Substituted Syndecan-2-Derived Mimetic Peptides Show Improved Antitumor Activity over the Parent Syndecan-2-Derived Peptide

We previously showed that a synthetic peptide (S2-P) corresponding to a portion of the human syndecan-2 (SDC2) sequence can bind to the pro-domain of matrix metalloproteinase-7 (MMP-7) to inhibit colon cancer activities. Since S2-P had a relatively weak binding affinity for the MMP-7 pro-domain, we herein modified the amino acid sequence of S2-P to improve the anticancer potential. On the basis of the interaction structure of S2-P and MMP-7, four peptides were generated by replacing amino acids near Tyr 51, which is critical for the interaction. The SDC2-mimetic peptides harboring an Ala-to-Asp substitution at the C-terminal side of Tyr 51 (S2-D) or with an Ala-to-Phe substitution at the N-terminal side of Tyr 51 and an Ala-to-Asp substitution at the C-terminal side of Tyr 51 (S2-FE) showed improved interaction affinities for the MMP-7 pro-domain. Compared to S2-P, S2-FE was better able to inhibit the SDC2–MMP-7 interaction, the cell surface localization of MMP-7, the gelatin degradation activity of MMP-7, and the cancer activities (cell migration, invasion, and colony-forming activity) of human HCT116 colon cancer cells in vitro. In vivo, S2-FE inhibited the primary tumor growth and lung metastasis of CT26 mouse colon cancer cells in a xenograft mouse model. Together, these data suggest that S2-FE could be useful therapeutic anticancer peptides for colon cancer.


Introduction
The syndecans (SDCs) comprise a family of highly conserved heparan sulfate proteoglycans that act as cell surface receptors; they bind to the extracellular matrix (ECM) and thereby regulate cell adhesion and related cell functions [1,2]. Although there is some functional redundancy, all syndecan family members play distinct functional roles. Among them, SDC2 has specific functions in colorectal cancer [3]. Previous studies have shown that SDC2 is highly expressed in colon cancer cell lines [4,5] compared to normal cell lines and in cancer tissues [6] compared to neighboring normal tissues, and that this plays an important role in regulating colon cancer activities [3,7]. As seen for other heparan sulfate proteoglycans, SDC2 communicates with the ECM via an extracellular heparan-binding domain [8]. This domain of SDC2 directly interacts with ECM proteins, such as fibronectin [9] and integrin α2 [10] in order to regulate intracellular signaling. Through this regulation, increased SDC2 promotes the cell adhesion and spreading of colon cancer cells [4]. SDC2 also provides docking sites for various ligands, such as growth factors and proteases, to regulate extracellular events. For example, matrix metalloproteinase-7 (MMP-7) binds to the extracellular domain of SDC2 and subsequently undergoes activation in colon cancer cells [11].
The MMPs are a large family of calcium-dependent zinc-containing endopeptidases that are responsible for tissue remodeling and ECM degradation [12]. This proteolytic activity of MMPs is crucial for the ability of cancer cells to exhibit increased cell growth, metastasis, and angiogenesis [13]. Therefore, many efforts have been made to develop agents that can inhibit the proteolytic activity of MMPs for use as anticancer drugs. Early efforts to develop MMP inhibitors mainly focused on compounds that mimicked the natural peptide substrate of the desired MMP by containing a catalytic zinc ion and a group that can chelate the backbone [14,15]. Although hydroxamate-based chelation inhibitors (e.g., batimastat) showed promising antitumor effects in animal models of various cancers, clinical trials have been unsuccessful due to severe side effects. MMP-7 is the smallest MMP; its expression is associated with the development and metastasis of colon cancer [16,17]. MMP-7 and the other MMPs are synthesized and secreted as a pro-enzyme (zymogen), and their cleavage-mediated activation is essential for their proteolytic activity. MMP-7 (and the other MMPs) contains two domains, the pro-domain and the catalytic domain. The pro-domain has free cysteine residues that interfere with the activity of latent pro-MMP-7 by interacting with a zinc ion in the catalytic domain, which exhibits enzymatic activity upon cleavage of the pro-domain [18,19]. The active form of MMP-7 degrades an ECM complex that includes a heparan sulfate proteoglycan [20] and perlecan [21] to promote tumor cell migratory phenotypes and metastasis. MMP-7 also cleaves cell surface receptors, such as E-cadherin, to promote cancer cell invasion [22][23][24].
In addition to their pro-domain-cleavage-mediated activation, the cell surface localization of MMPs is an important step for their activation and substrate access. For instance, pro-MMP-2 localizes to the cell surface by interacting with the membrane MMP, MT1-MMP, and tissue inhibitor of metalloproteinase-2 (TIMP-2). On the cell surface, pro-MMP-2-TIPM-2-MT1-MMP induces cleavage of the MMP-2 pro-domain, leading to the activation of MMP-2 [25]. Another surface-localized MMP is MMP-1, which localizes at the front of the invasive edge of the cancer cell and cleaves ECM components, such as type I collagen, in human lung carcinoma [26] and pancreatic cancer [27].
A previous study showed that SDC2 acts as a cell surface docking receptor of MMP-7 [6]. In the ECM, secreted pro-MMP-7 binds to the extracellular domain of SDC2 core protein at the cell surface [11]. Tyrosine residue 51 (Tyr 51) of the extracellular domain of SDC2 occupies the binding pocket of pro-MMP-7, which is formed by the α2 helix-loop-α3 loop of the MMP-7 pro-domain; through this direct interaction, pro-MMP-7 is converted to active MMP-7 [11]. Thus, the SDC2-MMP-7 interaction appears to be important for the enzymatic activity of MMP-7 in colon cancer cells. Interestingly, a synthetic peptide (S2-P) corresponding to a portion of the SDC2 sequence bound the pro-domain of MMP-7 to interrupt the SDC2-MMP-7 interaction and block the activation of pro-MMP-7, and this inhibited the tumorigenic activities of colon cancer cells [11]. These results suggested that MMP-7 activity could be decreased by inhibiting its cell surface localization (and thereby its activation), and that S2-P could potentially act as an anticancer peptide. In this context, we sought to further investigate the potential of S2-P-derived SDC2-mimetic peptides as anticancer therapeutics that could potentially replace hydroxamate-based chelating inhibitors. We have previously reported that that the N-terminal extracellular domain of SDC2 interacts with MMP-7 through its pro-domain [6]. In this study, we further determined whether this interaction is sufficient for cell surface localization of MMP-7 ( Figure 1). As previously reported [28], human HT29 colon adenocarcinoma cells stably expressing SDC2 (HT29-SDC2) showed upregulation of the mRNA expression of MMP-7 ( Figure S1), and their conditioned media showed increased MMP-7 enzymatic activity to cleave an exogenous substrate (Mca-Pro-Leu-Gly-Leu-Dap(Dnp)-Ala-Arg-NH2, Mca-PLA-Nva-Dap(Dnp)-AR-NH2) ( Figure 1A). Moreover, HT29-SDC2 cell showed increased cell surface localization of MMP-7 ( Figure 1B), confirming that SDC2 regulates the cell surface localization of MMP-7 in colon cancer cells. Data are shown as mean ± S.D. (n = 3); **, p < 0.01 versus MMP-7 activity of HT29-VEC. (B) Cells were incubated with anti-SDC2 antibody and stained with a Texas-Red-conjugated secondary antibody or incubated with anti-MMP-7 antibody and stained with a FITC-conjugated secondary antibody. Photographs were obtained under confocal microscopy. Scale bar: 10 µm. (C) HT29-SDC2 cells were plated on E-plates pre-coated with gelatin 10 µg (top) or 0.1% BSA (bottom) with the indicated amounts of His-tagged MMP-7 pro-domain (PDMMP-7). After 4 h, cells were photographed under phase-contrast microscopy, and cell attachment was monitored and plotted using the xCELLigence system. The mean of cell attachment ± S.D. is shown (n = 3); *, p < 0.05 versus 0 ng of PDMMP-7.
To further investigate whether SDC2 directly regulates the cell surface localization of MMP-7 through its interaction with pro-domain, xCelligence E-plate was coated with recombinant His-tagged pro-domain of MMP-7 (PDMMP-7) in the presence of either 10 µg of gelatin or 0.1 % BSA, and then cell adhesion of HT29-SDC2 cells seeded on the plates were analyzed ( Figure 1C). As expected, PDMMP-7 on the plate enhanced cell attachment on the E-plates in both cases ( Figure 1C), supporting the idea that MMP-7 directly interacts with cell surface SDC2 through the pro-domain.

Synthetic Peptide Derived from Syndecan-2 Inhibited Its Interaction with the MMP-7 Pro-Domain
It has been shown that the synthetic peptides (S2-P) corresponding to human SDC2 sequence bound to pro-domain of MMP-7 inhibits SDC2-mediated cancer activities [11], suggesting that S2-P could have potential as an anticancer peptide. However, S2-P had a relatively weak binding affinity for the MMP-7 pro-domain. Having found that Tyr 51 of S2-P was critical for the interaction [11], substitution of neighboring amino acid residues was applied in an effort to improve this interaction and possibly the anticancer activities ( Figure 2). In addition to Tyr 51, assessment of the electrostatic plus shape interaction scoring suggested that the S2-P peptide interacted with the pro-domain of MMP-7 through Asp 49 of S2-P, which is located in a cluster of acidic amino acid residues at the N-terminal side of Tyr 51 [11]. We thus hypothesized that the acidic residues near Tyr 51 might influence the interaction affinity. Therefore, we either added acidic amino acids or changed acidic amino acids to nonpolar amino acids near Tyr 51. We generated four modified peptides from the original sequence of S2-P (Figure 2A), on the basis of stabilizing energy score by the Single Amino Acid Mutation Change of Binding Energy (SAAMBE) method [29]. The acidic amino acid residue was added at the C-terminus of Tyr residue (S2-D, S2-FE, and S2-FLD) or the acidic amino acid at the N-terminus of Tyr 51 residue was replaced by non-polar amino acid (S2-L, S2-FE, and S2-FLD) to increase their interaction with the MMP-7 pro-domain ( Figure 2A). The ability of these peptides to physically interact with the MMP-7 pro-domain was subsequently evaluated using florescence spectrophotometer with purified recombinant SDC2 and a PDMMP-7 ( Figures 2B and S2). Compared to S2-P (Kd of 1.22 mM), S2-D, S2-FE, and S2-FLD showed greater binding activity to the pro-domain of MMP-7 (Kd of 0.55, 0.456, and 0.064 mM, respectively) ( Figure 2B). In contrast, the Asp-to-Leu replacement (S2-L) reduced binding. As the addition of an acidic amino acid at the C-terminal side of Tyr 51 improved the interaction with the MMP-7 pro-domain, these data suggest that this interaction is favored by the maintenance of acidic surroundings around Tyr 51. Since the synthetic peptides (e.g., S2-D, S2-FE, and S2-FLD) showed improved interaction with the MMP-7 pro-domain, we further examined whether these peptides competitively reduced the interaction of SDC2 with the MMP-7 pro-domain ( Figure 3). We first examined whether the modified peptides could inhibit the attachment of HT29-SDC2 cells to E-plates coated with the pro-domain ( Figure 3A). Among the four peptides, only S2-D and S2-FE reduced the attachment of HT29-SDC2 cells to the MMP-7 pro-domain to a degree stronger than that exhibited by S2-P ( Figure 3A). Thus, both S2-D and S2-FE appeared to have enhanced ability to block the interaction of SDC2 with MMP-7 pro-domain at the surface of HT29-SDC2 cells. Consistently, S2-D and S2-FE decreased the cell-surface level of MMP-7 ( Figure 3B) at the HT29-SDC2 cell surface in cells with similar expression levels of SDC2 and MMP-7 ( Figures 3C and S1). Together, these data suggest that the modified peptides, S2-D and S2-FE, efficiently inhibited the SDC2-mediated cell surface localization of MMP-7.
To further investigate the anticancer activity of S2-D and S2-FE, HCT116 human colon cancer cells, which are more invasive cancer cells than HT29, were treated with 50 nM of each peptide and monitored changes in various cancer activities ( Figure 5). As expected, the peptides did not exert cytotoxicity ( Figure 5A) or influence the proliferation of HCT116 cells ( Figure 5B). However, treatment of HCT116 cells with either S2-D or S2-FE decreased migration ( Figure 5C), invasion through Matrigel ( Figure 5D), and soft agar colony formation ( Figure 5E). While the inhibitory activities of S2-D and S2-FE on cell migration were similar to that of S2-P ( Figure 5C), S2-D showed inhibitory activities against invasion, and colony formation comparable to those of S2-P and S2-FE inhibited these parameters to an even greater degree in HCT116 cells ( Figure 5D,E). Consistently, treatment of HCT116 cells with either S2-D or S2-FE decreased their cell surface co-localization of SDC2 with MMP-7 ( Figure 5F). The values of Mander's overlap coefficient (MOC), which measures the degree of overlap between SDC2 and MMP-7, were ≈0.838 in untreated control cells, ≈0.834 in S2-P-treated cells, ≈0.815 in S2-D-treated cells, and ≈0.803 in S2-FEtreated cells. Since lower MOC values indicate less overlap between SDC2 and MMP-7, the results suggest that S2-FE has better inhibitory effects on SDC2 and MMP-7 colocalization than S2-P on the cell surface of HCT116 cells. Together, these data suggest that S2-FE has better anticancer activity than S2-P.

S2-D and S2-FE Inhibited MMP-7-Mediated ECM Degradation in Colon Cancer Cells
Since the interaction with SDC2 induces the processing of pro-MMP-7 into its active form [26], interrupting this interaction with a synthetic peptide would be expected to reduce MMP-7 activity. Therefore, we further investigated the effect of S2-FE on the MMP-7-mediated ECM degradation of HCT116 cells ( Figure 6). HCT116 cells were seeded on a plate coated with fluorescently labeled gelatin (a well-known substrate of MMP-7), and gelatin degradation was monitored under fluorescence microscopy. A total of 50 nM of the peptides were treated twice every 24 h. We did not detect any significant degradation of fluorescence-labeled gelatin by untreated HCT116 cells ( Figure 6, top left), but HCT116 cells pretreated with interleukin 1α (IL-1α), which induces MMP-7 expression [31], were found to promote gelatin degradation. This degradation was significantly reduced by the application of S2-D or S2-FE ( Figure 6, top right). These data suggest that S2-D and S2-FE can inhibit MMP-7 activation and thus MMP-7 activity at the cell surface of colon cancer cells.

The S2-FE Peptide Was Sufficient for Reducing Primary Tumor Growth and Metastasis
To further investigate anticancer activity of S2-D and S2-FE in vivo, a xenograft mouse model was employed (Figures 7 and 8). To assess the effect of the peptides on the primary tumor growth of colon cancer cells, luciferase-expressing mouse colon cancer cells (CT26luc) were injected subcutaneously into 6-week-old male BALB/c nude mice, and tumor growth was monitored up to 21 days after the injection of CT26-luc cells in either the absence or presence of the peptides (Figure 7). Subcutaneous images showed a significant decrease (14.03% reduction for S2-D, 64.58% reduction for S2-FE at day 21) in photon emission from the tumor sites in mice injected with either S2-D or S2-FE compared to S2-Ptreated control mice ( Figure 7A). S2-FE-treated mice exhibited slower tumor growth than S2-D-treated mice ( Figure 7A). Correspondingly, the average tumor weight of S2-FE-treated mice was much less than that of S2-P-treated mice ( Figure 7B). To further evaluate the effects of the synthetic peptide in a metastasis model, we injected CT26-luc cells pre-incubated with the peptide into mouse tail veins. Analysis of lung samples revealed that both S2-D and S2-FE inhibited the pulmonary metastasis of CT-26 cells, but S2-FE showed a significant inhibitory effect ( Figure 8A). Similar to the subcutaneous model, mice given S2-FE exhibited significantly slower tumor growth than S2-P-treated mice ( Figure 8B). Together, these in vivo results suggest that S2-D and S2-FE have cancer-inhibiting effects, and that S2-FE exhibits more effective inhibition of primary tumor growth and metastasis than S2-P.

Discussion
We previously reported that the extracellular domain of SDC2 interacts with the MMP-7 pro-domain [32], and that a synthetic peptide corresponding to a portion of the human SDC2 sequence (S2-P) bound to the pro-domain of MMP-7 and inhibited SDC2-mediated cancer activities [11]. In this study, we generated modified peptides (S2-D, S2-L, S2-FE, and S2-FLD) from the original sequence of S2-P to improve the anticancer activity of S2-P and then investigated the anticancer activities of the peptides. Our data showed that, among four mimetic peptides, S2-D and S2-FE showed the higher binding affinity to the MMP-7 pro-domain ( Figure 2B) and greater inhibitory effects on the SDC2-MMP-7 pro-domain interaction ( Figure 2B) compared to S2-P, suggesting that S2-D and S2-FE could replace the endogenous MMP-7 interaction with cell surface SDC2, a cell surface docking receptor of MMP-7 [6,26]. Indeed, both S2-D and S2-FE decreased cell surface localization of MMP-7 ( Figure 3B), which supports the ability of these peptides to prevent pro-MMP-7 from docking with SDC2. Given that MMP-7 activity critically regulates the cancer activity of colon cancer cells [11,32], it is highly possible that both S2-D and S2-FE could act as anticancer agents. As expected, S2-D and S2-FE showed a cancer-inhibiting effect against human colon cancer cells, as evidenced by the abilities of these peptides to reduce the migration, invasion, and soft-agar colony formation activity of HT29-SDC2 and HCT116 cells (Figures 4 and 5). Since the interaction of SDC2 with MMP-7 induced the activation of MMP-7 [11], we hypothesized that disruption of their interaction could decrease MMP-7 activity. Indeed, S2-D and S2-FE reduced the MMP-7-mediated degradation of gelatin substrate ( Figure 6). However, while S2-D showed similar inhibitory activity compared to S2-P in all cancer activities tested, S2-FE showed an improved anticancer activity in comparison with S2-P against HCT116 cells ( Figure 5). In addition, S2-FE, but not S2-D, showed better inhibitory effect on primary tumor growth ( Figure 7) and lung metastasis (Figure 8) of CT26 cells than S2-P peptide. Together, these data suggest that modifying S2-P by replacing an acidic amino acid in the vicinity of Tyr 51 enhanced the interaction with pro-MMP-7 and thereby inhibited MMP-7 activity and the cancer activities of colon cancer cells. These findings further suggest that S2-FE could be effective anticancer peptides against colon cancer.
Many efforts have been made to develop anticancer therapeutics targeting MMPs, with a particular focus on their proteolytic activities. The developed agents include hydroxamatebased chelation inhibitors (e.g., batimastat [33], marimatat [34], prinomastat [35]). Although such agents have exhibited promising antitumor effects in animal models of various cancers, the clinical trials have been largely unsuccessful due to severe side effects [36]. For instance, batimastat showed malignant pleural effusion and ascites in a phase I study [37,38] and marimastat was canceled in phase III clinical trials because it had significant musculoskeletal toxicity [35,39]. Since the various MMPs are widely distributed throughout the body, structurally conserved, and involved in many different aspects of cell functions, agents that block MMP protease activity can have huge impacts on cell functions. This may limit the usefulness of strategies directly targeting the enzymatic domain of MMPs. Interestingly, this study implies that MMP-7 activity can be specifically regulated through a direct interaction with SDC2 on the cancer cell surface, and that blocking this activating interaction could be a means to specifically inhibit the enzymatic activity of MMP-7.
MMPs are secreted as catalytically inactive pro-MMPs [12]. To obtain the enzymatic activity of MMPs, cleavage of pro-domain is necessary. It is well known that the docking of pro-MMP-2 with TIMP-2 facilitates the activation of MMP-2 [25]. This interaction elevates the density of pro-MMP-2, and MT1-MMP forms a complex with pro-MMP-2 and TIMP-2 to enable the activating cleavage of the MMP-2 pro-domain [25,40]. To inhibit the activity of MMP-2 in human melanoma cells, researchers previously used a cyclic peptide to target the protein-protein interaction of pro-MMP-2 and TIMP-2 [41]. Although a previous study showed that the docking of pro-MMP-7 to SDC2 elevated the activation of MMP-7 [11], the exact underlying mechanism remained unknown. Similar to the case of MMP-2, the interaction of SDC2 with pro-MMP-7 may cause molecules of the latter to cluster via SDC2mediated homodimerization [42], and the clustered pro-MMP-7 molecules may cleave one another's pro-domains in a cross-cutting manner. Further studies are needed to clarify the mechanism underlying the SDC2-prompted activating cleavage of pro-MMP-7 at the colon cancer cell surface.
Indeed, the abilities of S2-FE to inhibit MMP-7 activity and the tumorigenic activities of colon cancer cells suggest that these peptides could potentially serve as effective anticancer drugs for colon cancer. However, since therapeutic peptides composed of natural amino acid sequences are known to be easily degraded, relatively unstable, and sensitive to proteases of the digestive system and blood plasma [43], further studies will be required to improve the stability of the peptides. Possible strategies for this include cyclization, bioisosteric replacement of peptide bonds, and changing the stereochemistry of amino acids [44].
In summary, we herein reveal that mimetic peptides derived from SDC2 (e.g., S2-FE) can interrupt the interaction of SDC2 with pro-MMP-7 and thus suppress colon cancer cell activities. Since S2-FE have improved anticancer activity over the parent peptide S2-P, S2-FE could be promising anticancer peptides against colon cancer. Although future studies will be required to clarify the underlying inhibitory mechanism and improve the stability of these peptides before they can be developed as new anticancer drugs, the present study lays new groundwork for the specific targeting of MMP-7 via SDC2 in colon cancer.

Expression and Purification of Recombinant Pro-Domain of MMP-7
The gene encoding the pro-domain sequence (PDMMP-7) was cloned into the pET32a vector, which has a TRX-His tag at the N-terminus, for protein expression and purification. Between the TRX-His tag and the pro-domain, a TEV cleavage sequence was inserted. Proteins were overexpressed with 1 mM IPTG and isolated using Ni-NTA affinity chromatography. The purified TRX-His tag-PDMMP-7 was incubated with TEV protease at a 1:1 molar ratio for 12 h. The reaction mixture was further purified in phosphate-buffered saline (PBS) buffer at pH 7.4 using a HiLoad TM superdex TM 75 column (GE Healthcare, Little Chalfont, UK) to remove tags. The final protein sample was concentrated to 14 mg/mL for the fluorescence experiment.

Fluorescence Assay
The binding affinity between the PDMMP-7 and each synthetic peptide was measured at 298 K (25 • C) using an LS55 fluorescence spectrophotometer (Perkin Elmer, Waltham, MA, USA) at wavelengths of 280 nm (excitation) and 300-450 nm (emission). PDMMP-7 and peptides were prepared in phosphate-buffered saline (PBS) buffer at pH 7.4. The concentration of PDMMP-7 was set at 10 µM, and the input amount of each peptide versus PDMMP-7 was titrated at a molar ratio up to 1:80 using a thermostat cuvette. The dissociation constant (Kd) of the PDMMP-7/peptide complex was calculated using the equation log[(Fo − F)/F] = log(1/Kd) + nlog [ligand], where Fo and F represent the fluorescence intensity of PDMMP-7 at 347 nm in the absence and presence of the peptide, respectively, and n represents the number of binding sites.

Monitoring of Cell Attachment
Cell adhesion was monitored in real time using an xCELLigence system (Roche Diagnostics GmbH, Basel, Switzerland). The bottom of an E16 xCelligence plate (ACEA Biosciences, Santa Clara, CA, USA) was coated with a mixture of gelatin and PDMMP-7 at 37 • C for 1 h. The plate was washed with PBS, loaded with serum-free medium (50 µL/well), and incubated at 37 • C in 5% CO 2 for 15 min. The background was measured using an RTCA DP Analyzer (RTCA software version 1.2, ACEA Biosciences). HT29-SDC2 cells (2 × 10 4 cells/well) were added to each well with SDC2 mimetic peptides (50 nM), and the plate was incubated at 37 • C in 5% CO 2 for 15 min. After 15 min, the plate was assembled onto the RTCA DP Analyzer, and cell adhesion was assessed at 2-minute intervals for 1 h at 37 • C under 5% CO 2 . The data obtained were analyzed using the provided RTCA software.

Immunofluorescence Analysis
Cells were seeded onto glass coverslips in 12-well plates, incubated for 48 h (HT29 3 × 10 5 cells/well, HCT116 2 × 10 5 cells/well), and treated with 50 nM of SDC2 peptide. After 24 h from peptide treatment, cells were fixed with 3.5% paraformaldehyde in PBS at room temperature for 10 min. The cells were rinsed three times with PBS, blocked with 0.5% bovine serum albumin (BSA) in PBS for 1 h, washed, and stained with the appropriate primary antibody overnight at 4 • C. The cells were then washed with PBS and incubated with FITC-conjugated mouse antibodies (Thermo Fisher Scientific, Waltham, MA, USA) and Texas-Red-conjugated rabbit antibodies (Thermo Fisher Scientific) for 1 h at room temperature. The coverslips were washed with PBS and mounted on glass slides with mounting solution containing 4 ,6-diamidino-2-phenylindole (DAPI), and the results were imaged under a confocal fluorescence microscope (Carl Zeiss, Gottingen, Germany) [11].

Cell Proliferation Assay
Cell proliferation was measured by a colorimetric assay using 3-(4,5-dimethythiazol-2yl) 2,5-diphenyltetrazolium bromide (MTT; Amresco, Solon, OH, USA) according to the manufacturer's instructions. Briefly, cells were harvested with 0.05% trypsin/EDTA and seeded to 48-well plates (Thermo Fisher Scientific). After cells were allowed to attach to the plate for 24 h, medium containing 0.5 mg/mL MTT was added to each plate and incubation was continued for 1 h. The medium was then removed, and 200 µL of dimethyl sulfoxide (DMSO) was added to each plate for 30 min at room temperature. The mean concentration of absorbance at 570 nm in each set of samples was measured using a 96-well microtiter plate reader (Dynatech, Chantilly, VA, USA).

Cell Cycle Analysis
Cells were seeded to 6-well plates (HT29, 3 × 10 5 cells/well; HCT116, 2 × 10 5 cells/well) and incubated for 48 h. Peptides were applied for 24 h. The cells were then washed with PBS, released, and fixed with 5 mL of cold 70% ethanol in PBS at 4 • C. The ethanol was discarded, and the cells were washed with cold PBS and incubated with 0.1 mg/mL of propidium iodide (Sigma-Aldrich) and 0.6% Triton-X 100 in PBS with RNase at 25 • C for 45 min. Cell cycle parameters were analyzed by flow cytometry.

Invasion and Migration Assay
For the migration assay, gelatin (10 µg/mL) was added to each well of a Transwell plate (8 µm pore size; Costar, Corning, NY, USA), and the membranes were allowed to dry at 25 • C for 1 h. The Transwell plates were assembled into a 24-well plate, and the lower chamber was filled with McCoy's 5A medium containing 10% FBS, 1% BSA, and basic fibroblast growth factor (10 µg/mL). Cells (HT29, 1 × 10 6 cells/well; HCT116, 5 × 10 5 cells/well) were added to each upper chamber, and the plate was incubated at 37 • C in a 5% CO 2 incubator. The cells that had migrated to the lower surface of the filters were stained with hematoxylin and 0.5% eosin and counted. For the in vitro invasion assay, Transwell plates were coated with gelatin (10 µg/mL) on the lower side of the membrane and with Matrigel (BD Biosciences, Mississauga, Canada) (3 mg/mL) on the upper side of the membrane.

Anchorage-Independent Growth in Soft Agarose
Each well of a 6-well culture plate was coated with 3 mL of bottom agar mixture (McCoy's 5A containing 10% FBS and 0.6% agar). After the bottom layer had solidified, 1 mL of top agar mixture (McCoy's 5A containing 10% FBS and 0.3% agar) containing cells (HT29, 2 × 10 5 cells/well; HCT116, 1 × 10 5 cells/well) was added to each well, and the cultures were incubated at 37 • C in a 5% CO 2 atmosphere. Colony formation was monitored daily with a light microscope. After 14 days, the colonies were stained with 0.005% crystal violet and photographed with a digital camera.

Gelatin Degradation Assay
Oregon-Green-488-conjugated gelatin from pig skin was dissolved at 1 mg/mL in 2% sucrose. Coverslips in 24-well plates were coated with 0.2 mg/mL Oregon-Green-488conjugated gelatin at 4 • C for 20 min in the presence of 40 µL of 0.5% glutaraldehyde diluted in 1 mL PBS. The coverslips were washed three times in PBS at room temperature and incubated with 5 mg/mL NaBH 4 (dissolved in PBS) at room temperature for 3 min. The coverslips were washed as above, sterilized in 70% ethanol for 1 min, and then dried. After being incubating in McCoy's 5A medium at 37 • C for 1 h, 1.25 × 10 5 HCT116 cells mixed with or without 1 ng/mL of IL-1α and 50 nM of SDC2 peptide were seeded onto coverslips in 24-well plates (Costar, Corning, NY, USA) and cultured for 20-72 h at 37 • C in a 5% CO 2 atmosphere. After 24 h, 50 nM of SDC2 peptide were added in the treated 24-well plates. The cells were fixed with 3.5% paraformaldehyde for 10 min and washed, and the coverslips were mounted on microscopic slides and imaged by fluorescent microscopy.

Mouse Model
For the subcutaneous model, CT26-luc cells (5 × 10 6 cells/mouse) incubated with synthetic peptide (final 250 nM) at 37 • C for 30 min in 100 µL of PBS were subcutaneously injected below the dorsal flank into 6-week-old male BALB/c mice (n = 5 per group). For imaging, mice were intraperitoneally injected with 150 mg/kg D-luciferin and anesthetized with 1% isoflurane. At 10-20 min after D-luciferin injection, mice were placed in an IVIS Imaging System (IVIS SPECTRUM; Caliper Life Sciences, Waltham, MA, USA) and imaged dorsally. Tumor growth was monitored weekly by the IVIS Imaging System, and external caliper measurements were taken (L × W × D) for 21 days. For the in vivo experimental pulmonary metastasis assay, CT26-luc cells (1 × 10 5 cells/mouse) incubated with synthetic SDC2 peptide (final 250 nM) at 37 • C for 30 min in 200 µL of PBS were injected via the tail veins into 6-week-old male BALB/c mice (n = 3 per group; Orient Bio Co., Seoul, Korea). Growth of metastatic lung tumors was monitored weekly by the IVIS Imaging System; images were captured on days 6, 16, and 19. On day 19, mice were sacrificed, lungs were excised, and metastatic nodules were photographed and counted. Animal study was conducted in accordance with the Institutional Animal Care and Use Committee (IACUC) of the National Cancer Center Research Institute (NCCRI), and the IACUC approval number is NCC-21-680 (the date of approval 4 August 2021)

Statistical Analysis
All data are presented as mean ± S.D. Differences between groups were tested for statistical significance using Student's t-test and were considered significant at * p < 0.05 or ** p < 0.01.