Tumor Inhibitory Effect of IRCR201, a Novel Cross-Reactive c-Met Antibody Targeting the PSI Domain

Hepatocyte growth factor receptor (HGFR, c-Met) is an essential member of the receptor tyrosine kinase (RTK) family that is often dysregulated during tumor progression, driving a malignant phenotypic state and modulating important cellular functions including tumor growth, invasion, metastasis, and angiogenesis, providing a strong rationale for targeting HGF/c-Met signaling axis in cancer therapy. Based on its protumorigenic potentials, we developed IRCR201, a potent antagonistic antibody targeting the plexin-semaphorin-integrin (PSI) domain of c-Met, using synthetic human antibody phage libraries. We characterized and evaluated the biochemical properties and tumor inhibitory effect of IRCR201 in vitro and in vivo. IRCR201 is a novel fully-human bivalent therapeutic antibody that exhibits cross-reactivity against both human and mouse c-Met proteins with high affinity and specificity. IRCR201 displayed low agonist activity and rapidly depleted total c-Met protein via the lysosomal degradation pathway, inhibiting c-Met-dependent downstream activation and attenuating cellular proliferation in various c-Met-expressing cancer cells. In vivo tumor xenograft models also demonstrated the superior tumor inhibitory responsiveness of IRCR201. Taken together, IRCR201 provides a promising therapeutic agent for c-Met-positive cancer patients through suppressing the c-Met signaling pathway and tumor growth.


IRCR201 Exhibits High Affinity to Both Human and Mouse c-Met
IRCR201 is a cross-reactive c-Met-targeting human immunoglobulin G1 (IgG1), and was developed through panning techniques using synthetic human antibody phage libraries. To evaluate whether IRCR201 specifically binds to the c-Met extracellular domain (ECD)-fragment crystallizable region (Fc), we performed enzyme-linked immunosorbent assay (ELISA) using recombinant human c-Met and recepteur d'origine nantais (RON) proteins. The RON receptor tyrosine kinase is a member of the c-Met family, and has similar biochemical and structural properties to c-Met [30][31][32]. We confirmed that IRCR201 specifically binds to human c-Met in a dose-dependent manner, but not to the RON protein, indicating the selectivity of IRCR201 against the c-Met protein (Figure 1a). The binding kinetics of IRCR201 to recombinant human and mouse c-Met ECD was determined using Biacore™ T100 based on surface plasmon resonance (SPR). Our analysis showed that IRCR201 binds to human c-Met ECD-Fc with a K D of 0.7207 nM, and mouse c-Met ECD-Fc with a K D of 0.8448 nM (Figure 1b,c and Table 1). However, IRCR201 does not bind to bovine serum albumin (BSA) (Figure 1d). Additionally, IRCR201 demonstrates strong cross-reactivity to both human and mouse c-Met, representing differential binding properties compared to other c-Met-targeting antibodies such as 5D5 and huOA5D5.v2 (onartuzumab) (Figure 1b-e). To confirm the expression of c-Met in various cancer cell lines, we performed immunoblot analysis using commercial c-Met antibody (Figure 1f). To assess whether IRCR201 binds to the c-Met on the cell surface, the reactivity of IRCR201 against the cell surface c-Met of A549 (c-Met+) or MCF7 (c-Met−) was confirmed by flow cytometry. Overall, our results showed that IRCR201 specifically binds to the A549 cell surface in a dose-dependent manner, but does not engage to MCF7 lacking c-Met (Figure 1g). Int. J. Mol. Sci. 2017, 18,1968 3 of 22

IRCR201 Exhibits High Affinity to Both Human and Mouse c-Met
IRCR201 is a cross-reactive c-Met-targeting human immunoglobulin G1 (IgG1), and was developed through panning techniques using synthetic human antibody phage libraries. To evaluate whether IRCR201 specifically binds to the c-Met extracellular domain (ECD)-fragment crystallizable region (Fc), we performed enzyme-linked immunosorbent assay (ELISA) using recombinant human c-Met and recepteur d'origine nantais (RON) proteins. The RON receptor tyrosine kinase is a member of the c-Met family, and has similar biochemical and structural properties to c-Met [30][31][32]. We confirmed that IRCR201 specifically binds to human c-Met in a dose-dependent manner, but not to the RON protein, indicating the selectivity of IRCR201 against the c-Met protein (Figure 1a). The binding kinetics of IRCR201 to recombinant human and mouse c-Met ECD was determined using Biacore™ T100 based on surface plasmon resonance (SPR). Our analysis showed that IRCR201 binds to human c-Met ECD-Fc with a KD of 0.7207 nM, and mouse c-Met ECD-Fc with a KD of 0.8448 nM (Figure 1b,c and Table 1). However, IRCR201 does not bind to bovine serum albumin (BSA) ( Figure  1d). Additionally, IRCR201 demonstrates strong cross-reactivity to both human and mouse c-Met, representing differential binding properties compared to other c-Met-targeting antibodies such as 5D5 and huOA5D5.v2 (onartuzumab) (Figure 1b-e). To confirm the expression of c-Met in various cancer cell lines, we performed immunoblot analysis using commercial c-Met antibody (Figure 1f). To assess whether IRCR201 binds to the c-Met on the cell surface, the reactivity of IRCR201 against the cell surface c-Met of A549 (c-Met+) or MCF7 (c-Met−) was confirmed by flow cytometry. Overall, our results showed that IRCR201 specifically binds to the A549 cell surface in a dose-dependent manner, but does not engage to MCF7 lacking c-Met (Figure 1g).

IRCR201 Specifically Binds to the PSI Domain of c-Met
To determine the binding epitope of IRCR201 against c-Met, we performed epitope mapping using the IRCR201 single-chain variable fragment (scFv) format ( Figure 2a, Table 2). The binding intensity of the epitope mapping analysis was quantified using Multi Gauge software V3.0, an image analysis program (Figure 2b, Table 2). Our results indicated that the scFv binds to E-7, F-7, and G-7 peptides spanning the Phe523-Cys545 amino acids, demonstrating that the scFv specifically recognizes an epitope within the PSI domain ( Figure 2a,b and Table 2). Key residues of c-Met and scFv interaction were further validated through the SAPPFVQ (Ser531-Gln537) amino acid sequence (Figure 2a,b and Table 2). Although the binding activity is mediated by a small number of contacts, such types of interactions are extremely specific. Furthermore, we conducted binding pattern analysis using recombinant human c-Met domain proteins for additional verification. Our result showed that IRCR201 exclusively binds to the recombinant proteins with the PSI domain ( Figure 2c). To confirm whether IRCR201 prevents molecular interaction between HGF and c-Met, competitive ELISA was conducted at fixed HGF concentration (2.5 µg/mL). The results showed that huOA5D5.v2 significantly inhibited the interaction between HGF and c-Met, whereas IRCR201 did not exhibit such behavior (Figure 2d).

IRCR201 Specifically Binds to the PSI Domain of c-Met
To determine the binding epitope of IRCR201 against c-Met, we performed epitope mapping using the IRCR201 single-chain variable fragment (scFv) format ( Figure 2a, Table 2). The binding intensity of the epitope mapping analysis was quantified using Multi Gauge software V3.0, an image analysis program (Figure 2b, Table 2). Our results indicated that the scFv binds to E-7, F-7, and G-7 peptides spanning the Phe523-Cys545 amino acids, demonstrating that the scFv specifically recognizes an epitope within the PSI domain (Figure 2a,b and Table 2). Key residues of c-Met and scFv interaction were further validated through the SAPPFVQ (Ser531-Gln537) amino acid sequence (Figure 2a,b and Table 2). Although the binding activity is mediated by a small number of contacts, such types of interactions are extremely specific. Furthermore, we conducted binding pattern analysis using recombinant human c-Met domain proteins for additional verification. Our result showed that IRCR201 exclusively binds to the recombinant proteins with the PSI domain ( Figure 2c). To confirm whether IRCR201 prevents molecular interaction between HGF and c-Met, competitive ELISA was conducted at fixed HGF concentration (2.5 µg/mL). The results showed that huOA5D5.v2 significantly inhibited the interaction between HGF and c-Met, whereas IRCR201 did not exhibit such behavior (Figure 2d). The reactivity of IRCR201 against a panel of overlapping peptides representing the c-Met extracellular domain was determined by peptide array. A peptide library spanning amino acids 1-932 of the c-Met extracellular domain was synthesized (JPT Peptide Technologies GmbH, Berlin, Germany). The library was prepared as overlapping linear peptides covalently bound to a cellulose membrane. The results show dot blots of specific epitope sequences for IRCR201 (yellow frame). The quantified levels of dot intensity were determined by Multi Gauge V3.0 program; (c) Binding pattern analysis using domain proteins of c-Met; (d) Hepatocyte growth factor (HGF)/c-Met competitive ELISA. After the HGF (2.5 µg/mL) was pre-treated with the human c-Met protein-immobilized plates, IRCR201 or huOA5D5.v2 was added to confirm whether HGF and each antibody were competitively bound. IPT: immunoglobulin-plexin-transcription; PSI: plexinsemaphorin-integrin domain. The reactivity of IRCR201 against a panel of overlapping peptides representing the c-Met extracellular domain was determined by peptide array. A peptide library spanning amino acids 1-932 of the c-Met extracellular domain was synthesized (JPT Peptide Technologies GmbH, Berlin, Germany). The library was prepared as overlapping linear peptides covalently bound to a cellulose membrane. The results show dot blots of specific epitope sequences for IRCR201 (yellow frame). The quantified levels of dot intensity were determined by Multi Gauge V3.0 program; (c) Binding pattern analysis using domain proteins of c-Met; (d) Hepatocyte growth factor (HGF)/c-Met competitive ELISA. After the HGF (2.5 µg/mL) was pre-treated with the human c-Met protein-immobilized plates, IRCR201 or huOA5D5.v2 was added to confirm whether HGF and each antibody were competitively bound. IPT: immunoglobulin-plexin-transcription; PSI: plexin-semaphorin-integrin domain.

IRCR201 Docks onto the PSI Domain of c-Met in Computational Modeling Analysis
To expand the structural understanding of the docking of IRCR201 onto c-Met, computational modeling was conducted using a published c-Met structure (PDB accession: 1SHY) [33]. A three-dimensional (3D) structural model of IRCR201 scFv in the docking region of the c-Met structure was generated using the Rosetta-based computational homology modeling technique [34] (Figure 3a). Our results showed that IRCR201 possesses an elongated complementarity-determining region (CDR)-H3 consisting of 18 amino acids, and the extended CDR-H3 loop is stabilized through interchain disulfide bond between the cysteines of CDR-H3 ( Figure 3a). We used ZDOCK-a protein-protein docking program-to aid in the structural understanding of the interaction between c-Met and IRCR201 and to observe whether IRCR201 scFv docks onto the specific epitope (yellow) within the PSI domain (dark gray) [35] (Figure 3b). The binding site of IRCR201 was on the PSI domain (dark gray), whereas the serine proteinase homology domain (SPHD) of HGF binds to the Sema domain (light gray). The distance between the two binding sites was considerably remote, proving that IRCR201 does not interfere with the interaction between HGF and c-Met in the 3D structure analysis. In this docking model, the CDR-H3, CDR-L1, and CDR-L3 of IRCR201 showed strong interaction with the epitope (Figure 3c).

IRCR201 Exhibits Low Agonistic Activity
In the development of c-Met inhibitory antibodies, the bivalent nature of the c-Met antibodies enabled the antibodies to mimic the role of HGF and thus activate the c-Met signaling pathway [18]. Because Akt acts as the main mediator of the c-Met signaling pathway [3], we investigated Akt signaling activity to evaluate the agonist activity of c-Met antibodies in the Caki1 renal cell carcinoma cell line (Figure 4a). In our assay, IRCR201 was compared with HGF and 5D5 as positive controls. As a result, we confirmed that 5D5-a c-Met agonistic antibody-phosphorylates Akt to a similar extent as HGF. huOA5D5.v2 is a monovalent antibody engineered to minimize agonistic activity, and showed Akt phosphorylation levels at 27.3% compared to the phosphate-buffered saline (PBS)-treated group, and IRCR201 exhibited a similar Akt phosphorylation level at 21.5% (Figure 4a). In addition, we further confirmed IRCR201-triggered agonistic effect by investigating the HGF/c-Met signaling pathway in Caki1 renal cell carcinoma cell line (Figure 4b). As a result, IRCR201 exhibited low agonist activity compared to HGF or 5D5 (Figure 4b). Although IRCR201 is a bivalent antibody, it demonstrated a lower agonistic effect compared to the modified monovalent antibody with minimal agonist activity.

IRCR201 Impedes Tumor Growth and Induces Cellular Apoptosis
In vitro functional inhibition activity of IRCR201 was analyzed in various cancer types. IRCR201 did not alter cancer cell growth of the low c-Met-expressing cell line MCF7 (Figures 1f and 5a). In U87MG (an HGF-dependent GBM cell line), IRCR201 showed more potent cancer cell growth inhibition compared to huOA5D5.v2, which has previously been reported to suppress only HGFdependent cellular proliferation ( Figure 5b) [19,20]. To determine whether IRCR201 could inhibit the

IRCR201 Impedes Tumor Growth and Induces Cellular Apoptosis
In vitro functional inhibition activity of IRCR201 was analyzed in various cancer types. IRCR201 did not alter cancer cell growth of the low c-Met-expressing cell line MCF7 (Figures 1f and 5a). In U87MG (an HGF-dependent GBM cell line), IRCR201 showed more potent cancer cell growth inhibition compared to huOA5D5.v2, which has previously been reported to suppress only HGF-dependent cellular proliferation ( Figure 5b) [19,20]. To determine whether IRCR201 could inhibit the tumor growth of cancer cells with c-Met amplification and constitutively phosphorylate c-Met activity in the absence of HGF, an MKN45 gastric cancer cell line was employed and showed significantly reduced IRCR201-mediated cellular proliferation compared to huOA5D5.v2 ( Figure 5c). Additionally, we used an A549 lung cancer cell line, which has been previously demonstrated to secrete no detectable level of HGF but could potentially promote c-Met activation in response to HGF, for further assessment of cellular proliferation in the presence of IRCR201. In the absence of HGF, IRCR201 displayed more potent cancer cell growth inhibition in A549 compared to huOA5D5.v2 ( Figure 5d). Furthermore, the inhibitory effect of HGF-induced cancer cell growth of IRCR201 was superior to that of huOA5D5.v2 (Figure 5e). We also confirmed that IRCR201 could dramatically induce cellular apoptosis in c-Met-expressing cancer cell lines except for MCF7 (c-Met expression: low) (Figure 5f-i).
for further assessment of cellular proliferation in the presence of IRCR201. In the absence of HGF, IRCR201 displayed more potent cancer cell growth inhibition in A549 compared to huOA5D5.v2 ( Figure 5d). Furthermore, the inhibitory effect of HGF-induced cancer cell growth of IRCR201 was superior to that of huOA5D5.v2 (Figure 5e). We also confirmed that IRCR201 could dramatically induce cellular apoptosis in c-Met-expressing cancer cell lines except for MCF7 (c-Met expression: low) (Figure 5f-i).

IRCR201 Induces Rapid c-Met Internalization and Lysosomal Degradation
To elucidate the cellular growth inhibitory mechanism of IRCR201 in c-Met-expressing cancer cell lines, we measured the change in total c-Met protein levels according to antibody treatment time in various cancer cell lines using ELISA. In multiple cancer cell lines, the total c-Met protein level was rapidly attenuated after 15 min from the initial treatment with IRCR201 (Figure 6a-c). Four hours Figure 5. In vitro potency of IRCR201. All results are shown as the mean ± standard error of mean (SEM) from triplicate treatments. (a-d) Inhibitory effect of IRCR201 on cancer cell proliferation. MCF7, U87MG, MKN45, and A549 cells were treated with IRCR201, huOA5D5.v2, or human IgG control for 72 h. Cell proliferation was measured using CellTiter Glo ® (Promega); (e) HGF-induced growth inhibition by IRCR201. The inhibitory effect of IRCR201 and huOA5D5.v2 on cell growth was examined under the condition of the addition of 50 ng/mL HGF in A549, an HGF-dependent cell. After 72 h of antibody treatment, the number of cells was measured with CellTiter Glo ® (Promega); (f-i) Apoptosis assay. MCF7, U87MG, MKN45, and A549 cells were treated with IRCR201 or human IgG for 24 h. Apoptosis activity was detected with caspase-3/7 activity.

IRCR201 Induces Rapid c-Met Internalization and Lysosomal Degradation
To elucidate the cellular growth inhibitory mechanism of IRCR201 in c-Met-expressing cancer cell lines, we measured the change in total c-Met protein levels according to antibody treatment time in various cancer cell lines using ELISA. In multiple cancer cell lines, the total c-Met protein level was rapidly attenuated after 15 min from the initial treatment with IRCR201 (Figure 6a-c). Four hours after the IRCR201 treatment, the total c-Met level was reduced to 52.4, 60.0, and 56.8% in U87MG, A549, and MKN45, respectively (Figure 6a-c). We further analyzed the molecular interaction between IRCR201 and c-Met receptors on the cell surface in various c-Met-overexpressed cancer cell lines. Our results highlighted the decomposition of cell surface c-Met via IRCR201 treatment (Figure 6d-f). The resulting surface c-Met levels 4 h post-treatment were 35.4%, 37.0%, and 44.5% in U87MG, A549, and MKN45, respectively. Immunocytochemistry analysis was performed on MKN45 to clarify whether the reduced level of c-Met on the cell surface was due to the degradation of antibody-mediated internalization. The results showed that IRCR201 had progressed into the cytoplasmic area of MKN45, eventually exhibiting co-localization with lysosomal-associated membrane protein 1 (LAMP1, a lysosomal marker), inducing the efficient degradation of cell surface c-Met (Figure 6g). To investigate the underlying mechanism behind the IRCR201-mediated c-Met degradation, we analyzed the c-Met degradation patterns of IRCR201 or 5D5 in the presence of a proteasome inhibitor or lysosome inhibitor. Numerous studies have reported that the binding of HGF or 5D5 with c-Met induces internalization of the ligand/receptor complex and initiates the degradation of receptor tyrosine kinase (RTK) through the ubiquitin-proteasome pathway [24,[36][37][38]. IRCR201 dramatically decomposed c-Met in A549 under specific conditions where MG132 and lactacystin (proteasome inhibitors) were first treated, but 5D5 failed to degrade c-Met (Figure 7a,b). Treatment with concanamycin A (a lysosome inhibitor) failed to induce IRCR201-mediated c-Met degradation in A549, confirming that IRCR201-induced c-Met depletion is mediated by the lysosomal degradation pathway (Figure 7c,d). Our results further showed that the 5D5 antibody could induce c-Met degradation, whether in the presence or absence of concanamycin A (Figure 7c,d). In conclusion, these results indicate that IRCR201 mediates endogenous c-Met depletion through the lysosomal degradation pathway.
resulting surface c-Met levels 4 h post-treatment were 35.4%, 37.0%, and 44.5% in U87MG, A549, and MKN45, respectively. Immunocytochemistry analysis was performed on MKN45 to clarify whether the reduced level of c-Met on the cell surface was due to the degradation of antibody-mediated internalization. The results showed that IRCR201 had progressed into the cytoplasmic area of MKN45, eventually exhibiting co-localization with lysosomal-associated membrane protein 1 (LAMP1, a lysosomal marker), inducing the efficient degradation of cell surface c-Met (Figure 6g). To investigate the underlying mechanism behind the IRCR201-mediated c-Met degradation, we analyzed the c-Met degradation patterns of IRCR201 or 5D5 in the presence of a proteasome inhibitor or lysosome inhibitor. Numerous studies have reported that the binding of HGF or 5D5 with c-Met induces internalization of the ligand/receptor complex and initiates the degradation of receptor tyrosine kinase (RTK) through the ubiquitin-proteasome pathway [24,[36][37][38]. IRCR201 dramatically decomposed c-Met in A549 under specific conditions where MG132 and lactacystin (proteasome inhibitors) were first treated, but 5D5 failed to degrade c-Met (Figure 7a,b). Treatment with concanamycin A (a lysosome inhibitor) failed to induce IRCR201-mediated c-Met degradation in A549, confirming that IRCR201-induced c-Met depletion is mediated by the lysosomal degradation pathway (Figure 7c,d). Our results further showed that the 5D5 antibody could induce c-Met degradation, whether in the presence or absence of concanamycin A (Figure 7c,d). In conclusion, these results indicate that IRCR201 mediates endogenous c-Met depletion through the lysosomal degradation pathway.  HGF-dependent lung cancer cell line. In the absence of HGF, IRCR201 exhibited exceptional c-Met degradation, subsequently inhibiting phosphorylation of c-Met downstream pathway molecules including Akt and Erk1/2 (Figure 7f). In the presence of HGF, IRCR201 treatment not only promoted c-Met degradation, but also inhibited the phosphorylation of c-Met, Akt, and Erk1/2 ( Figure 7f). Overall, our results demonstrated that IRCR201 can efficiently interfere with the HGF/c-Met axis, regardless of HGF-dependent or -independent growth mechanisms.

IRCR201 Impedes Tumor Growth In Vivo
In vivo anti-tumor activity of IRCR201 was observed in an A549 NSCLC subcutaneous model. Our results showed that 3 mg/kg IRCR201 treatment (48.4% tumor inhibition) could potently inhibit tumor growth, while 3 mg/kg huOA5D5.v2 treatment (13.3% tumor inhibition) showed a similar tumor growth rate compared to the PBS-treated group (Figure 8a). In addition, we further analyzed whether IRCR201 could significantly inhibit c-Met-amplified tumor models. The c-Met-amplified gastric cancer cell line MKN45 was subcutaneously inoculated into BALB/c-nu mice. IRCR201 was injected intravenously twice a week at 3 or 15 mg/kg, and significant inhibition in tumor growth was

IRCR201 Suppresses the c-Met Signaling Pathway via the Degradation of c-Met
Immunoblot analysis of HGF-dependent (U87MG) and HGF-independent c-Met-amplified (MKN45) cell lines were conducted to analyze the inhibitory effect of IRCR201 on the c-Met signaling pathway. In both models, IRCR201 efficiently depleted total c-Met levels, and inhibited c-Met phosphorylation and its downstream signaling pathway (Figure 7e). In addition, we further confirmed the inhibition of the c-Met pathway by IRCR201 under the presence of HGF in A549, an HGF-dependent lung cancer cell line. In the absence of HGF, IRCR201 exhibited exceptional c-Met degradation, subsequently inhibiting phosphorylation of c-Met downstream pathway molecules including Akt and Erk1/2 (Figure 7f). In the presence of HGF, IRCR201 treatment not only promoted c-Met degradation, but also inhibited the phosphorylation of c-Met, Akt, and Erk1/2 (Figure 7f). Overall, our results demonstrated that IRCR201 can efficiently interfere with the HGF/c-Met axis, regardless of HGF-dependent or -independent growth mechanisms.

IRCR201 Impedes Tumor Growth In Vivo
In vivo anti-tumor activity of IRCR201 was observed in an A549 NSCLC subcutaneous model. Our results showed that 3 mg/kg IRCR201 treatment (48.4% tumor inhibition) could potently inhibit tumor growth, while 3 mg/kg huOA5D5.v2 treatment (13.3% tumor inhibition) showed a similar tumor growth rate compared to the PBS-treated group (Figure 8a). In addition, we further analyzed whether IRCR201 could significantly inhibit c-Met-amplified tumor models. The c-Met-amplified gastric cancer cell line MKN45 was subcutaneously inoculated into BALB/c-nu mice. IRCR201 was injected intravenously twice a week at 3 or 15 mg/kg, and significant inhibition in tumor growth was observed compared to the PBS-treated group (Figure 8b). The 3 mg/kg IRCR201-treated group (56.3% tumor inhibition) showed significant reduction in tumor volume compared to the 3 mg/kg huOA5D5.v2-treated group (29.1% tumor inhibition) (Figure 8b). When we increased the IRCR201 dose to 15 mg/kg (81.5% tumor inhibition), the tumor growth rate was significantly diminished compared to the 3 mg/kg IRCR201-treated group, portraying a dose-dependent reduction of tumor volume compared to the vehicle-treated group (Figure 8b). 18,1968 10 of 22 observed compared to the PBS-treated group (Figure 8b). The 3 mg/kg IRCR201-treated group (56.3% tumor inhibition) showed significant reduction in tumor volume compared to the 3 mg/kg huOA5D5.v2-treated group (29.1% tumor inhibition) (Figure 8b). When we increased the IRCR201 dose to 15 mg/kg (81.5% tumor inhibition), the tumor growth rate was significantly diminished compared to the 3 mg/kg IRCR201-treated group, portraying a dose-dependent reduction of tumor volume compared to the vehicle-treated group (Figure 8b).

IRCR201 Inhibits Tumor Cell Proliferation and Angiogenesis through the Downregulation of c-Met
Based on the inhibitory effects of IRCR201 in vitro and in vivo, the cellular mitotic index was further evaluated through immunohistochemical analysis of Ki-67-positive nuclei in the paraffin sections of MKN45 xenograft models that had been treated with either PBS (vehicle) or 10 mg/kg IRCR201 and harvested after 48 h. Immunohistochemistry (IHC) staining showed that Ki-67 positive cells were significantly reduced in the IRCR201-treated group compared to the PBS-treated group (Figure 9a). Additionally, immunohistochemical analysis in MKN45 tumor tissues demonstrated a significant increase of IRCR201-triggered terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL)-positive apoptotic cells (Figure 9b). To investigate whether IRCR201 could induce c-Met degradation and phospho-c-Met inhibition in vivo, both total c-Met and phospho-c-Met levels were evaluated. Our results showed reduced expression of both total c-Met and phospho-c-Met when administered with IRCR201 at 10 mg/kg (Figure 9c,d). To investigate the status of downstream mediators in HGF/c-Met signaling pathway by IRCR201 treatment, we evaluated phosphorylation of Akt and Erk1/2 in PBS-or 15 mg/kg IRCR201-treated MKN45 tumor sections through immunohistochemical analysis (Figure 9e,f). The data demonstrated that IRCR201 significantly abrogated the phosphorylation of Akt and Erk1/2 in vivo (Figure 9e,f). To assess the anti-angiogenic effects of IRCR201, paraffin sections of MKN45 tumors were immuno-stained with platelet endothelial cell adhesion molecule 1 (PECAM1, also known as CD31), an endothelial cell marker (Figure 9g). The number of PECAM1-positive cells was drastically reduced in MKN45 tumors treated with 15 mg/kg IRCR201 compared to the PBS-treated group (Figure 9g).

IRCR201 Inhibits Tumor Cell Proliferation and Angiogenesis through the Downregulation of c-Met
Based on the inhibitory effects of IRCR201 in vitro and in vivo, the cellular mitotic index was further evaluated through immunohistochemical analysis of Ki-67-positive nuclei in the paraffin sections of MKN45 xenograft models that had been treated with either PBS (vehicle) or 10 mg/kg IRCR201 and harvested after 48 h. Immunohistochemistry (IHC) staining showed that Ki-67 positive cells were significantly reduced in the IRCR201-treated group compared to the PBS-treated group (Figure 9a). Additionally, immunohistochemical analysis in MKN45 tumor tissues demonstrated a significant increase of IRCR201-triggered terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL)-positive apoptotic cells (Figure 9b). To investigate whether IRCR201 could induce c-Met degradation and phospho-c-Met inhibition in vivo, both total c-Met and phospho-c-Met levels were evaluated. Our results showed reduced expression of both total c-Met and phospho-c-Met when administered with IRCR201 at 10 mg/kg (Figure 9c,d). To investigate the status of downstream mediators in HGF/c-Met signaling pathway by IRCR201 treatment, we evaluated phosphorylation of Akt and Erk1/2 in PBS-or 15 mg/kg IRCR201-treated MKN45 tumor sections through immunohistochemical analysis (Figure 9e,f). The data demonstrated that IRCR201 significantly abrogated the phosphorylation of Akt and Erk1/2 in vivo (Figure 9e,f). To assess the anti-angiogenic effects of IRCR201, paraffin sections of MKN45 tumors were immuno-stained with platelet endothelial cell adhesion molecule 1 (PECAM1, also known as CD31), an endothelial cell marker (Figure 9g).

Discussion
Aberrant activation of the HGF/c-Met signaling pathway has been reported in numerous human cancers [1,2]. In c-Met dysregulated malignancy, tumor progression is facilitated largely by three mechanisms: ligand-dependent c-Met activation; genomic amplification; and oncogenic mutations [5][6][7][8][9]. Additionally, c-Met amplification has been associated with acquired resistance to epidermal growth factor receptor (EGFR)-and vascular endothelial growth factor (VEGF)-targeted therapies [39,40]. Therefore, targeting the HGF/c-Met signaling axis could offer a promising therapeutic approach for the treatment of c-Met-expressing tumors. In the present study, we demonstrated the development and characterization of IRCR201-novel fully-human anti-c-Met IgG1 which

Discussion
Aberrant activation of the HGF/c-Met signaling pathway has been reported in numerous human cancers [1,2]. In c-Met dysregulated malignancy, tumor progression is facilitated largely by three mechanisms: ligand-dependent c-Met activation; genomic amplification; and oncogenic mutations [5][6][7][8][9]. Additionally, c-Met amplification has been associated with acquired resistance to epidermal growth factor receptor (EGFR)-and vascular endothelial growth factor (VEGF)-targeted therapies [39,40]. Therefore, targeting the HGF/c-Met signaling axis could offer a promising therapeutic approach for the treatment of c-Met-expressing tumors. In the present study, we demonstrated the development and characterization of IRCR201-novel fully-human anti-c-Met IgG1 which specifically binds to a distinct epitope on c-Met and effectively disrupts the c-Met signaling pathway. In contrast to previously developed c-Met-targeting antibodies with lack of mouse cross-reactivity, IRCR201 binds to both human and mouse c-Met with high affinity. Cross-reactivity with the mouse c-Met ortholog enables the precise evaluation of the tumor inhibitory efficacy of IRCR201 in mouse xenograft models during preclinical studies.
IRCR201 does not mimic the role of HGF and exhibits lower agonist activity compared to huOA5D5.v2 as it binds to the SAPPFVQ amino acid sequence of the PSI domain, as distinct from the Sema domain, which is an HGF binding site of c-Met. The PSI domain containing the SAPPFVQ sequence-the IRCR201 epitope-acts as a domain that gives flexibility to c-Met, promoting molecular interaction between c-Met and HGF [41,42]. Although the biological insights of the PSI domain remain relatively unexplored, IRCR201 interacts with the PSI domain of c-Met, thus resulting in an inhibitory effect without agonist activity. Basilico and colleagues also developed specific antibodies that bind to the PSI domain and disrupt the interaction between HGF and c-Met [43], resulting in the inhibition of HGF-induced c-Met autophosphorylation, distinct from IRCR201, which does not impair the interaction between c-Met and HGF. Previous studies have reported that bivalent c-Met-targeting antibodies with different epitopes could elicit different levels of biochemical functional activity [18,43,44]. The differential epitope of c-Met antibodies may account for the distinct functional activity of IRCR201.
IRCR201 possesses a different binding epitope from previously developed c-Met inhibitory antibodies, and rapidly promoted c-Met depletion primarily through the lysosomal degradation pathway, thereby abrogating downstream signaling cascades in U87MG, A549, and MKN45. To further ascertain the effect of IRCR201 at a cell surface receptor level, a flow cytometry assay was configured to detect cell surface c-Met of U87MG, A549, and MKN45 following the antibody treatment. Similar results were observed throughout various cell lines, as the rapid degradation of cell surface c-Met was observed at the 15-min mark. Total receptor expression levels in the membrane were decreased by approximately 40% post IRCR201 treatment. Furthermore, we found that IRCR201 resulted in c-Met co-localization within the lysosome, suggesting an underlying mechanism behind IRCR201-induced c-Met degradation. The c-Met depletion activity of IRCR201 was inhibited by concanamycin A, supporting the notion that IRCR201 deteriorates c-Met through the lysosomal degradation pathway. These findings demonstrate that IRCR201 induces receptor-mediated internalization and lysosomal degradation, providing a potential underlying mechanism behind the loss of total c-Met protein.
IRCR201 does not inhibit the interaction between HGF and c-Met, but it exhibited excellent growth inhibition capacity in various types of c-Met-expressing cancer cell lines compared to huOA5D5.v2 in vitro. IRCR201-induced c-Met degradation demonstrated a therapeutic effect mediated by the downregulation of tyrosine kinase activity and the subsequent inhibition of cellular proliferation in U87MG, A549, and MKN45. IRCR201 also showed significant antitumor effect in an A549 NSCLC xenograft tumor model. Since c-Met amplification leads to shorter survival in patients with gastric cancer and NSCLC [45,46], we assessed the tumor inhibitory capability of IRCR201 in a c-Met-amplified tumor model to provide clinical benefit in patients. Treatment with IRCR201 dramatically decreased tumor growth in the c-Met-amplified MKN45 gastric xenograft model, suggesting that it may portray antitumor activity in a broader range of cancer classes with constitutive c-Met activation via genomic amplification or mutations. These results suggest that IRCR201 could be a promising therapeutic agent to inhibit tumor growth driven by constitutively active c-Met through overexpression, gene amplification, or genomic mutations.
The HGF/c-Met signaling pathway is a pivotal component of tumor-associated angiogenesis that leads to tumor progression and metastasis via a sufficient supply of oxygen and nutrients through blood vessels [1,2]. Immunohistochemistry analysis also revealed that IRCR201 significantly inhibited the proliferation of tumor cells and angiogenesis of tumor-associated blood vessels, suggesting that IRCR201 suppresses the HGF/c-Met pathway in the tumor microenvironment.
By impairing both HGF-dependent and c-Met-amplified downstream activation in various cancer types, IRCR201 is differentiated from other therapeutic antibodies by securing a wide range of drug response groups. The anti-cancer activity of IRCR201 may be further enhanced by antibody-mediated cell cytotoxicity (ADCC), as the antibody's isotype is human IgG1, which binds to the Fc receptors on immune cells. IRCR201 is also applicable to antibody-drug conjugate (ADC) platforms for toxin delivery, as it demonstrated the lysosomal degrading traits of c-Met.
This study currently has an accompanying research program to investigate the specific degradation pathway responsible for IRCR201-induced c-Met depletion. IRCR201 is also being used in toxicological studies using mouse models to confirm that its mouse cross-reactivity has an adverse effect on mouse health.
In summary, the newly identified traits of IRCR201 allow the evaluation of precise therapeutic efficacy in mouse models. Our comprehensive results show that IRCR201 is capable of inhibiting a variety of cancer types with HGF-dependent c-Met activation or gene amplification-driven constitutive c-Met activation. Taken together, IRCR201 represents a promising therapeutic antibody to treat cancer patients suffering from dysregulation of the HGF/c-Met signaling pathway.

Cells and Cell Cultures
The

Surface Plasmon Resonance Assay
The binding affinities and kinetics of IRCR201 for human c-Met and mouse c-Met were measured using a Biacore™ T100 (GE Healthcare Life Sciences, Uppsala, Sweden). Human c-Met, mouse c-Met, or BSA were immobilized on a CM5 sensor chip (GE Healthcare Life Sciences, BR100530, Uppsala, Sweden) using an Amine coupling kit (GE Healthcare Life Sciences, BR100050, Uppsala, Sweden). Different concentrations of purified IRCR201 were injected into immobilized human and mouse c-Met to determine K D values. To obtain the kinetic and affinity constants of IRCR201 to human and mouse c-Met, Biacore™ T100 evaluation software (GE Healthcare Life Sciences, Uppsala, Sweden) was used.

Epitope Mapping
The epitope mapping of IRCR201 to c-Met was performed using the scFv format of IRCR201. The scFv of IRCR201 was produced using TOP10F' Escherichia coli (E. coli) (Invitrogen, C303003, Carlsbad, CA, USA) and purified from the bacterial lysates by immobilized metal affinity chromatography (IMAC) using nickel-nitrilotriacetic acid (Ni-NTA) agarose (QIAGEN, 30210, Hilden, Germany) according to the instructions of the manufacturer. The c-Met 15-mer peptides corresponding to 1-932 of the c-Met extracellular domain were synthesized at 5 nmol/spot and covalently bound to a Whatman ® 50 cellulose support by the C-terminus (JPT Peptide Technologies GmbH, Berlin, Germany). The membrane of c-Met 15-mer peptides was blocked with 5% skim milk for 3 h at room temperature with shaking. The membrane was incubated with the scFv of IRCR201 in 5% skim milk at 4 • C overnight. The scFv of IRCR201 bound to the c-Met 15-mer peptides was transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, IPVH00010, Billerica, MA, USA) using the electroblotting method. The transferred PVDF membranes were washed in 1× TBST (tris-buffered saline with Tween 20) twice for 10 min. The PVDF membrane was blocked with 5% skim milk for 3 h and incubated with anti-HA-peroxidase (Roche Diagnostics, 12013819001, Mannheim, Germany) for 2 h with agitation. The membrane was washed three times with 1× TBST for 5 min and incubated with enhanced chemiluminescence (ECL) detection reagents (GE Healthcare Life Sciences, RPN2106, Little Chalfont, UK) for about 1 min. A sheet of film on the membrane was placed into a film cassette in a darkroom, according to the manufacturer's instructions. The intensity of the bands was quantified by Multi Gauge software V3.0 (Fuji photo film Co., Ltd., Tokyo, Japan).

Protein Modeling and Docking
The sequences of the V H and V L segment of IRCR201 were assembled in the scFv format by homology modeling using the Rosetta-based computational homology modeling technique [31]. The IRCR201/c-Met docking model was made using the ZDOCK server [32]. Structure models were analyzed using PyMOL (DeLano Scientific LLC, Palo Alto, CA, USA).

Agonism Analysis
Wells of 96-well tissue culture plates (Costar #3595, Corning, NY, USA) were seeded with 5000 Caki1 cells in RPMI1640 medium supplemented with 10% (v/v) FBS. After culture for 24 h, cells were starved in serum-free RPMI1640 medium for another 24 h. The cells were then cultured in the presence of anti-c-Met antibodies and HGF in serum-free medium for 30 min at 37 • C. Next, the medium was removed and the cells were washed once with 1× PBS (pH 7.4). Cells were lysed and p-Akt levels were quantified by PathScan ® phospho-Akt1 (Ser473) chemiluminescent sandwich ELISA kit (Cell Signaling Technology, #7134, Danvers, MA, USA) according to the manufacturer's protocol.
To clarify agonistic activity of IRCR201, Caki1 cells were starved in serum-free RPMI 1640 medium for 24 h and then treated with HGF or c-Met antibodies for 30 min. Cells were lysed and then the Western blotting experiments were proceeded. To analyze the total c-Met degradation of c-Met-expressing cells by IRCR201, an ELISA-based quantification method was used. When U87MG, A549, and MKN45 cells in 96-well cell culture plates (Costar, #3595, Corning, NY, USA) reached approximately 70% confluency, cells were treated with 100 nM of IRCR201 for 0 min, 15 min, 30 min, 60 min, 90 min, and 120 min. After washing with 1× PBS (pH 7.4), cells were resuspended with lysis buffer (Roche Diagnostics, 04719956001, Mannheim, Germany) supplemented with protease inhibitor cocktail tablets (Roche Diagnostics, 04719956001, Mannheim, Germany) and phosphatase inhibitor (Roche Diagnostics, 4906845001, Mannheim, Germany). The changes in c-Met protein were analyzed in a timely manner by a human HGFR/c-MET DuoSet ® ELISA kit (R&D systems, DY358, Minneapolis, MN, USA) according to the manufacturer's protocol.

Proliferation Assay
MCF7, U87MG, A549, and MKN45 cells were seeded into 96-well white plates (Costar, #3610, Corning, NY, USA) at a density of 3,000 cells/well and cultured for 72 h with 0-100 nM of IRCR201, huOA5D5.v2, or human IgG control at 37 • C in a humidified 5% CO 2 atmosphere. The number of viable cells was estimated by a CellTiter-Glo ® luminescent cell viability assay kit (Promega, G7573, Madison, WI, USA). The luminescent signal intensity of the plates was read by an Infinite ® m200 pro (Tecan, Männedorf, Switzerland) and normalized to the untreated group.

Caspase 3/7 Activity Assay
The caspase 3/7 activity was detected using the Caspase-Glo ® 3/7 Luciferase assay kit (Promega, G8091, Madison, WI, USA) in accordance with the manufacturer's recommendations. Cells in appropriate complete medium were incubated for 24 h with the untreated group, Human IgG, or IRCR201. After the addition of Caspase-Glo ® 3/7 reagent, the luminescent signal intensity was measured by an Infinite ® m200 pro (Tecan, Männedorf, Switzerland) and normalized to the untreated group.

In Vivo Therapeutic Efficacy
The efficacy of IRCR201 was evaluated in various tumor xenograft models. A549 and MKN45 single cells suspended in Hank's Balanced Salt Solution (HBSS; Gibco, 24020117, Carlsbad, CA, USA) with Matrigel ® (Corning, 354234, Lowell, MA, USA) were subcutaneously injected into the right flank region of female BALB/c-nu mice (1 × 10 6 cells per mouse in 100 µL). When the mean tumor volume reached approximately 150 mm 3 (day 0), animals were randomized according to tumor volume to minimize intragroup and intergroup variation. After regrouping, IRCR201, huOA5D5.v2, or vehicle was intravenously administered twice per week. Each treatment group consisted of six mice. Tumor volumes were measured using three-dimensional calipers. Body weights were measured as general confirmation of toxicity.

Immunohistochemistry
For the immunostaining of Ki-67, c-Met, and phospho-c-Met, phospho-Akt, phospho-Erk1/2, and PECAM1, antigen retrieval was processed in formalin-fixed and subsequently paraffin-embedded MKN45 tumor tissues. After blocking with 5% bovine serum albumin in 1× PBS (pH 7.4) for 1 h at room temperature, the sections were incubated overnight with primary antibodies at 4 . Detection and visualization of the tumor sections were performed using the avidin/biotin/peroxidase complex (Vector Laboratories, PK-4000, Burlingame, CA, USA) and 3,3 -diaminobenzidine (Invitrogen, 750118, Carlsbad, CA, USA). For TUNEL assay, the ApopTag ® Peroxidase In Situ Apoptosis Detection Kit (Millipore, S7100, Billerica, MA, USA) was used according to the manufacturer's protocol. The IHC images of the tumor sections were obtained using the Aperio imaging system (Leica Biosystems, Wetzlar, Germany).

Statistical Analysis
All data were analyzed with GraphPad Prism ® V5.01 (GraphPad Software, Inc., La Jolla, CA, USA) and expressed as the mean ± SEM unless otherwise noted. The statistical significance in tumor growth between different groups was analyzed by one-tailed two sample t-test.

Conclusions
IRCR201 exhibits tumor growth inhibitory activity in various cancer types harboring HGF-dependent or HGF-independent/c-Met-amplified activation.
Author Contributions: Hyunkyu Park conceived and designed the experiments; Hyunkyu Park, Donggeon Kim, Eunmi Kim, Suji Yu, and Jiwon Oh performed the experiments; Hyunkyu Park and Donggeon Kim analyzed the data; Hyunkyu Park and Eunmi Kim contributed antibody screening/production/purification; Hyunkyu Park wrote and revised the manuscript; Jason K. Sa and Hee Won Lee revised the manuscript; Seok-Hyung Kim reviewed the manuscript; Do-Hyun Nam and Yeup Yoon supervised the experiments. All authors read and approved the final version of the manuscript.

Conflicts of Interest:
The authors declare no conflict of interest.