769-P human renal tumor cells and A549 human lung tumor cells were cultured in RMPI-1640 medium containing 10% fetal bovine serum (FBS). ACHN and A498 human renal tumor cells, HepG2 human liver tumor cells, and HEK-293 human embryonic kidney cells were grown in Dulbecco’s modified Eagle’s medium containing 10% FBS. All culture media were supplemented with 100 U/mL penicillin and 100 μg/mL streptomycin.

Enzyme-linked immunosorbent assay (ELISA) plates were coated with recombinant KIM-1 (Cloud-Clone Corp., Katy, TX) in 100 μL of a 10 μg/mL solution in 0.1 M NaHCO₃ pH 9.6 at 4°C overnight or with bovine serum albumin (BSA; Bovogen, Melbourne, Australia) as a negative control. After washing the wells six times with Tris-buffered saline (TBS) containing 0.1% Tween (TBS-T), the plates were blocked with 1% of BSA in TBS at room temperature (RT) for 1 hour. Each biopanning round consisted of subtraction of phages nonspecifically bound to BSA and subsequent selection of phages bound to KIM-1, followed by amplification of the eluted phages. In the first round, an M13 phage library (New England Biolabs, Massachusetts, UK) at a titer of 1×10¹¹ was added to BSA-coated wells and incubated at 4°C for 1 hour with gentle shaking for subtraction of phages that bound BSA. Following incubation, the unbound phages in the supernatant were collected and incubated with KIM-1–coated plates at 4°C for 1 hour with gentle shaking. Following incubation, the unbound phages were washed out using phosphate-buffered saline (PBS) containing 0.1% Tween (PBS-T). The concentration of Tween was gradually increased from 0.1% in the first round up to 0.5% in the fifth round, with a 0.1% increase in each round, in order to minimize the non-specific binding of phages. The KIM-1–bound phages were eluted by incubating the plates with 0.1 M glycine HCl (pH 3.1) at RT for 5 minutes and eluates were immediately neutralized with 1 M Tris-HCl (pH 9.1). Phage titration was performed by plaque assays. The phages were amplified using Escherichia coli ER2738 as a host and subjected to consecutive rounds of biopanning. The same amount of input was maintained in subsequent rounds.

Following five rounds of biopanning, phage clones were randomly picked and the inserted DNA was amplified by polymerase chain reaction for DNA sequencing. DNA samples of twenty individual clones were sequenced by Macrogen (Seoul, Korea). Peptide sequences corresponding to the DNA sequences were analyzed using the Clustal W program in order to align amino acid sequences and shared motifs and were also analyzed using the NCBI BLAST search to identify proteins with motifs homologous to the peptide sequences.

All peptides were synthesized by standard Fmoc method using a peptide synthesizer and purified by high-performance liquid chromatography to > 90% purity, and their mass were confirmed by matrix assisted laser desorption ionization-time of flight by Peptron Inc. (Daejeon, Korea). For fluorescence labeling, peptides were conjugated at the amino terminus with fluorescein isothiocyanate (FITC) dye, or with FPR675 near-infrared (NIR) fluorescence dye (BioActs, Incheon, Korea). The NSSSVDK peptide, a sequence present in T7 phage coat protein, was used as a control.

Cells (3×10⁴ cells) were plated on the wells of 4-well chamber slides. After overnight culture, cells were fixed with 4% paraformaldehyde at RT for 10 minutes. Following washing, the cells were incubated with an anti–KIM-1 antibody (Novus Biologicals, Littleton, CO) at a 1:200 dilution in blocking buffer at RT for 2 hours and then with an Alexa594- or Alexa488-cojugated anti-rabbit secondary antibody (Thermo Fisher Scientific, Waltham, MA) at a 1:200 dilution in blocking buffer at RT for 1 hour. Cells were incubated with 4,6-diamidino-2- phenylindole (DAPI; Sigma-Aldrich, St. Louis, MO) for nuclear staining. The chamber slides were then mounted with a ProLong anti-fade reagent (Thermo Fisher Scientific) and visualized with a fluorescent microscope (Zeiss, Oberkochen, Germany).

5. Phage protein- and cell-binding ELISA

For phage protein-binding ELISA, plates were coated with 100 μL of 10 μg/mL KIM-1 or BSA in 0.1 M NaHCO₃ pH 9.6 at 4°C overnight. The plates were blocked with 1% BSA in TBS at RT for 1 hour. After washing the plates with TBS-T, the protein was incubated with 100 μL of each phage clone (1×108 plaque forming unit [pfu]/well) at RT for 1 hour. For phage cell-binding ELISA, cells (1×10⁴ cells) were seeded in the wells of 96-well plates and allowed to grow overnight. Cells were blocked with culture medium containing 1% BSA at RT for 30 minutes. After washing the plates with PBS, cells were incubated with each phage clone (1×10⁹ pfu/well) at 4°C for 1 hour, washed, and then incubated with the horseradish peroxidase-conjugated anti-M13 antibody (GE Healthhcare, Chicago, IL) diluted 1:3,000 in blocking buffer at 4°C for 1 hour. After washing the plates, 3,3',5,5'-tetramethylbenzidine substrate (Pierce, Rockland, IL) was added and incubated at RT for 10 minutes. The reaction was stopped using 2 M H₂SO₄ and plates were read at 450 nm using a microplate reader (Tecan, Zurich, Switzerland).

Cells were grown to 70% to 80% confluence, harvested, and suspended in a culture medium. Cells were blocked with culture medium containing 1% BSA at RT for 30 minutes and then incubated with 10 μM of a FITC-conjugated peptide in blocking buffer at 4°C for 1 hour. After washing, cells were subjected to flow cytometry (BD Bioscience, San Jose, CA).

To form the NeutrAvidin/biotin-peptide multimer, FITCconjugated NeutrAvidin (Thermo Fisher Scientific) at a concentration of 82.7 μM was incubated with biotin-conjugated peptide at a concentration of 330 μM at RT for 1 hour with rotation. The mixture was then dialyzed using a 15 kDa Tube-O-Dialyzer (G-biosciences, St. Louis, MO) in PBS at 4°C overnight. For immunofluorescence analysis, 3×10⁴ cells were plated onto the wells of 4-well chamber slides. Following overnight incubation, cells were blocked with culture medium containing 1% BSA at RT for 30 minutes and then incubated with 10 μM of the NeutrAvidin/biotin-peptide multimer in blocking buffer at 4°C for 1 hour. Cells were fixed with 4% paraformaldehyde at RT for 10 minutes. For co-localization, cells were incubated with the anti–KIM-1 antibody at a 1:200 dilution in blocking buffer at RT for 2 hours and then incubated with an Alexa594-conjugated antirabbit secondary antibody at a 1:200 dilution in blocking buffer at RT for 1 hour. Cells were incubated with DAPI for nuclear staining and with an anti-fade reagent prior to observation using a fluorescence microscope (Zeiss). For flow cytometry analysis, cells were harvested, blocked, and then incubated with 10 μM of the NeutrAvidin/biotin-peptide multimer in blocking buffer at 4°C for 1 hour. For competition of binding, different concentrations of the anti–KIM-1 antibody were pre-incubated at RT for 2 hours prior to incubation with the NeutrAvidin/biotin-peptide multimer. Following incubation, cells were subjected to flow cytometry (BD Bioscience).

Black colored microplate wells (SPL Life Science, Pocheon, Korea) were coated with 100 μL of 10 μg/mL BSA or KIM-1 in 0.1 M NaHCO₃, pH 9.6 and incubated at 4°C, overnight. The plates were blocked at RT for 1 hour with a blocking buffer (1% BSA in PBS) and incubated with 100 μL of 5 μM FITC-NeutrAvidin/biotin-peptide multimer in blocking buffer at RT for 1 hour. After washing the plates, fluorescence intensity was measured using a fluorometer.

Mouse blood was collected and allowed to clot. Serum was obtained by centrifuging the blood clots twice at 4°C followed by filtration through a 0.22 μm filter. Each peptide (100 μg in 50 μL of PBS) was incubated with 50 μL of filtered serum at 37°C for the indicated time periods. The incubated samples were diluted 100-fold and fractionated by C18 reverse phase fast performance liquid chromatography with a linear gradient of acetonitrile. The peptide peak was collected and analyzed.

Cells (1×10⁴ cells) were seeded in the wells of 96-well plates and allowed to grow overnight. Cells were incubated with the indicated concentration of peptide at 37°C for 24 hours. Cells were then incubated with 10 μL of the CCK-8 dye (Dojindo, Kumamoto, Japan) mixed with 100 μL of culture medium. Absorbance was measured with a microplate reader at 450 nm.

Tumor cells (5×10⁶ cells) were subcutaneously injected into the right flank of 6-week-old BALB/c nu/nu female mice. Tumor-bearing mice were injected via the tail vein with 20 nmol of flamma675 NIR dye (BioActs, Incheon, Korea)-conjugated peptide. In vivo fluorescence images were obtained using an IVIS imaging system (Perkin Elmer, Waltham, MA) at 2, 4, 8, and 24 hours after injection. At the end of in vivo imaging, tumor and other organs were isolated and ex vivo fluorescence images were taken. After taking the ex vivo images, the tumors and organs were fixed with 4% paraformaldehyde for 2 hours and frozen for cryosection. Tissue slides were incubated with the anti–KIM-1 antibody (1:200 dilution) at RT for 2 hours and with the Alexa594 anti-rabbit secondary antibody (1:200 dilution) at RT for 1 hour. Tissues were mounted using mounting medium containing DAPI for nuclear staining and observed with a fluorescence microscope.

Statistical analysis was performed by Student’s t-test and a p-value of < 0.05 was considered as statistically significant.

All animal experiments were approved by the Kyungpook National University Animal Experiment Ethics Committee.

An M13 phage library displaying CX7C random peptides was screened in vitro to identify phage clones that bind to KIM-1 protein-coated plates. The overall screening scheme is shown in Fig. 1A. Following five rounds of screening, the phage titer of the fifth round was enriched approximately 10-fold compared to that of the first round (Fig. 1B). Twenty phage clones were randomly picked from the third, fourth, and fifth rounds and the peptide-coding DNA inserts of the phage clones were sequenced (Table 1). Four phage clones displaying the CNWMINKEC, CVPSKPGLC, CPKGDENTC, and CPTSQRDNC sequences were chosen for further evaluation, as they occurred more frequently than other phage clones.

The binding activity of individual phage clones to the KIM-1 protein was examined by phage-binding ELISA. Compared to the other phage clones, the CNWMINKEC peptide phage clone showed higher levels of binding to the KIM-1 protein relative to BSA (Fig. 2A). Next, we examined the binding of the phage clones to KIM-1–overexpressing cells. Immunofluorescence analysis showed that 769-P renal cancer cells and A549 lung cancer cells expressed higher levels of KIM-1 on the cell surface than HEK-293 normal cells (Fig. 2B). The CNWMINKEC-phage clone and CVPSKPGLC-phage clone demonstrated higher levels of binding to KIM-1–overexpressing 769-P cells than to KIM-1–low expressing HEK-293 cells compared to the other phage clones and the control M13 insertless phage clone (Fig. 2C).

To examine the binding of the CNWMINKEC and CVPSKPGLC peptides to the KIM-1 protein and KIM-1–overexpressing cells, the peptides were synthesized and conjugated with a FITC green fluorescent dye at the amino terminus. Analysis of peptide binding using a fluorometer showed that the CNWMINKEC peptide bound to the KIM-1 protein at higher levels than to BSA (Fig. 3A). The binding of the CVPSKPGLC and control peptides to the KIM-1 protein were similar to their binding to BSA and were much lower than the CNWMINKEC peptide (Fig. 3A). Flow cytometry analysis demonstrated that the percent binding of the CNWMINKEC peptide to 769-P cells was approximately four-fold higher than to HEK-293 cells (21% and 5%, respectively) (Fig. 3B), while the percent binding of CVPSKPGLC to 769-P cells was similar to the binding of HEK-293 cells (10% and 8%, respectively) (Fig. 3B and C). In addition, the percent binding of CNWMINKEC to A549 lung tumor cells was approximately three-fold higher than to HEK-293 cells (approximately 20% and 6%, respectively) (Fig. 3D and E). These results indicate that the CNWMINKEC peptide preferentially binds to the KIM-1 protein and KIM-1–overexpressing cells compared to albumin and KIM-1–low expressing cells, respectively.

Biotin-labeled CNWMINKEC peptide was incubated with NeutrAvidin to form a multimeric complex using the biotinavidin interaction. The binding of the NeutrAvidin/biotin-CNWMINKEC multimer to KIM-1 protein was higher than to the BSA control, while NeutrAvidin itself showed only minimal levels of binding to both proteins (Fig. 4A). Immunofluorescence analysis demonstrated that the binding of the NeutrAvidin/biotin-CNWMINKEC multimer to 769-P cells was higher than to HEK-293 cells, while the binding of NeutrAvidin to 769-P cells was negligible (Fig. 4B). The percent binding of the NeutrAvidin/biotin-CNWMINKEC multimer to 769-P cells was approximately four-fold higher than to HEK-293 cells, as determined by flow cytometry (38% and 9%, respectively) (Fig. 4C). The NeutrAvidin/biotin-CNWMINKEC multimer co-localized with KIM-1 on 769-P cells, as observed by confocal fluorescence microscopy following co-staining with the anti–KIM-1 antibody (Fig. 4D). Furthermore, pre-treatment of 769-P cells with the anti–KIM-1 antibody reduced the subsequent binding of the NeutrAvidin/biotin-CNWMINKEC multimer in a dose-dependent manner, while minimal effect was observed in HEK-293 cells (Fig. 4E). These results suggest that the binding of CNWMINKEC to 769-P cells is specifically mediated by KIM-1.

To examine the non-specific cytotoxicity of the CNWMINKEC peptide, 769-P and HEK-293 cells were incubated with different concentrations of the CNWMINKEC peptide for 24 hours. Treatment of the cells with the CNWMINKEC peptide did not significantly affect cell viability (Fig. 5A), suggesting that the CNWMINKEC peptide itself has no cytotoxicity. To examine the serum stability of the CNWMINKEC peptide, it was incubated with mouse serum for up to 24 hours and the amount of the peptide present over a period of time was analyzed. The peptide peak was separable from the non-specific serum peaks and was not significantly reduced up to 24 hours (Fig. 5B). These results suggest that the CNWMINKEC peptide is stable in the serum for up to 24 hours.

As 769-P cells do not form tumors in mice, we used A498 cells, another renal tumor cell line, as a KIM-1–overexpressing tumor model. In addition, we used HepG2 liver tumor cells as a KIM-1–low expressing tumor model. Immunofluorescence analysis showed that A498 and ACHN renal tumor cells highly express KIM-1, while HepG2 cells little express KIM-1 (Fig. 6A). The percent binding of the CNWMINKEC peptide to A498 and ACHN cells was approximately six-fold higher than that to HepG2 cells (23% and 26% vs. 4%) (Fig. 6B). Mice bearing A498 or HepG2 tumor were intravenously injected with the flamma675 NIR fluorescence dye-labelled CNWMINKEC and control peptide. Whole body fluorescence imaging showed enhanced fluorescence signals at KIM-1–overexpressing A498 tumor by the homing and accumulation of the CNWMINKEC peptide at 2-, 4-, and 8-hour post injection with a decline at 24 hours, while minimal signals were detected at A498 tumor by the control peptide (Fig. 6C). Only minimal or background signals were detected with either the CNWMINKEC or control peptide at KIM-1–low expressing HepG2 tumor (Fig. 6C). Ex vivo fluorescence images of isolated organs further demonstrated higher levels of fluorescence signals at A498 tumor in mice injected with the CNWMINKEC peptide compared to control peptide, whereas little signals were detected at HepG2 tumor (Fig. 6D and E). Compared to A498 tumor, much lower levels of fluorescence signals were observed at the liver, lungs, and kidneys of mice injected with either the CNWMINKEC or control peptide (Fig. 6D and E), suggesting selective homing of the CNWMINKEC peptide to KIM-1–overexpressing tumor. In addition, immunohistochemical examination of tissues demonstrated that the CNWMINKEC peptide co-localized with KIM-1 at tumor tissues, as detected by staining with an anti–KIM-1 antibody (Fig. 6F). Accumulation of either the CNWMINKEC or control peptide in kidney tissues was not detected (Fig. 6F), suggesting that the ex vivo fluorescence signals at the kidney were not due to accumulation in the kidney, but rather to urinary excretion of the peptides.