Further studies revealed that anoikis resistance is closely related to migration and invasion. than one-fourth of all cancer-related deaths worldwide1. Nearly sixty percent of lung cancer patients are diagnosed at late stages with metastasis, and their 5-year survival is less than 5%1. Thus, identifications of novel therapeutic targets against lung cancer metastasis are urgently needed to improve patients survival. Cellular retinoic acid-binding proteins, Crabp1 and Crabp2, are small cytosolic proteins that belong to a family of two isotypes2. CRABP1 has been found to promote tumorigenicity of transformed mesenchymal cells3. In breast cancer, CRABP1 is correlated with poor prognosis4. CRABP1 also plays a promoting role in metastasis of transformed hamster fibroblasts3. The overexpression of CRABP2 has been reported in tumor tissues of non-small cell lung cancer (NSCLC)5C7. However, the role of Crabp2 in metastasis of lung cancer is still unclear. Metastasis is a multi-step process termed invasion-metastasis cascade, which requires multiple capabilities of cancer cells including migration and invasion8. Resistance to cell death induced by loss of anchorage (anoikis) has also been recognized as an essential ability for metastasis9,10. Further studies revealed that anoikis resistance is closely related to migration and invasion. Selection of anoikis-resistant pancreatic cancer cells results TC-S 7010 (Aurora A Inhibitor I) in enhanced cell migration and invasion11. Elevated migration and invasion were also found in anoikis-resistant prostate cancer cells12. It has been reported that activation of integrin signaling molecules including FAK and ERK is known to promote anoikis resistance, migration, invasion, and metastasis of cancer cells13C16, and both FAK and ERK are thus suggested as therapeutic targets17, 18 while side effects disturbing normal cell C5AR1 functions have also been reported19. Thus, identification of tumor-overexpressing molecules mediating the activation of integrin signaling and promotion of lung cancer metastasis is needed. In this study, we selected the high-metastatic C10F4 lung cancer cells from low-metastatic C9F6 lung adenocarcinoma cells. Further analyses identified Crabp2 as an overexpressed gene in C10F4 cells in comparison with C9F6 cells and mouse lung cells. Multiple cohorts of lung cancer patients were analyzed to reveal the correlation of CRABP2 with tumor progression and clinical outcomes. We further explored the role of Crabp2 in migration, invasion, anoikis resistance, and metastasis. The signaling regulated by Crabp2 was investigated, and their roles in Crabp2-mediated pro-metastatic features were examined. We then addressed the potential implication of Crabp2 knockdown in inhibiting the TC-S 7010 (Aurora A Inhibitor I) growth of cancer cells as compared with that by gemcitabine or irinotecan alone. We also explored the potential upstream regulating factors leading to the upregulation of Crabp2 in lung cancer cells. Overall, our findings reveal the promoting role of Crabp2 in migration, invasion, anoikis resistance, and metastasis of lung cancer. CRABP2 could be a useful prognostic biomarker and a target against lung cancer metastasis. Results Establishment of high-metastatic C10F4 lung cancer cells We initially used tail vein injection selection to obtain a high-metastatic subline. Three cycles of tail vein injection selection yielded the highly metastatic C10F4 cells from low-metastatic C9F6 cells. We further compared metastatic behaviors, including migration and invasion, in C10F4 and C9F6 cells. The C10F4 cells displayed significantly enhanced migration and invasion ability compared to C9F6 cells (Fig.?1a,b). The BALB/c mice tail vein injection model showed that C10F4 cells exhibited higher lung and liver metastatic abilities than C9F6 cells (Fig.?1c). Thus C10F4 line provides us with a valuable tool for exploring metastasis-related signaling pathways and molecules. Open in a separate window Figure 1 Crabp2 is overexpressed in high-metastatic C10F4 cells. (a) Migration assay of C9F6 and C10F4 cells for 12?hours. Cells migrated into the lower compartment of Boyden chamber were photographed (left) and quantified (right). (b) Matrigel cell invasion assay of C9F6 and C10F4 for 15?hours. Cells invaded through the matrigel were photographed (left) and quantified (right). (c) Metastasis of C9F6 (n?=?3) and C10F4 (n?=?3) cells. One million cells were injected into tail veins TC-S 7010 (Aurora A Inhibitor I) of each BALB/c mouse. Twelve days later, mouse lungs and livers were harvested, and tumor regions were visualized by H&E staining (left). Metastasis index was calculated as tumor area over lung/liver area (right). T: tumor region. (d,e) Venn diagram comparing two differential expression analyses. For the up red circle, 124 genes were expressed more than 2-fold-higher in C10F4.