Evaluation of 99mTc-HYNIC-VCAM-1scFv as a Potential Qualitative and Semiquantitative Probe Targeting Various Tumors

Vascular cell adhesion molecule 1 (VCAM-1) is overexpressed in varieties of cancers. This study aimed to evaluate the application of a single chain variable fragment (scFv) of anti-VCAM-1 antibody labeled with 99mTc as a possible imaging agent in several tumors. VCAM-1 scFv was labeled with 99mTc using succinimidyl 6-hydrazinium nicotinate hydrochloride, and 99mTc-HYNIC-VCAM-1scFv was successfully synthesized with a high radiolabeling yield. VCAM-1 expression was evaluated in six cell lines by immunofluorescence staining. In vitro binding assays showed different binding affinities of 99mTc-HYNIC-VCAM-1scFv in different tumor cell lines, with high uptake in B16F10 melanoma and HT1080 fibrosarcoma cells, which was consistent with immunofluorescence staining results. In vivo SPECT planar imaging demonstrated that B16F10 and HT1080 tumors could be clearly visualized. Less intense uptake was observed in human SKOV3.ip ovarian tumor, and weak uptake was observed in human A375m melanoma, MDA-MB-231 breast cancer, and 786-O renal tumors. These findings were confirmed by biodistribution and immunofluorescence studies. High uptake by B16F10 tumors was inhibited by excess unlabeled VCAM-1scFv. 99mTc-HYNIC-VCAM-1scFv, which selectively binds to VCAM-1, can provide a qualitative and semiquantitative method for noninvasive evaluation of VCAM-1 expression by tumors.


Introduction
Metastasis, one of the hallmarks of malignancy, remains a significant clinical obstacle to a favorable prognosis. Early, accurate diagnosis and targeted therapy are crucial. A prognostic tumor biomarker is very helpful for the diagnosis and targeted therapy of cancers [1]. Vascular cell adhesion molecule 1 (VCAM-1) is an immunoglobulin-(Ig-) like adhesion molecule with seven extracellular Ig domains. VCAM-1 is believed to be responsible for tumor proliferation and metastasis, and its levels correlate with prognosis [2]. It is expressed by multiple types of aggressive neoplasms, including those involving lung, prostate, breast, ovaries, and colon [3]. VCAM-1 has emerged as a target for therapy of these tumors [4]. Considering the aberrant expression of VCAM-1 in tumor biology, the development of noninvasive molecular imaging for VCAM-1 is crucial for better tumor diagnosis, prognosis, and therapy planning.
Multiple new techniques for targeting VCAM-1 have been developed in the past decades, including monoclonal antibodies, nanobodies, peptides, and single chain variable fragments (scFvs) [5][6][7][8]. As we know, intact monoclonal antibodies have strong binding affinity, but their large molecular weight leads to their slow clearance from blood and poor tissue penetration into tumors. [9]. In contrast, small peptides have the advantages of prompt excretion of unbound tracer and allow prompter imaging, but at the cost of low binding affinity [10]. Given all these considerations, small antibody 2 Contrast Media & Molecular Imaging fragments of moderate size and sufficient targeting ability are becoming attractive candidates for clinical application [11].
We previously prepared the scFv of anti-VCAM-1 (VCAM-1 scFv ) using the phage display method, which is a widely used process to obtain scFvs with high specificity [12,13]. Due to the small molecular size (∼28 kDa), scFv has the advantage of rapid clearance through renal excretion, lower concentration in liver, and stronger penetration into tumor tissues [14].
The aim of this study was to explore the possibility of a noninvasive and semiquantitative method for targeting VCAM-1 in tumors, which may allow early cancer diagnosis, more precise prognosis, and targeted treatment options. In this study, we radiolabeled VCAM-1 scFv with 99m Tc using succinimidyl 6-hydrazinium nicotinate hydrochloride (SHNH) to detect levels of VCAM-1 in several tumor models in vivo.

Biodistribution Study.
For biodistribution studies, five B16F10 tumor-bearing mice were injected with 99m Tc-HYNIC-VCAM-1 scFv (1.85 MBq) via tail vein and sacrificed at 1, 2, and 4 h postinjection. For the blocking study, B16F10 tumor-bearing mice ( = 5) were sacrificed at 1 h after the injection. The biological tissues of interest (i.e., blood, brain, myocardium, liver, spleen, lung, kidney, stomach, intestine, muscle, bone, and tumor) were removed, washed, and weighed, and their radioactivity was measured with decay correction using an automatic well-type gamma counter. The uptake in the tissues was expressed as a percentage of the injected dose per gram of tissue (% ID/g). Five groups of five mice, each group grafted with one of the five other tumors, were sacrificed 4 h postinjection of 99m Tc-HYNIC-VCAM-1 scFv (1.85 MBq), and the % ID/g was calculated as described above. Portions of the tumors, livers, and kidneys in the 4 h biodistribution groups were removed to check the expression of VCAM-1 with immunofluorescence staining. The results were analyzed using ImageJ software (version 1.46r, Wayne Rasband, National Institutes of Health, Bethesda, MD, USA).

Statistical Analysis.
Statistical Package for the Social Sciences (SPSS) software (version 13.0, SPSS Inc., Chicago, IL, USA) and GraphPad Prism (version 5.0, San Diego, CA, USA) were applied in statistical analysis. All data are presented as the mean ± standard deviation (SD). Means were compared using Student's -test with < 0.05 being statistically significant.

SPECT Planar
These results indicate that 99m Tc-HYNIC-VCAM-1 scFv can specifically target VCAM-1-positive tumors. In addition, the images of the kidneys and the livers were visualized clearly, which confirmed that in vivo clearance of the probe is mainly through the renal and hepatic routes.
As shown in Figure 7, immunofluorescence staining of the kidneys and liver showed relatively low signals, indicating that these tissues did not express VCAM-1 significantly, again showing that high uptake in the kidneys and liver was unrelated to specific binding to VCAM-1 in these organs and largely attributed to the clearance of the probe. The immunofluorescence intensities of the tumor tissues (Figure 7), which were extracted from different tumor-bearing mice of the 4 h Contrast Media & Molecular Imaging

Discussion
Recently, strategies for targeting VCAM-1 [5][6][7][8], such as 18 F labeled nanobodies and 111 In labeled peptides, have been investigated. VCAM-1 expressed in atherosclerosis has been the main target of research. For the availability and low costs, 99m Tc is an ideal radionuclide for radiopharmaceutical synthesis and has been used more widely than 18 F and 111 In in clinical applications. Therefore, we used 99m Tc-HYNIC-VCAM-1 scFv to target VCAM-1 in various tumors to assess the binding affinity and characteristics of VCAM-1 scFv .  99m Tc has a half-life of 6 h, which is well matched with the relatively short physiological blood half-life of VCAM-1 scFv , making the probe's clinical translation feasible in future [15]. In addition, HYNIC-NHS, as a bifunctional chelating agent, participates in the radiolabeling of 99m Tc with peptides or antibodies, which is easily prepared, with a high radiolabeling yield, radiochemical purity, and stability [16,17]. As an alternative to PET, we sought to evaluate the feasibility of a 99m Tc-HYNIC-VCAM-1 scFv as a SPECT tracer to target VCAM-1, which is much cheaper and more easily available. In the uptake studies, six cell lines with different VCAM-1 expression levels confirmed by immunofluorescence staining showed corresponding binding affinities of the radiolabeled VCAM-1 scFv . The uptake value in B16F10 cells (VCAM-1 positive) was effectively blocked by an excess of unlabeled VCAM-1 scFv , further verifying the specificity of 99m Tc-HYNIC-VCAM-1 scFv to VCAM-1 in vitro.
The uptake pattern and blocking studies in the different cell lines closely correlated with the SPECT planar imaging and biodistribution study of xenograft models. In the imaging of B16F10 and HT1080 xenograft mice, tumors were visualized clearly from normal tissues as early as 1 h after injection of 99m Tc-HYNIC-VCAM-1 scFv . Based on the high tissue penetrability of small antibody fragments, the probe can reach the tumor site more quickly [18]. It is significantly different from tumor imaging with intact monoclonal antibodies, mainly due to slow clearance of the larger tracers from the blood pool [19]. Moderate uptake was seen in SKOV3.ip tumors, measuring 2.99 ± 0.44% ID/g at 4 h. In contrast with VCAM-1-111 In peptide distribution in omentum of SKOV3ip1 cells (about 2% ID/g), 99m Tc-HYNIC-VCAM-1 scFv has a relatively higher binding affinity with SKOV3.ip tumors [8]. No obvious tracer uptake could be seen in the imaging of A375m, MDA-MB-231, and 786-O tumor models, which are consistent with the lack of VCAM-1 expression in these tumor tissues and also confirmed by immunofluorescence staining of the tissues. High T/M ratios in B16F10 and HT1080 xenografts (8.47 ± 1.05 and 6.89 ± 0.64) were observed at 4 h postinjection, showing ideal contrast for imaging of these tumors. These results suggest the targeting ability and specificity of the probe in vivo.
As the main excretory organs, the kidneys showed the highest accumulation in the biodistribution study, which is in agreement with Broisat et al. [20]. This is due to the small  molecular weight (∼28 kDa) of scFv, which is below the cut-off for renal clearance. Immunofluorescence staining showed liver and kidneys do not express significant VCAM-1, demonstrating that the accumulations in the kidneys and livers is mainly related to the excretion mode. The immunofluorescence intensity of VCAM-1 has a relationship with the radioactivity accumulated in tumors, which indicates the feasibility of achieving semiquantitative evaluation of VCAM-1 expression in vivo.
There are various methods to detect protein expression using molecular biological techniques [21], such as immunohistochemistry, immunofluorescence, western blotting, and ELISA. Among them, ELISA is mainly applied for the detection of secretory proteins [22]. Although the specificity of western blotting is high, the procedure is complex. This research focuses on the study of immunofluorescence and radioimmunoassay, which are both based on the antigenantibody reaction. VCAM-1 expression detected by 99m Tc-HYNIC-VCAM-1 scFv relies on the binding of the radioactive antibody fragment (VCAM-1 scFv ) with the antigen. After the binding, the radioactivity uptake is measured to quantitate the antigen expression [23]. This technique is easy to perform with high sensitivity. Immunofluorescence, which makes use of labeling antibodies with fluorescent substances, has gained more and more attention due to its high sensitivity and superiority of obtaining anatomic and physiological information. However, it needs to get specimen which is invasive and hard to repeat in the living body, and it also has some shortcomings, such as inadequate penetration depth [24]. In contrast, PET and SPECT imaging with the ability to image the living human body deeply, are more advantageous in clinical applications. The results of immunofluorescence in our study are consistent with radioimmunoassay, indicating that 99m Tc-HYNIC-VCAM-1 scFv has the potential to be used to detect VCAM-1 noninvasively and repeatedly in vitro and in vivo.
There are several issues that need to be pointed out in this study. First, high activity in blood was also observed, which renders T/B ratios (1.25 ± 0.08) not ideal in VCAM-1 positive tumor models at 1 h. Fortunately, the accumulation in blood decreased rapidly (from 4.54 ± 0.13% ID/g at 1 h to 2.67 ± 0.32% ID/g at 4 h) and the T/B value increased steadily. Second, although lower accumulation was seen in liver than that with intact monoclonal antibodies [25], the liver retained high amounts of activity after 99m Tc-HYNIC-VCAM-1 scFv administration. This will interfere with imaging of lesions in 8 Contrast Media & Molecular Imaging the liver and surrounding tissues. Further studies will focus on modifying and optimizing the probe to minimize its blood and liver accumulation.

Conclusion
We successfully labeled an scFv-based probe, 99m Tc-HYNIC-VCAM-1 scFv , that specifically binds to VCAM-1. We identified different expression levels of VCAM-1 with SPECT planar imaging of corresponding tumor lesions, which potentially provides a qualitative and semiquantitative method for noninvasive evaluation of VCAM-1 expression in vivo.

Conflicts of Interest
The authors declare no potential conflicts of interest.