¹²⁵I–F56 Peptide as Radioanalysis Agent Targeting VEGFR1 in Mice Xenografted with Human Gastric Tumor

Abstract

¹²⁵I-Radiolabeled F56 peptide was designed as a radioactive analogue of F56 (peptide WHSDMEWWYLLG) to bind with VEGFR1 receptor. It was synthesized in high radiochemical yield and specific activity. The in vitro stability of ¹²⁵I–F56 was tested, and the bioactivity of ¹²⁵I–F56 was confirmed by both cell uptake and binding affinity measurement in VEGFR1 positive BGC-823 cells. The time–radioactivity relationship and biodistribution of ¹²⁵I–F56 tracer were conducted using nude mice bearing human gastric carcinoma BGC-823, by noninvasive micro-SPECT/CT imaging. The tracer’s tumor uptake was further confirmed by autoradiography and HE stain of ¹²⁵I–F56 in tumor tissues ex vivo. Those results demonstrated that ¹²⁵I–F56 holds great potential as a diagnostic agent in both molecular imaging and radioanalysis probe for gastric cancer.

Gastric cancer (GCa) is the most prevalent type of cancer and the second leading cause of cancer-related mortality in Asian countries.^1,2 Early detection of GCa remains a great challenge due to the limit of clinical symptoms. Almost 90% of GCa patients were diagnosed in advanced stage, and even worse, the 5-year survival rate of those patients was less than 20%.

Early detection of recurrent and metastasis lesions can provide critical clinical value for GCa patients. Up to now, proactive intervention by surgical resection is still the predominate treatment method for GCa. After surgery, there are always still small lesions left. In addition, based on the 2016 National Comprehensive Cancer Networks (NCCN) guidelines (version 3.0), clinical stage of GCa relies on an extensive diagnostic modalities, such as endoscopic ultrasound (EUS), computed tomography (CT), positron emission tomography/computed tomography (PET/CT), magnetic resonance imaging (MRI), and laparoscopic, respectively.³ The diagnostic accuracy of EUS is always operator dependent, ranging from 57% to 88% for T staging and 30% to 90% for N staging.⁴ Ultrasonography or CT-guided needle biopsy is invasive with uncontrollable risk.

Noninvasive molecular imaging can provide in vivo visualization and measurement of biological processes. ¹⁸F-Fluorodeoxyglucose (¹⁸F-FDG) is the typical PET probe, which has been extensively applied for the diagnosis of diseases, especially in tumors.⁵ However, the limited sensitivity and specificity of ¹⁸F-FDG may lead to confounding diagnosis.⁶¹⁸F-FDG PET is reported to has low detection rate (47%) in diffused and mucinous tumor types, which are frequent in GCa, comparing with the EUS, because of low tracer accumulation.⁷

Radionuclide imaging techniques, including PET and single photon emission computed tomography (SPECT), are critical in the clinical diagnosis and treatment of GCa. VEGFR 1 is confirmed to have overexpressed in GCa and also in lots of advanced cancers.^8,9 Up to now, several PET tracers targeting VEGFRs have been designed by mimic the natural/mutate VEGF isoforms. We previously reported a novel VEGFR1 specific 12-mer F56 peptide, which was isolated from phage display screening technology, and serves as an effective competitor against the VEGF binding to VEGFR1.¹⁰ The VEGFR1 specific radiotracer ⁶⁴Cu-DOTA-F56 has also been designed and evaluated for noninvasive GCa PET imaging.¹¹

The advanced radioanalysis technique that employs a radiolabeled protein or peptide that can specifically target a tumor antigen gives great clinical convenience.¹² This approach gives a full view for the detection of tumor recurrence and metastasis. In the present study, we designed ¹²⁵I–F56 and evaluated its application in human gastric cancer cell line BGC-823 in vitro and in vivo. Those results warrant the further application of ¹²⁵I–F56 in advanced gastric tumor radioanalysis agent.

The radiolabeling of F56 by ¹²⁵I was performed as reported with some modification (Scheme 1).¹³

Scheme 1. Schematic Illustration of ¹²⁵I-Radiolabeling of F56 Peptide by Iodogen Vessel.

The blue dotted circle indicates the reactive amino acids where radioactive ¹²⁵I may be attached.

The radiolabeled conjugate, ¹²⁵I–F56, was synthesized by the reaction of F56 peptide with [¹²⁵I]NaI at room temperature in 15 min. The radiolabeling yield of ¹²⁵I was generally over 90%. After purification by C18 sep-pak column, the radiochemical purity of ¹²⁵I–F56 was >94% with the retention time at 8.5 min, which was different from that of F56 at 7.8 min (Figure 1). The specific activity of ¹²⁵I–F56 up to 5 GBq per μmol was obtained, which was calculated from a previously reported method.¹¹

Figure 1.

HPLC chromatograms of ¹²⁵I–F56 performed with an RP-18 column. HPLC analysis revealed that the radiochemical purity of ¹²⁵I–F56 was 94%. (A) UV detection showed that the chromatographic peak of unreacted F56 peptide appeared at 7.8 min. (B) Radiation detection showed that the radiation peak of ¹²⁵I–F56 appeared at 8.5 min.

¹²⁵I–F56 peptide showed excellent in vitro stability in phosphate-buffered saline (PBS) and sodium acetate (NaOAc) at 37 °C up to 48 h, and >90% of the conjugates retained their original structures. It also had excellent resistance to proteolysis and transchelation in 5.0% human serum albumin (HSA) with >90% of the probe intact after 4 h incubation, and with >80% after 24 h incubation, as monitored by radio-HPLC at the retention time of 8.6 min. The log P value of ¹²⁵I–F56 was determined to be 2.92 ± 0.11 from the octanol–water partition coefficient measurements, indicating high lipophilicity of the radiolabeled peptide.

The cell uptake of ¹²⁵I–F56 in BGC-823 cells increased over time, and the uptake was blocked using excess cold F56 (P < 0.01, Figure 2). The cell uptake values of ¹²⁵I–F56 at 10, 30, and 60 min at 37 °C were 2.04 ± 0.56%, 2.83 ± 0.88%, and 5.84 ± 0.90%, respectively. When blocked, the uptake was 3.46 ± 0.26% at 60 min. The 68% uptake reduction of [¹²⁵I–F56] by the block of F56 indicates that the two ligands occupy the same binding site at cell surface, and the binding sites are saturable (P < 0.01).

Figure 2.

In vitro BGC-823 cell uptake. The cell uptake data were expressed as the percentage of the radioactivity measured with respect to the total radioactive compound used. **P < 0.01 from student’s t test.

Saturation studies demonstrated that ¹²⁵I–F56 binds specifically and with nanomolar affinity to BGC-823 cells (Figure 3A, Figure S1). After fitting the curve with an one site saturable model, the B_max was 423 fmol per 6000 BGC 823 cells, and binding constant, K_d, was 66 nM for specific binding of ¹²⁵I–F56.

Figure 3.

Binding affinity assay of ¹²⁵I–F56 to BGC-823 cells. (A) BGC-823 cells were exposed to increasing concentrations of ¹²⁵I–F56 peptide for 4 h at 37 °C, showing saturation binding. The binding constant was 66 nM, according to a one-site binding model. (B) BGC-823 cells were incubated with increasing concentrations of F56 peptide for 2 h at 37 °C, and the F56 peptide concentration was determined by the antibody against F56 peptide, but conjugated HRP label, measured at 490 nm absorbance.

The binding of F56 peptide to BGC-823 cells with cell ELISA assay also showed one saturable bond (Figure 3B). This is consistent with our previously study, where we demonstrated the binding of F56 peptide to BGC-823 cells and, specifically, VEGFR1 with immunofluorescence and immunohistochemistry assay.^11,14

The time–radioactivity relationship of ¹²⁵I–F56 in mouse model showed rapid blood clearance (Figure 4), with 4.22% ID%/g at 1 h postinjection of 1.11 MBq ¹²⁵I–F56, 1.63% ID/g at 4 h, and 0.07% ID/g at 24 h postinjection, respectively. The biodistribution within a mouse showed that the kidney had relatively high uptake, with 2.81% ID/g at 4 h and 3.33% ID/g at 24 h. The uptake of ¹²⁵I–F56 displayed tumor accumulation, with 2.37 ± 0.1% ID/g at 1 h and 2.15 ± 0.2% ID/g at 4 h. With the metabolism of ¹²⁵I–F56, the tumor uptake was reduced to 0.45 ± 0.15%ID/g at 24 h postinjection. As a result, radioactivity ratios of tumor to nontumor tissues (T/NT) increased markedly in a time-dependent manner.

Figure 4.

Biodistribution of ¹²⁵I–F56 in mice xenografted with BGC-823 gastric cancer cells.

The tumor-targeting efficacy of ¹²⁵I–F56 was further evaluated by static images of micro-SPECT scans in a mouse model bearing BGC-823 tumor. Representative decay-corrected three-dimensional images including coronal (left) and sagittal (middle) SPECT/CT imaging obtained at 4 h postinjection are shown in Figure 5. The tumor location was clearly visible. The uptake of ¹²⁵I–F56 in kidney was at the highest, but that in the tumor, the second. No radioactivity uptake was observed in the region of thyroid, due to in vivo stability of ¹²⁵I–F56. The coronal (right in Figure 5) SPECT image obtained at 5 h postcoinjection with 0.1 mgof F56 peptide as block agent was also shown in Figure 5. It showed that the signal in tumor tissues was greatly decreased by blocking, but the kidney uptake remained the same. Those results indicate that the coadministration of excess F56 peptide can block the ¹²⁵I–F56 probe in mice tumor model specifically.

Figure 5.

Micro-SPECT images of ¹²⁵I–F56 in human gastric cancer cell BGC-823 bearing nude mice at 4 h postinjection and 5 h postcoinjection with 0.1 mg of F56 peptide as block agent. Left, coronal imaging; middle, sagittal imaging; right, coronal imaging of blocked group. The arrow indicates the location of the tumor.

Better contrast images of ¹²⁵I–F56 and block were obtained by maximum intensity projection (MIP) image (Figure 6). In the left of Figure 6, the bone system can be clearly visible, and the tumor and kidney still showed the highest uptakes of ¹²⁵I–F56. In the right of Figure 6, tumor uptake were greatly decreased by excess F56 peptide. The micro-SPECT/CT images in both static picture and MIP view were consistent with each other. The autoradiography analysis was performed as previously reported with some modification.¹⁵ The autoradiography of ¹²⁵I–F56 in tumor tissues confirmed the tracer’s high uptake in tumor tissues (Figure 6).

Figure 6.

MIP Micro-SPECT/CT imaging of ¹²⁵I–F56 (left) and SPECT images of ¹²⁵I–F56 coadministration with 0.1 mg of F56 peptide as block agent (right) in human gastric cancer cell BGC-823 bearing nude mice. P means prone position. The red arrow indicates the location of the tumor. The white dotted square (and oval) shows the position and the extension of the tumor, which were further examined by autoradiography imaging in dissected tumor tissue.

Monitoring of VEGFR1 status in GCa is not only possible but also necessary to avoid or prevent disease relapse. The present micro-SPECT/CT imaging study of ¹²⁵I–F56 and coadministration of F56 peptide block agent in mice xenografted with BGC-823 tumor demonstrated that the technique can give both functional and anatomic information from organism, if it is used with high binding radiotracer at nanomolar range.¹⁶ Recent studies revealed that overexpression of VEGFR1 may cause poor disease outcome in various cancer types, including GCa.¹⁷⁻²⁰ Furthermore, there is also a consistent association between high VEGFR1 level and the relapse of surgically treated patients.¹⁷

After dissecting the tumor tissue, HE staining was preformed to show different structures of the tissue. To stain the microvasculature, the sections were incubated with an antibody against CD31 or Factor VIII antibody. We can see that the vessel was rare (Figure S2). We have demonstrated the binding of F56 peptide to VEGFR1 positive BGC-823 cells in vitro and in vivo.¹⁵ These findings suggest that F56-based radiotracer binds not only to new blood vessels but also to tumor cells that express VEGFR1.

In summary, this study was designed to evaluate radioanalysis potential of ¹²⁵I–F56 as a molecular probe for VEGFR1 overexpressed gastric cancer. The radiolabeling was achieved using Iodogen method, obtaining the radiotracer in both high radiolabeling yield and suitable specific activity. ¹²⁵I–F56 exhibited good in vitro stability and cell binding affinity. Micro-SPECT/CT image of ¹²⁵I–F56 showed clear imaging of tumor in mice xenografted with BGC-823 cells, and this uptake also can be blocked by excess cold F56 peptide. The tumor imaging was further confirmed by ex vivo autoradiography. To our knowledge, ¹²⁵I–F56 is the first peptide based VEGFR1 specific radiotracer for tumor-targeting radioanalysis. All of these results demonstrated that ¹²⁵I–F56 peptide has potential for radioanalysis agent guided surgery in advanced gastric cancer.

Glossary

ABBREVIATIONS

HE stain

hematoxylin and eosin stain

VEGFR1

vascular endothelial growth factor receptor 1

GCa

gastric cancer

SPECT

single photo emission computed tomography

HPLC

high performance liquid chromatographic

ELISA

enzyme linked immunosorbent assay

HRP

horseradish peroxidase

Supporting Information Available

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsmedchemlett.6b00498.

• Details regarding materials and methods, chemistry, radiochemistry, octanol–water partition coefficient, in vitro stability study, animal models, microSPECT/CT, and autoradiography protocols (PDF)

Author Contributions

^§ These authors contributed equally. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

This work was supported by Major Project on Drug Research and Development for the 12th Five-Year Plan of China (2012ZX09103-301-013), the National 973 Program of China (2015CB553906), the National Natural Science Foundation of China (No. 81371592, 81401467, 81501519, 81571705, 81301966, and 81571705), Beijing Natural Science Foundation (No. 7154188 and 7162041), and Beijing Municipal Commission of Health and Family Planning (215 backbone program).

The authors declare no competing financial interest.