Fig 2: CXCR7 is transported along the microtubules to the pericentrosomal Golgi apparatus.(A) Microtubule destabilization impairs CXCR7 colocalization with a-tubulin and accumulation at the Golgi apparatus. C4-2B cells with stable CXCR7 overexpression were treated with either vehicle control or nocodazole, a microtubule polymerization inhibitor, at 10µg/mL for 1 hour and then subjected to IF costaining of CXCR7 and a-tubulin (top 2 rows) or GM130 (bottom 2 rows). Scale bar: 10 µm(B and C) In situ PLA of C4-2B cells that were pretreated with DMSO or 10 µg/mL of nocodazole for 1 hour shows decreased interaction between CXCR7 and AURKA in drug-treated cells, as shown by the red speckles (B) that were quantified in C. Scale bar: 40 µm The number of speckles was counted and normalized to the number of nuclei from 3 individual imaged field views (n=3, 2-tailed unpaired Student’s t test, **P <0.01). (D) CXCR7 enhances ARRB2-AURKA interaction. 293T cells were transiently transfected with a combination of indicated plasmids for 48 hours and then subjected to co-IP with anti-HA antibodies, followed by Western blot. (E) CXCR7 cofractionates with membrane, ARRB2, AURKA, and a-tubulin. C4-2B cells with stable CXCR7 overexpression were subjected to subcellular fractionation for cytoplasmic (Cyto), membrane (Memb), and nuclear (Nuc) fractions. The purity of the fractionation was validated by GAPDH (Cyto), EGFR (Memb), and HDAC3 (Nuc) markers.

Fig 3: The CXCR7-ARRB2 protein complex interacts with AURKA.(A) The whole lysate of C4-2B-EnzR cells was subjected to anti-CXCR7 and IgG co-IP, followed by Western blot analyses. (B) ARRB2 increases CXCR7 and AURKA interaction. The 293T cells overexpressing Myc-AURKA, CXCR7-FLAG, and/or ARRB2-HA were subjected to co-IP with anti-FLAG antibodies, followed by Western blot. WCL, whole-cell lysate. (C) AURKA binds to the C-terminal domain (CTD) of ARRB2. 293T cells were cotransfected with Myc-AURKA and an empty vector, full-length (FL), N-terminal domain (NTD), or CTD of ARRB2. Co-IP was performed with an anti-HA antibody, followed by Western blot. (D) GST pull-down assay shows direct interaction between ARRB2 and AURKA. GST-ARRB2 FL, GST-ARRB2 CTD, or GST-GFP control proteins were incubated with FLAG-AURKA protein for 2 hours, separated by GSH-Sepharose, and resolved for Western blot. #, GST-GFP control band. (E) The CTD of ARRB2 restores CXCR7-AURKA interaction. The 293T-CXCR7 cells were transfected with the indicated plasmids and then subjected to co-IP using an anti-FLAG antibody, followed by Western blot analysis. (F) The kinase domain of AURKA binds to ARRB2. The 293T cells were cotransfected with ARRB2-HA and Myc-AURKA, FL, its regulatory (Reg), or kinase domain (Kin). Co-IP was performed with an anti-HA antibody, followed by WB analysis. (G) The kinase domain of AURKA forms a complex with CXCR7. CXCR7-FLAG construct was cotransfected in 293T cells with Myc-AURKA FL, regulatory domain (Reg), or kinase domain (Kin). Co-IP was performed with an anti-FLAG antibody, followed by Western blot. (H) ARRB2 KD decreases AURKA activation. C4-2B-CXCR7 cells were transfected with 2 independent siARRB2 for 48 hours and then analyzed by Western blot. Actin was used as a loading control. (I) A model depicting the interaction among the CXCR7-ARRB2-AURKA protein complex. Both the CTD and NTD of ARRB2 can interact with CXCR7, while only the CTD of ARRB2 binds to the kinase (Kin) domain of AURKA. The image was generated in BioRender. *, non-specific bands.

Fig 4: Aurora-A overexpression correlates with poor prognosis of endometrial cancer. (A) Real-time quantitative PCR showing increased Aurora-A mRNA expression in EC tissues compared with normal endometrium. GAPDH is used as an internal control. Data are expressed as means ± S.E.M., Student's t-test, ***P < 0.001. (B) Representative western blots showing increased Aurora-A protein level in EC tissues compared with normal endometrium. β-actin is used as loading control. (C) Quantification of relative Aurora-A level. The images of western blots were quantified using Image J software. Data are expressed as means ± S.E.M., Student's t-test, ***P < 0.001. (D) Representative IHC staining showing Aurora-A expression. Scale bar, 100 μm. (E) H-Scores of Aurora-A levels were presented as scatter diagram in (D). Data are expressed as means ± S.E.M., Student's t-test, **P < 0.01. (F) Kaplan–Meier curves are made to show the survival rate of 552 EC patients with high and low Aurora-A mRNA expression levels. The cut-off value of Aurora-A expression is 7.22.

Fig 5: CXCR7 increases PCa growth, which is abolished by AURKA inhibition.(A) CXCR7 KD reduces C4-2B cell proliferation. WST1 assay was performed to measure cell proliferation. CXCR7 KD was confirmed by qRT-PCR. Representative proliferation data from 3 repeated experiments are shown. The data were analyzed by 2-way ANOVA followed by Tukey-corrected multiple-comparison test (mean ± SD, n =3, ****P <0.0001). qRT-PCR data were normalized to GAPDH and then the control condition (mean ± SEM, n =3). Statistical test is based on 1-way ANOVA paired with Dunnett’s multiple comparison test,****P <0.0001. (B) CXCR7 overexpression, confirmed by Western blot, increases 22Rv1 cell proliferation measured by WST-1 assay. Representative proliferation data from 3 repeated experiments are shown here. The data were analyzed by 2-way ANOVA followed by Bonferroni-corrected multiple-comparison test (mean ± SD, n =3, **P <0.01, ****P <0.0001). (C and D) AURKA inhibitor decreases CXCR7-driven cell proliferation in LNCaP (C) and 22Rv1 (D) measured by WST-1 assay. Representative proliferation data from 3 repeated experiments were analyzed by 2-way ANOVA combined with Tukey-corrected multiple-comparison test (mean ± SD, n=3, **P<0.01 ***P<0.001). Western blot confirms a reduction of AURKA and TACC3 phosphorylation under alisertib treatment. (E) AURKA-targeting delays CXCR7-driven tumor growth in vivo. NSG mice were injected s.c. with 1 × 106 of 22Rv1-GFP control or 22Rv1-CXCR7 cells. Tumor size (mm3) was monitored by caliper measurements twice a week. Once tumors reached 100 mm3, mice were randomized to receive either vehicle or alisertib (30 mg/kg) once a day for 21 days for a total of 4 treatment groups, n =4 (GFP-veh), n =5 (GFP-alisertib), n =5 (CXCR7-veh), n =5 (CXCR7-alisertib). Tumor growth data are shown as mean ± SEM. The statistical test is based on 2-way ANOVA combined with a Bonferroni’s-corrected multiple-comparison test (***P <0.001). (F) IHC staining of CXCR7 and pAURKA (T288) of representative tumor samples collected at the endpoint. Scale bars: 20 µm (inset); 100 µm (larger image).