International Immunopharmacology 19 (2014) 119–126
Contents lists available at ScienceDirect
International Immunopharmacology journal homepage: www.elsevier.com/locate/intimp
Recombinant soluble CD226 protein directly inhibits cancer cell proliferation in vitro Shengke Hou a, Xiaodong Zheng a, Haiming Wei a,b, Zhigang Tian a,b, Rui Sun a,b,⁎ a b
Department of Immunology, School of Life Sciences, University of Science and Technology of China, Hefei, Anhui 230027, China Hefei National Laboratory for Physical Sciences at Microscale, Hefei, Anhui 230027, China
a r t i c l e
i n f o
Article history: Received 17 December 2013 Received in revised form 12 January 2014 Accepted 13 January 2014 Available online 24 January 2014 Keywords: Soluble CD226 Tumor Cell proliferation
a b s t r a c t Interactions between CD155 and nectins on tumor cells have been reported to potentially inhibit tumor growth. CD226, a receptor that recognizes CD155 and CD112, is an activation receptor of NK and T cells by which immune cells may attack a tumor. The purpose of this study is to explore whether soluble CD226 (sCD226) directly inhibits tumor growth by binding CD155 or CD112 on tumor cells. We expressed, purified and confirmed the identity of recombinant sCD226 (19 aa–248 aa) and then examined the effect of sCD226 on tumor cell growth using CD226 ligand (CD155 and CD112)-expressing cancer cell lines (K562, HeLa). After 3 days of co-culture with sCD226, we found that the numbers of K562 and HeLa cells were significantly reduced but those of a CD226-blocking mAb specifically attenuated the inhibitory effects of sCD226. We also noted that the sCD226 protein could compete with a PE-conjugated anti-CD112 antibody in flow cytometric analysis and block the binding of the PE-conjugated anti-CD112 antibody to tumor cells. Mechanistic studies using flow cytometric analysis demonstrated that sCD226 inhibited the division of CFSE (carboxyfluorescein diacetate succinimidyl ester)-labeled K562 cells by delaying the cell cycle. In addition, we observed that sCD226 might have an impact on the metastatic potential of solid tumors in vitro. These results demonstrated that sCD226 molecule might be a potential biotherapy against tumor for further development. © 2014 Elsevier B.V. All rights reserved.
1. Introduction CD226 is a transmembrane glycoprotein that was first discovered by Burns GF's group in 1985 [1]. Because it was first found to be expressed in platelets and T cells, it is also called platelet and T-cell antigen 1 (PTA1). In 1996, Shibuya first revealed the role of this molecule in T-cell cytotoxicity and called it DNAM-1 [2]. CD226 is also expressed on NK and NKT cells [1–4] and is now considered to be an important NK cell activation receptor that mediates NK cell cytotoxicity against tumor and infected cells. Many immunotherapies for malignant diseases are based on NK cells [5]. CD226 can recognize ligands on cancer cells and mediate NK cell cytotoxicity. The CD226 ligands CD155 and CD112 belong to the immunoglobulinlike superfamily, which is a type I transmembrane molecular superfamily that contains CD226 [6,7]. CD155 is also called PVR and nectin like-5 (Necl-5) because it functions as a poliovirus receptor and as an adhesion protein, respectively [6,8]. CD112 is an adhesion protein that is also called nectin-2 [9]. Previous studies have shown that trans interactions between nectins usually activate the RAC/Cdc42 protein complex [10–12]. Rac and Cdc42 are members of the Rho-GTPase family and ⁎ Corresponding author at: Department of Immunology, School of Life Sciences, University of Science and Technology of China, Hefei, Anhui 230027, China. Tel.: +86 551 6360 7977; fax: +86 551 6360 0832. E-mail address: [email protected] (R. Sun). 1567-5769/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.intimp.2014.01.012
play key signaling roles in cytoskeleton reorganization, membrane trafficking, transcriptional regulation, cell growth and development [10]. A recent study found that sCD226 is present in the serum of normal individuals and cancer patients, but at significantly higher levels in cancer patients suggested that the up-regulation of sCD226 might be a mechanism for tumor escape [13]. In this study, using recombinant sCD226 (extracellular segment), we found that sCD226 could directly reduce tumor cell growth in a dose-dependent manner by combining with and activating CD155 and CD112. The phosphorylation of Rac/ Cdc42 and cyclin D, which are two key signaling molecules of CD155 and CD122, was up-regulated and delayed the cell cycle of tumor cells. These results suggested that the up-regulation of sCD226 might be a protective compensation response of the organism to cancer cells.
2. Materials and methods 2.1. Antibodies and cell lines The antibodies that were used in this study are as follows: the PEconjugated CD112 (R2.525), PE-conjugated CD155 (SKII.4) and purified anti-CD226 mAbs (DX11) were purchased from BD Biosciences (San Diego, CA, USA); the anti-Rac/Cdc42 antibody, phospho-Rac1/Cdc42 (Ser71) antibody, cyclin D1 (92G2) rabbit mAb and phospho-cyclin D1 (Thr286) (D29B3) XP® rabbit mAb were purchased from Cell
120
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
Signaling (Beverly, MA, USA); and the anti-β actin antibody was purchased from Wuhan Boster Company (China). K562 tumor cells were cultured in RPMI 1640 medium (Gibco) supplemented with 10% heat-inactivated FBS, 100 IU/ml penicillin and 100 μg/ml streptomycin. The HeLa tumor cell line was cultured in RPMI 1640 medium (Gibco) supplemented with 10% heat-inactivated FBS, 100 IU/ml penicillin and 100 μg/ml streptomycin. A non-tumor cell line, CHO-K1, was cultured in DMEM medium (Gibco) supplemented with 10% heat-inactivated FBS, 100 IU/ml penicillin and 100 μg/ml streptomycin. Cell culturing was performed at 37 °C in a 5% CO2 humidified atmosphere. Cells from a mid-log phase culture were harvested. The human NK cell line NK92 was cultured according to the specifications outlined by the American Type Culture Collection (ATCC). 2.2. Expression, purification and preparation of soluble CD226 extracellular domain The amino acid sequence of this protein can be accessed through the NCBI Protein Database under NCBI Accession # NP_006557.2 [14]. The sequences of the primers are as follows: forward primer, 5′-cgcggatccgaagaggtgctttggcatac-3′, and reverse primer, 5′-ccgctc gagttaaactctagtctttggtc-3′. The extracellular segment of human CD226 (19–248 aa) (CD226 ECD) was amplified by PCR from the total mRNA of NK-92 cells (a human NK cell line) and inserted into the pET32a vector (Novagen) to construct the expression vector pET32a/His–MBP–CD226. Escherichia coli Rosetta (DE3) cells were successfully transformed with the constructed expression plasmid and were grown in Luria–Bertani broth containing kanamycin (100 mg/l) and chloromycetin (34 mg/l). The expression of the recombinant protein was induced during the exponential phase with 1 mM isopropyl-β-D -thiogalactoside (IPTG). The E. coli cells were cultured for another 4 h at 37 ºC, harvested and then lysed. The target proteins were found mainly in the supernatant and were fused to His and MBP tags. The protein was purified using Ni Sepharose (Amersham Biosciences) and Superdex-200 (GE Healthcare). The TEV protease with a His tag was prepared from pGEX4T2 vector-induced expression in E. coli BL21 (DE3) and purified using Ni Sepharose. The TEV protease was mixed with the His–MBP–CD226 recombinant protein at an equal dose and incubated overnight at 4 °C. The mixture was purified using Ni Sepharose to remove the MBP tag and TEV protease, which both possessed a His tag. The protein was purified using Superdex200 to obtain the pure, soluble CD226 extracellular domain protein. 2.3. Flow cytometry The cells were washed with PBS twice, blocked and stained with saturating concentrations of the appropriate fluorochrome-conjugated mAbs for 30 min at 4 °C. Thereafter, the cells were washed with PBS twice and analyzed using a FACSCalibur (Becton Dickinson). For the binding analysis of CD226, the sCD226 protein was added as blocking antibody with its final concentration at 1 μg/ml, and the cells were then washed with PBS twice and incubated with PE-conjugated CD112 mAb. 2.4. Western blot analysis K562 or HeLa cells were cultured in RPMI 1640 medium containing 10% FBS and equal doses of sCD226 or BSA proteins for 1 h. The cells were harvested and boiled in Laemmli sample buffer. Then, proteins from 106 cells were resolved by SDS-PAGE (from 7.5% to 12.5% acrylamide), transferred to PVDF (polyvinylidene fluoride) membranes and probed with primary antibodies. The proteins were visualized using a secondary antibody that was conjugated to HRP (horseradish peroxidase) and chemiluminescence detection (enhanced chemiluminescence).
2.5. Cell staining with CFSE to detect the cell proliferation K562 cells were stained with CFSE (carboxyfluorescein diacetate succinimidyl ester). Briefly, the cells were resuspended in 1 ml PBS/1% BSA at a final concentration of 5 × 106/ml and then incubated with CFSE (2 μM, Sigma) for 10 min at 37 °C. Quench staining was performed on ice for 5 min by adding five volumes of ice-cold RPMI 1640/10% FBS. Then, the cells were washed three times with ice-cold PBS/1% BSA and cultured under the appropriate conditions. 2.6. Cell migration assay HeLa or CHO-K1 cells were planted at a cell density of 105 per well. A scratch was made in the cell monolayer using a pipette tip. Then, the cells were washed with the appropriate medium containing 2% FBS to remove the suspended cells from the scratch. The appropriate medium containing 2% FBS was added to the cells to eliminate any effects due to cell proliferation, and the cell migration distances were equal to the mean width of the scratch at 0 h minus the mean width of the scratch at 36 h. 2.7. Statistical analysis Statistical analyses were performed using Student's t test. All p values were two-tailed, and p b 0.05 was considered statistically significant. 2.8. Bioinformatics tools We used the ExPASy online tools to predict the structure of the CD226 protein. The SMART tool (http://smart.embl-heidelberg.de/ smart/set_mode.cgi?NORMAL=1) was used to predict the domains of CD226, and the PSIpred tool (http://bioinf.cs.ucl.ac.uk/psipred/) was used to predict the secondary structure of the CD226 protein sequence. 3. Results 3.1. Recombinant expression of the ectodomain of the human CD226 receptor To explore the function of soluble CD226 (sCD226), we constructed a vector containing the extracellular segment of CD226. Because CD226 is a transmembrane protein and the sCD226 molecule might be a cleavage product of proteases that target different sites of the intact molecule, we predicted the structure of the CD226 molecule (Fig. 1A). Based on the predicted secondary structure result, we selected a hydrophilic amino acid, Q248, that was not in the β-sheet (yellow arrows in Fig. 1B) to create the 19 aa–248 aa soluble CD226 extracellular segment. To obtain the soluble protein, we constructed the recombinant protein with 6His and MBP tags at its N terminus (Fig. 1C) and soluble CD226 protein was expressed in E. coli. We purified the recombinant protein by using Ni Sepharose. The recombinant protein had a molecular weight of around 68 kD on an SDSPAGE gel, which is consistent with its estimated size (Fig. 2A). Moreover, the peak of the recombinant protein was observed in the molecular sieve chromatogram (Fig. 2B). After being digested by TEV protease, there were three main bands on the SDS-PAGE gel: the undigested protein band (68 kD), the 6His–MBP tag band (40 kD) and the TEV protease (27 kD)/CD226 segment (CD226 ECD, 28 kD) mixed band (Fig. 2C). To obtain purer CD226 protein (no tags CD226 ECD, 28 kD), we used Ni Sepharose and molecular sieve chromatography to remove the 6His–MBP segments and the TEV protease with His tag. The CD226 ECD was observed at 28 kD band in the Ni Sepharose effluent lane on the SDSPAGE gel and the peak of CD226 ECD in the chromatogram (Fig. 2D). The protein were identified by LC–MS. Eight different peptides were
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
121
Fig. 1. Expression vector construction for the ectodomain protein of the human CD226 receptor. (A) Prediction of the tertiary structure of the CD226 protein. According to the protein sequence of CD226, we predicted the tertiary structure of the CD226 protein using the ExPASy SMART tool. (B) Prediction of the secondary structure of the CD226 protein. Based on the protein sequence of CD226, we predicted the secondary structure of the CD226 protein using the ExPASy PSIpred tool. Gln248 was involved in the coil conformation. (C) The strategy for the construction of the pET-32a(+)–6His–MBP–CD226 extracellular segment (19 aa–248 aa) expression vector. Using total RNA from the CD226-positive, human NK-92 cell line as a template, PCR was performed to amplify the target fragment. The gene encoding the two extracellular domains of CD226 was then inserted into the pET32a(+) vector that contained a 6×His–MBP-tag on its N terminus.
found, covering 47.62% of the total CD226 residue sequences and nearly all the extracellular regions of CD226 (Fig. 2E). 3.2. sCD226 inhibits the proliferation of CD226 ligand-positive tumor cell lines We then assessed the effect of sCD226 on tumor cell growth using CD226 ligand (CD155 and CD112)-expressing cancer cell lines and a CD226 ligand-negative CHO-K1 cell line as a control (Fig. 3A). After 3 days of co-culturing the cells with sCD226, we counted the number of tumor cells. The results suggested that the numbers of K562 and HeLa cells, but not the number of control CHO-K1 cells, were significantly reduced. The addition of the CD226 blocking monoclonal antibody to the sCD226 co-culture system attenuated the inhibitory effect of sCD226 on tumor cell growth (Fig. 3B). To confirm the dose-dependent inhibitory effect of sCD226 on tumor growth, we examined the binding of the CD226 protein to tumor cells. The flow cytometry results demonstrated that the sCD226 proteins could compete with the PE-conjugated anti-CD112 antibody and block the binding of the PE-conjugated anti-CD112 antibody to tumor cells (Fig. 4A and B). Furthermore, in a tumor cell growth inhibition experiment, we observed that sCD226 dose-dependently inhibited tumor cell growth during a 72-h culture (Fig. 4C). 3.3. sCD226 inhibits the proliferation of leukemia cells by delaying the cell cycle The reduction in cell number could be due to either cell death or cell cycle delay. We used both FACS and Western blot to test cell death. FACS analysis revealed that Hela and K562 cell deaths were not increased with Annexin V–PI (propidium iodide) double stained after incubated with sCD226 for 72 h (data not shown). We also detected the Caspase8 precursors by Western blot in Hela cells and K562 cells after being incubated with the sCD226 for 30 min. It was not degraded in all the experimental groups (data not shown). The results from two methods were coincident. Therefore, we examined the cell cycle of the tumor cells. We selected and stained K562 cells with CFSE. Because CFSE is equally distributed to the daughter cell during cell division, the fluorescence intensity of the tumor cell lines represents the number of times the tumor cell lines have divided. We detected the CFSE fluorescence intensity of the stained
tumor cells at 24, 48, 72 and 96 h. The results showed that the CFSE fluorescence intensity decreased with cell division (Fig. 5A) and that in the sCD226 co-culture systems, tumor cell division was delayed dramatically, with the tumors having higher numbers of undivided cells (cells with higher CFSE fluorescence intensity) (Fig. 5B and C). The inhibitory effect of sCD226 on the cell cycle began at 24 h and became more obvious at 48 and 72 h (Fig. 5B and C). 3.4. sCD226 induces the phosphorylation of cyclin D and the Rac/Cdc42 complex A previous study demonstrated that the trans interactions of nectins and nectin-Necl 5 (CD155) usually activate the Rac/Cdc42 protein complex [15], and the activation of Cdc42 enables the cells to avoid contact inhibition [16]. The Akt phosphorylation site at Ser71 of Rac1/Cdc42 has been identified, and phosphorylation at this site may inhibit GTP binding of Rac1, thereby attenuating the signal transduction pathway downstream of Rac1 [17]. Therefore, we examined the phosphorylation at the Ser71 site of the Rac/Cdc42 complex in HeLa cells after coculturing with sCD226 for 1 h. The phosphorylation levels of the Rac/ Cdc42 protein were up-regulated (Fig. 6A and B), and although the expression level of cyclin D was slightly up-regulated, the phosphorylation at Thr286 was significantly up-regulated (Fig. 6C and D). These results suggest that serum sCD226 may inhibit the activation of Rac/Cdc42 and phosphorylation-dependent degradation of cyclin D, leading to tumor cell cycle delay. 3.5. sCD226 might impact the metastatic potential of solid tumors in vitro Because some reports suggest that Rac/Cdc42 promotes cell migration [18,19]. We also examined the metastatic potential of solid tumor cells (HeLa cells) after sCD226 treatment. In this experiment, we cultured cells in 2% low-serum culture medium to exclude proliferation effects. The results showed that sCD226 could inhibit the metastatic potential of HeLa cells. After a CD226 blocking antibody was added to the sCD226 co-culture system, the metastatic potential of the tumor cells increased to a certain extent (Fig. 7A and B). To confirm this result, we selected the CD226 ligand-negative cell line CHO-K1 and performed the same test. The results showed that the metastatic potential of the CHO-K1 cells was not affected by sCD226 (Fig. 7A and C).
122
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
Fig. 2. Purification and identification of the extracellular domains of the soluble CD226 protein (CD226 ECD). (A) The recombinant 6His–MBP–CD226 extracellular segment (19 aa–248 aa) protein was detected using SDS-PAGE analysis. The supernatant (lane 2) and pellet (lane 3) of the lysates from Rosetta (DE3)/pET32a(+)–6His–CD226 cells were incubated with 1 mM IPTG at 37 °C for 6 h. The flow-through solution (lane 4), wash solution containing 50 mM imidazole (lane 5) and elution solution containing 500 mM imidazole (lane 6) from Ni Sepharose binding were also detected using SDS-PAGE. The proteins were separated by 12% SDS-PAGE. The arrowheads indicate the position of the recombinant protein. (B) Purification of the recombinant protein via a Superdex-200 molecular sieve. The protein was collected by the no. 29 to no. 43 collection tubes and detected using SDS-PAGE. The arrowheads indicate the peak of the recombinant protein. (C) Removal of the 6His and MBP tags from the recombinant protein from the N terminus using TEV protease. The CD226 extracellular segment (19 aa–248 aa) with no tags that was purified using Ni Sepharose (lane 1), protein marker (lane 2), the protein solution after digestion and before purification (lane 3 and lane 4, respectively) and the recombinant 6His–MBP–CD226 extracellular segment (19 aa–248 aa) protein before digestion (lane 5) were detected using SDS-PAGE analysis. (D) Ni Sepharose and Superdex-200 were used to purify the soluble CD226 extracellular segment (19 aa–248 aa) protein that had been digested by TEV protease with no tags. We purified the digested proteins via Ni Sepharose to remove the TEV proteins, MBP tags and undigested 6His–MBP–CD226 extracellular segment (19 aa–248 aa) proteins with a His tag. The soluble CD226 extracellular segment (19 aa– 248 aa) protein with no tag flow-through solution from Ni Sepharose binding was collected and purified using Superdex-200. The soluble CD226 extracellular segment (19 aa–248 aa) protein was collected in the no. 53 to no. 66 collection tubes and then detected by SDS-PAGE. The arrowheads indicate the peak of the CD226 extracellular segment (19 aa–248 aa) protein. (E) Sequencing of the CD226 extracellular segment (19 aa–248 aa) protein by LC–MS. The underlined amino acid residue sequences were identified using an LTQ Mass Spectrometer.
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
123
Fig. 3. sCD226 inhibits the proliferation of CD226 ligand (CD155/CD112)-positive tumor cell lines. (A) Flow cytometric analysis of CD226 ligand (CD155/CD112) expression by tumor cells. K562, HeLa and CHO-K1 cells were incubated with the PE-conjugated anti-CD155 mAb (SKII.4) or PE-conjugated anti-CD112 mAb (R2.525) for 30 min at 4 °C and then examined using FACS. (B) Cell proliferation assay. K562 cells or HeLa cells (104) were cultured with 40 μg of BSA, 40 μg of sCD226 or 40 μg of sCD226, followed by either 40 μg of anti-CD226 mAb or 40 μg of sCD226, and last with 40 μg of IgG isotype for 72 h. CHO-K1 cells (104) were cultured with 40 μg of BSA or 40 μg of sCD226 for 72 h. The cell numbers were then counted and recorded. The data are representative of three independent experiments.
4. Discussion In this study, we explored the in vitro function of recombinant sCD226, which mimics serum sCD226. We found that sCD226 could block tumor cell proliferation via the phosphorylation-dependent
degradation of cyclin D to delay the cell cycle. sCD226 led to the phosphorylation of the RAC/Cdc42 complex and promoted cell contact inhibition effects on the cell cycle. In a previous study, CD226 was found to be a membrane protein that mediates the cytotoxicity and adhesion of NK cells and CTLs [13].
Fig. 4. sCD226 dose-dependently inhibits the proliferation of K562 leukemia cells. (A) The binding of sCD226 to K562 cells. K562 cells were incubated with PE-conjugated anti-CD112 mAb (R2.525) and sCD226 at concentrations of 20, 40, 80 and 160 μg/ml for 30 min at 4 °C. The fluorescence intensity of the K562 cells was detected using FACS. (B) Statistical analysis of the percentage of the CD112 mAbs that bound to the K562 cells in A. (C) Proliferation inhibitory effect of sCD226. K562 cells were cultured with sCD226 or BSA at concentrations of 10, 20, 40 or 80 μg/ml for 72 h, and the cell numbers were counted. The data are representative of three independent experiments.
124
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
Fig. 5. sCD226 inhibits the proliferation of K562 cells by delaying the cell cycle. (A) Flow cytometric analysis of tumor cell proliferation. K562 cells were stained with CFSE and cultured in RPMI-1640 containing 10% FBS. The CFSE fluorescence intensity of the K562 cells at 24 h (black), 48 h (green), 72 h (blue) and 96 h (purple) is shown. (B) Flow cytometric analysis of the tumor cell proliferating inhibition by sCD226. K562 cells were stained with CFSE and cultured in RPMI-1640 with 10% FBS with an equal volume of protein buffer (control), 40 μg of BSA or 40 μg of sCD226. The CFSE fluorescence intensity of K562 cells in each experimental group at 24 h, 48 h and 72 h is shown. (C) Statistical analysis of the mean fluorescence intensity of K562 cells in B, *p b 0.05 and **p b 0.01.
Soluble CD226 was discovered in 2003; however, its function is unclear. Because CD226 is an activation receptor, its soluble form is usually considered to inhibit cytotoxicity. As reported by Jin et al., sCD226 might potentially block the cytotoxicity of NK cells through blocking CD155 or CD112 on tumor cells, which might promote tumor cell escape from NK or CTL cytolysis [13]. We also detected the impaction of the
sCD226 on PBMC cytotoxicity to K562 cells. We found that the sCD226 could partly reduce the PBMC cytotoxicity, however, the reduction was limited and does not exceed 20% of the PBMC original cytotoxicity (data not shown). We think that the reason why sCD226 function is limited is that there are several activation receptors, such as NKG2D, NKp30, and so on, leading to the PBMC cytotoxicity that can't be
Fig. 6. sCD226 induces the phosphorylation of cyclin D and the Rac/Cdc42 complex. (A) Western blot analysis of the Rac/Cdc42 complex in HeLa cells. Western blot analysis of phosphorylation at Ser71 of the Rac/Cdc42 complex in HeLa cells was performed. HeLa cells were incubated with an equal volume of protein buffer (control), 40 μg of BSA or 40 μg of sCD226 for 1 h. (B) Statistical analysis of the mean gray values of the phosphorylation of the Rac/Cdc42 complex in the three experimental groups. The data were collected from three independent experiments and analyzed using a t test, *p b 0.05. (C) Western blot analysis of the cyclin D complex in HeLa cells. Western blot analysis of the phosphorylation at the Thr286 site of the cyclin D complex in HeLa cells. HeLa cells were incubated with an equal volume of protein buffer (control), 40 μg of BSA or 40 μg of sCD226 for 1 h. (D) Statistical analysis of the mean gray values of the phosphorylation of cyclin D in the three experimental groups. The data were collected from three independent experiments and analyzed using a t test, *p b 0.05.
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
125
Fig. 7. sCD226 inhibits the metastatic potential of tumor cells. (A) The cell scratch test was used to assess cell migration. Cells were planted at a high cell density. A scratch was made on the plate's surface using a pipette tip, and the HeLa cells were cultured in RPMI-1640 with 2% FBS containing the protein buffer, 40 μg of BSA, 40 μg of sCD226, 40 μg of sCD226 or 40 μg of CD226 blocking antibody for 36 h. CHO-K1 cells were cultured in DMEM with 2% FBS containing the protein buffer, 40 μg of BSA, 40 μg of sCD226, 40 μg of sCD226 or 40 μg of CD226 blocking antibody for 36 h. The scratches for each experimental group at 0 and 36 h are shown. (B) Statistical analysis of the mean HeLa cell migration distance. The cell migration distances were equal to the mean width of the scratch at 0 h minus the mean width of the scratch at 36 h. The data were collected from three independent experiments and analyzed using a t test, *p b 0.05. (C) Statistical analysis of the mean CHO-K1 cell migration distance. The cell migration distances were equal to the mean width of the scratch at 0 h minus the mean width of the scratch at 36 h. The data were collected from three independent experiments and analyzed using a t test.
completely inhibited. Moreover, in our study, we found that sCD226 directly inhibits the proliferation of the cancer cells without co-culture with PBMC. And the reduction of the tumor cell number is nearly 50% compared with the control group to maximum extent (Figs. 3B and 4C). Therefore, we think that the used sCD226 in our study, different from the one working as blocker in Jin's work, may directly combine with its ligand on tumor cells in a manner to deliver proliferation–inhibition signaling. Soluble CD155 is also found in the serum. Because CD155 is a poliovirus receptor, serum sCD155 is considered to reduce poliovirus entry mediated by membrane-bound CD155 [20,21]; however, soluble CD155 also blocks the recognition of cancer cells by the immune system, which helps tumors escape from immune attack. Therefore, sCD226 might be a means to neutralize sCD155, but the biological and pathological importance of this mechanism requires further investigation. We detected cell apoptosis via PI–Annexin V staining and FACS assays and the expression levels of Caspase-8 by Western blotting (data not shown). In this study, we discovered that CD226 did not induce cell apoptosis. CD226, as an adhesion molecule and member of the immunoglobulin superfamily, cannot activate cell apoptosis pathways. Therefore, this adhesion molecule is associated with the cytoskeleton, cell migration and proliferation [6]. Because of the strong proliferation ability and metastatic potential of cancer cells, the treatment of cancer is still a challenge, and the recognition and killing of cancer cells still requires long-term research. Many immunotherapies focused on finding a method to activate the immune system and kill the tumor cells [22]. However, from another perspective, if the proliferation ability and metastatic potential of cancer cells could be inhibited, the harm of a cancer cell to human health will be reduced to a minimum. In our study, we discovered that the soluble
CD226 protein could inhibit the proliferation of cancer cells in vitro and might inhibit cancer cells' metastatic potential. If soluble CD226 can be used clinically as an intravenous drug, it would accelerate our quest to conquer cancer. These results suggest that soluble CD226 is worthy to develop an antitumor drug.
Acknowledgment This work was supported by the Ministry of Science & Technology of China (973 Basic Science Project; 2010CB911901, 2013CB530506), the Natural Science Foundation of China (91029303, 31021061) and the National Science & Technology Major Projects (2012ZX10002014, 2012ZX10002006).
References [1] Burns GF, Triglia T, Werkmeister JA, Begley CG, Boyd AW. TLiSA1, a human T lineagespecific activation antigen involved in the differentiation of cytotoxic T lymphocytes and anomalous killer cells from their precursors. J Exp Med 1985;161:1063–78. [2] Shibuya A, Campbell D, Hannum C, Yssel H, Franz-Bacon K, McClanahan T, et al. DNAM-1, a novel adhesion molecule involved in the cytolytic function of T lymphocytes. Immunity Jun 1996;4(6):573–81. [3] Scott JL, Dunn SM, Jin B, Hillam AJ, Walton S, Berndt MC, et al. Characterization of a novel membrane glycoprotein involved in platelet activation. J Biol Chem 1989;264:13475–82. [4] Sherrington PD, Scott JL, Jin B, Simmons D, Dorahy DJ, Lloyd J, et al. TLiSA1 (PTA1) activation antigen implicated in T cell differentiation and platelet activation is a member of the immunoglobulin superfamily exhibiting distinctive regulation of expression. J Biol Chem 1997;272:21735–44. [5] Cheng M, Chen Y, Xiao W, Sun R, Tian Z. NK cell-based immunotherapy for malignant diseases. Cell Mol Immunol 2013;10:230–52. [6] Takai Yoshimi, Ikeda Wataru, Ogita Hisakazu, Rikitake Yoshiyuki. The immunoglobulin-like cell adhesion molecule nectin and its associated protein afadin. Annu Rev Cell Dev Biol 2008;24:309–42.
126
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
[7] Xu Zhuwei, Jin Boquan. A novel interface consisting of homologous immunoglobulin superfamily members with multiple functions. Cell Mol Immunol 2010;7:11–9. [8] Mendelsohn CL, Wimmer E, Racaniello VR. Cellular receptor for poliovirus: molecular cloning, nucleotide sequence, and expression of a new member of the immunoglobulin superfamily. Cell Mar 10 1989;56(5):855–65. [9] Fabre S, Reymond N, Cocchi F, Menotti L, Dubreuil P, Campadelli-Fiume G, et al. Prominent role of the Ig-like V domain in trans-interactions of nectins. Nectin3 and nectin 4 bind to the predicted C–C′–C″–D beta-strands of the nectin1 V domain. J Biol Chem Jul 26 2002;277(30):27006–13. [10] Ogita H, Takai Y. Activation of Rap1, Cdc42, and rac by nectin adhesion system. Methods Enzymol 2006;406:415–24. [11] Nakanishi H, Takai Y. Roles of nectins in cell adhesion, migration and polarization. Biol Chem Oct 2004;385(10):885–92. [12] Irie K, Shimizu K, Sakisaka T, Ikeda W, Takai Y. Roles and modes of action of nectins in cell–cell adhesion. Semin Cell Dev Biol Dec 2004;15(6):643–56. [13] Xu Z, Zhang T, Zhuang R, Zhang Y, Jia W, Song C, et al. Increased levels of soluble CD226 in sera accompanied by decreased membrane CD226 expression on peripheral blood mononuclear cells from cancer patients. BMC Immunol Jun 2 2009;10:34. [14] Bottino C, Castriconi R, Pende D, Rivera P, Nanni M, Carnemolla B, et al. Identification of PVR (CD155) and Nectin-2 (CD112) as cell surface ligands for the human DNAM-1 (CD226) activating molecule. J Exp Med Aug 18 2003;198(4):557–67.
[15] Nheu T, He H, Hirokawa Y, Walker F, Wood J, Maruta H. PAK is essential for RAS-induced up regulation of cyclin D1 during the G1 to S transition. Cell Cycle Jan 2004;3(1):71–4. [16] Appledorn DM, Dao KH, O'Reilly S, Maher VM, McCormick JJ. Rac1 and Cdc42 are regulators of HRasV12-transformation and angiogenic factors in human fibroblasts. BMC Cancer Jan 12 2010;10:13. http://dx.doi.org/10.1186/1471-2407-10-13. [17] Kwon T, Kwon DY, Chun J, Kim JH, Kang SS. Akt protein kinase inhibits Rac1-GTP binding through phosphorylation at serine 71 of Rac1. J Biol Chem Jan 7 2000;275(1):423–8. [18] Valderrama F, Thevapala S, Ridley AJ. Radixin regulates cell migration and cell–cell adhesion through Rac1. J Cell Sci Jul 15 2012;125(Pt. 14):3310–9. [19] Lai FP, Szczodrak M, Oelkers JM, Ladwein M, Acconcia F, Benesch S, et al. Cortactin promotes migration and platelet-derived growth factor-induced actin reorganization by signaling to Rho-GTPases. Mol Biol Cell Jul 2009;20(14):3209–23. [20] Koike S, Horie H, Ise I, Okitsu A, Yoshida M, Iizuka N, et al. The poliovirus receptor protein is produced both as membrane-bound and secreted forms. EMBO J Oct 1990;9(10):3217–24. [21] Baury B, Masson D, McDermott Jr BM, Jarry A, Blottière HM, Blanchardie P, et al. Identification of secreted CD155 isoforms. Biochem Biophys Res Commun Sep 12 2003;309(1):175–82. [22] Fu Yang-Xin. New immune therapy targets tumor-associated environment: from bone marrow to tumor site. Cell Mol Immunol 2012;9:1–2.
Contents lists available at ScienceDirect
International Immunopharmacology journal homepage: www.elsevier.com/locate/intimp
Recombinant soluble CD226 protein directly inhibits cancer cell proliferation in vitro Shengke Hou a, Xiaodong Zheng a, Haiming Wei a,b, Zhigang Tian a,b, Rui Sun a,b,⁎ a b
Department of Immunology, School of Life Sciences, University of Science and Technology of China, Hefei, Anhui 230027, China Hefei National Laboratory for Physical Sciences at Microscale, Hefei, Anhui 230027, China
a r t i c l e
i n f o
Article history: Received 17 December 2013 Received in revised form 12 January 2014 Accepted 13 January 2014 Available online 24 January 2014 Keywords: Soluble CD226 Tumor Cell proliferation
a b s t r a c t Interactions between CD155 and nectins on tumor cells have been reported to potentially inhibit tumor growth. CD226, a receptor that recognizes CD155 and CD112, is an activation receptor of NK and T cells by which immune cells may attack a tumor. The purpose of this study is to explore whether soluble CD226 (sCD226) directly inhibits tumor growth by binding CD155 or CD112 on tumor cells. We expressed, purified and confirmed the identity of recombinant sCD226 (19 aa–248 aa) and then examined the effect of sCD226 on tumor cell growth using CD226 ligand (CD155 and CD112)-expressing cancer cell lines (K562, HeLa). After 3 days of co-culture with sCD226, we found that the numbers of K562 and HeLa cells were significantly reduced but those of a CD226-blocking mAb specifically attenuated the inhibitory effects of sCD226. We also noted that the sCD226 protein could compete with a PE-conjugated anti-CD112 antibody in flow cytometric analysis and block the binding of the PE-conjugated anti-CD112 antibody to tumor cells. Mechanistic studies using flow cytometric analysis demonstrated that sCD226 inhibited the division of CFSE (carboxyfluorescein diacetate succinimidyl ester)-labeled K562 cells by delaying the cell cycle. In addition, we observed that sCD226 might have an impact on the metastatic potential of solid tumors in vitro. These results demonstrated that sCD226 molecule might be a potential biotherapy against tumor for further development. © 2014 Elsevier B.V. All rights reserved.
1. Introduction CD226 is a transmembrane glycoprotein that was first discovered by Burns GF's group in 1985 [1]. Because it was first found to be expressed in platelets and T cells, it is also called platelet and T-cell antigen 1 (PTA1). In 1996, Shibuya first revealed the role of this molecule in T-cell cytotoxicity and called it DNAM-1 [2]. CD226 is also expressed on NK and NKT cells [1–4] and is now considered to be an important NK cell activation receptor that mediates NK cell cytotoxicity against tumor and infected cells. Many immunotherapies for malignant diseases are based on NK cells [5]. CD226 can recognize ligands on cancer cells and mediate NK cell cytotoxicity. The CD226 ligands CD155 and CD112 belong to the immunoglobulinlike superfamily, which is a type I transmembrane molecular superfamily that contains CD226 [6,7]. CD155 is also called PVR and nectin like-5 (Necl-5) because it functions as a poliovirus receptor and as an adhesion protein, respectively [6,8]. CD112 is an adhesion protein that is also called nectin-2 [9]. Previous studies have shown that trans interactions between nectins usually activate the RAC/Cdc42 protein complex [10–12]. Rac and Cdc42 are members of the Rho-GTPase family and ⁎ Corresponding author at: Department of Immunology, School of Life Sciences, University of Science and Technology of China, Hefei, Anhui 230027, China. Tel.: +86 551 6360 7977; fax: +86 551 6360 0832. E-mail address: [email protected] (R. Sun). 1567-5769/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.intimp.2014.01.012
play key signaling roles in cytoskeleton reorganization, membrane trafficking, transcriptional regulation, cell growth and development [10]. A recent study found that sCD226 is present in the serum of normal individuals and cancer patients, but at significantly higher levels in cancer patients suggested that the up-regulation of sCD226 might be a mechanism for tumor escape [13]. In this study, using recombinant sCD226 (extracellular segment), we found that sCD226 could directly reduce tumor cell growth in a dose-dependent manner by combining with and activating CD155 and CD112. The phosphorylation of Rac/ Cdc42 and cyclin D, which are two key signaling molecules of CD155 and CD122, was up-regulated and delayed the cell cycle of tumor cells. These results suggested that the up-regulation of sCD226 might be a protective compensation response of the organism to cancer cells.
2. Materials and methods 2.1. Antibodies and cell lines The antibodies that were used in this study are as follows: the PEconjugated CD112 (R2.525), PE-conjugated CD155 (SKII.4) and purified anti-CD226 mAbs (DX11) were purchased from BD Biosciences (San Diego, CA, USA); the anti-Rac/Cdc42 antibody, phospho-Rac1/Cdc42 (Ser71) antibody, cyclin D1 (92G2) rabbit mAb and phospho-cyclin D1 (Thr286) (D29B3) XP® rabbit mAb were purchased from Cell
120
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
Signaling (Beverly, MA, USA); and the anti-β actin antibody was purchased from Wuhan Boster Company (China). K562 tumor cells were cultured in RPMI 1640 medium (Gibco) supplemented with 10% heat-inactivated FBS, 100 IU/ml penicillin and 100 μg/ml streptomycin. The HeLa tumor cell line was cultured in RPMI 1640 medium (Gibco) supplemented with 10% heat-inactivated FBS, 100 IU/ml penicillin and 100 μg/ml streptomycin. A non-tumor cell line, CHO-K1, was cultured in DMEM medium (Gibco) supplemented with 10% heat-inactivated FBS, 100 IU/ml penicillin and 100 μg/ml streptomycin. Cell culturing was performed at 37 °C in a 5% CO2 humidified atmosphere. Cells from a mid-log phase culture were harvested. The human NK cell line NK92 was cultured according to the specifications outlined by the American Type Culture Collection (ATCC). 2.2. Expression, purification and preparation of soluble CD226 extracellular domain The amino acid sequence of this protein can be accessed through the NCBI Protein Database under NCBI Accession # NP_006557.2 [14]. The sequences of the primers are as follows: forward primer, 5′-cgcggatccgaagaggtgctttggcatac-3′, and reverse primer, 5′-ccgctc gagttaaactctagtctttggtc-3′. The extracellular segment of human CD226 (19–248 aa) (CD226 ECD) was amplified by PCR from the total mRNA of NK-92 cells (a human NK cell line) and inserted into the pET32a vector (Novagen) to construct the expression vector pET32a/His–MBP–CD226. Escherichia coli Rosetta (DE3) cells were successfully transformed with the constructed expression plasmid and were grown in Luria–Bertani broth containing kanamycin (100 mg/l) and chloromycetin (34 mg/l). The expression of the recombinant protein was induced during the exponential phase with 1 mM isopropyl-β-D -thiogalactoside (IPTG). The E. coli cells were cultured for another 4 h at 37 ºC, harvested and then lysed. The target proteins were found mainly in the supernatant and were fused to His and MBP tags. The protein was purified using Ni Sepharose (Amersham Biosciences) and Superdex-200 (GE Healthcare). The TEV protease with a His tag was prepared from pGEX4T2 vector-induced expression in E. coli BL21 (DE3) and purified using Ni Sepharose. The TEV protease was mixed with the His–MBP–CD226 recombinant protein at an equal dose and incubated overnight at 4 °C. The mixture was purified using Ni Sepharose to remove the MBP tag and TEV protease, which both possessed a His tag. The protein was purified using Superdex200 to obtain the pure, soluble CD226 extracellular domain protein. 2.3. Flow cytometry The cells were washed with PBS twice, blocked and stained with saturating concentrations of the appropriate fluorochrome-conjugated mAbs for 30 min at 4 °C. Thereafter, the cells were washed with PBS twice and analyzed using a FACSCalibur (Becton Dickinson). For the binding analysis of CD226, the sCD226 protein was added as blocking antibody with its final concentration at 1 μg/ml, and the cells were then washed with PBS twice and incubated with PE-conjugated CD112 mAb. 2.4. Western blot analysis K562 or HeLa cells were cultured in RPMI 1640 medium containing 10% FBS and equal doses of sCD226 or BSA proteins for 1 h. The cells were harvested and boiled in Laemmli sample buffer. Then, proteins from 106 cells were resolved by SDS-PAGE (from 7.5% to 12.5% acrylamide), transferred to PVDF (polyvinylidene fluoride) membranes and probed with primary antibodies. The proteins were visualized using a secondary antibody that was conjugated to HRP (horseradish peroxidase) and chemiluminescence detection (enhanced chemiluminescence).
2.5. Cell staining with CFSE to detect the cell proliferation K562 cells were stained with CFSE (carboxyfluorescein diacetate succinimidyl ester). Briefly, the cells were resuspended in 1 ml PBS/1% BSA at a final concentration of 5 × 106/ml and then incubated with CFSE (2 μM, Sigma) for 10 min at 37 °C. Quench staining was performed on ice for 5 min by adding five volumes of ice-cold RPMI 1640/10% FBS. Then, the cells were washed three times with ice-cold PBS/1% BSA and cultured under the appropriate conditions. 2.6. Cell migration assay HeLa or CHO-K1 cells were planted at a cell density of 105 per well. A scratch was made in the cell monolayer using a pipette tip. Then, the cells were washed with the appropriate medium containing 2% FBS to remove the suspended cells from the scratch. The appropriate medium containing 2% FBS was added to the cells to eliminate any effects due to cell proliferation, and the cell migration distances were equal to the mean width of the scratch at 0 h minus the mean width of the scratch at 36 h. 2.7. Statistical analysis Statistical analyses were performed using Student's t test. All p values were two-tailed, and p b 0.05 was considered statistically significant. 2.8. Bioinformatics tools We used the ExPASy online tools to predict the structure of the CD226 protein. The SMART tool (http://smart.embl-heidelberg.de/ smart/set_mode.cgi?NORMAL=1) was used to predict the domains of CD226, and the PSIpred tool (http://bioinf.cs.ucl.ac.uk/psipred/) was used to predict the secondary structure of the CD226 protein sequence. 3. Results 3.1. Recombinant expression of the ectodomain of the human CD226 receptor To explore the function of soluble CD226 (sCD226), we constructed a vector containing the extracellular segment of CD226. Because CD226 is a transmembrane protein and the sCD226 molecule might be a cleavage product of proteases that target different sites of the intact molecule, we predicted the structure of the CD226 molecule (Fig. 1A). Based on the predicted secondary structure result, we selected a hydrophilic amino acid, Q248, that was not in the β-sheet (yellow arrows in Fig. 1B) to create the 19 aa–248 aa soluble CD226 extracellular segment. To obtain the soluble protein, we constructed the recombinant protein with 6His and MBP tags at its N terminus (Fig. 1C) and soluble CD226 protein was expressed in E. coli. We purified the recombinant protein by using Ni Sepharose. The recombinant protein had a molecular weight of around 68 kD on an SDSPAGE gel, which is consistent with its estimated size (Fig. 2A). Moreover, the peak of the recombinant protein was observed in the molecular sieve chromatogram (Fig. 2B). After being digested by TEV protease, there were three main bands on the SDS-PAGE gel: the undigested protein band (68 kD), the 6His–MBP tag band (40 kD) and the TEV protease (27 kD)/CD226 segment (CD226 ECD, 28 kD) mixed band (Fig. 2C). To obtain purer CD226 protein (no tags CD226 ECD, 28 kD), we used Ni Sepharose and molecular sieve chromatography to remove the 6His–MBP segments and the TEV protease with His tag. The CD226 ECD was observed at 28 kD band in the Ni Sepharose effluent lane on the SDSPAGE gel and the peak of CD226 ECD in the chromatogram (Fig. 2D). The protein were identified by LC–MS. Eight different peptides were
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
121
Fig. 1. Expression vector construction for the ectodomain protein of the human CD226 receptor. (A) Prediction of the tertiary structure of the CD226 protein. According to the protein sequence of CD226, we predicted the tertiary structure of the CD226 protein using the ExPASy SMART tool. (B) Prediction of the secondary structure of the CD226 protein. Based on the protein sequence of CD226, we predicted the secondary structure of the CD226 protein using the ExPASy PSIpred tool. Gln248 was involved in the coil conformation. (C) The strategy for the construction of the pET-32a(+)–6His–MBP–CD226 extracellular segment (19 aa–248 aa) expression vector. Using total RNA from the CD226-positive, human NK-92 cell line as a template, PCR was performed to amplify the target fragment. The gene encoding the two extracellular domains of CD226 was then inserted into the pET32a(+) vector that contained a 6×His–MBP-tag on its N terminus.
found, covering 47.62% of the total CD226 residue sequences and nearly all the extracellular regions of CD226 (Fig. 2E). 3.2. sCD226 inhibits the proliferation of CD226 ligand-positive tumor cell lines We then assessed the effect of sCD226 on tumor cell growth using CD226 ligand (CD155 and CD112)-expressing cancer cell lines and a CD226 ligand-negative CHO-K1 cell line as a control (Fig. 3A). After 3 days of co-culturing the cells with sCD226, we counted the number of tumor cells. The results suggested that the numbers of K562 and HeLa cells, but not the number of control CHO-K1 cells, were significantly reduced. The addition of the CD226 blocking monoclonal antibody to the sCD226 co-culture system attenuated the inhibitory effect of sCD226 on tumor cell growth (Fig. 3B). To confirm the dose-dependent inhibitory effect of sCD226 on tumor growth, we examined the binding of the CD226 protein to tumor cells. The flow cytometry results demonstrated that the sCD226 proteins could compete with the PE-conjugated anti-CD112 antibody and block the binding of the PE-conjugated anti-CD112 antibody to tumor cells (Fig. 4A and B). Furthermore, in a tumor cell growth inhibition experiment, we observed that sCD226 dose-dependently inhibited tumor cell growth during a 72-h culture (Fig. 4C). 3.3. sCD226 inhibits the proliferation of leukemia cells by delaying the cell cycle The reduction in cell number could be due to either cell death or cell cycle delay. We used both FACS and Western blot to test cell death. FACS analysis revealed that Hela and K562 cell deaths were not increased with Annexin V–PI (propidium iodide) double stained after incubated with sCD226 for 72 h (data not shown). We also detected the Caspase8 precursors by Western blot in Hela cells and K562 cells after being incubated with the sCD226 for 30 min. It was not degraded in all the experimental groups (data not shown). The results from two methods were coincident. Therefore, we examined the cell cycle of the tumor cells. We selected and stained K562 cells with CFSE. Because CFSE is equally distributed to the daughter cell during cell division, the fluorescence intensity of the tumor cell lines represents the number of times the tumor cell lines have divided. We detected the CFSE fluorescence intensity of the stained
tumor cells at 24, 48, 72 and 96 h. The results showed that the CFSE fluorescence intensity decreased with cell division (Fig. 5A) and that in the sCD226 co-culture systems, tumor cell division was delayed dramatically, with the tumors having higher numbers of undivided cells (cells with higher CFSE fluorescence intensity) (Fig. 5B and C). The inhibitory effect of sCD226 on the cell cycle began at 24 h and became more obvious at 48 and 72 h (Fig. 5B and C). 3.4. sCD226 induces the phosphorylation of cyclin D and the Rac/Cdc42 complex A previous study demonstrated that the trans interactions of nectins and nectin-Necl 5 (CD155) usually activate the Rac/Cdc42 protein complex [15], and the activation of Cdc42 enables the cells to avoid contact inhibition [16]. The Akt phosphorylation site at Ser71 of Rac1/Cdc42 has been identified, and phosphorylation at this site may inhibit GTP binding of Rac1, thereby attenuating the signal transduction pathway downstream of Rac1 [17]. Therefore, we examined the phosphorylation at the Ser71 site of the Rac/Cdc42 complex in HeLa cells after coculturing with sCD226 for 1 h. The phosphorylation levels of the Rac/ Cdc42 protein were up-regulated (Fig. 6A and B), and although the expression level of cyclin D was slightly up-regulated, the phosphorylation at Thr286 was significantly up-regulated (Fig. 6C and D). These results suggest that serum sCD226 may inhibit the activation of Rac/Cdc42 and phosphorylation-dependent degradation of cyclin D, leading to tumor cell cycle delay. 3.5. sCD226 might impact the metastatic potential of solid tumors in vitro Because some reports suggest that Rac/Cdc42 promotes cell migration [18,19]. We also examined the metastatic potential of solid tumor cells (HeLa cells) after sCD226 treatment. In this experiment, we cultured cells in 2% low-serum culture medium to exclude proliferation effects. The results showed that sCD226 could inhibit the metastatic potential of HeLa cells. After a CD226 blocking antibody was added to the sCD226 co-culture system, the metastatic potential of the tumor cells increased to a certain extent (Fig. 7A and B). To confirm this result, we selected the CD226 ligand-negative cell line CHO-K1 and performed the same test. The results showed that the metastatic potential of the CHO-K1 cells was not affected by sCD226 (Fig. 7A and C).
122
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
Fig. 2. Purification and identification of the extracellular domains of the soluble CD226 protein (CD226 ECD). (A) The recombinant 6His–MBP–CD226 extracellular segment (19 aa–248 aa) protein was detected using SDS-PAGE analysis. The supernatant (lane 2) and pellet (lane 3) of the lysates from Rosetta (DE3)/pET32a(+)–6His–CD226 cells were incubated with 1 mM IPTG at 37 °C for 6 h. The flow-through solution (lane 4), wash solution containing 50 mM imidazole (lane 5) and elution solution containing 500 mM imidazole (lane 6) from Ni Sepharose binding were also detected using SDS-PAGE. The proteins were separated by 12% SDS-PAGE. The arrowheads indicate the position of the recombinant protein. (B) Purification of the recombinant protein via a Superdex-200 molecular sieve. The protein was collected by the no. 29 to no. 43 collection tubes and detected using SDS-PAGE. The arrowheads indicate the peak of the recombinant protein. (C) Removal of the 6His and MBP tags from the recombinant protein from the N terminus using TEV protease. The CD226 extracellular segment (19 aa–248 aa) with no tags that was purified using Ni Sepharose (lane 1), protein marker (lane 2), the protein solution after digestion and before purification (lane 3 and lane 4, respectively) and the recombinant 6His–MBP–CD226 extracellular segment (19 aa–248 aa) protein before digestion (lane 5) were detected using SDS-PAGE analysis. (D) Ni Sepharose and Superdex-200 were used to purify the soluble CD226 extracellular segment (19 aa–248 aa) protein that had been digested by TEV protease with no tags. We purified the digested proteins via Ni Sepharose to remove the TEV proteins, MBP tags and undigested 6His–MBP–CD226 extracellular segment (19 aa–248 aa) proteins with a His tag. The soluble CD226 extracellular segment (19 aa– 248 aa) protein with no tag flow-through solution from Ni Sepharose binding was collected and purified using Superdex-200. The soluble CD226 extracellular segment (19 aa–248 aa) protein was collected in the no. 53 to no. 66 collection tubes and then detected by SDS-PAGE. The arrowheads indicate the peak of the CD226 extracellular segment (19 aa–248 aa) protein. (E) Sequencing of the CD226 extracellular segment (19 aa–248 aa) protein by LC–MS. The underlined amino acid residue sequences were identified using an LTQ Mass Spectrometer.
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
123
Fig. 3. sCD226 inhibits the proliferation of CD226 ligand (CD155/CD112)-positive tumor cell lines. (A) Flow cytometric analysis of CD226 ligand (CD155/CD112) expression by tumor cells. K562, HeLa and CHO-K1 cells were incubated with the PE-conjugated anti-CD155 mAb (SKII.4) or PE-conjugated anti-CD112 mAb (R2.525) for 30 min at 4 °C and then examined using FACS. (B) Cell proliferation assay. K562 cells or HeLa cells (104) were cultured with 40 μg of BSA, 40 μg of sCD226 or 40 μg of sCD226, followed by either 40 μg of anti-CD226 mAb or 40 μg of sCD226, and last with 40 μg of IgG isotype for 72 h. CHO-K1 cells (104) were cultured with 40 μg of BSA or 40 μg of sCD226 for 72 h. The cell numbers were then counted and recorded. The data are representative of three independent experiments.
4. Discussion In this study, we explored the in vitro function of recombinant sCD226, which mimics serum sCD226. We found that sCD226 could block tumor cell proliferation via the phosphorylation-dependent
degradation of cyclin D to delay the cell cycle. sCD226 led to the phosphorylation of the RAC/Cdc42 complex and promoted cell contact inhibition effects on the cell cycle. In a previous study, CD226 was found to be a membrane protein that mediates the cytotoxicity and adhesion of NK cells and CTLs [13].
Fig. 4. sCD226 dose-dependently inhibits the proliferation of K562 leukemia cells. (A) The binding of sCD226 to K562 cells. K562 cells were incubated with PE-conjugated anti-CD112 mAb (R2.525) and sCD226 at concentrations of 20, 40, 80 and 160 μg/ml for 30 min at 4 °C. The fluorescence intensity of the K562 cells was detected using FACS. (B) Statistical analysis of the percentage of the CD112 mAbs that bound to the K562 cells in A. (C) Proliferation inhibitory effect of sCD226. K562 cells were cultured with sCD226 or BSA at concentrations of 10, 20, 40 or 80 μg/ml for 72 h, and the cell numbers were counted. The data are representative of three independent experiments.
124
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
Fig. 5. sCD226 inhibits the proliferation of K562 cells by delaying the cell cycle. (A) Flow cytometric analysis of tumor cell proliferation. K562 cells were stained with CFSE and cultured in RPMI-1640 containing 10% FBS. The CFSE fluorescence intensity of the K562 cells at 24 h (black), 48 h (green), 72 h (blue) and 96 h (purple) is shown. (B) Flow cytometric analysis of the tumor cell proliferating inhibition by sCD226. K562 cells were stained with CFSE and cultured in RPMI-1640 with 10% FBS with an equal volume of protein buffer (control), 40 μg of BSA or 40 μg of sCD226. The CFSE fluorescence intensity of K562 cells in each experimental group at 24 h, 48 h and 72 h is shown. (C) Statistical analysis of the mean fluorescence intensity of K562 cells in B, *p b 0.05 and **p b 0.01.
Soluble CD226 was discovered in 2003; however, its function is unclear. Because CD226 is an activation receptor, its soluble form is usually considered to inhibit cytotoxicity. As reported by Jin et al., sCD226 might potentially block the cytotoxicity of NK cells through blocking CD155 or CD112 on tumor cells, which might promote tumor cell escape from NK or CTL cytolysis [13]. We also detected the impaction of the
sCD226 on PBMC cytotoxicity to K562 cells. We found that the sCD226 could partly reduce the PBMC cytotoxicity, however, the reduction was limited and does not exceed 20% of the PBMC original cytotoxicity (data not shown). We think that the reason why sCD226 function is limited is that there are several activation receptors, such as NKG2D, NKp30, and so on, leading to the PBMC cytotoxicity that can't be
Fig. 6. sCD226 induces the phosphorylation of cyclin D and the Rac/Cdc42 complex. (A) Western blot analysis of the Rac/Cdc42 complex in HeLa cells. Western blot analysis of phosphorylation at Ser71 of the Rac/Cdc42 complex in HeLa cells was performed. HeLa cells were incubated with an equal volume of protein buffer (control), 40 μg of BSA or 40 μg of sCD226 for 1 h. (B) Statistical analysis of the mean gray values of the phosphorylation of the Rac/Cdc42 complex in the three experimental groups. The data were collected from three independent experiments and analyzed using a t test, *p b 0.05. (C) Western blot analysis of the cyclin D complex in HeLa cells. Western blot analysis of the phosphorylation at the Thr286 site of the cyclin D complex in HeLa cells. HeLa cells were incubated with an equal volume of protein buffer (control), 40 μg of BSA or 40 μg of sCD226 for 1 h. (D) Statistical analysis of the mean gray values of the phosphorylation of cyclin D in the three experimental groups. The data were collected from three independent experiments and analyzed using a t test, *p b 0.05.
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
125
Fig. 7. sCD226 inhibits the metastatic potential of tumor cells. (A) The cell scratch test was used to assess cell migration. Cells were planted at a high cell density. A scratch was made on the plate's surface using a pipette tip, and the HeLa cells were cultured in RPMI-1640 with 2% FBS containing the protein buffer, 40 μg of BSA, 40 μg of sCD226, 40 μg of sCD226 or 40 μg of CD226 blocking antibody for 36 h. CHO-K1 cells were cultured in DMEM with 2% FBS containing the protein buffer, 40 μg of BSA, 40 μg of sCD226, 40 μg of sCD226 or 40 μg of CD226 blocking antibody for 36 h. The scratches for each experimental group at 0 and 36 h are shown. (B) Statistical analysis of the mean HeLa cell migration distance. The cell migration distances were equal to the mean width of the scratch at 0 h minus the mean width of the scratch at 36 h. The data were collected from three independent experiments and analyzed using a t test, *p b 0.05. (C) Statistical analysis of the mean CHO-K1 cell migration distance. The cell migration distances were equal to the mean width of the scratch at 0 h minus the mean width of the scratch at 36 h. The data were collected from three independent experiments and analyzed using a t test.
completely inhibited. Moreover, in our study, we found that sCD226 directly inhibits the proliferation of the cancer cells without co-culture with PBMC. And the reduction of the tumor cell number is nearly 50% compared with the control group to maximum extent (Figs. 3B and 4C). Therefore, we think that the used sCD226 in our study, different from the one working as blocker in Jin's work, may directly combine with its ligand on tumor cells in a manner to deliver proliferation–inhibition signaling. Soluble CD155 is also found in the serum. Because CD155 is a poliovirus receptor, serum sCD155 is considered to reduce poliovirus entry mediated by membrane-bound CD155 [20,21]; however, soluble CD155 also blocks the recognition of cancer cells by the immune system, which helps tumors escape from immune attack. Therefore, sCD226 might be a means to neutralize sCD155, but the biological and pathological importance of this mechanism requires further investigation. We detected cell apoptosis via PI–Annexin V staining and FACS assays and the expression levels of Caspase-8 by Western blotting (data not shown). In this study, we discovered that CD226 did not induce cell apoptosis. CD226, as an adhesion molecule and member of the immunoglobulin superfamily, cannot activate cell apoptosis pathways. Therefore, this adhesion molecule is associated with the cytoskeleton, cell migration and proliferation [6]. Because of the strong proliferation ability and metastatic potential of cancer cells, the treatment of cancer is still a challenge, and the recognition and killing of cancer cells still requires long-term research. Many immunotherapies focused on finding a method to activate the immune system and kill the tumor cells [22]. However, from another perspective, if the proliferation ability and metastatic potential of cancer cells could be inhibited, the harm of a cancer cell to human health will be reduced to a minimum. In our study, we discovered that the soluble
CD226 protein could inhibit the proliferation of cancer cells in vitro and might inhibit cancer cells' metastatic potential. If soluble CD226 can be used clinically as an intravenous drug, it would accelerate our quest to conquer cancer. These results suggest that soluble CD226 is worthy to develop an antitumor drug.
Acknowledgment This work was supported by the Ministry of Science & Technology of China (973 Basic Science Project; 2010CB911901, 2013CB530506), the Natural Science Foundation of China (91029303, 31021061) and the National Science & Technology Major Projects (2012ZX10002014, 2012ZX10002006).
References [1] Burns GF, Triglia T, Werkmeister JA, Begley CG, Boyd AW. TLiSA1, a human T lineagespecific activation antigen involved in the differentiation of cytotoxic T lymphocytes and anomalous killer cells from their precursors. J Exp Med 1985;161:1063–78. [2] Shibuya A, Campbell D, Hannum C, Yssel H, Franz-Bacon K, McClanahan T, et al. DNAM-1, a novel adhesion molecule involved in the cytolytic function of T lymphocytes. Immunity Jun 1996;4(6):573–81. [3] Scott JL, Dunn SM, Jin B, Hillam AJ, Walton S, Berndt MC, et al. Characterization of a novel membrane glycoprotein involved in platelet activation. J Biol Chem 1989;264:13475–82. [4] Sherrington PD, Scott JL, Jin B, Simmons D, Dorahy DJ, Lloyd J, et al. TLiSA1 (PTA1) activation antigen implicated in T cell differentiation and platelet activation is a member of the immunoglobulin superfamily exhibiting distinctive regulation of expression. J Biol Chem 1997;272:21735–44. [5] Cheng M, Chen Y, Xiao W, Sun R, Tian Z. NK cell-based immunotherapy for malignant diseases. Cell Mol Immunol 2013;10:230–52. [6] Takai Yoshimi, Ikeda Wataru, Ogita Hisakazu, Rikitake Yoshiyuki. The immunoglobulin-like cell adhesion molecule nectin and its associated protein afadin. Annu Rev Cell Dev Biol 2008;24:309–42.
126
S. Hou et al. / International Immunopharmacology 19 (2014) 119–126
[7] Xu Zhuwei, Jin Boquan. A novel interface consisting of homologous immunoglobulin superfamily members with multiple functions. Cell Mol Immunol 2010;7:11–9. [8] Mendelsohn CL, Wimmer E, Racaniello VR. Cellular receptor for poliovirus: molecular cloning, nucleotide sequence, and expression of a new member of the immunoglobulin superfamily. Cell Mar 10 1989;56(5):855–65. [9] Fabre S, Reymond N, Cocchi F, Menotti L, Dubreuil P, Campadelli-Fiume G, et al. Prominent role of the Ig-like V domain in trans-interactions of nectins. Nectin3 and nectin 4 bind to the predicted C–C′–C″–D beta-strands of the nectin1 V domain. J Biol Chem Jul 26 2002;277(30):27006–13. [10] Ogita H, Takai Y. Activation of Rap1, Cdc42, and rac by nectin adhesion system. Methods Enzymol 2006;406:415–24. [11] Nakanishi H, Takai Y. Roles of nectins in cell adhesion, migration and polarization. Biol Chem Oct 2004;385(10):885–92. [12] Irie K, Shimizu K, Sakisaka T, Ikeda W, Takai Y. Roles and modes of action of nectins in cell–cell adhesion. Semin Cell Dev Biol Dec 2004;15(6):643–56. [13] Xu Z, Zhang T, Zhuang R, Zhang Y, Jia W, Song C, et al. Increased levels of soluble CD226 in sera accompanied by decreased membrane CD226 expression on peripheral blood mononuclear cells from cancer patients. BMC Immunol Jun 2 2009;10:34. [14] Bottino C, Castriconi R, Pende D, Rivera P, Nanni M, Carnemolla B, et al. Identification of PVR (CD155) and Nectin-2 (CD112) as cell surface ligands for the human DNAM-1 (CD226) activating molecule. J Exp Med Aug 18 2003;198(4):557–67.
[15] Nheu T, He H, Hirokawa Y, Walker F, Wood J, Maruta H. PAK is essential for RAS-induced up regulation of cyclin D1 during the G1 to S transition. Cell Cycle Jan 2004;3(1):71–4. [16] Appledorn DM, Dao KH, O'Reilly S, Maher VM, McCormick JJ. Rac1 and Cdc42 are regulators of HRasV12-transformation and angiogenic factors in human fibroblasts. BMC Cancer Jan 12 2010;10:13. http://dx.doi.org/10.1186/1471-2407-10-13. [17] Kwon T, Kwon DY, Chun J, Kim JH, Kang SS. Akt protein kinase inhibits Rac1-GTP binding through phosphorylation at serine 71 of Rac1. J Biol Chem Jan 7 2000;275(1):423–8. [18] Valderrama F, Thevapala S, Ridley AJ. Radixin regulates cell migration and cell–cell adhesion through Rac1. J Cell Sci Jul 15 2012;125(Pt. 14):3310–9. [19] Lai FP, Szczodrak M, Oelkers JM, Ladwein M, Acconcia F, Benesch S, et al. Cortactin promotes migration and platelet-derived growth factor-induced actin reorganization by signaling to Rho-GTPases. Mol Biol Cell Jul 2009;20(14):3209–23. [20] Koike S, Horie H, Ise I, Okitsu A, Yoshida M, Iizuka N, et al. The poliovirus receptor protein is produced both as membrane-bound and secreted forms. EMBO J Oct 1990;9(10):3217–24. [21] Baury B, Masson D, McDermott Jr BM, Jarry A, Blottière HM, Blanchardie P, et al. Identification of secreted CD155 isoforms. Biochem Biophys Res Commun Sep 12 2003;309(1):175–82. [22] Fu Yang-Xin. New immune therapy targets tumor-associated environment: from bone marrow to tumor site. Cell Mol Immunol 2012;9:1–2.
