A novel CXCR4-targeted near-infrared (NIR) fluorescent probe (Peptide R-NIR750) specifically detects CXCR4 expressing tumors

C-X-C chemokine receptor 4 (CXCR4) is over-expressed in multiple human cancers and correlates with tumor aggressiveness, poor prognosis and increased risk for distant metastases. Imaging agents for CXCR4 are thus highly desirable. We developed a novel CXCR4-targeted near-infrared (NIR) fluorescent probe (Peptide R-NIR750) conjugating the new developed CXCR4 peptidic antagonist Peptide R with the NIR fluorescent dye VivoTag-S750. Specific CXCR4 binding was obtained in cells overexpressing human CXCR4 (B16-hCXCR4 and human melanoma cells PES43), but not in CXCR4 low expressing cells (FB-1). Ex vivo evaluation demonstrated that PepR-NIR750 specifically detects B16-hCXCR4-derived subcutaneous tumors and lung metastases. Fluorescence Molecular Tomography (FMT) in vivo imaging was performed on mice carrying subcutaneous CHO and CHO-CXCR4 tumors. PepR-NIR750 accumulates only in CXCR4-positive expressing subcutaneous tumors. Additionally, an intense NIR fluorescence signal was detected in PES43-derived lung metastases of nude mice injected with PepR-NIR750 versus mice injected with VivoTag-S750. With a therapeutic intent, mice bearing PES43-derived lung metastases were treated with Peptide R. A the dramatic reduction in PES43-derived lung metastases was detected through a decrease of the PepR-NIR750 signal. PepR-NIR750 is a specific probe for non-invasive detection of human high CXCR4-expressing tumors and metastatic lesion and thus a valuable tool for cancer molecular imaging.

osteosarcoma and hepatocellular carcinoma progression through CXCR4 activation 10 . Furthermore, CXCL12 can attract CXCR4-positive immune cells or fibroblasts to the tumor sites to assist in tumor development. High CXCL12 in tumors attract CXCR4-positive inflammatory, vascular and stromal cells that support tumor by secreting growth factors, cytokines, chemokines and pro-angiogenic factors 9 . CXCR4 is also expressed on normal stem cells 11 and in prostate and pancreatic cancer progenitors 12,13 . Recently, we reported that CXCR4 and CD133 expression identified a discrete population with stem cell properties in human ovarian cancer cells that might be critical for tumor development and chemo-resistance 14 . Thus early detection of CXCR4 positive cancer cells may identify and target an aggressive cellular cancer component 15 . A new class of rationally designed CXCR4 cyclic peptide antagonists was recently developed by us. Three novel peptides impaired CXCR4 function in vitro and in vivo with peptide R being the most efficient in "in vivo" studies 16 . With the intent to develop a CXCR4-targeted NIR fluorescent imaging agent Peptide R was labelled with Vivo Tag-S 750 dye (Peptide R-NIR750). CXCR4 selectivity and sensitivity was evaluated in in vitro and in vivo CXCR4 primary and secondary tumors. Furthermore, Peptide R was developed as anti-metastatic agents and its potential of theranostic agent in cancer was demonstrated.

PepR-NIR750 specifically binds CXCR4 expressing cancer cells. Peptide R was conjugated with
Vivo-Tag S750 NIR-dye according to the manufacture instruction (see methods section) (PepR-NIR750). The ability of PepR-NIR750 to bind CXCR4 was evaluated on cancer cell lines differentially expressing the receptor: FB1, human anaplastic thyroid cancer cells, known to express low level of CXCR4; CHO, Chinese hamster ovarian cells and CHO cells transfected with human CXCR4; PES43, human melanoma cell line and B16 mouse melanoma cell lines transfected with human CXCR4 (B16-CXCR4) ( Figure S1) 17,18 . As shown in Fig. 1, PepR-NIR750 binds CXCR4 on B16-CXCR4 and PES43 cells but not on FB-1 cells. No signal was detected in the presence of the dye VivoTag-S 750 alone, demonstrating the specificity of PepR-NIR750 in visualizing CXCR4 expressing cells.
In vivo binding of CXCR4 expressing subcutaneous tumor and lung metastasis by PepR-NIR750. To evaluate PepR-NIR750 capacity to bind CXCR4 expressing cells in vivo, FB-1 cells and B16-CXCR4 cells were s.c. inoculated in CD-1 nu/nu athymic mice. When tumors reached 100 mm 3 , mice were i.v. injected with PepR-NIR750 and after 1 hour of tracer biodistribution mice were euthanized and tumors explanted. Tumor sections were stained with a-CXCR4/AlexaFluor488 and DAPI. As shown in Fig. 2,   Figure 1. PepR-NIR750 specifically binds CXCR4 expressing cells. B16-CXCR4 and PES43 tumor cell lines expressing high CXCR4 levels and FB1 cells expressing very low CXCR4 levels were plated on glass coverslips and labeled with PepR-NIR750 or VivoTag-S 750 alone (100 nM) for 1 h at 37 °C; nuclei were labeled with DAPI and the fluorescence was observed by confocal microscope (LSM 510 Zeiss). PepR-NIR750 binds to B16-CXCR4 and PES43 cells but not to FB-1 cells whereas VivoTag-S 750 signal is not detected in all cell lines tested.
To evaluate the capability of Peptide R to reduce the development of metastatic lung lesions, human melanoma cells PES43 cells were i.v. injected in nude mice and the day after treatment with Peptide R started. Peptide R (2 mg/Kg) or PBS (control) were i.p. administrated once a day for 10 days. Three weeks later mice were i.v. inoculated with PepR-NIR750 and visualized by FMT4000. As shown in Fig. 6A an intense NIR fluorescence signal was detected in lung metastases 24 hours post-injection of PepR-NIR750, while no signal was revealed in mice injected with VivoTag-S 750 alone. In addition, PepR-NIR750 signal was not detectable in healthy lungs. Peptide R treatment determined a dramatic reduction of PepR-NIR750 signal (65%) mainly detected 24 hours after injection (Fig. 6B). These evidence were confirmed by the ex vivo study conducted on lung sections (Fig. 6C). As expected, tissue analysis confirmed the presence of lung metastases (Fig. 7A) and their reduction after Peptide R treatment (Fig. 7B).

Discussion
A novel CXCR4 targeted NIR fluorescent imaging agent (Peptide R-NIR750) was described. Peptide R belongs to a new family of CXCR4 peptides developed and characterized as potent CXCR4 antagonists with anti-metastatic potential in in vivo models 16 . CXCR4 antagonist peptide binds tumor cells overexpressing the receptor; in addition CXCR4 modulation regulates the access of T effector cells to tumor microenvironment affecting the efficacy of immunotherapy 19,20 . With the intent to magnify the checkpoints inhibitors (CI) response ongoing clinical trials couple CXCR4 antagonists to CI (NCT02823405, NCT02680782, NCT02737072, NCT02826486 NCT02472977). The PepR-NIR750 CXCR4 tracer will allow to visualize in vivo the CXCR4 modulating drugs on tumor microenvironment through labeling of CXCR4 positive T and immunoregulatory cells. BM-MSCs (bone marrow-mesenchymal cells) are recruited to the tumor microenvironment contributing to build the metastatic niches. We previously showed that BM-MSCs potentiate osteosarcoma and hepatocellular carcinoma cells growth and migration and that treatment with CXCR4 antagonists, AMD3100 and Peptide R, reduced migration and invasion BM-MSCs mediated 10 . Here we show that Peptide R-NIR750 specifically binds to CXCR4 overexpressing tumor cells in vitro and in vivo and discriminates the effect of Peptide R treatment on the development of lung metastases through fluorescence molecular tomography. Fluorescence molecular tomography has been demonstrated to detect and quantify fluorescent biomarkers in sites of cancer, inflammation, and infectious disease. This technology allows few centimeters deep imaging, high sensitivity, low background and is radioactive-free 21 . A novel peptide CXCR4-targeted-NIR fluorescence imaging agent (CXCR4-IR-783) was recently described to detect primary osteosarcoma and lung micro-metastasis 22 although the authors are seeking improvements of the probe aimed to reduce the high signal-to-noise ratio and improve specificity toward neoplastic cells. Our tracer conjugates a new peptide CXCR4 antagonist belonging to a rational-designed patented new family of CXCR4 antagonists (PCT/IB2011/000120/EP 2 528 936 B1/US2013/0079292A1) to the near infrared probe. Peptide R itself has an inhibitory effect on CXCR4 coupling double effect of labeling and inhibiting the CXCR4 receptor 16 . In addition, Meinke et al. synthesized conjugates of the CXCR4 ligand, CXCL12, with the NIR fluorescent dye IRDye 800CW to visualizes breast and glioma xenografts through CXCR4 and CXCR7 targeting 23 . The unique CXCR4 antagonist clinically approved is AMD3100 indicated for hematopoietic stem cell mobilization 24 . Other receptors are also druggable to induce hematopoietic stem cell mobilization. Targeting α9β1/α4β1 integrins with a single dose of a small molecule antagonist (BOP (N-(benzenesulfonyl)-L-prolyl-L-O-(1-pyrrolidinylcarbonyl) tyrosine) rapidly mobilizes long-term multi-lineage reconstituting hematopoietic stem cells (HSC); nevertheless synergistic engraftment is observed when BOP is co-administered with AMD3100. The authors, with a related fluorescent analogue of BOP (R-BC154), showed that this class of antagonists preferentially bind human and mouse HSC and progenitors within the endosteal niche 25 . PepR-NIR750 will bind hematopioeic precursors in vivo allowing to follow and understand the mobilization process as again, Peptide R itself was demonstrated to mobilize mouse hematopoietic precursor cells (Portella L. et al. AACR 2011;number 394). We are currently evaluating Peptide R-NIR750 to study how the receptor dynamic and CXCR4 transport is regulated in the presence of CXCR4 antagonists. Moreover the recruitment and retention of progenitor cells in ischemic tissue are regulated by CXCR4 and hypoxia 26,27 ; thus tracing migration of endothelial precursors through Peptide R-NIR750 is feasible and may add insights into the CXCR4 antagonists mechanism of action. It was previously demonstrated that Peptide R inhibits recruitment of intratumoral bone marrow derived cells in B16-hCXCR4-lung metastases in CXCR4−/+ mice. A reduction in LY6G-positive myeloid/granulocytic cells and in p38 MAPK activation was detected in lungs from CXCR4(+/−) mice compared to CXCR4(+/+) mice suggesting that CXCR4 reduction on myeloid-derived cells decreased their recruitment to the lung, consequently impairing lung metastases 28 . In ongoing studies the Peptide R-NIR750 traces noninvasively BMDC migration.
In conclusion, the fluorescent probe Peptide R-NIR750 is a theranostic tool specifically targeting CXCR4 expressing cells in in vivo models. Peptide R has a therapeutic efficacy in preventing the development of lung metastases. This tool will allow pharmacodynamics and mechanistic studies on CXCR4 antagonists on neoplastic cells and immune cells shedding further insights and optimizing therapies targeting CXCR4 alone and in combination with immunomodulatory drugs.
Cell culture. Tumor cell lines were grown in appropriate medium with 10% fetal bovine serum (FBS) 2 mM glutamine, 50 mg/mL penicillin, 50 mg/mL streptomycin (complete medium) at 37 °C in 5%CO2. PES43 human melanoma cells derived from a lung metastases of melanoma patient were grown in Iscove's Modified Dulbecco's Medium (IMDM). FB-1 human anaplastic thyroid cancer cell line (kindly provided by Dr. Melillo, University of Naples, Federico II, Italy) were grown in Dulbecco's Modified Eagle's Medium (DMED). B16-CXCR4 murine melanoma cells were obtained by transfection of B16 with pYF1-fusin plasmid containing human CXCR4 gene (kindly provided by Dr Aloj, NCI "Pascale", Naples, Italy) using Fugen 6 (Roche Applied Science, Indianapolis, IN) in according to manufacturer's instruction and grown in complete medium supplemented with 100 µg/mL G418. CHO and CHO cells transfected with human CXCR4 (CHO-CXCR4) were kindly provided by Dr. David McDermott (NIH, Bethesda, USA) and grown in complete medium supplemented with 1 mg/mL G418.
Immunofluorescence microscopy. For in vitro binding experiments, cells were seeded (1 × 10 4 cells/well) on coverslips in 24-well plates in 1,5 ml growth complete medium. After five days the medium was replaced with 1 ml of fresh medium and the cells were incubated for 1 hour at 37 °C with Peptide R-VivoTag-S 750 (PepR-NIR750) conjugate or VivoTag-S 750 alone at 100 nM concentration. Cells were washed three times with PBS, and incubated in 4% BSA in this buffer. Cells were fixed with 4% paraformaldehyde for 15 min at room temperature (RT) and DAPI nuclei staining performed. The coverslips were then mounted on glass slides and fluorescence was observed by confocal microscope (LSM 510 Zeiss). CXCR4 immunofluorescence was conducted on tumor derived frozen tissues with a CXCR4 antibody clone 44716 (R&D Biosystem) and anti-mouse AlexaFluor488 (Jackson) according to manufacture instruction. Slides were imaged at fluorescence microscope Axioscope A.1 equipped with proper optical filter.  (authorization number 2013/0100808). All efforts were made to minimize animal suffering and the number of animals necessary to produce reliable results. All experimental procedures described were performed under general anesthesia with 2% isoflurane in 100% oxygen at 0.8 L/min. Subcutaneous models were realized with 3 × 10 5 B16-CXCR4 cells and 10 × 10 6 FB-1 cells, respectively, injected in the flank of CD-1 nu/nu athymic mice. Alternatively, 10 × 10 6 CHO and CHO-CXCR4 cells were injected in right and left flank of CD-1 nu/ nu athymic mice. Once tumors became palpable (established), approximately 100 mm 3 [volume = 0.5 × long diameter × (short diameter) 2 ], nude mice were i.v. injected with PepR-NIR750 (5 nM) or VivoTag-S 750 alone. B16-CXCR4 syngenic tumor model of lung metastases were realized with 3,5 × 10 5 B16-CXCR4 injected in the tail vein of 6 CD-1 nu/nu athymic mice. After 3 weeks mice were divided in 2 groups (3 mice groups), i.v. injected with PepR-NIR750 (5 nM) or VivoTag-S 750 alone and sacrificed after 1 hour. To obtain human model of lung metastases 8 CD-1 nu/nu athymic mice were injected in the tail vein with 2 × 10 6 PES43 human melanoma cells and treated with PBS or Peptide R (2 mg/Kg) 5 days/week for 2 weeks; after 3 weeks mice were i.v. injected with PepR-NIR750 or VivoTag-S 750 alone.

Mouse Tumor
Fluorescence tomography (FMT). For FMT4000 Quantitative Tomography Imaging In Vivo Imaging System (PerkinElmer, Inc.) studies, tumor-bearing mice were maintained on a diet with a purified, alfalfa-free rodent chow for 15 days before fluorescent imaging to minimize fluorescence in the gut. Mice were placed in a biplanar imaging cassette supplied with the instrument and trans illuminated with laser light. Resulting transmission and fluorescence patterns were captured with a thermoelectrically cooled CCD camera, and the position and intensity of fluorescence sources were reconstructed in 3D using the TrueQuant software package (Perkin Elmer, Inc.), supplied with the FMT4000. 3D regions of interest (ROIs) were drawn around tumor regions, and a threshold was applied equal to 30% of the maximum value of fluorescence in the adjoining non-tumor area. The total amount (pmoles) of fluorochrome was automatically calculated relative to internal standards generated with known concentrations of the appropriate dye. After imaging studies mice were euthanized. Statistical analysis. All data were presented as mean ± SD. Significance of difference was analyzed with a two-tailed Student's t test and p < 0.05 were considered statistically significant. The metastasis number was compared among two groups of mice (control group and Peptide R treated group) using Kruskal-Wallis test.