The lack of molecular targets and targeting probes remains a major drawback for targeted imaging and drug delivery in lung cancer. In this study, we exploited in vivo phage display to identify a novel targeting probe that homes to the tumor in a K-rasLA2 mutant mouse lung cancer model. Compared with other candidate peptides selected from 5 rounds of phage display, the CRQTKN peptide homed to tumor nodules in the lung of mutant mice at higher levels. Photoacoustic tomography of mutant mice detected lung tumors via tumor homing of the near-infrared fluorescence dye-labeled CRQTKN peptide. Ex vivo photoacoustic images of isolated organs further demonstrated tumor homing of the CRQTKN peptide, whereas minimal accumulation was observed in control organs, such as the liver. Compared with untargeted liposomes and doxorubicin, doxorubicin-loaded liposomes whose surface was modified with the CRQTKN peptide more efficiently delivered doxorubicin and reduced the number or size of tumor lesions in K-rasLA2 mutant mice. Analysis of hematologic parameters and liver and kidney function showed no significant systemic side effects by the treatments. Affinity-based identification was used to detect TNF receptor superfamily member 19L (TNFRSF19L), which was upregulated in lung tumors of mutant mice, as the receptor for the CRQTKN peptide. In conclusion, these results suggest that the CRQTKN peptide is a promising targeting probe for photoacoustic-guided detection and drug delivery to lung cancer, and acts by binding to TNFRSF19L.

Significance:

These findings present a new tumor-targeting probe for photoacoustic-guided detection and drug delivery.

Lung cancer is the most common cause of cancer-related mortality, and the 5-year survival rate is only approximately 18% (1). Currently, nonsurgical treatment of non–small cell lung cancer, which accounts for 80% to 85% of all cases of lung cancer, includes chemotherapy, radiation therapy, and targeted therapy. Chemotherapy with cisplatin and paclitaxel has been used to effectively inhibit tumor growth, but its systemic side effects on normal tissues including the bone marrow and intestine have limited its use. Targeted therapy with gefitinib, an epidermal growth factor receptor kinase inhibitor, has been used to treat lung cancer with the L858R activation mutation with less systemic side effects. However, it is frequently accompanied by the T790M mutation that causes an adaptive resistance to the gefitinib treatment and thus may require other targeted drugs that can act on T790M mutant cells such as osimertinib (2). Aside from adaptive resistance to targeted therapeutic agents, the heterogeneity of the tumors requires multiple targeted therapeutics for an efficient cancer therapy, which eventually results in an increased cost for therapy (3).

Targeted delivery of conventional chemotherapeutic drugs to the tumor while saving normal tissues could be an approach to overcome these issues. It can increase the therapeutic efficacy of drugs even if given at a dose similar with the conventional dose while decreasing the systemic side effects. In addition, chemotherapeutic drugs are generally less expensive than targeted drugs. However, despite growing understanding of the biology of lung cancer, the lack of well-established molecular targets and targeting probes for lung cancer limits such a targeted delivery of chemotherapeutic drugs. Thus, novel target molecules and probes for targeted imaging (or detection) and drug delivery to lung cancer need to be identified.

Photoacoustic tomography (PAT) is a premier volumetric imaging modality that can provide strong optical contrast and high ultrasonic resolution in live animals and humans in vivo (4, 5). Particularly, molecular PAT using targeted exogenous contrast agents has become increasingly important to delineate the diseased areas with molecular specificities. In addition, photoacoustic-guided drug delivery has been widely used in neuroscience, cancer physiopathology, and pharmacology (6–9).

Peptides are suitable for deep tissue penetration due to their small size and are easy to be chemically conjugated with therapeutic drugs and imaging agents (10, 11). These features enable them to be useful tumor-homing probes for targeted drug delivery and in vivo imaging of cancer (12–14). Many tumor-homing peptides have been identified by phage display (15, 16). Unlike in vitro phage display, in vivo phage display is performed via systemic administration of phage library into animal models, and its main advantage is that the binding condition to target molecules or cells is closer to tumor microenvironment (17–19). In addition to targeting probes, peptides are also used as a ligand for the identification of the corresponding receptor, which may be a novel molecular target of tumors. For example, the NGR peptide homes to the tumor by binding to aminopeptidase N in angiogenic tumor blood vessels and tumors (20). The CRGRRST tumor-homing peptide identified via in vivo phage display binds to platelet-derived growth factor receptor-β in transgenic pancreatic cancer mouse model (19). The CRKRLDRNC peptide homes to the tumor by binding to interleukin-4 receptor overexpressed in tumors (14). Using an affinity-based purification, we have previously reported that the CQRPPR peptide recognizes apoptotic cells by binding to histone H1 as a novel target exposed on apoptotic cell surface (21).

This study aimed to identify a novel targeting probe that homes to lung tumor in vivo. Herein, we exploited in vivo phage display of peptide library in the K-rasLA2 mutant mouse lung cancer model. This model has K-ras oncogene activation (G12D mutation in exon 1) that is frequently seen in human lung cancer and shows a pathologic similarity with human lung cancer (22, 23).

K-rasLA2 mutant mouse model of lung cancer

All animal experiments were performed in compliance with institutional guidelines approved by the Institutional Animal Care and Use Committee of Kyungpook National University (permission number, KNU 2014-27). K-rasLA2 mutant mice were provided by the NIH (Bethesda, MD). Although mutations in K-rasLA2 gene are common in human pancreatic and colon cancers, this K-rasLA2 transgenic mice show a high incidence of lung cancer, whereas they do not develop pancreatic and colon cancers (22). K-rasLA2 mutant mice were mated with wild-type C57BL/6N mice. For genotyping, complementary DNA was purified via Qiaquick columns (Qiagen) and amplified via PCR with an annealing temperature of 60°C for 35 rounds. The PCR primers used were as follows: K1/K2 for wild-type K-ras and K1/K3 for mutant K-rasLA2 (K1: TGC ACA GCT TAG TGA GAC CC, K2: GAC TGC TCT CTT TCA CCT CC, K3: GGA GCA AAG CTG CTA TTG GC).

In vivo phage display for selection of tumor-homing peptides

A T7 phage library displaying CX7C peptides (Merck), with a diversity of 5 × 108 plaque-forming unit (pfu), was used. K-rasLA2 mutant mice bearing lung tumors (18–20 weeks old) were intravenously injected with a solution containing phages at a total of 1 × 1011 pfu. After 15 minutes of circulation, mice were sacrificed and tumor nodules from lungs were excised and weighed. Cell suspension of tumor nodules was prepared in culture media by chopping with a knife, homogenizing using Medimachine (DAKO), and passing through a 70 μm cell strainer. The cell-bound phages were eluted by lysing cells with 100 μL of 1% NP-40 for 10 minutes on ice and then by adding BL21 host bacteria. Phages were amplified in the host bacteria and used for next round of selection. After 5 rounds of selection, individual phage clones were randomly picked, and their DNA inserts were amplified by PCR and sequenced.

Peptide synthesis

All peptides were synthesized via standard Fmoc method using a peptide synthesizer (Peptron Inc.). Peptides were purified via high-performance liquid chromatography (HPLC) to >90% purity (Supplementary Fig. S1A), and their mass (M.W. 748.28) was confirmed via MALDI-TOF (Supplementary Fig. S1B). For fluorescence labeling, peptides were site-specifically conjugated at the amino terminus with fluorescein isothiocyanate (FITC) dye or with FPR675 or FPI774 near-infrared (NIR) fluorescence dye by a dye company (BioActs). The NSSSVDK peptide, a sequence present in the coat protein of T7 phage, was used as a control.

In vivo and ex vivo PAT

The output laser beam from an Nd:YAG laser (SLIII-10; Continuum; 532 nm) pumped a tunable optical parabolic oscillator laser (Surelite OPO PLUS; Continuum; wavelength tuning range, 680–2,500 nm) to produce a tunable optical light beam. The pulse repetition rate of the pulsed laser was 10 Hz, and the wavelengths of 774 and 870 nm were mainly used for spectroscopic photoacoustic (PA) imaging. Captured PA signals were amplified through an amplifier (5072PR; Olympus) and were acquired using a mixed-signal oscilloscope (MSO5024; Tektronix). Upon one laser shot, one-dimensional depth-resolved images were formed (called A-line). By moving the system along the x-axis, two-dimensional depth sensitive images were acquired (called B-mode). Additional laser scanning along the y-axis allowed us to obtain 3-dimensional images. The image acquisition time for one volumetric image was 42 minutes, with a field of view of 6 cm × 4 cm along the x and y axes. The acquired 3-dimensional images were processed through a maximum amplitude projection, which projected the maximum amplitude along 1 axis data onto the corresponding imaging plane (i.e., the imaging plane standing on the other 2 axes). The maximum amplitude projection process was applied to all imaging planes including the sagittal, coronal, and transverse planes.

Measurement of ex vivo fluorescence intensity of tumor

FPR675- or FPI774-labeled CRQTKN peptide (100 μL of 0.2 mmol/L solution) was injected via tail vein into mice and circulated for 2 hours. After circulation, lung tumors of mutant mice and normal lung tissue of wild-type mice were isolated. Ex vivo fluorescence intensities were measured using the eXplore Optix and Analysis Workstation software (ART Inc.).

Preparation of doxorubicin-loaded liposomes

Liposomes were prepared as previously described (24). Briefly, egg L-R-phosphatidylcholine (PC), egg L-R-phosphatidylglycerol (PG), cholesterol, 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE), and polyethylene glycol2000 (PEG2000)-1,2-distearoyl-sn-glycero-3-phosphoethanolamine (DSPE) were dissolved at a molar ratio of 5:5:5:0.1:0.2 (μmol/mL) in chloroform. The CRQTKN peptide-conjugated PEG2000-DSPE and FPR675-conjugated DOPE were used to prepare the targeted and fluorescent liposomes, respectively. Vesicles were sized via repeated extrusion through polycarbonate membrane filters with the pore size of 200 nm. To load doxorubicin into the liposomes, 500 μg of doxorubicin was mixed with 1 mL of the liposome solutions. Unloaded doxorubicin in the liposome suspension was removed through a PD10 column (GE Healthcare Bio-Sciences). Loaded amounts of doxorubicin were measured by absorbance at 490 nm (Molecular Devices Cooperation). Loading efficiency was calculated by dividing the loaded amounts of doxorubicin with the amounts of initially added doxorubicin. The size of liposomes was measured by dynamic light scattering using an ELS-Z 1000 instrument (Photal).

In vivo tumor therapy

K-rasLA2 mutant mice at 10 or 18 weeks of age were treated with intravenous injection of the CRQTKN peptide-targeted or untargeted liposomes containing doxorubicin (4 mg of doxorubicin/kg body weight), or free doxorubicin twice a week for 3 weeks. At the end of the treatment, blood was collected and was analyzed for hematologic parameters and liver as well as kidney function (DGMIF). Mice were euthanized, and tumors and control organs were isolated. The isolated tumor-bearing lungs were then weighed, and the total numbers of tumor lesions and legions with a diameter 3 mm or above were counted. Tissue samples were subjected to analysis by the immunofluorescence and IHC.

Immunofluorescence and IHC

Tissue samples were fixed in 4% paraformaldehyde, followed by dehydration in 15% to 30% sucrose, and then frozen quickly in Tissue-Tek O.C.T. embedding medium (Sakura Finetechnical). Frozen tissue sections (6 μm) were stained with hematoxylin and eosin or a rabbit antimouse TNFRSF19L antibody (Antibodies-online Inc.). Apoptosis levels of tissues were examined with ApopTag Red In Situ Apoptosis Detection Kit utilizing terminal deoxynucleotidyl terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling (TUNEL) method (Millipore). Nucleus was counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Stained tissue samples were observed under a light microscope or confocal microscope (Carl Zeiss). The accumulation of doxorubicin in tumor tissue was examined using an autofluorescence of doxorubicin (excitation, 530–560 nm/emission, 573–647 nm) under a fluorescence microscope.

Affinity-based receptor purification

The CRQTKN peptide was conjugated with sulfo-N-hydroxysuccinimidyl-2-(6-[biotinamido]-2-(p-azido benzamido)-hexanoamido) ethyl-1,3′-dithioproprionate (sulfo-SBED; Pierce), which is a biotinylated and photoreactive cross-linker, at room temperature in the dark for 60 minutes. Peptide-conjugated cross-linker was incubated with cells isolated from lung tumor tissues of K-rasLA2 mutant mice and exposed to ultraviolet light at 365 nm for 15 minutes, which induced the formation of a covalent bonding between the sulfo-SBED and a target protein. Cells were lysed, and the CRQTKN peptide was removed after cleavage of a disulfide bond between the peptide and cross-linker with 50 mmol/L dithiothreitol. The cross-linker and target protein complex was subjected to affinity purification using a monomeric avidin column (Pierce) at 4°C for 1 hour. Proteins bound to the cross-linker were eluted with 2 mmol/L D-biotin and were resolved via SDS-PAGE followed by silver staining compatible with mass spectrometry. Protein bands were excised from the gel and subjected to LC/MS-MS analysis as previously described (21).

Pull-down assays

Cell lysates of cells isolated from lung tumor tissues were prepared using Mem-PER Plus Membrane Protein Extraction Kit (Thermo Fisher Scientific). The cell lysates were incubated with 10 μmol/L of biotin-labeled CRQTKN peptide or control peptide and were subjected to a pull-down using monomeric avidin beads. The pull-down samples were subjected to electrophoresis, transferred to a PVDF membrane, and immunoblotted with a rabbit anti-mouse TNFRSF19L antibody (Antibodies-online Inc.). An excess concentration (50 μmol/L) of unlabeled CRQTKN was added to compete the binding of the biotin-labeled CRQTKN peptide to TNFRSF19L.

Competition and knockdown assays of peptide binding to cells

Cells isolated from lung tumor tissues were incubated with 10 μmol/L of FITC-labeled CRQTKN or control peptide, and the peptide cell binding was analyzed using flow cytometry. For competition of binding, cells were preincubated with the indicated concentrations of the antimouse TNFRSF19L antibody or IgG and incubated with the CRQTKN peptide. For knockdown of TNFRSF19L, H226 human lung tumor cells (ATCC) were used. Cells were tested for mycoplasma contamination using e-Myco plus Mycoplasma PCR Detection Kit (iNtRON) every 6 months and were used within 15 passages after thawing. H226 cells were transfected with control or TNFRSF19L siRNA (Bioneer). The sequences of TNFRSF19L siRNA were CUCUGUACAGGGCCCUAGA(dTdT) for sense and UCUAGGGCCCUGUACAGAG(dTdT) for antisense. After transfection, cells were incubated with FITC-CRQTKN and subjected to flow cytometry. TNFRSF19L knockdown was confirmed via Western blotting using rabbit antihuman TNFRSF19L and anti-glyceraldehyde 3-phosphate dehydrogenase antibodies (Thermo Fisher Scientific). For colocalization assays, H226 tumor cells were incubated with 10 μmol/L of FITC-labeled CRQTKN peptide and then with the rabbit anti-human TNFRSF19L antibody and Alexa Fluor-594-labeled secondary antibody. After nuclear staining with DAPI, cells were observed using a confocal microscopy.

Proximity ligation assays

H226 human lung tumor cells (12,000 cells per well of 8-well chamber slides) were incubated with 10 μmol/L of biotin-labeled CRQTKN peptide followed by incubation with streptavidin and a rabbit anti-streptavidin antibody. Cells were then incubated with a mouse anti-human TNFRSF19L antibody (Abcam). After incubation with primary antibodies, cells were incubated with secondary antibodies conjugated with oligonucleotides (anti-rabbit PLA probe PLUS and anti-mouse PLA probe MINUS) and subjected to ligation for 30 minutes at 37°C and amplification for 100 minutes at 37°C (Duolink In Situ Fluorescence; Merck). Nucleus was counterstained with DAPI, and cells were observed under a confocal microscope.

Statistical analysis

Data were presented as mean ± SE. Statistical significance was determined by one-way ANOVA analysis followed by Tukey multiple post hoc test using GraphPad Prism (GraphPad Software). Differences were considered statistically significant when P value was less than 0.05.

In vivo phage display for the selection of peptides that home to lung tumors in K-rasLA2 mutant mice

A phage library containing CX7C peptides (C, cysteine; X7, random 7 amino acids) was systemically administered into 18- to 20-week-old K-rasLA2 mutant mice bearing autochthonous lung tumor. After 15 minutes of circulation, phages that homed to tumor nodules were collected (Fig. 1A). After 5 rounds of selection, the phage titer of the fifth round was enriched approximately by 160-fold compared with that of the first round (Fig. 1B). Sequencing of peptide-coding DNA inserts of phage clones that are randomly picked at the fourth and fifth round revealed that 4 phage clones displaying CKSRKDGAC, CRGTAEG, CMPKRPSSC, and CRQTKN peptide more frequently occurred than other phage clones (Table 1).

Figure 1.

In vivo biopanning of phage library for peptides that home to lung tumors in K-rasLA2 mutant mice. A, Biopanning scheme. K-rasLA2 mutant mice bearing lung tumors were intravenously injected with a phage library (1 × 1011 pfu) displaying random peptides. After 15 minutes of circulation, mice were sacrificed and lung tumor nodules were isolated. Phages that homed to lung tumor nodules were collected and amplified for next round of biopanning. After 5 rounds of biopanning, the DNA inserts of selected phage clones were sequenced. B, Enrichment of phage titers. Data are phage titers (×104 pfu) per 100 mg tumor tissues determined after each round. C, Tumor homing of peptides. Four candidate peptides labeled with FPR675 NIR fluorescence dye were intravenously injected into tumor-bearing mice (PBS, n = 10; control peptide, n = 8; CKSRKDGAC, n = 3; CRGTAEG, n = 6; CMPKRPSSC, n = 6; CRQTKN, n = 14) or wild-type mice (n = 3 per group). After 4 hours of circulation, lungs were isolated, and ex vivo fluorescence intensities of lungs of tumor-bearing and wild-type mice were measured. Data are presented as mean ± SE. **, P < 0.01; n.s., not significant.

Figure 1.

In vivo biopanning of phage library for peptides that home to lung tumors in K-rasLA2 mutant mice. A, Biopanning scheme. K-rasLA2 mutant mice bearing lung tumors were intravenously injected with a phage library (1 × 1011 pfu) displaying random peptides. After 15 minutes of circulation, mice were sacrificed and lung tumor nodules were isolated. Phages that homed to lung tumor nodules were collected and amplified for next round of biopanning. After 5 rounds of biopanning, the DNA inserts of selected phage clones were sequenced. B, Enrichment of phage titers. Data are phage titers (×104 pfu) per 100 mg tumor tissues determined after each round. C, Tumor homing of peptides. Four candidate peptides labeled with FPR675 NIR fluorescence dye were intravenously injected into tumor-bearing mice (PBS, n = 10; control peptide, n = 8; CKSRKDGAC, n = 3; CRGTAEG, n = 6; CMPKRPSSC, n = 6; CRQTKN, n = 14) or wild-type mice (n = 3 per group). After 4 hours of circulation, lungs were isolated, and ex vivo fluorescence intensities of lungs of tumor-bearing and wild-type mice were measured. Data are presented as mean ± SE. **, P < 0.01; n.s., not significant.

Close modal
Table 1.

Selected peptide sequences that home to lung tumors in K-rasLA2 mutant mice

Phage clonePeptide sequenceFrequency (4R)Frequency (5R)
4R-3 CKSRKDGAC 4/27 (14.8%) 3/22 (13.6%) 
4R-16 CMPKRPSSC 3/27 (11.1%) 1/22 (4.5%) 
4R-21 CRGTAEG 3/27 (11.1%) 6/22 (27.3%) 
5R-7 CRQTKN 0/27 (0%) 9/22 (40.9%) 
Phage clonePeptide sequenceFrequency (4R)Frequency (5R)
4R-3 CKSRKDGAC 4/27 (14.8%) 3/22 (13.6%) 
4R-16 CMPKRPSSC 3/27 (11.1%) 1/22 (4.5%) 
4R-21 CRGTAEG 3/27 (11.1%) 6/22 (27.3%) 
5R-7 CRQTKN 0/27 (0%) 9/22 (40.9%) 

NOTE: A total of 27 and 22 phage clones from the fourth (4R) and fifth (5R) round, respectively, were randomly picked, and the peptide-coding DNA inserts of the phage clones were sequenced. Frequency is the number of the clones with the same sequence out of the number of whole sequenced clones. The number in parenthesis represents the frequency in percent.

To examine tumor homing of the candidate peptides, FPR675 NIR fluorescence dye-labeled peptides were intravenously injected into tumor-bearing K-rasLA2 mutant mice. After 4 hours of circulation, lungs were isolated, and fluorescence intensities of the lungs by the homing of the fluorescent peptides to lung tumors in K-rasLA2 mutant and wild-type mice were measured ex vivo. The CRQTKN peptide showed 2.9-fold higher levels of fluorescence intensity in the lungs of K-rasLA2 mutant mice than the NSSSVDK control peptide that was a sequence present in the phage coat protein (Fig. 1C). Meanwhile, fluorescence intensities by the homing of peptides to normal lungs of wild-type mice were low and were not significantly different among groups (Fig. 1C). Representative ex vivo fluorescence images showed a selective homing of the CRQTKN peptide to tumor-bearing lung and minimal accumulation at control organs including normal lungs of wild-type mice as well as the liver and kidney of wild-type and mutant mice (Supplementary Fig. S2). Given that the CRQTKN peptide showed the highest homing activity among candidate peptides, it was chosen for further study.

In vivo PAT of lung tumors targeted by the CRQTKN peptide in K-rasLA2 mutant mice

To examine whether PAT targeted by the CRQTKN peptide can detect lung tumors in mice, we performed in vivo PAT (Supplementary Fig. S3) using FPI774 NIR fluorescence dye-labeled CRQTKN peptide as an imaging probe. The minimum detectable concentration of FPI774-labeled CRQTKN peptide in the PAT was 3.9 μmol/L in vitro, with a signal-to-noise ratio of 10.9 ± 0.8 (Supplementary Fig. S4A). The peak PA amplitude was obtained at 774 nm, which matched well with the optical spectrum, whereas no apparent PA amplitude was observed at 850 nm even at the highest concentration (1 mmol/L) of the peptide examined (Supplementary Fig. S4B). FPI774-labeled CRQTKN peptide was intravenously injected into K-rasLA2 mutant and wild-type mice, and PA images were taken in 3 different imaging planes (i.e., coronal, sagittal, and transverse). An optical wavelength of 774 nm was used up to 270 minutes postinjection, and then the wavelength was switched to 870 nm for spectroscopic analyses up to 315 minutes postinjection.

Prior to injection, the whole-body PA images at 770 nm showed that no distinct PA signals were detected within lung regions in all imaging planes of all experimental groups (Fig. 2A). The PA images at 774 nm in the 3 imaging planes taken at 270 minutes postinjection showed that the CRQTKN peptide was accumulated at tumor-bearing lung of mutant mice (Fig. 2B, a1–a3). The PA image contrast, calculated using the formula (PAlung – PAbackground)/PAbackground, obtained from the sagittal imaging plane of the lung region in mutant mice injected with the CRQTKN peptide, was enhanced by 4.1-folds at 270 minutes postinjection (4.50 ± 3.66; Fig. 2B, a2) compared with that at preinjection (1.11 ± 0.95; Fig. 2A, a2).

Figure 2.

In vivo PAT of lung tumors targeted by the CRQTKN peptide in K-rasLA2 mutant mice. A, PA images at 774 nm before the injection of FPI774-labeled CRQTKN and control peptides into K-rasLA2 mutant and wild-type mice. B, PA images at 774 nm taken 270 minutes postinjection of the peptides into mice. C, PA images at 870 nm taken 315 minutes postinjection of the peptides into mice. D, Subtraction PA images (PA270 min postinjection – PApreinjection) and (PA774 nm – PA870 nm) in the sagittal and transverse imaging planes. E, Imaging planes. F, Quantified PA amplitude enhancement in the regions of interest in tumor-bearing lungs at various time points. *, P < 0.015 (n = 3 per group). Dotted circles indicate the regions of interest in tumor-bearing lungs. C, coronal; S, sagittal; T, transverse; a.u., arbitrary unit.

Figure 2.

In vivo PAT of lung tumors targeted by the CRQTKN peptide in K-rasLA2 mutant mice. A, PA images at 774 nm before the injection of FPI774-labeled CRQTKN and control peptides into K-rasLA2 mutant and wild-type mice. B, PA images at 774 nm taken 270 minutes postinjection of the peptides into mice. C, PA images at 870 nm taken 315 minutes postinjection of the peptides into mice. D, Subtraction PA images (PA270 min postinjection – PApreinjection) and (PA774 nm – PA870 nm) in the sagittal and transverse imaging planes. E, Imaging planes. F, Quantified PA amplitude enhancement in the regions of interest in tumor-bearing lungs at various time points. *, P < 0.015 (n = 3 per group). Dotted circles indicate the regions of interest in tumor-bearing lungs. C, coronal; S, sagittal; T, transverse; a.u., arbitrary unit.

Close modal

Once the wavelength was switched to 870 nm, at which the optical absorption of FPI774 dye was very low, the spectral PA amplitudes were lowered. The PA image contrast obtained from the sagittal imaging plane of the lung region in mutant mice injected with the CRQTKN peptide was 1.30 ± 0.06 (Fig. 2C, a2), which was equivalent to that of preinjection (Fig. 2A, a2). By contrast, both PA amplitudes at 774 nm and spectral PA amplitudes at 870 nm in the lung regions were hardly observed in the control groups: K-rasLA2 mutant mice injected with a control peptide (Fig. 2B, b1–b3 for 774 nm and Fig. 2C, b1–b3 for 870 nm), wild-type mice injected with the CRQTKN peptide (Fig. 2B, c1–c3 for 774 nm and Fig. 2C, c1–c3 for 870 nm), and wild-type mice injected with the control peptide (Fig. 2B, d1–d3 for 774 nm and Fig. 2C, d1–d3 for 870 nm).

To further confirm the accumulation of the CROTKN peptides, we processed the PA images using the following formula: PApostinjection – PApreinjection and PA774 nm –PA870 nm. The subtraction PA amplitude between pre- and postinjection in the sagittal imaging plane (i.e., Fig. 2A, a2 and B, a2) was 26.94 ± 0.09 (Fig. 2D, a1), and that between 774 and 870 nm (i.e., Fig. 2B, a2 and C, a2) was 48.51 ± 0.10 (Fig. 2D, a2). These results show distinct accumulation of the CRQTKN peptide at the lung tumor in K-rasLA2 mutant mice. The subtraction PA images in the transverse imaging plane also clearly visualized the specific accumulation of the CRQTKN peptide (Fig. 2D, a3–a4). Figure 2E shows the coronal, sagittal, and transverse imaging planes. The PA amplitude enhancement, calculated using the formula [(PApostinjection – PApreinjection)/PApostinjection × 100], taken at various time points within the lung regions of mutant mice injected with the CRQTKN peptide significantly (P < 0.015) increased by 130 ± 29% at 270 minutes postinjection, whereas that of the control groups increased only by 30 ± 5% (Fig. 2F).

Ex vivo PA imaging of organs isolated from the mice injected with the CRQTKN peptide showed that the PA amplitudes at 774 nm in the tumor-bearing lung were higher than those in the control organs such as the heart, liver, and kidney (Fig. 3A). Subtraction PA amplitude (PA774 nm − PA870 nm) of the tumor-bearing lung of the mutant mice injected with the CRQTKN peptide was 4.8 ± 0.9, which was 4.7-, 3.9-, and 2.2-fold higher than those of the heart (1.0 ± 0.3), liver (1.3 ± 0.4), and kidney (2.2 ± 0.4), respectively, from the same mice (Fig. 3B). Meanwhile, it was 2.5-, 3.0-, and 2.2-fold higher than that of tumor-bearing lung of mutant mice injected with the control peptide (1.9 ± 0.3), lung of wild-type mice injected with the CRQTKN peptide (1.6 ± 0.5), and lung of wild-type mice injected with the control peptide (2.2 ± 0.3), respectively (Fig. 3B).

Figure 3.

Ex vivo PAT of lung tumors and control organs in K-rasLA2 mutant mice. A, Representative ex vivo PA and spectroscopic PA images at 774 and 870 nm, respectively, and subtraction PA images (PA774 nm – PA870 nm) of lung tumors and control organs (heart, liver, and kidney) isolated from K-rasLA2 mutant mice after injection with FPI774-labeled CRQTKN peptide. Arrows, regions of interest in the tumor-bearing lungs. B, Average subtraction PA amplitudes of lung tumors and control organs (heart, liver, and kidney) isolated from K-rasLA2 mutant and wild-type mice after injection with FPI774-labeled CRQTKN and control peptides (n = 3 per group). a.u., arbitrary unit. **, P < 0.01; n.s., not significant.

Figure 3.

Ex vivo PAT of lung tumors and control organs in K-rasLA2 mutant mice. A, Representative ex vivo PA and spectroscopic PA images at 774 and 870 nm, respectively, and subtraction PA images (PA774 nm – PA870 nm) of lung tumors and control organs (heart, liver, and kidney) isolated from K-rasLA2 mutant mice after injection with FPI774-labeled CRQTKN peptide. Arrows, regions of interest in the tumor-bearing lungs. B, Average subtraction PA amplitudes of lung tumors and control organs (heart, liver, and kidney) isolated from K-rasLA2 mutant and wild-type mice after injection with FPI774-labeled CRQTKN and control peptides (n = 3 per group). a.u., arbitrary unit. **, P < 0.01; n.s., not significant.

Close modal

Enhanced drug delivery and antitumor growth activity of liposomes targeted by the CRQTKN peptide in K-rasLA2 mutant mice

To examine whether the CRQTKN peptide can guide and enhance drug delivery, we used doxorubicin-loaded liposomes as a nanocarrier and surface modified them with the CRQTKN peptide. Dynamic light scattering analysis revealed that doxorubicin loading did not significantly affect the size of liposomes, which were 160 to 170 nm (Supplementary Fig. S5A). Moreover, the peptide modification of liposomal surfaces did not influence the loading efficiency of doxorubicin into untargeted and CRQTKN-targeted liposomes, which were 94.5 ± 0.1% and 97.4 ± 0.9%, respectively (Supplementary Fig. S5B). Both untargeted and targeted liposomes loaded with doxorubicin showed the characteristic absorption peak of doxorubicin at 490 nm (Supplementary Figs. S5C and S5D, respectively).

FPI774-labeled liposomes were circulated for 4 hours after intravenous injection, and organs were isolated from K-rasLA2 mutant mice. Ex vivo fluorescence images and the fluorescence intensities of tumor-bearing lungs and control organs showed that the accumulation at the tumor by liposomes targeted by the CRQTKN peptide was 5.4-fold higher than that by untargeted liposomes and was 5-fold higher than the accumulation of targeted liposomes at the liver, whereas the uptake of both targeted and untargeted liposomes in the liver were higher than that in heart, kidney, and spleen (Supplementary Fig. S6A–S6C). Enhanced tumor accumulation of the targeted liposomes was also observed by microscopic analysis of tumor tissues (Supplementary Fig. S6D).

To examine antitumor growth activity of the CRQTKN-targeted liposomes, K-rasLA2 mutant mice were treated with doxorubicin-loaded liposomes untargeted (L-D) or targeted by the CRQTKN peptide (CRQTKN-L-D). It has been reported that, when examined after sacrifice, the number of tumor nodules (or lesions) in the lungs of K-rasLA2 mutant mice increased in both number and size with age, and those values in mice at the same week of age were similar (22, 23). We treated K-rasLA2 mutant mice at 10 weeks of age with untargeted and targeted liposomes and doxorubicin alone as control. Lung weights were not significantly different among groups after treatments, which were 242 ± 9 mg, 191 ± 29 mg, 220 ± 30 mg, and 200 ± 30 mg in groups treated with saline, doxorubicin, L-D, and CRQTKN-L-D, respectively (Fig. 4A). Total number of tumor nodules in the lungs significantly decreased to 6 ± 2 after treatment with CRQTKN-L-D, whereas those were 25 ± 2 in groups treated with saline, doxorubicin, and L-D (Fig. 4B). The number of large tumor nodules (3 mm or above in diameter) did not decrease after the treatments, which were 3 to 5 ± 1 in all groups (Fig. 4C).

Figure 4.

Antitumor growth activity of untargeted and the CRQTKN-targeted liposomes in K-rasLA2 mutant mice. A–C,K-rasLA2 mutant mice at 10 week of age were treated with intravenous injection of PBS, doxorubicin (DOX), and L-D or CRQTKN-L-D. Weights of tumor-bearing lungs (A), total number of tumor nodules (B), and the number of tumor nodules measuring 3 mm or above in diameter (C) in the lungs after the treatments. Data are mean ± SE (n = 3 for each group). D–F,K-rasLA2 mutant mice at 18 week of age were treated with intravenous injection of PBS, L-D, and CRQTKN-L-D. Weights of tumor-bearing lungs (D), total number of tumor nodules (E), and the number of tumor nodules measuring 3 mm or above in diameter (F) in the lungs after the treatments. Horizontal bars represent mean values for each group (n = 5). G, The levels of doxorubicin (red) and TUNEL-stained apoptosis (green) in tumor tissues after the treatments. Nuclei were stained with DAPI (blue). H, Hematoxylin and eosin (H&E) and TUNEL (green) staining of the liver tissues in K-rasLA2 mutant after the treatments. Nuclei were stained with DAPI (blue). Scale bars, 20 μm. I, Serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in K-rasLA2 mutant mice after the treatments. *, P < 0.05 compared with healthy control; ***, P < 0.001; n.s., not significant.

Figure 4.

Antitumor growth activity of untargeted and the CRQTKN-targeted liposomes in K-rasLA2 mutant mice. A–C,K-rasLA2 mutant mice at 10 week of age were treated with intravenous injection of PBS, doxorubicin (DOX), and L-D or CRQTKN-L-D. Weights of tumor-bearing lungs (A), total number of tumor nodules (B), and the number of tumor nodules measuring 3 mm or above in diameter (C) in the lungs after the treatments. Data are mean ± SE (n = 3 for each group). D–F,K-rasLA2 mutant mice at 18 week of age were treated with intravenous injection of PBS, L-D, and CRQTKN-L-D. Weights of tumor-bearing lungs (D), total number of tumor nodules (E), and the number of tumor nodules measuring 3 mm or above in diameter (F) in the lungs after the treatments. Horizontal bars represent mean values for each group (n = 5). G, The levels of doxorubicin (red) and TUNEL-stained apoptosis (green) in tumor tissues after the treatments. Nuclei were stained with DAPI (blue). H, Hematoxylin and eosin (H&E) and TUNEL (green) staining of the liver tissues in K-rasLA2 mutant after the treatments. Nuclei were stained with DAPI (blue). Scale bars, 20 μm. I, Serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in K-rasLA2 mutant mice after the treatments. *, P < 0.05 compared with healthy control; ***, P < 0.001; n.s., not significant.

Close modal

In another set of experiments, K-rasLA2 mutant mice were treated with untargeted and targeted liposomes at 18 week of age as a model of an advanced stage of tumor. Although the weights of the lungs slightly decreased after treatment with CRQTKN-L-D, there were no significant differences among groups, which were 693 ± 427 mg, 552 ± 238 mg, and 306 ± 50 mg in groups treated with saline, L-D, and CRQTKN-LD, respectively (Fig. 4D). Unlike to the treatment of mice at 10 week of age, total number of tumor nodules in the lungs did not decrease after treatment with CRQTKN-L-D, which were 80 ± 25 in all groups (Fig. 4E). On the other hand, the number of large tumor nodules (3 mm or above in diameter) significantly decreased to 2 ± 1 after treatment with CRQTKN-LD, while those were 8 ± 2 and 6 ± 2 after treatment with saline and L-D, respectively (Fig. 4F).

Microscopic analysis of tumor tissues at the end of treatment showed that the accumulation of doxorubicin, as determined by its auto-fluorescence, and the levels of apoptosis were higher at tumor tissues of mutant mice treated with the CRQTKN-targeted liposomes than untargeted liposomes (Fig. 4G). Toxicity to the liver tissue, as determined via hematoxylin and eosin (H & E) and apoptosis staining, was not observed from treatments with either the CRQTKN-targeted or untargeted liposomes (Fig. 4H). Serum levels of the liver function enzymes, such as alanine aminotransferase and aspartate aminotransferase, were not significantly different from those of healthy control (Fig. 4I). In addition, analysis of hematologic parameters and kidney function in mouse blood at the end of treatments showed no significant systemic side effects by the treatments (Supplementary Fig. S7A-J).

Identification of TNF receptor superfamily 19-like (TNFRSF19L) as the receptor for the CRQTKN peptide in lung tumors of K-rasLA2 mutant mice

To identify the receptor for the CRQTKN peptide, we performed affinity purification using a biotinylated and photoactivatable cross-linker after conjugation with the CRQTKN peptide (Supplementary Fig. S8). Pull-down of proteins covalently bound to the cross-linker using avidin beads and mass spectrometry analysis of the proteins revealed TNFRSF19L, a member of TNF receptor (TNFR) superfamily, as a candidate receptor. IHC showed the upregulation of TNFRSF19L in lung tumor tissues of K-rasLA2 mutant mice compared with normal lung tissues in wild-type mice (Fig. 5A).

Figure 5.

Identification of TNFRSF19L as the receptor for the CRQTKN peptide in lung tumor. A, Immunohistochemistry of lung tumors from K-rasLA2 mutant mice and normal lung tissues of wild-type mice. Magnifications, ×400. B, Pull-down assays. Cell lysates of cells isolated from lung tumor tissues were incubated with 10 μmol/L of biotin-labeled CRQTKN or control peptide and then with avidin beads for pull-down (PD), followed by immunoblotting (IB) with the antimouse TNFRSF19L antibody. C, Competition assays. Cells isolated from lung tumors were incubated with 10 μmol/L of FITC-labeled CRQTKN and control peptide with or without pretreatment with the indicated concentrations (μg/mL) of the antimouse TNFRSF19L antibody or IgG control, followed by flow cytometry. D, Knockdown assays. H226 human lung tumor cells were transfected with TNFRSF19L or control siRNA and then incubated with 10 μmol/L of FITC-labeled CRQTKN or control peptide, followed by flow cytometry. E, Colocalization assays. H226 cells were incubated with 10 μmol/L of FITC-CRQTKN peptide (green), antihuman TNFRSF19L antibody (red), and DAPI for nuclear staining (blue) and were observed under a confocal microscope. F, Proximal ligation assays (PLA). H226 cells were incubated with 10 μmol/L of biotin-labeled CRQTKN or control peptide, followed by streptavidin and then with PLA PLUS probe. Cells were also incubated with the anti-TNFRSF19L antibody and then with PLA MINUS probe. Samples were subjected to ligation and amplification reaction for detection (red) under a confocal microscope. Nuclei were stained with DAPI (blue). Scale bars, 20 μm. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 5.

Identification of TNFRSF19L as the receptor for the CRQTKN peptide in lung tumor. A, Immunohistochemistry of lung tumors from K-rasLA2 mutant mice and normal lung tissues of wild-type mice. Magnifications, ×400. B, Pull-down assays. Cell lysates of cells isolated from lung tumor tissues were incubated with 10 μmol/L of biotin-labeled CRQTKN or control peptide and then with avidin beads for pull-down (PD), followed by immunoblotting (IB) with the antimouse TNFRSF19L antibody. C, Competition assays. Cells isolated from lung tumors were incubated with 10 μmol/L of FITC-labeled CRQTKN and control peptide with or without pretreatment with the indicated concentrations (μg/mL) of the antimouse TNFRSF19L antibody or IgG control, followed by flow cytometry. D, Knockdown assays. H226 human lung tumor cells were transfected with TNFRSF19L or control siRNA and then incubated with 10 μmol/L of FITC-labeled CRQTKN or control peptide, followed by flow cytometry. E, Colocalization assays. H226 cells were incubated with 10 μmol/L of FITC-CRQTKN peptide (green), antihuman TNFRSF19L antibody (red), and DAPI for nuclear staining (blue) and were observed under a confocal microscope. F, Proximal ligation assays (PLA). H226 cells were incubated with 10 μmol/L of biotin-labeled CRQTKN or control peptide, followed by streptavidin and then with PLA PLUS probe. Cells were also incubated with the anti-TNFRSF19L antibody and then with PLA MINUS probe. Samples were subjected to ligation and amplification reaction for detection (red) under a confocal microscope. Nuclei were stained with DAPI (blue). Scale bars, 20 μm. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

To examine the specific binding of the CRQTKN peptide to TNFRSF19L, cell lysates of lung tumor tissues of mutant mice were incubated with a biotin-labeled CRQTKN peptide, subjected to a pull-down using monomeric avidin beads, and immunoblotted with an anti-TNFRSF19L antibody. The pull-down assays using the CRQTKN peptide resulted in the precipitation of TNFRSF19L, which was not observed in the assays using biotin-labeled control peptide or biotin-labeled CRQTKN peptide competed with unlabeled (biotin-free) CRQTKN (Fig. 5B). Preincubation of cells isolated from lung tumor tissues of mutant mice with an anti-TNFRSF19L antibody inhibited the cellular binding of the CRQTKN peptide by 50% in a dose-dependent manner (Fig. 5C). Knockdown of TNFRSF19L expression using a siRNA in H226 human lung tumor cells reduced the cellular binding of the CRQTKN peptide by 60% (Fig. 5D). Confocal microscopy showed that the CRQTKN peptide was closely colocalized with TNFRSF19L on the surface of H226 cells (Fig. 5E). Moreover, proximity ligation assays demonstrated the direct interaction between the CRQTKN peptide and TNFRSF19L in H226 cells (Fig. 5F). Collectively, these results suggest that the TNFRSF19L is upregulated in lung tumors of K-rasLA2 mutant mice and is the binding receptor for the CRQTKN peptide.

Using in vivo phage display, we identified the CRQTKN peptide that homes to autochthonous lung tumors in K-rasLA2 mutant mice. In vivo PAT targeted by the fluorescent CRQTKN peptide detected lung tumors in mutant mice. Compared with L-D, CRQTKN-L-D peptide more efficiently reduced the number of tumor nodules or size of large tumor nodules (3 mm or above in diameter) in tumor-bearing lung of mutant mice. Also, using affinity purification, we identified that TNFRSF19L is the receptor for the homing of CRQTKN peptide to lung tumors in K-rasLA2 mutant mice.

Liposomal encapsulation of doxorubicin has been shown to reduce its systemic toxicity, particularly cardiac toxicity (25, 26). It also contributes to preferential accumulation of the payload (i.e., doxorubicin) at the tumor tissue through the enhanced permeability and retention (EPR) effect, which takes advantage of the extravasation of nanoparticles from leaky tumor blood vessels but little extravasation from normal blood vessels (27). Antibodies have been used for targeted delivery of liposomes to tumor. Compared with nontargeted liposomes, liposomes labeled with anti-β1 integrin antibody or anticarbonic anhydrase IX antibody and anti-Her2 antibody-labeled liposomes more efficiently inhibited tumor growth in mice bearing human lung and breast tumor xenografts, respectively (28–31). However, to our knowledge, there have been no reports on the therapeutic efficacy of antibody-labeled liposomes in K-rasLA2 mutant mice. In addition, labeling with anti-Her2 antibody did not increase tumor homing but did increase the internalization of liposomes (31). Moreover, compared with anti-Her2 antibody with a low affinity (3.2 × 10−7 M), anti-Her2 antibody with a high affinity (1.5 × 10−11 M) did not deeply penetrate into tumor tissues but rather localized at the perivascular area (32). As such, surface modification of liposomes with homing peptides that have a smaller size and weaker affinity than antibodies may have advantages for delivery of liposomes. Enhanced delivery of liposomes targeted by a homing peptide to lung tumors has been previously reported in human lung tumor xenografts models (13, 14). In this study, we demonstrated an enhanced antitumor growth activity by the CRQTKN peptide-targeted liposomes, compared with doxorubicin alone and untargeted (naïve) liposomes that are already available in the clinic, in autochthonous lung tumors in K-rasLA2 mutant mice. Given that lung tumors in K-rasLA2 mutant mice closely mimic human lung cancer (23), the CRQTKN peptide would be a promising probe for targeted drug delivery to human lung cancer.

In normal human tissues, TNFRSF19L, also known as receptor expressed in lymphoid tissue (RELT), is expressed in the spleen, lymph node, blood leukocytes, and bone marrow (33). TNFRSF19L induces apoptosis by activating p38 through a pathway different from TNFR1 (34). In this study, however, we found that the CRQTKN peptide homes to lung tumors by targeting TNFRSF19L that is upregulated in lung tumor tissues of K-rasLA2 mutant mice. The accumulation of the CRQTKN peptide as well as the peptide-targeted liposomes in normal tissues, such as heart, liver, and kidney, were lower than that in tumor. Moreover, analysis of hematologic parameters and liver and kidney function after treatment with the CRQTKN-targeted liposomes did not show significant systemic side effects. EPR effect might also contribute to the selective homing of the targeted liposomes to tumor over normal tissues. In support of our study, the gene expression levels of TNFRSF19L in human primary lung cancer have been previously reported to be 2-fold higher than those in adjacent normal bronchial tissue and decrease close to normal levels after treatment with romidepsin, a histone deacetylase inhibitor (35). In addition, analysis of the human protein atlas dataset (https://www.proteinatlas.org) reveals that high expression of TNFRSF19L is correlated with poor survival rate of renal cancer and lung cancer. Autoantibodies against TNFRSF19L are a potential biomarker for the detection of breast cancer (36). Moreover, TNFR superfamily members, particularly their extracellular domains, are evolutionary conserved (37). The amino acid sequences of TNFRSF19L in rat and macaque are highly identical to the human sequence (33). An NCBI database search reveals that the mouse and human TNFRSF19L has 86% similar identity. These findings may explain why the CRQTKN peptide is able to bind to H226 human lung tumor cells as well as mouse primary cells isolated from lung tumors of K-rasLA2 mutant mice. Collectively, these findings suggest that TNFRSF19L holds a potential as a novel receptor for targeted therapy of lung cancer.

Homology search using NCBI BLAST protein database (https://blast.ncbi.nlm.nih.gov) revealed that the CRQTKN sequence is homologous to the 797CRQTK801 motif of Malrd1 (also known as Diet1) in mouse. This protein has been reported to affect FGF15/19 synthesis and regulate FGF15/19-dependent bile acid synthesis (38). In addition, the CRQTKN sequence is homologous to 464CRQEKN469 of the disintegrin and metallopeptidase domain 34-like protein. Interestingly, TNFRSF19L is an orphan receptor whose corresponding TNF ligand is not known (39). Further studies on the interaction between TNFRSF19L and those proteins containing a motif homologous to the CRQTKN sequence may provide a clue for identifying a bona fide ligand of TNFRSF19L.

Absorption of photons by biomolecules produces thermoelastic pressure waves due to the PA effect, which can be detected and converted to images by PAT (4). This conversion from optical to ultrasonic energy by PAT enables multiscale high-resolution imaging of objects ranging in size from organelles to organs and higher imaging depth than fluorescence imaging (4). PAT has been used to detect a hypoxic status in orthotopic lung tumors of mice (40). PAT of melanoma has been reported using gold nanoparticles targeted by α-melanocyte stimulating hormone (α-MSH) that selectively binds to α-MSH receptor upregulated on melanomas (6). As such, we exploited PAT, instead of fluorescence imaging, for in vivo imaging of autochthonous lung tumors in K-rasLA2 mutant mice. The peak absorption wavelength of FPI774 NIR fluorescence dye was obtained at 774 nm. Indocyanine green, an FDA-approved fluorescence dye, is known to have a similar peak absorption wavelength at 790 nm (7). Enhanced PA amplitude at 774 nm by tumor homing of the FPI774-CRQTKN peptide enabled us to detect lung tumors. To our knowledge, this is the first study on the in vivo detection of orthotopic lung tumors in mice by PAT targeted by a tumor-homing peptide.

Taken together, the results of this study are expected to provide a promising peptide probe for early detection and targeted drug delivery to the tumor and a novel molecular target of the tumor in patients with lung cancer. However, it needs further development for PAT using the fluorescent CRQTKN peptide to be applied in the clinic. In the meantime, the PET using a radioisotope-labeled CRQTKN peptide is a more practical imaging modality for the early detection of lung cancer in the clinic.

No potential conflicts of interest were disclosed.

Conception and design: H. Jung, S. Park, G. Rangaswamy Gunassekaran, M. Jeon, C. Kim, B. Lee

Development of methodology: H. Jung, G. Rangaswamy Gunassekaran, M. Jeon, M.-C. Baek, C. Kim, B. Lee

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): H. Jung, S. Park, G. Rangaswamy Gunassekaran, M. Jeon, J.Y. Park, G. Shim, Y.-K. Oh, C. Kim, B. Lee

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): H. Jung, S. Park, G. Rangaswamy Gunassekaran, M. Jeon, Y.-E. Cho, G. Shim, C. Kim, B. Lee

Writing, review, and/or revision of the manuscript: H. Jung, S. Park, G. Rangaswamy Gunassekaran, M. Jeon, C. Kim, B. Lee

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): H. Jung, M. Jeon, M.-C. Baek, C. Kim, B. Lee

Study supervision: M. Jeon, I.-S. Kim, C. Kim, B. Lee

This work was supported by the grants from the National Research Foundation funded by the Ministry of Science & ICT (NRF-2014R1A5A2009242, NRF-2018R1A2B200837, and 2017M3A9G8083382 to B. Lee), ICT Consilience Creative Program (IITP-2018-2011-1-00783 to C. Kim) funded by the Ministry of Science and ICT, and the Korea Health Technology R&D Project (HI15C1817 to C. Kim) funded by the Ministry of Health & Welfare.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Siegel
RL
,
Miller
KD
,
Jemal
A
. 
Cancer statistics, 2018
.
CA Cancer J Clin
2018
;
68
:
7
30
.
2.
Chong
CR
,
Jänne
PA
. 
The quest to overcome resistance to EGFR-targeted therapies in cancer
.
Nat Med
2013
;
19
:
1389
400
.
3.
Gerlinger
M
,
Rowan
AJ
,
Horswell
S
,
Larkin
J
,
Endesfelder
D
,
Gronroos
E
, et al
Intratumor heterogeneity and branched evolution revealed by multiregion sequencing
.
N Engl J Med
2012
;
366
:
883
92
.
4.
Wang
LV
,
Hu
S
. 
Photoacoustic tomography: in vivo imaging from organelles to organs
.
Science
2012
;
335
:
1458
62
.
5.
Wang
LV
,
Yao
J
. 
A practical guide to photoacoustic tomography in the life sciences
.
Nat Methods
2016
;
13
:
627
38
.
6.
Kim
C
,
Cho
EC
,
Chen
J
,
Song
KH
,
Au
L
,
Favazza
C
, et al
In vivo molecular photoacoustic tomography of melanomas targeted by bioconjugated gold nanocages
.
ACS Nano
2010
;
4
:
4559
64
.
7.
Kim
C
,
Favazza
C
,
Wang
LV
. 
In vivo photoacoustic tomography of chemicals: high-resolution functional and molecular optical imaging at new depths
.
Chem Rev
2010
;
110
:
2756
82
.
8.
Razansky
D
,
Buehler
A
,
Ntziachristos
V
. 
Volumetric real-time multispectral optoacoustic tomography of biomarkers
.
Nat Protoc
2011
;
6
:
1121
9
.
9.
Pu
K
,
Shuhendler
AJ
,
Jokerst
JV
,
Mei
J
,
Gambhir
SS
,
Bao
Z
, et al
Semiconducting polymer nanoparticles as photoacoustic molecular imaging probes in living mice
.
Nat Nanotechnol
2014
;
9
:
233
9
.
10.
Ladner
RC
,
Sato
AK
,
Gorzelany
J
,
de Souza
M
. 
Phage display-derived peptides as therapeutic alternatives to antibodies
.
Drug Discov Today
2004
;
9
:
525
9
.
11.
Ruoslahti
E
. 
Peptides as targeting elements and tissue penetration devices for nanoparticles
.
Adv Mater
2012
;
24
:
3747
56
.
12.
Arap
W
,
Pasqualini
R
,
Ruoslahti
E
. 
Cancer treatment by targeted drug delivery to tumor vasculature in a mouse model
.
Science
1998
;
279
:
377
80
.
13.
Chang
D-K
,
Lin
C-T
,
Wu
C-H
,
Wu
H-C
. 
A novel peptide enhances therapeutic efficacy of liposomal anti-cancer drugs in mice models of human lung cancer
.
PLoS One
2009
;
4
:
e4171
.
14.
Chi
L
,
Na
M-H
,
Jung
H-K
,
Vadevoo
SMP
,
Kim
C-W
,
Padmanaban
G
, et al
Enhanced delivery of liposomes to lung tumor through targeting interleukin-4 receptor on both tumor cells and tumor endothelial cells
.
J Control Release
2015
;
209
:
327
36
.
15.
Scott
JK
,
Smith
GP
. 
Searching for peptide ligands with an epitope library
.
Science
1990
;
249
:
386
90
.
16.
Kolonin
MG
,
Bover
L
,
Sun
J
,
Zurita
AJ
,
Do
K-A
,
Lahdenranta
J
, et al
Ligand-directed surface profiling of human cancer cells with combinatorial peptide libraries
.
Cancer Res
2006
;
66
:
34
40
.
17.
Pasqualini
R
,
Ruoslahti
E
. 
Organ targeting in vivo using phage display peptide libraries
.
Nature
1996
;
380
:
364
6
.
18.
Hoffman
JA
,
Giraudo
E
,
Singh
M
,
Zhang
L
,
Inoue
M
,
Porkka
K
, et al
Progressive vascular changes in a transgenic mouse model of squamous cell carcinoma
.
Cancer Cell
2003
;
4
:
383
91
.
19.
Joyce
JA
,
Laakkonen
P
,
Bernasconi
M
,
Bergers
G
,
Ruoslahti
E
,
Hanahan
D
. 
Stage-specific vascular markers revealed by phage display in a mouse model of pancreatic islet tumorigenesis
.
Cancer Cell
2003
;
4
:
393
403
.
20.
Pasqualini
R
,
Koivunen
E
,
Kain
R
,
Lahdenranta
J
,
Sakamoto
M
,
Stryhn
A
, et al
Aminopeptidase N is a receptor for tumor-homing peptides and a target for inhibiting angiogenesis
.
Cancer Res
2000
;
60
:
722
7
.
21.
Wang
K
,
Purushotham
S
,
Lee
J-Y
,
Na
M-H
,
Park
H
,
Oh
S-J
, et al
In vivo imaging of tumor apoptosis using histone H1-targeting peptide
.
J Control Release
2010
;
148
:
283
91
.
22.
Johnson
L
,
Mercer
K
,
Greenbaum
D
,
Bronson
RT
,
Crowley
D
,
Tuveson
DA
, et al
Somatic activation of the K-ras oncogene causes early onset lung cancer in mice
.
Nature
2001
;
410
:
1111
6
.
23.
Sweet-Cordero
A
,
Mukherjee
S
,
Subramanian
A
,
You
H
,
Roix
JJ
,
Ladd-Acosta
C
, et al
An oncogenic KRAS2 expression signature identified by cross-species gene-expression analysis
.
Nat Genet
2005
;
37
:
48
55
.
24.
Shim
G
,
Kim
D
,
Lee
S
,
Chang
RS
,
Byun
J
,
Oh
YK
. 
Staphylococcus aureus-mimetic control of antibody orientation on nanoparticles
.
Nanomedicine
2019
;
16
:
267
77
.
25.
Rahman
AM
,
Yusuf
SW
,
Ewer
MS
. 
Anthracycline induced cardiotoxicity and cardiac-sparing effect of liposomal formulation
.
Int J Nanomed
2007
;
2
:
567
83
.
26.
Davis
ME
,
Shin
DM
. 
Nanoparticle therapeutics: an emerging treatment modality for cancer
.
Nat Rev Drug Discov
2008
;
7
:
771
82
.
27.
Iyer
AK
,
Khaled
G
,
Fang
J
,
Maeda
H
. 
Exploiting the enhanced permeability and retention effect for tumor targeting
.
Drug Discov Today
2006
;
11
:
812
8
.
28.
Sugano
M
,
Egilmez
NK
,
Yokota
SJ
,
Chen
FA
,
Harding
J
,
Huang
SK
, et al
Antibody targeting of doxorubicin-loaded liposomes suppresses the growth and metastatic spread of established human lung tumor xenografts in severe combined immunodeficient mice
.
Cancer Res
2000
;
60
:
6942
9
.
29.
Lin
C
,
Wong
BCK
,
Chen
H
,
Bian
Z
,
Zhang
G
,
Zhang
X
, et al
Pulmonary delivery of triptolide-loaded liposomes decorated with anti-carbonic anhydrase IX antibody for lung cancer therapy
.
Sci Rep
2017
;
7
:
1097
108
.
30.
Park
JW
,
Hong
K
,
Kirpotin
DB
,
Colbern
G
,
Shalaby
R
,
Baselga
J
, et al
Anti-HER2 immunoliposomes enhanced efficacy attributable to targeted delivery
.
Clin Cancer Res
2002
;
8
:
1172
81
.
31.
Kirpotin
DB
,
Drummond
DC
,
Shao
Y
,
Shalaby
MR
,
Hong
K
,
Nielsen
UB
, et al
Antibody targeting of long-circulating lipidic nanoparticles does not increase tumor localization but does increase internalization in animal models
.
Cancer Res
2006
;
66
:
6732
40
.
32.
Adams
GP
,
Schier
R
,
McCall
AM
,
Simmons
HH
,
Horak
EM
,
Alpaugh
RK
, et al
High affinity restricts the localization and tumor penetration of single-chain fv antibody molecules
.
Cancer Res
2001
;
61
:
4750
5
.
33.
Sica
GL
,
Zhu
G
,
Tamada
K
,
Liu
D
,
Ni
J
,
Chen
L
. 
RELT, a new member of the tumor necrosis factor receptor superfamily, is selectively expressed in hematopoietic tissues and activates transcription factor NF-kappaB
.
Blood
2001
;
97
:
2702
7
.
34.
Moua
P
,
Checketts
M
,
Xu
LG
,
Shu
HB
,
Reyland
ME
,
Cusick
JK
. 
RELT family members activate p38 and induce apoptosis by a mechanism distinct from TNFR1
.
Biochem Biophys Res Commun
2017
;
491
:
25
32
.
35.
Schrump
DS
,
Fischette
MR
,
Nguyen
DM
,
Zhao
M
,
Li
X
,
Kunst
TF
, et al
Clinical and molecular responses in lung cancer patients receiving Romidepsin
.
Clin Cancer Res
2008
;
14
:
188
98
.
36.
Zhong
L
,
Ge
K
,
Zu
JC
,
Zhao
LH
,
Shen
WK
,
Wang
JF
, et al
Autoantibodies as potential biomarkers for breast cancer
.
Breast Cancer Res
2008
;
10
:
R40
.
37.
Anderson
DM
,
Maraskovsky
E
,
Billingsley
WL
,
Dougall
WC
,
Tometsko
ME
,
Roux
ER
, et al
A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function
.
Nature
1997
;
390
:
175
9
.
38.
Reue
K
,
Lee
JM
,
Vergnes
L
. 
Diet1 is a regulator of fibroblast growth factor 15/19-dependent bile acid synthesis
.
Dig Dis
2015
;
33
:
307
13
.
39.
Bossen
C
,
Ingold
K
,
Tardivel
A
,
Bodmer
JL
,
Gaide
O
,
Hertig
S
, et al
Interactions of tumor necrosis factor (TNF) and TNF receptor family members in the mouse and human
.
J Biol Chem
2006
;
281
:
13964
71
.
40.
Raes
F
,
Sobilo
J
,
Le Mee
M
,
Retif
S
,
Natkunarajah
S
,
Lerondel
S
, et al
High resolution ultrasound and photoacoustic imaging of orthotopic lung cancer in mice: new perspectives for onco-pharmacology
.
PLoS One
2016
;
11
:
e0153532
.