by F98 npEGFRvIII and U87vIII cells, moreover unlabeled 4G1 blocked this specific binding ( Figure 1C, 1D). The specificity was also confirmed by western blotting, immunofluorescence, and flow cytometric analysis. In western blot analyses, 4G1 exclusively recognized EGFRvIII expressed by F98 npEGFRvIII and U87vIII cells but not wildtype EGFR expressed by F98 npEGFR and U87MG cells (Figure 2A). Immunofluorescence and immunohistochemistry confirmed that 4G1 exclusively bound to EGFRvIII-positive cells and tumor tissues ( Figure 2B-2D). Flow cytometry results showed that the positive rate of F98 npEGFRvIII and U87vIII cells stained with 4G1 was 92.5% and 83.4%, respectively ( Figure 3B, 3C), whereas 4G1 did not bind to F98 npEGFR cells ( Figure 3A). Furthermore, Flow cytometric analysis showed that 4G1 could not block the binding of Erbitux (a mAb against EGFR) to EGFRvIII on F98 npEGFRvIII cells, indicating that 4G1 had no common binding sites with Erbitux ( Figure 3D).
Oncotarget 6366 www.impactjournals.com/oncotarget Elimination of blood radioactivity in mice with F98 npEGFRvIII and F98 npEGFR tumors ranged from 4.75 ± 0.81%ID/g and 4.38 ± 0.22%ID/g at 24 h p.i. to 1.28 ± 0.13%ID/g and 2.51 ± 0.08 %ID/g at 72 h p.i., respectively. For most organs, there was progressive washout of radioactivity over time, which was concomitant with the clearance of radioactivity from the blood. The uptake of 125 I-4G1 in all organs was almost less than 3%ID/g at day 3 p.i.

SPECT/CT Imaging of 125 I-4G1
Representative SPECT/CT images of F98 npEGFR and F98 npEGFRvIII tumor-bearing BALB/c nude mice are shown in Figure 6. Based on the imaging results, the uptake of 125 I-4G1 in F98 npEGFRvIII tumors was significantly higher than that in F98 npEGFR tumors at each time point. A small amount of radioactivity was also found in bloodrich organs, such as lung and heart, as well as liver at 4 h p.i. The background radioactivity decreased over time, which was consistent with the ex vivo biodistribution data. Images with a higher contrast were obtained at 24 h p.i. Moreover, the SPECT/CT images of F98 npEGFR and F98 npEGFRvIII intracranial tumor models as shown in Figure 7 showed that 125 I-4G1 was markedly higher accumulated in EGFRvIII positive glioma than EGFRvIII negative glioma, which indicates that 4G1 might be used for real-time imaging studies in patients with gliomas in future.

DISCUSSION
Preclinical investigations have shown that EGFRvIII stimulates unregulated growth, survival and invasion of glioma cells, as well as angiogenesis [26,27]. In addition, the prognostic value of EGFRvIII has been studied extensively. Current viewpoints support the notion that the presence of EGFRvIII is an independent and significant unfavorable prognostic indicator of survival [16,17,28,29]. Furthermore, EGFRvIII has been implicated in resistance to chemotherapy and radiotherapy [19][20][21]. Therefore, early detection of EGFRvIII expression might provide an early prognosis and suggest chemotherapeutic drug resistance. There have been several reports of EGFRvIII detection, but their results remain controversial, Figure 1: In vitro inhibition of 125 I-4G1 binding to EGFRvIII on F98 npEGFRvIII cells by unlabeled 4G1 showed that the IC 50 value was 1.83 ± 0.03 nmol/L (n = 3, mean ± SD) (A). Saturation binding of 125 I-4G1 to EGFRvIII on F98 npEGFRvIII cells showed that the Kd value was 4.83 ± 0.12 nmol/L. Bmax was calculated to be approximately 1.21 ± 0.61 × 10 6 sites/cell (B). Cell binding assays showed that 125 I-4G1 specifically bound to F98 npEGFRvIII and U87vIII cells, but not F98 npEGFR and U87MG cells that express wild-type EGFR (C, D).
Oncotarget 6367 www.impactjournals.com/oncotarget partly because of the lack of a consistently used EGFRvIIIspecific antibody and detection assays. Moreover, these methods can not be applied to in vivo and real-time detection of EGFRvIII. Hence, a more sensitive, accurate and effective detection method should be developed and implemented.  To address this issue, we adopted a recently developed research methodology, molecular imaging, which enables noninvasive in vivo imaging and real-time monitoring of molecular biological processes at cellular and sub-cellular levels. Thus far, molecular imaging has been successfully applied to individualized therapies by assessment of disease progression, drug resistance, and disease prognosis [30][31][32].
In this study, we developed a novel mAb, named 4G1, to detect EGFRvIII expression through molecular imaging. The affinity assay results indicated that affinity of 4G1 to EGFRvIII was as good as other similar mAbs reported by previous studies [33,34]. A series of specificity tests showed that 4G1 specifically recognized EGFRvIII and had no cross reactivity with wild-type EGFR. These results indicated that 4G1 had the potential to be a suitable targeting molecule for EGFRvIII-positive tumor imaging. To confirm this potential, we performed biodistribution and SPECT/CT imaging in xenografted tumor models. The results of biodistribution showed that EGFRvIII-positive tumors had a higher uptake of 125 I-4G1 than EGFRvIII-negative tumors. Furtherly, in the biodistribution experiment of U87vIII tumor-bearing mice, a group of 125 I-mIgG was done to confirm the specific uptake of 125 I-4G1 in tumors. Since mIgG could not bind to EGFRvIII, the accumulation of 125 I-mIgG in U87vIII tumors should be lower than that of 125 I-4G1 and thus regarded as the background attributed to tumor nonspecific uptake of antibody.
Furthermore, SPECT/CT imaging of 125 I-4G1 showed that EGFRvIII-positive and -negative tumors could be distinguished easily. Thus, radiolabeled 4G1 may be a promising probe targeting EGFRvIII for tumor imaging.
We also examined potential common binding sites for 4G1 and Erbitux. Erbitux has been granted full approval by the U.S. Food and Drug Administration for the first-line treatment of patients with EGFRexpressing metastatic colorectal cancer [35]. Numerous studies have shown that Erbitux recognizes both EGFR and EGFRvIII [36,37]. Our flow cytometric analyses revealed that 4G1 had no common binding sites with Erbitux. We also confirmed high affinity, specificity, and in vivo tumor accumulation of the 4G1 mAb. Considering that EGFRvIII is a prognostic and chemo/radio resistant marker, 4G1 can be easily labeled with PET or SPECT radioisotopes for potential clinical imaging of EGFRvIII expression. Moreover, 4G1 can be used as a delivery vehicle for tumor-targeted therapeutics, because EGFRvIII is exclusively expressed in tumor tissues.
Since EGFRvIII is a very promising therapeutic and diagnostic target that is exclusively expressed in tumors, it has attracted much attention. Over the past 10 years, some polyclonal [38][39][40] or monoclonal 125 I-4G1 uptake in F98 npEGFRvIII tumors was significantly higher than that in F98 npEGFR tumors. Data are expressed as the mean ± SD (n = 4 per group). **P < 0.01 and ***P < 0.001.
Oncotarget 6369 www.impactjournals.com/oncotarget antibodies against EGFRvIII have been developed. In consideration of potential immunological cross reaction or difference among batches of polyclonal antibody, we excluded it in the first place. For monoclonal antibodies against EGFRvIII, the mAb 806 [41] and CH12 [33] were generated by immunization of mice with EGFRvIIIoverexpressing fibroblasts, whereas L8A4 [42][43][44] was prepared by immunization of mice with peptides that were very similar to the peptides used in this study. These three mAbs have been evaluated in detail and applied to a range of research fields including EGFRvIII-xenograft therapy, immune-detection in pathological sections, drug delivery, and synergic treatment. In addition, mAb D2C7 [45], Erbitux and mAb 528 [46] have been confirmed to bind to both wild-type EGFR and EGFRvIII, which partly limits their applications and research value. To the best of our knowledge, no antibody has been applied to nuclear medical imaging to observe EGFRvIII expression in vivo so far. Thus, our study provides the first systematic evaluation of immunoSPECT imaging using an isotopelabeled EGFRvIII mAb ( 125 I-4G1) to detect EGFRvIII expression, which provides a new real-time approach for the detection of EGFRvIII.

Antibody preparation
A series of anti-EGFRvIII mAbs were prepared by AbMax Biotechnology Co., Ltd, (Beijing, China). Briefly, three BALB/c mice were immunized subcutaneously three times with the above three peptides at 2-week intervals. The titer of antibodies against EGFRvIII in blood samples was determined by ELISAs using the three peptide-BSA conjugates or F98 npEGFRvIII cells as the coating antigens. Next, mouse spleen cells were collected and fused with SP2/0 myeloma cells. The fused cells were cultured in 10% FBS/DMEM containing hypoxanthineaminopterin-thymidine (Sigma, St. Louis, MO, USA). Hybridoma colonies were maintained in 10% FBS/DMEM containing hypoxanthine-thymidine (Sigma). Hybridoma supernatants were repeatedly screened by ELISA. Finally, ascites were prepared by intraperitoneal injection of mice with 2 × 10 6 hybridoma cells. Ascites titers were also determined by ELISA. Furthermore, analysis of the mAb subtype was performed with standard procedures provided by the protocol of the SBA Clonotyping™ System/HRP (Southern Biotechnology Associates, Inc., Birmingham, UK). The mAb with highest titer was named 4G1.

Affinity determination
For 125 I radiolabeling, 10 µg 4G1 and 37 MBq Na 125 I (Beijing Atom High Tech, Beijing, China) in 0.2 M (pH 7.4) phosphate-buffered (PB) were added to a vial coated with 40 µg iodogen (Sigma). After 7 min of reacting at room temperature, the reaction mixture was purified by a PD-10 column (Amersham, Piscataway, NJ, USA).
The binding affinity of 125 I-4G1 for EGFRvIII was evaluated by a saturation binding assay. Briefly, Increasing concentrations of 125 I-4G1 (0-15 nmol/L) were added to F98 npEGFRvIII cells. The total volume was adjusted to 300 µL. Nonspecific binding was determined in the presence of an excess (> 500-fold) of unlabeled 4G1. After incubation at 4°C for 4 h, the cells were washed three times with cold PBS and then solubilized with 2 mol/L NaOH. The cell-associated radioactivity was determined using a γ-counter (Wallac 1470-002, PerkinElmer, Finland). A saturation binding curve and Scatchard transformation were obtained by nonlinear regression analysis, and the binding affinity (Kd) value of 125 I-4G1 was determined using GraphPad Prism 4.0 (GraphPad Software, San Diego, CA). The experiment was performed twice and each data point represents the average value from triplicate wells.
The binding specificity of 4G1 to EGFRvIII was evaluated in F98 npEGFRvIII cells by the competitive binding assay. F98 npEGFRvIII cells seeded in a 48-well plate (5 × 10 4 cells per well) were incubated at 37°C in a humidified atmosphere with 5% CO 2 overnight to allow adherence. After washing with PBS, 3.7 kBq 125 I-4G1 were added to each well in the presence of increasing concentrations of unlabeled 4G1. The cells were then incubated at 4°C for 4 h. After washed three times with cold PBS and solubilized with 2 mol/L NaOH, specific binding of 125 I-4G1 to F98 npEGFRvIII cells was analyzed by measuring the radioactivity using the γ-counter. The experiment was performed twice with triplicate samples.

Specificity determination
Cell binding assay F98 npEGFR , F98 npEGFRvIII , U87MG and U87vIII cells were trypsinized, and a single cell suspension of 1 × 10 6 cells were prepared in 1 ml of 1% BSA/PBS (w/v). Then, 3.7 kBq 125 I-4G1 was added to cell suspensions, followed by incubation at 4°C for 4 h. Nonspecific binding was determined in the presence of an excess (2 µg) of unlabeled 4G1. After washing with PBS, the specific binding of 125 I-4G1 to cells was analyzed by measuring the radioactivity using γ-counter.

Western blot analysis
Cells were harvested, washed three times with PBS, centrifuged, and then lysed. The protein concentrations of the lysates were determined using BCA Protein Assay Reagent (Pierce Chemical, Rockford, IL, USA). Total protein (50 µg) was separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred onto nitrocellulose membranes at 250 V for 2 h. Membranes were blocked for 1 h with 5% non-fat milk powder in TBS/0.1% Tween 20 and then incubated with 4G1 or rabbit anti-GAPDH polyclonal IgG (Cell Signaling Technology, Beverly, MA, USA) at 4°C overnight. The membranes were washed and incubated with a corresponding secondary antibody conjugated with Dylight 800 (EarthOx, San Francisco, CA, USA). After washing, immunoblotted proteins were detected using an Odyssey Western Blotting Detection System (Gene, Hong Kong, China).

Immunofluorescence assay
Cells grown on LabTek chamber slides or frozen sections of tumor tissues were fixed with 2% paraformaldehyde/PBS (v/v) for 15 min on ice. After blocking for 30 min with 2% BSA and 1% normal goat serum, the cells or sections were incubated with 4G1 at 4°C overnight and then the corresponding fluorescently Oncotarget 6372 www.impactjournals.com/oncotarget conjugated secondary antibody. Mouse IgG was used as an isotype control. DAPI was used for nuclear staining. Fluorescence signals were detected using a confocal microscope (TCS SP5; Leica, Germany). Immunohistochemistry was performed according to previous study [18].

Flow cytometric analysis
Cells were trypsinized, and a single cell suspension of 1 × 10 6 cells was prepared in 100 μL of 1% BSA/ PBS (w/v) and immediately stained for 2 h at room temperature with relevant primary antibodies and then incubated with the corresponding fluorescently conjugated secondary antibodies. The primary antibodies included 4G1 and Erbitux-FITC. Human IgG (Becton-Dickinson, Rutherford, NJ, USA) was used as an isotype control.

Subcutaneous and orthotopic tumor models
All animal experiments were performed in accordance with the Guidelines of the Peking University Health Science Center Animal Care and Use Committee. For the subcutaneous tumor model, cells from each cell line (2 × 10 6 ) in 100 µl PBS were injected into the right upper flanks of female BALB/c nude mice. When the tumor size reached 200-300 mm 3 , the animals were used for biodistribution and imaging studies. As for orthotopic brain tumor model, a total of 2 × 10 5 cells of each cell line in 20 µl PBS were intracranially injected into the area between the cerebral cortex and dorsal hippocampus CA3 of experimental mice. When symptoms of hydrocephalus were observed, the mice immediately underwent SPECT/CT imaging [18].

Biodistribution of 125 I-4G1
F98 npEGFR , F98 npEGFRvIII , U87MG and U87vIII tumorbearing nude mice were randomly divided into groups (n = 4 per group). Each mouse was injected intravenously with 185 kBq 125 I-4G1. Mice were euthanized, and samples or organs of interest were removed, weighed, and analyzed in the γ-counter at 24, 48, and 72 h post-injection (p.i.). The results were calculated as a percentage of the injected dose per gram of tissue (%ID/g). To determine the specific accumulation of 125 I-4G1 in U87vIII tumors, 125 I-mIgG (185 kBq) was injected intravenously into four tumorbearing mice, and then the biodistribution of 125 I-mIgG at 24 h p.i. was determined as described above.

Small animal SPECT/CT imaging
For small animal SPECT/CT imaging studies, each tumor-bearing nude mouse was injected intravenously with 55.5 MBq 125 I-4G1 via the tail vein after anesthetization. The mice were placed in the prone position and imaged with a NanoScan (SPECT/CT) camera (Mediso Ltd, Budapest, Hungary) for 40 min. Imaging was performed at 4 and 24 h p.i. After completion of imaging, the mice were euthanized.

Statistical analysis
Quantitative data are expressed as means ± SD. Means were compared using one-way analysis of variance (ANOVA) and Student'st test. P values < 0.05 were considered statistically significant.

CONCLUSIONS
In summary, we generated an anti-EGFRvIII mAb, 4G1, which shows high receptor specificity, affinity and tumor uptake. Our results suggest that the radiolabeled 4G1 mAb is a promising probe to detect EGFRvIII expression in vivo by nuclear medicine imaging.