Overexpression of ANXA1 and EphA2 has been linked to various cancers and both proteins have attracted considerable attention for the development of new anticancer drugs. Here we report that ANXA1 competes with Cbl for binding EphA2 and increases its stability by inhibiting Cbl-mediated EphA2 ubiquitination and degradation in nasopharyngeal carcinoma (NPC). Binding of ANXA1 to EphA2 promoted NPC cell growth and metastasis in vitro and in vivo by elevating EphA2 levels and increasing activity of EphA2 oncogenic signaling (pS897-EphA2). Expression of ANXA1 and EphA2 was positively correlated and both were significantly higher in NPC tissues than in the normal nasopharyngeal epithelial tissues. Patients with high expression of both proteins presented poorer disease-free survival and overall survival relative to patients with high expression of one protein alone. Furthermore, amino acid residues 20-30aa and 28-30aa of the ANXA1 N-terminus bound EphA2. An 11 amino acid–long ANXA1-derived peptide (EYVQTVKSSKG) was developed on the basis of this N-terminal region, which disrupted the connection of ANXA1 with EphA2, successfully downregulating EphA2 expression and dramatically suppressing NPC cell oncogenicity in vitro and in mice. These findings suggest that ANXA1 promotes NPC growth and metastasis via binding and stabilization of EphA2 and present a strategy for targeting EphA2 degradation and treating NPC with a peptide. This therapeutic strategy may also be extended to other cancers with high expression of both proteins.

Significance:

These findings show that EphA2 is a potential target for NPC therapeutics and an ANXA1-derived peptide suppresses NPC growth and metastasis.

Annexin A1 (ANXA1) is a Ca2+ and phospholipid-binding protein (1). It plays a role in the regulation of inflammation and immunity, and cell proliferation, apoptosis, and differentiation (2–6). ANXA1 expression is deregulated in various cancers, which has been linked to tumor development and metastasis (7–11), and it is a potential target for novel therapeutic intervention (12, 13).

Eph receptors belong to a large family of receptor tyrosine kinases (RTK), are key regulators of both normal development and disease (14, 15). Perturbation of Eph receptor and ligand system has been observed in the various human cancers. Particularly, EphA2 is the most frequently affected Eph receptor in the human cancers (15). EphA2 is overexpressed in many human cancers, where it promotes tumor growth, metastasis, and cancer stem properties through a ligand-independent mechanism (16–21). As EphA2 is an important oncogenic protein and emerging drug target, approaches for targeting downregulation of EphA2 have attracted a considerable interest as anticancer strategies (22, 23).

Protein–protein interaction (PPI) controls various cellular functions through modulating protein posttranslational modification, stability, subcellular location, etc. Aberrant PPI is associated with cancers, representing a pivotal target for chemicobiological interventions (24–26). Numerous studies have indicated that inhibition of PPI by peptides is an efficient anticancer approach (27–29). We recently used immunoprecipitation and mass spectrometry (IP-MS) analysis to search proteins interacted with EphA2 in nasopharyngeal carcinoma (NPC) cells, and found that ANXA1 and Cbl, an E3 ubiquitin ligase of EphA2 (30–32), are interactors of EphA2. However, the physiologic and pathologic significances of ANXA1–EphA2 interaction are completely unclear.

NPC is a common head and neck cancer in southern China and Southeast Asia, and remains one of the leading lethal malignancies in these areas (33). NPC is a highly malignant cancer, and often invades adjacent regions and metastasizes to distant organs, which is a major cause for NPC lethality (34). Understanding the mechanisms of NPC development and metastasis will allow for the development of effective approaches. Our previous studies found that both ANXA1 and EphA2 are overexpressed and promote NPC growth and metastasis (35, 36).

In this work, we investigated the roles of ANXA1–EphA2 interaction in NPC growth and metastasis. On the basis of data from mapping PPI between these two proteins, we developed a peptide that disrupts ANXA1–EphA2 interaction and possesses anti-NPC effects.

Human NPC specimens

NPC and normal nasopharyngeal epithelial tissues used in this study were collected from Xiangya Hospital of Central South University (Changsha, China). Detailed clinicopathologic features are presented in the Supplementary Table S1. The use of human tissues was approved by the Ethics Committee, Xiangya Hospital of Central South University (Changsha, China). As only archived tumor specimens were included in this study, the ethics committee waived the need for consent, and patient records/information were analyzed anonymously. For details, see the Supplementary Materials and Methods section.

Animal experiments

Four-week-old nude male mice (BALB/c nu/nu) were obtained from the Laboratory Animal Center of Central South University and maintained in pathogen-free conditions. All animal experimental procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals of Xiangya Hospital, Central South University (Changsha, China), with the approval of the Institutional Animal Ethics Committee. For details, see the Supplementary Materials and Methods section.

Cell lines

Human NPC cell line HK1 was kindly gifted by Dr. Tao of the Chinese University of Hong Kong, and human NPC cell lines 5-8F and 6-10B, and colon cancer cell lines SW480 and SW620, and HEK293 cell line have been described previously by us (35–37). Cells were cultured in RPMI1640 medium supplemented with 10% FBS at 37°C in 5% CO2. The cell lines were authenticated by short tandem repeat profiling prior to use, and were routinely tested negative for Mycoplasma contamination using 4,6-diamidino-2-phenylindole staining.

IP-MS

IP-MS was preformed to search proteins that interact with EphA2 in the NPC cells as described previously by us (38). The detailed procedures are described in the Supplementary Materials and Methods section.

Establishment of NPC cell lines with the expression changes of ANXA1 and EphA2

NPC cells were infected with the indicated lentiviral particle following the manufacturer's instruction, or transfected with the indicated plasmid using Lipofectamine 2000, and then selected using antibiotics for 2 weeks. NPC cell lines with the stable knockdown of endogenous ANXA1 and EphA2, and with stable knockdown or overexpression of ANXA1 or EphA2 were established.

Immunoblotting

Immunoblotting was performed as described previously by us (35, 36). Briefly, proteins were exacted from cells using RIPA lysis buffer. An equal amount of protein in each sample was subjected to SDS-PAGE separation, followed by blotting onto a polyvinylidene difluoride membrane. After blocking, blots were incubated with primary antibodies overnight at 4°C, followed by incubation with horseradish peroxidase–conjugated secondary antibody for 2 hours at room temperature. The signal was visualized with an enhanced chemiluminescence detection reagent (Roche).

Immunoprecipitation and immunoblotting

Immunoprecipitation and immunoblotting (co-IP) was performed to detect PPI and EphA2 ubiquitination. In brief, whole cell lysates were precleared with Protein A/G-Sepharose 4B, and incubated with indicated antibodies and Protein G/A-Sepharose 4B overnight at 4°C. After five times wash with RIPA buffer, beads were boiled in 2 × SDS-PAGE loading buffer for 5 minutes to elute protein complexes, followed by SDS-PAGE separation and immunoblotting with specific antibodies.

GST pull-down assay

GST fusion EphA2 protein was immobilized on glutathione agarose, and incubated with Histidine fusion ANXA1 protein purified by Ni affinity chromatography in GST pull-down buffer for 4 hours at 4°C. After washing five times, the bound proteins were dissolved in SDS sample buffer, separated by SDS-PAGE, and subjected to immunoblotting with antibodies against GST or His.

Biotin pull-down assay

Biotin pull-down assay was performed to detect the interaction of peptide and EphA2 as described previously (39). In brief, 1 mg of whole cell lysates were incubated with 0–60 nmol/L biotin-labeled ANXA1-derived peptide overnight at 4°C, and then incubated with 30 μL streptavidin agarose beads for 4 hours at 4°C. After five times wash with PBS buffer, beads were boiled in 2 × SDS-PAGE loading buffer for 5 minutes, followed by SDS-PAGE separation and immunoblotting with EphA2 antibodies.

qRT-PCR

qRT-PCR was performed to detect the expression of ANXA1 and EphA2 in the indicated cells as described previously by us (35, 36). The primers are presented in the Supplementary Table S2.

Immunofluorescent staining

Immunofluorescent staining of EphA2, ANXA1, and Cbl in the indicated cells was performed as described previously by us (35, 36). The detailed procedures are described in the Supplementary Materials and Methods section.

IHC and staining evaluation

IHC and staining evaluation of ANXA1 and EphA2 were performed on the formalin-fixed and paraffin-embedded tissue sections as described previously by us (40). The detailed procedures are described in the Supplementary Materials and Methods section.

Cell counting kit-8 assay

Cell proliferation was measured using a cell counting kit-8 (CCK-8) kit as described previously by us (37). The detailed procedures are described in the Supplementary Materials and Methods section.

5-Ethynyl-2′-deoxyuridine incorporation assay

Cell proliferation was measured using 5-Ethynyl-2′-deoxyuridine (EdU) incorporation assay as described previously by us (37). The detailed procedures are described in the Supplementary Materials and Methods section.

Soft agar colony formation assay

Soft agar colony formation assay was performed to detect cell anchorage–independent growth as described previously by us (37). The detailed procedures are described in the Supplementary Materials and Methods section.

Scratch wound-healing and Matrigel invasion assay

Scratch wound-healing and Matrigel invasion assay was performed to detect cell migration and invasion as described previously by us (35, 36). The detailed procedures are described in the Supplementary Materials and Methods section.

Molecular docking

The docking models EphA2–ANXA1 and EphA2–Cbl complexes were generated using the ClusPro web server for protein–protein docking (41) based on the crystal structures of EphA2 (Protein Data Bank code 1MQB), Cbl (2Y1M), and ANXA1 (1MCX). Structural illustrations were prepared using the PyMOL Molecular Graphic Systems (version 0.99, Schrödinger LLC; http://www.pymol.org/).

Statistical analysis

Statistical analysis was performed using IBM SPSS statistical software package 22. Data are presented as means ± SD. Qualitative variables were compared by two-tailed unpaired Student t test or one-way ANOVA test; classification variables were compared by χ2 test. Kaplan–Meier survival analysis was used to compare patient survival by the log-rank test. Cox proportional hazards regression analysis was used to analyze the effect of clinical variables on patient survival. P values <0.05 were considered statistically significant.

Data availability

The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD015242.

Binding of ANXA1 to EphA2 increases EphA2 stability by ubiquitin proteasome pathway

To search proteins that interact with EphA2, EphA2 interactors were coimmunoprecipitated with anti-EphA2 antibody from NPC cell extracts, separated on SDS-PAGE and stained with Coomassie blue (Supplementary Fig. S1A). All protein bands were excised and subjected to LC/MS-MS analysis. As a result, a total of 1,352 proteins including ANXA1 and Cbl were identified (Supplementary Fig. S1B–S1D), proteomic data of which are available via ProteomeXchange with identifier PXD015242. As ANXA1 interacts with RTKs such as EGFR (42–44) and insulin receptor (45) to modulate their abundance or activity, ANXA1 was selected to further investigation. co-IP confirmed that ANXA1 physically interacted with EphA2 in the three NPC cell lines (Fig. 1A), and exogenous ANXA1 also interacted with exogenous EphA2 in HEK293 cells cotransfected with ANXA1 and EphA2 expression plasmids (Fig. 1B). In vitro GST pull-down assay using purified proteins showed that ANXA1 directly bound with EphA2 (Fig. 1C). Immunofluorescent staining showed that ANXA1 and EphA2 were colocalized in the NPC cells (Fig. 1D). We also observed that ANXA1 interacted with EphA2 in the two colonic cancer cell lines by co-IP (Supplementary Fig. S1E), indicating ANXA1–EphA2 interaction is not specific to NPC cells. Collectively, these data provide strong biochemical evidence that ANXA interacts with EphA2.

Next, we analyzed the effect of ANXA1 on EphA2 protein stability after blocking protein synthesis with cycloheximide. EphA2 was rapidly degraded in the ANXA1 knockdown NPC cells whereas had a much longer half-life in the ANXA1 overexpression NPC cells (Fig. 1E), but had not changes in its mRNA level (Supplementary Fig. S2A and S2B), indicating that ANXA1 tightly controlled EphA2 protein stability. Moreover, the decrease of EphA2 protein in the ANXA1 knockdown cells was specifically reversed by treatment with proteasome inhibitor MG132 (Fig. 1F), indicating a proteasome-dependent mechanism in EphA2 destabilization by ANXA1 knockdown.

We then investigated how ANXA1 enhances EphA2 stability. As ubiquitination is a key mechanism driving the proteasomal degradation of EphA2 (30–32, 46), we assessed the capacity of ANXA1 to modulate EphA2 ubiquitination, and observed that ANXA1 knockdown increased while ANXA1 overexpression decreased EphA2 polyubiquitination in the NPC cells (Fig. 1G), indicating that ANXA1 enhances EphA2 stabilization by inhibiting its polyubiquitination degradation. K48-linked and K63-linked polyubiquitination are the two main types of protein ubiquitination; the former leads to proteasomal degradation and the latter usually regulates protein function, signal transduction, and DNA damage response. Therefore, we examined the composition of the ANXA1-inhibited EphA2 polyubiquitin chains in the HEK293 cells transfected with the various combinations of plasmids expressing EphA2, ANXA1, Ub, Ubk48, or Ubk63. As shown in Fig. 1H, the total and K48-linked ubiquitination but not the K63-linked ubiquitination of EphA2 was obviously downregualted by ANXA1, suggesting that it inhibits the K48-linked polyubiquitination of EphA2. Together, these results indicate that binding of ANXA1 to EphA2 increases EphA2 stability by ubiquitin proteasome pathway in the NPC cells.

ANXA1 competes with Cbl for binding EphA2 to increase its stability

Our IP-MS also identified Cbl as a protein interacted with EphA2 in the NPC cells (Supplementary Fig. S1C and S1D). Cbl, a well-characterized E3-ubiquitin ligase, mediates RTKs to undergo ubiquitination (47, 48), and Cbl-mediated ubiquitination represents a key mechanism driving EphA2 proteasomal degradation (30–32, 46). Therefore, we analyzed whether ANXA1 competes with Cbl for binding EphA2 to inhibit Cbl-mediated EphA2 ubiquitination degradation in the NPC cells. We observed that ANXA1 knockdown dramatically increased Cbl bound to EphA2, and decreased EphA2 abundance, whereas ANXA1 overexpression had the opposite effects (Fig. 2A); knockdown of Cbl by siRNAs restored EphA2 levels in the ANXA1 knockdown NPC cells (Fig. 2B), being accompanied by the decrease of EphA2 polyubiquitination (Fig. 2C). Moreover, we transfected the various combinations of plasmids expressing EphA2, Cbl, or ANXA1 into HEK293 cells, and observed that Cbl decreased EphA2 levels in a dose-dependent manner, and increased its polyubiquitination, and ANXA1 antagonized Cbl affection on the levels and polyubiquitination of EphA2 (Fig. 2D and E). EphA2 internalization is a typical process for Cbl-mediated its degradation (30–32). We also observed that ANXA1 knockdown dramatically increased EphA2 internalization and colocalization of EphA2 and Cbl in the NPC cells by confocal microscopy (Fig. 2F). Together, these results suggest that ANXA1 increases EphA2 stability by competing with Cbl for binding EphA2 in the NPC cells.

Mapping of the binding region of ANXA1 and EphA2

To map the regions of ANXA1 and EphA2 responsible for their interaction, we constructed a series of deletion mutants of ANXA1 and EphA2 (Fig. 3A), and cotransfected each of ANXA1 deletion mutants with wild-type (WT) EphA2, or each of EphA2 deletion mutants with WT ANXA1 into HEK293 cells following co-IP analysis. The results showed that EphA2 D606-906 and D808-812 could not bind to ANXA1 (Fig. 3B and C), indicating that EphA2 kinase domain (KD) and its five amino acid residues (808-812aa) responsible for binding ANXA1; ANXA1 D1-40 could not bind to EphA2 (Fig. 3D), indicating ANXA1 N-terminal responsible for binding EphA2. As ANXA1 N-terminal regulates the level and activity of EGFR (42–44), we mapped its domain bound to EphA2, and found that D1-19 and D31-40 but not D20-40 and D20-30 could bind to EphA2 (Fig. 3E), indicating its 11 amino acid residues (20-30aa) responsible for binding EphA2. We further mapped the region binding EphA2, and found the amino acid residues (28-30aa) responsible for binding EphA2 (Fig. 3F). Moreover, we constructed the point mutants of the four amino acid residues (27-30aa), and observed that point mutations at S28, G30, and S28K29G30 could abolish the binding of ANAX1 to EphA2 (Fig. 3G).

Next we analyzed whether the three amino acid residues (28-30aa) have functional relevance with EphA2 stability. We transfected the plasmid expressing short hairpin (shRNA)-resistant ANXA1 or D28-30 into the NPC cells with knockdown of endogenous ANXA1 by shRNA, and observed that ANXA1 but not D28-30 could rescue EphA2 levels (Supplementary Fig. S3A). Moreover, we also observed that ANXA1 but not D28-30 could antagonize Cbl-degrading exogenous EphA2 in the HEK293 cells (Supplementary Fig. S3B). The result indicates that the three amino acid residues (28-30aa) of ANXA1 N-terminal are important for EphA2 stability.

Structural modeling of ANXA1 competing with Cbl for EphA2 binding

Our result showed that ANXA1 bound to EphA2 KD in the 808-812 amino acid residues (Fig. 3B and C). Previous reports have indicated that the SH2 domain of Cbl binds to EphA2 KD in the consensus Cbl-docking sequence (Y813XXXP; refs. 30–32). Because both proteins bind to EphA2 KD in the contiguous sites, ANXA1 may compete with Cbl for binding EphA2, which was validated by our experimental results (Fig. 2A and B). Modeling of the structure of the Cbl–EphA2 complex and ANXA1–EphA2 complex showed the EphA2-binding interface of Cbl and ANXA1 is mutually exclusive (Supplementary Fig. S4), supporting that ANXA1 competes with Cbl for binding EphA2.

Binding of ANXA1 to EphA2 promotes NPC growth and metastasis in vitro and in vivo

To test the function of ANXA1–EphA2 interaction in the NPC, we examined the effects of D28-30, which loses the ability of binding and stabilizing EphA2 (Fig. 3F; Supplementary Fig. S3), on NPC growth and metastasis. Various combinations of plasmids expressing ANXA1, D28-30, or EphA2 were cotransfected into the NPC cells with knockdown of endogenous ANXA1 and EphA2 (Supplementary Fig. S5). The results showed that reintroduction of D28-30 and EphA2 could not rescue in vitro cell growth (Fig. 4A–C), and migration and invasion (Fig. 4D and E), and in vivo cell growth (Fig. 5A and B) and lung metastasis (Fig. 5C and D) in mice as reintroduction of ANXA1 and EphA2 did. Interestingly, IHC showed that EphA2 expression was dramatically lower in the subcutaneous xenografts carrying EphA2 and D28-30 as compared with subcutaneous xenografts carrying EphA2 and ANXA1 (Fig. 5E), suggesting that ANXA1 promotes in vivo NPC cell growth by binding and stabilizing EphA2. These results indicate that binding of ANXA1 to EphA2 promotes NPC growth and metastasis in vitro and in vivo.

We also observed that reintroduction of D28-30 had the same tumor promotion as reintroduction of ANXA1 in the absence of rescued EphA2 expression, indicating that D28-30 does not affect the functions and activities of ANXA1 (Figs. 4 and 5). Moreover, reintroduction of D28-30 and EphA2 had slightly strong tumor promotion relative to reintroduction of ANXA1 or 28–30 alone, but without statistical significance (Figs. 4 and 5), supporting that NPC promotion of EphA2 needs binding of ANXA1 to EphA2. But it could not be excluded that NPC promotion of ANXA1 by other targets except EphA2.

Binding of ANXA1 to EphA2 promotes EphA2 oncogenic signaling pathway

As phosphorylation of EphA2 at S897 (S897-EphA2) and AKT, a key oncogenic signaling pathway, contributes to EphA2-promoting tumor growth and metastasis (17, 18, 21), we analyzed the effect of D28-30 on p-S897-EphA2 and p-AKT in the NPC cells. The results showed that ANXA1 knockdown reduced while overexpression increased pS897-EphA2 and p-AKT in the absence and presence of its ligand Ephrin A1 (Supplementary Fig. S6A), and reintroduction of D28-30 could not rescue pS897-EphA2 and p-AKT in the NPC cells with endogenous ANXA1 knockdown as ANXA1 did in the absence and presence of Ephrin A1 (Supplementary Fig. S6B). Moreover, we observed a similar change in the levels of EphA2 and pY772-EphA2 (Supplementary Fig. S6A and S6B). The results indicate that binding of ANXA1 to EphA2 promotes pS897-EphA2/AKT oncogenic signaling pathway by stabilizing EphA2 in the NPC cells.

ANXA1-derived peptide inhibits the in vitro and in vivo oncogenicity of NPC cells by targeting EphA2 degradation

Because ANXA1 promotes NPC growth and metastasis through binding EphA2, one potentially effective targeting therapeutic approach is to cut its connection with EphA2. Therefore, we reasonably speculate that the ANXA1-derived 11 amino acid–long peptide (20-30aa) occupies the ANXA1-binding site on EphA2, and disturbs binding of ANXA1 to EphA2, leading to ANXA1 not able to compete with Cbl for binding EphA2. For this reason, ANXA1 (20-30aa; EYVQTVKSSKG) peptide was synthesized in fusion to previously characterized cell-penetrating peptide (CPP; YGRKKRRQRRR; ref. 29), thereafter named as A11 (ANXA1-derived 11 amino acid–long peptide), and CPP was used as control. Efficient cellular uptake of both peptides was confirmed by immunofluorescent labeling with FITC (Fig. 6A). As expected, A11 dramatically decreased ANXA1 bound to EphA2 and increased Cbl bound to EphA2 (Fig. 6B); A11 efficiently decreased EphA2 level in a dose-dependent manner (Fig. 6C), and substantially increased EphA2 ubiquitination (Fig. 6D). Biotin pull-down assay showed that A11 efficiently pulled down EphA2 in the NPC cells (Fig. 6E), confirming that A11 could bind EphA2. Moreover, we also observed that A11 dramatically increased EphA2 internalization and colocalization of EphA2 and Cbl in the NPC cells (Fig. 6F). These results demonstrate that A11 blocks the competition of ANXA1 with Cbl for binding EphA2, and targets EphA2 degradation.

Next, we tested the tumor suppression function of A11. The results showed that A11 inhibited the in vitro NPC cell proliferation (Fig. 6G), and migration and invasion (Fig. 6H). We further tested the in vivo tumor suppression function of A11 via peritoneal injection into mice carrying the xenograft tumors of NPC cells. The results showed that A11 dramatically inhibited the growth of NPC cells in nude mice (Fig. 6I). IHC showed that EphA2 expression in the xenograft tumors were markedly decreased in the animals received A11 (Fig. 6J), indicating that A11 inhibiting NPC cell growth in mice also by targeting EphA2 degradation.

Finally, we tested the impact of A11 on the capillary-like structure formation of human umbilical vascular endothelial cells (HUVEC) grown in the conditioned medium from NPC cells treated with A11 peptide, and observed that medium from A11-treated cells inhibited the tube formation of HUVECs, indicating that A11 inhibits angiogenesis in vitro (Supplementary Fig. S7A and S7B). Moreover, we detected the impact of A11 on the activity of pS897-EphA2/AKT signaling pathway in the NPC cells, and observed that A11-reduced EphA2 was accompanied by a similar decrease in pS897-EphA2 and p-AKT (Supplementary Fig. S8A). IHC also showed that A11 dramatically decreased pS897-EphA2 and p-AKT in the xenografts (Supplementary Fig. S8B), indicating that A11 impairs the oncogenic mechanism of ANXA1–EphA2 interaction. Collectively, the results demonstrate that A11 appears to have dramatic anti-NPC effect in vitro and in vivo.

Association of ANXA1 and EphA2 expression with outcomes of patients with NPC

To investigate the clinical significance of ANXA1–EphA2 interaction, we detected expression of ANXA1 and EphA2 in the 184 NPC tissues, and 30 normal nasopharyngeal mucosal tissues (NNMT) by IHC. The results showed that expression levels of both ANXA1 and EphA2 were significantly higher in NPC tissues than in the NNMT (Fig. 7A), and positively correlated in NPC tissues (Fig. 7B and C). We also observed that expression levels of both ANXA1 and EphA2 were positively correlated with lymphonode and distant metastasis and clinical tumor–node–metastasis stage (Supplementary Table S3). Interestingly, patients with high level of both proteins presented poorer disease-free survival and overall survival relative to patients with high level of one protein alone (Fig. 7D and E). A univariate and multivariate Cox regression analysis showed that a combination of ANXA1 with EphA2 was an independent predictor for both DFS and OS (Supplementary Table S4). Together, these results indicate that ANXA1-stablizing EphA2 might present in the clinical NPC tissues, and a combination of ANXA1 with EphA2 could be considered as an important marker for NPC prognosis.

ANXA1 is not only a regulator of inflammation and immunity (2–6), but also deregulates in the multiple cancers, where it promotes tumor development and metastasis (7–11). Although numerous reporters have investigated how ANXA1 promotes tumor progression, the underlying mechanisms are poorly understood. In this study, we first time report that ANXA1 promotes NPC growth and metastasis by stabilizing EphA2, in which ANXA1 competes with Cbl for binding EphA2, and then inhibits Cbl-mediated EphA2 ubiquitination degradation. Our data reveal a novel mechanism on ANXA1-promoting tumor as well as regulation of EphA2 stability. Our previous study indicates that ANXA1-suppressed autophagy promotes NPC invasion and metastasis by activating PI3K/AKT signaling (35). In this study, we reveal that binding of ANXA1 to EphA2 not only promotes NPC metastasis by stabilizing EphA2, but also activates pS897-EphA2/AKT signaling. Therefore, ANXA1-suppressed autophagy promotes NPC metastasis possibly by stabilizing EphA2 and then activating pS897-EphA2/AKT signaling, which needs to be validated.

EphA2 promotes invasion and metastasis (21, 36), induces epithelial–mesenchymal transition (49), and maintains cancer stem properties (19, 20, 36). As EphA2-dependent oncogenic properties are conferred by EphA2 expression levels, EphA2 is an ideal drug target in malignancies (22, 23). Our results reveal that disrupting connection of ANXA1 and EphA2 downregulates EphA2 and inhibits NPC growth and metastasis, which may become a promising therapeutic strategy for NPC. Indeed, we found that an ANXA1-derived peptide, A11, prevents ANXA1 from competition with Cbl for binding EphA2, dramatically downregulates EphA2 expression, and inhibits NPC growth and metastasis. Therefore, targeting ANXA1–EphA2 interaction by A11 may represent a novel strategy to impair oncogenicity of EphA2, and antagonize NPC growth and metastasis. Considering the anti-NPC activity of A11 in vivo, it is worth further evaluating its safety and stability with the aim of proceeding to clinical trials in the future.

Why does A11 prevent ANXA1 from competition with Cbl for binding EphA2? As ANXA1 and Cbl bind to the contiguous sites in the EphA2 KD, and competition of ANXA1 with Cbl for binding EphA2 requires full-length ANXA1 protein. A11 binds to EphA2, and occupies the ANXA1-binding site in the EphA2 KD, leading to ANXA1 not able to binding EphA2, but a short peptide, A11, loses competition with Cbl for binding EphA2, and then increases EphA2 degradation.

We found that binding and stabilization of EphA2 by ANXA1 are important for NPC-promotion potentials of ANXA1 in vitro and in vivo. Consistent with this observation, our results showed that patients with NPC with high level of both proteins presented poorer DFS and OS relative to patients with high level of one protein alone, indicating that ANXA1-stablizing EphA2 might present in the clinical NPC tissues, and a combination of ANXA1 and EphA2 should be considered as an important prognostic marker for personalized NPC treatment.

In summary (Fig. 7F), our study identifies ANXA1 as a novel interactor of EphA2, uncovers that ANXA1 inhibits Cbl-mediated EphA2 ubiquitination degradation by competition with Cbl for binding EphA2, and demonstrates tumor-promotion function and pathologic relevance of ANXA1–EphA2 interaction in NPC. We also develop an ANXA1-derived peptide blocking ANXA1–EphA2 interaction, which downregulates EphA2 with anti-NPC effects.

No potential conflicts of interest were disclosed.

J. Feng: Data curation, investigation, methodology. S.-S. Lu: Resources, investigation, methodology. T. Xiao: Data curation, investigation, methodology. W. Huang: Data curation, investigation, methodology. H. Yi: Data curation, supervision, investigation. W. Zhu: Resources, investigation. S. Fan: Data curation, methodology. X.-P. Feng: Data curation, methodology. J.-Y. Li: Investigation, methodology. Z.-Z. Yu: Resources, data curation. S. Gao: Software, methodology. G.-H. Nie: Data curation, supervision. Y.-Y. Tang: Data curation, supervision. Z.-Q. Xiao: Conceptualization, supervision, funding acquisition, writing-original draft, project administration, writing-review and editing.

This work was supported by grants from National Key Basic Research Program (973 Program) of China (2013CB910502 to Z.-Q. Xiao), National Natural Science Foundation of China (81230053, 81874132, and 81672687 to Z.-Q. Xiao), and Shenzhen Science and Technology Program of China (KQTD20170810160226082 to Z.-Q. Xiao). We thank Dr. Jin-Yu Yang for assistance with structural modelling of Cbl–EphA2 complex and ANXA1–EphA2 complex.

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

1.
Rescher
U
,
Gerke
V
. 
Annexins–unique membrane binding proteins with diverse functions
.
J Cell Sci
2004
;
117
:
2631
39
.
2.
Perretti
M
,
D'Acquisto
F
. 
Annexin A1 and glucocorticoids as effectors of the resolution of inflammation
.
Nat Rev Immunol
2009
;
9
:
62
70
.
3.
Senchenkova
EY
,
Ansari
J
,
Becker
F
,
Vital
SA
,
Al-Yafeai
Z
,
Sparkenbaugh
EM
, et al
Novel role for the AnxA1-Fpr2/ALX signaling axis as a key regulator of platelet function to promote resolution of inflammation
.
Circulation
2019
;
140
:
319
35
.
4.
Girol
AP
,
Mimura
KK
,
Drewes
CC
,
Bolonheis
SM
,
Solito
E
,
Farsky
SH
, et al
Anti-inflammatory mechanisms of the annexin A1 protein and its mimetic peptide Ac2-26 in models of ocular inflammation in vivo and in vitro
.
J Immunol
2013
;
190
:
5689
701
.
5.
Bist
P
,
Shu
S
,
Lee
H
,
Arora
S
,
Nair
S
,
Lim
JY
, et al
Annexin-A1 regulates TLR-mediated IFN-β production through an interaction with TANK-binding kinase 1
.
J Immunol
2013
;
191
:
4375
82
.
6.
Williams
SL
,
Milne
IR
,
Bagley
CJ
,
Gamble
JR
,
Vadas
MA
,
Pitson
SM
, et al
A proinflammatory role for proteolytically cleaved annexin A1 in neutrophil transendothelial migration.
J Immunol
2010
;
185
:
3057
63
.
7.
Moraes
LA
,
Kar
S
,
Foo
SL
,
Gu
T
,
Toh
YQ
,
Ampomah
PB
, et al
Annexin-A1 enhanced breast cancer growth and migration by promoting alternative macrophage polarization in the tumour microenvironment
.
Sci Rep
2017
;
7
:
17925
.
8.
Boudhraa
Z
,
Rondepierre
F
,
Ouchchane
L
,
Kintossou
R
,
Trzeciakiewicz
A
,
Franck
F
, et al
Annexin A1 in primary tumors promotes melanoma dissemination
.
Clin Exp Metastasis
2014
;
31
:
749
60
.
9.
Bist
P
,
Leow
SC
,
Phua
QH
,
Shu
S
,
Zhuang
Q
,
Loh
WT
, et al
Annexin-1 interacts with NEMO and RIP1 to constitutively activate IKK complex and NF-κB: implication in breast cancer metastasis
.
Oncogene
2011
;
30
:
3174
85
.
10.
de Graauw
M
,
van Miltenburg
MH
,
Schmidt
MK
,
Pont
C
,
Lalai
R
,
Kartopawiro
J
, et al
Annexin A1 regulates TGF-beta signaling and promotes metastasis formation of basal-like breast cancer cells
.
Proc Natl Acad Sci USA
2010
;
107
:
6340
5
.
11.
Sakaguchi
M
,
Murata
H
,
Sonegawa
H
,
Sakaguchi
Y
,
Futami
J
,
Kitazoe
M
, et al
Truncation of annexin A1 is a regulatory lever for linking epidermal growth factor signaling with cytosolic phospholipase A2 in normal and malignant squamous epithelial cells
.
J Biol Chem
2007
;
282
:
35679
86
.
12.
Moraes
LA
,
Ampomah
PB
,
Lim
LHK
. 
Annexin A1 in inflammation and breast cancer: a new axis in the tumor microenvironment
.
Cell Adh Migr
2018
;
12
:
417
23
.
13.
Biaoxue
R
,
Xiguang
C
,
Shuanying
Y
. 
Annexin A1 in malignant tumors: current opinions and controversies
.
Int J Biol Markers
2014
;
29
:
e8
20
.
14.
Pasquale
EB
. 
Eph-ephrin bidirectional signaling in physiology and disease
.
Cell
2008
;
133
:
38
52
.
15.
Pasquale
EB
. 
Eph receptors and ephrins in cancer: bidirectional signalling and beyond
.
Nat Rev Cancer
2010
;
10
:
165
80
.
16.
Markosyan
N
,
Li
J
,
Sun
YH
,
Richman
LP
,
Lin
JH
,
Yan
F
, et al
Tumor cell-intrinsic EPHA2 suppresses anti-tumor immunity by regulating PTGS2 (COX-2)
.
J Clin Invest
2019
;
130
:
3594
609
.
17.
Zhou
Y
,
Yamada
N
,
Tanaka
T
,
Hori
T
,
Yokoyama
S
,
Hayakawa
Y
, et al
Crucial roles of RSK in cell motility by catalysing serine phosphorylation of EphA2
.
Nat Commun
2015
;
6
:
7679
.
18.
Miao
H
,
Gale
NW
,
Guo
H
,
Qian
J
,
Petty
A
,
Kaspar
J
, et al
EphA2 promotes infiltrative invasion of glioma stem cells in vivo through cross-talk with Akt and regulates stem cell properties
.
Oncogene
2015
;
34
:
558
67
.
19.
Song
W
,
Ma
Y
,
Wang
J
,
Brantley-Sieders
D
,
Chen
J
. 
JNK signaling mediates EPHA2-dependent tumor cell proliferation, motility, and cancer stem cell-like properties in non-small cell lung cancer
.
Cancer Res
2014
;
74
:
2444
54
.
20.
Binda
E
,
Visioli
A
,
Giani
F
,
Lamorte
G
,
Copetti
M
,
Pitter
KL
, et al
The EphA2 receptor drives self-renewal and tumorigenicity in stem-like tumor-propagating cells from human glioblastomas
.
Cancer Cell
2012
;
22
:
765
80
.
21.
Miao
H
,
Li
DQ
,
Mukherjee
A
,
Guo
H
,
Petty
A
,
Cutter
J
, et al
EphA2 mediates ligand-dependent inhibition and ligand-independent promotion of cell migration and invasion via a reciprocal regulatory loop with Akt
.
Cancer Cell
2009
;
16
:
9
20
.
22.
Tandon
M
,
Vemula
SV
,
Mittal
SK
. 
Emerging strategies for EphA2 receptor targeting for cancer therapeutics
.
Expert Opin Ther Targets
2011
;
15
:
31
51
.
23.
Amato
KR
,
Wang
S
,
Hastings
AK
,
Youngblood
VM
,
Santapuram
PR
,
Chen
H
, et al
Genetic and pharmacologic inhibition of EphA2 promotes apoptosis in NSCLC
.
J Clin Invest
2014
;
124
:
2037
49
.
24.
Akram
ON
,
DeGraff
DJ
,
Sheehan
JH
,
Tilley
WD
,
Matusik
RJ
,
Ahn
JM
, et al
Tailoring peptidomimetics for targeting protein-protein interactions
.
Mol Cancer Res
2014
;
12
:
967
78
.
25.
Iyer
VV
. 
A review of stapled peptides and small molecules to inhibit protein-protein interactions in cancer
.
Curr Med Chem
2016
;
23
:
3025
43
.
26.
Ferreira
LG
,
Oliva
G
,
Andricopulo
AD
. 
Protein-protein interaction inhibitors: advances in anticancer drug design
.
Expert Opin Drug Discov
2016
;
11
:
957
68
.
27.
Jafary
F
,
Ganjalikhany
MR
,
Moradi
A
,
Hemati
M
,
Jafari
S
. 
Novel peptide inhibitors for Lactate Dehydrogenase A (LDHA): a survey to inhibit LDHA activity via disruption of protein-protein interaction
.
Sci Rep
2019
;
9
:
4686
.
28.
Yeon
M
,
Byun
J
,
Kim
H
,
Kim
M
,
Jung
HS
,
Jeon
D
, et al
CAGE binds to beclin1, regulates autophagic flux and CAGE-derived peptide confers sensitivity to anti-cancer drugs in non-small cell lung cancer cells
.
Front Oncol
2018
;
8
:
599
.
29.
Liang
L
,
Wang
H
,
Shi
H
,
Li
Z
,
Yao
H
,
Bu
Z
, et al
A designed peptide targets two types of modifications of p53 with anti-cancer activity
.
Cell Chem Biol
2018
;
25
:
761
74.e5
.
30.
Walker-Daniels
J
,
Riese
DJ
 2nd
,
Kinch
MS
. 
c-Cbl-dependent EphA2 protein degradation is induced by ligand binding
.
Mol Cancer Res
2002
;
1
:
79
87
.
31.
Sabet
O
,
Stockert
R
,
Xouri
G
,
Brüggemann
Y
,
Stanoev
A
,
Bastiaens
PI
. 
Ubiquitination switches EphA2 vesicular traffic from a continuous safeguard to a finite signalling mode
.
Nat Commun
2015
;
6
:
8047
.
32.
Wang
YJ
,
Ota
S
,
Kataoka
H
,
Kanamori
M
,
Li
ZY
,
Band
H
, et al
Negative regulation of EphA2 receptor by Cbl
.
Biochem Biophys Res Commun
2002
;
296
:
214
20
.
33.
Lo
KW
,
To
KF
,
Huang
DP
. 
Focus on nasopharyngeal carcinoma
.
Cancer Cell
2004
;
5
:
423
8
.
34.
Chan
AT
. 
Current treatment of nasopharyngeal carcinoma.
Eur J Cancer
2011
;
47
:
S302
3
.
35.
Zhu
JF
,
Huang
W
,
Yi
HM
,
Xiao
T
,
Li
JY
,
Feng
J
, et al
Annexin A1-suppressed autophagy promotes nasopharyngeal carcinoma cell invasion and metastasis by PI3K/AKT signaling activation
.
Cell Death Dis
2018
;
9
:
1154
.
36.
Li
JY
,
Xiao
T
,
Yi
HM
,
Yi
H
,
Feng
J
,
Zhu
JF
, et al
S897 phosphorylation of EphA2 is indispensable for EphA2-dependent nasopharyngeal carcinoma cell invasion, metastasis and stem properties
.
Cancer Lett
2019
;
444
:
162
74
.
37.
Xiao
T
,
Zhu
W
,
Huang
W
,
Lu
SS
,
Li
XH
,
Xiao
ZQ
, et al
RACK1 promotes tumorigenicity of colon cancer by inducing cell autophagy
.
Cell Death Dis
2018
;
9
:
1148
.
38.
Huang
WG
,
Cheng
AL
,
Chen
ZC
,
Peng
F
,
Zhang
PF
,
Li
MY
, et al
Targeted proteomic analysis of 14–3–3sigma in nasopharyngeal carcinoma
.
Int J Biochem Cell Biol
2010
;
42
:
137
47
.
39.
Mozayan
A
,
Khaled
A
. 
Elucidation of therapeutic peptide binding partners from isolated mitochondria
.
Cureus
2018
;
10
:
e2898
.
40.
Yi
HM
,
Yi
H
,
Zhu
JF
,
Xiao
T
,
Lu
SS
,
Guan
YJ
, et al
A five-variable signature predicts radioresistance and prognosis in nasopharyngeal carcinoma patients receiving radical radiotherapy
.
Tumour Biol
2016
;
37
:
2941
9
.
41.
Kozakov
D
,
Hall
DR
,
Xia
B
,
Porter
KA
,
Padhorny
D
,
Yueh
C
, et al
The ClusPro web server for protein-protein docking
.
Nat Protoc
2017
;
12
:
255
78
.
42.
White
IJ
,
Bailey
LM
,
Aghakhani
MR
,
Moss
SE
,
Futter
CE
. 
EGF stimulates annexin 1-dependent inward vesiculation in a multivesicular endosome subpopulation
.
EMBO J
2006
;
25
:
1
12
.
43.
Radke
S
,
Austermann
J
,
Russo-Marie
F
,
Gerke
V
,
Rescher
U
. 
Specific association of annexin 1 with plasma membrane- resident and internalized EGF receptors mediated through the protein core domain
.
FEBS Lett
2004
;
578
:
95
8
.
44.
Poeter
M
,
Radke
S
,
Koese
M
,
Hessner
F
,
Hegemann
A
,
Musiol
A
, et al
Disruption of the annexin A1/S100A11 complex increases the migration and clonogenic growth by dysregulating epithelial growth factor (EGF) signaling
.
Biochim Biophys Acta
2013
;
1833
:
1700
11
.
45.
Melki
V
,
Hullin
F
,
Mazarguil
H
,
Fauvel
J
,
Ragab-Thomas
JM
,
Chap
H
. 
Annexin I as a potential inhibitor of insulin receptor protein tyrosine kinase
.
Biochem Biophys Res Commun
1994
;
203
:
813
9
.
46.
Naudin
C
,
Sirvent
A
,
Leroy
C
,
Larive
R
,
Simon
V
,
Pannequin
J
, et al
SLAP displays tumour suppressor functions in colorectal cancer via destabilization of the SRC substrate EPHA2
.
Nat Commun
2014
;
5
:
3159
.
47.
Thien
CB
,
Langdon
WY
. 
Cbl: many adaptations to regulate protein tyrosine kinases
.
Nat Rev Mol Cell Biol
2001
;
2
:
294
307
.
48.
Schmidt
MH
,
Dikic
I
. 
The Cbl interactome and its functions
.
Nat Rev Mol Cell Biol
2005
;
6
:
907
18
.
49.
Huang
J
,
Xiao
D
,
Li
G
,
Ma
J
,
Chen
P
,
Yuan
W
, et al
EphA2 promotes epithelial-mesenchymal transition through the Wnt/β-catenin pathway in gastric cancer cells
.
Oncogene
2014
;
33
:
2737
47
.