Fig 2: The proposed mechanism of how CCL2 enhances cell motility of HNSCC.The proposed models described the effect of CCL2 on cell migration and tumor metastasis in HNSCC. From a macroscopic perspective, HNSCC can metastasize to distant sites through lymphatics and blood vessels, among which cervical lymph node metastasis was most frequent. From a microscopic perspective, HNSCC cells can secrete CCL2 which could bind to the membrane receptor CCR4 by autocrine or paracrine function. Then, the PI3K pathway transduces extracellular signals into the cell and activates the Vav2-Rac1-MLC signal axis. Finally, CCL2 promotes HNSCC cytoskeletal remodeling and pseudopodia formation, enhances the mobility of HNSCC cells, and promotes the metastasis of HNSCC.

Fig 3: Vav2 is essential for CCL2-CCR4 signaling induced Rac1 activation and cell migration.A CCL2 stimulated the formation of the Vav2-Rac1 complex. HNSCC cells were subjected to treatment with CCL2 for 2 h. The cell lysates were subjected to immunoprecipitation (IP) with Rac1 or Vav2 antibodies, respectively. The immunoprecipitates were subjected to Western blot assay with the indicated antibodies. Results revealed that CCL2 did not upregulate the expression of Vav2 or Rac1 but significantly induce the Vav2-Rac-1 complex formation in HNSCC cells treated with CCL2 (**P < 0.01). B CCL2 stimulated the colocalization of Vav2 and Rac1 via CCR4. HNSCC cells were treated with or without exogenous CCL2 (100 ng/mL) for 2 h and stained with green fluorescence (for Rac1), red fluorescence (for Vav2), and DAPI (nuclear stain). Confocal images showed that CCL2 promoted the colocalization of Vav2 and Rac1 in the cytoplasm (red arrows); however, it was abolished with siCCR4 (Original magnification, ×400; Bar: 5 μm). C Vav2-silencing short interfering RNA (siVav2) could inhibit the activation of Rac1 induced by CCL2. HNSCC cells were either untreated or treated with negative control siRNA or siVav2 for 48 h. The cells were then treated with exogenous CCL2 (100 ng/mL) for 2 h. The amount of GTP-bound Rac1 was significantly reduced in HNSCC cells transfected with siVav2 compared with the NC group (*P < 0.05). D CCL2 induced the upregulation of p-Vav2 through PI3K signaling. HNSCC cells were incubated in a culture medium with or without PI3K inhibitors (LY294002, 50 μM; Wortmannin 2 μM, respectively) for 24 h. The cells were then treated with exogenous CCL2 (100 ng/mL) for 2 h. The cell lysates were subjected to Western blot assay with antibodies against p-Vav2 and total Vav2. Results revealed that PI3K inhibition could inhibit the upregulation of p-Vav2 induced by CCL2 (*P < 0.05). E CCL2 induced the formation of the Vav2-Rac1 complex in a PI3K-dependent manner. HNSCC cells were incubated in a culture medium with or without the PI3K inhibitor (Wortmannin, 2 μM) for 24 h. The cells were then treated with exogenous CCL2 (100 ng/mL) for 2 h. HNSCC cell lysates were subjected to immunoprecipitation (IP) with Rac1 or Vav2 antibodies, respectively. The immunoprecipitates were subjected to Western blot assay with the indicated antibodies. Results revealed that CCL2 upregulates the phosphorylation level of Vav2 and significantly induces the Vav2-Rac1 complex formation in HNSCC cells. But the PI3K inhibitor could abolish the enhancement of Vav2-Rac1 complex formation and phosphorylation level of Vav2 induced by CCL2 (**P < 0.01). F Differential expression of CCR4, p-Vav2, and p-MLC between normal tissue and HNSCC tissue. IHC staining was used to detect the presence of CCR4, p-Vav2, and p-MLC expression in normal tissue and HNSCC tissue. The results revealed that expression of CCR4, p-Vav2, and p-MLC was significant higher in HNSCC tissues compared with normal tissues (Magnification, ×400; Bar: 50 μm; *P < 0.05; **P < 0.01).

Fig 4: Role of DOCK180 in the effects of PTP1B inhibitor on EC motility.(A) Western blots showing the endogenous expression levels of Vav2, Tiam1 and DOCK180. IP, immunoprecipitation; WB, western blot. (B,C) Effects of gene silencing for DOCK180 with different siRNA constructs on the stimulatory effects of PTPI22 on EC spreading and migration. (D) Effects of gene silencing for Vav2 with different siRNA constructs on the stimulatory effect of PTPI22 on EC spreading. Data were mean ± S.E.M.; *P < 0.05 vs vehicle control, one-way ANOVA, n = 3–5.

Fig 5: CCR4 monoclonal antibody blocks the activation of Vav2-Rac1-MLC signaling induced by CCL2 and inhibits HNSCC growth and invasion in vivo.A Experimental design of nude mice in vivo. Twenty-four nude mice were randomly divided into four groups, with six mice in each group. Each nude mouse was injected with 2.5 million SCC15 cells labeled with luciferase under the skin of the axillary for subcutaneous tumor formation (0 W). In 1–3 weeks, four groups of nude mice were treated as indicated in the flow chart, respectively. All treatments were terminated at 3–6 weeks, and all mice were euthanized at the 7th week. The axillary lymph nodes and tumors were collected for IHC staining and weight measurement. B, C Biofluorescence analysis of nude mice. Bioluminescence images were obtained once a week to analyze the tumor growth and local invasion from 1st to the 6th week. There was no significant difference in tumor size among the four groups before treatment (1st week). After 2 weeks of treatment, tumor sizes of three treatment groups were significantly smaller than that of the control group (3rd week). With the suspension of treatment for 3 weeks, the tumor sizes of the CCL2 neutralizing antibody group progressed particularly rapidly, even approaching to control group. However, the tumor progression in the CCR4 monoclonal antibody group and combination group were flatted and significantly slower than that of tumors of the control group (6th weeks) (*P < 0.05; NS no statistical significance). D CCR4 monoclonal antibody inhibited HNSCC progression without rapid relapse after treatment withdrawal. All mice were euthanized at the 7th week, and the tumor sizes were measured. It was similar to the results of the bioluminescence analysis; the tumor weight in the CCL2 neutralizing antibody group exhibited no significant difference between that of the control group. But the tumor weight of both the CCR4 monoclonal antibody group and combination group was lighter than that of the control group (**P < 0.01; NS no statistical significance). E CCR4 monoclonal antibody inhibited the lymph node metastasis of HNSCC. Representative images of HE and Luciferase staining (Luc IHC) in the axillary lymph nodes. The number of metastatic lymph nodes (Positive LN) of both the CCR4 monoclonal antibody group and combination group was lesser than that of the control group, but the CCL2 neutralizing antibody group was an exception (Original magnification, ×400; Bar: 20 μm; *P < 0.05; NS no statistical significance). F CCR4 monoclonal antibody inhibited the activation of Vav2-Rac1-MLC signaling in HNSCC. Representative images of CCR4, p-Vav2, and p-MLC staining in the xenograft tumors. Although there was no significant difference in CCR4 staining in each group, p-Vav2 and p-MLC level of both the CCR4 monoclonal antibody group and combination group were weaker than that of the CCL2 neutralizing antibody group and control group. In addition, the expression levels of mouse CCL2 (mCCL2) in the transplanted tumors and mouse lymph nodes (LN) of each group were not statistically different, it was considered that the expression of mCCL2 would not affect the mobility of HNSCC cells in the in vivo xenograft model. (Magnification, ×400; Bar: 50 μm; *P < 0.05; NS no statistical significance).