Figure 5.

Pro-adhesive glycoproteins on tumor cells with BZM-sensitive adhesion

(A) Cell-surface CD24 and CD44 levels on the indicated tumor cell lines. (B) shRNA-mediated depletion of CD44 and numbers of adhesive events of control and CD44 knockdown tumor cells on ECs. (C) CD44 WB of total protein extracts from the annotated tumor cell lines. Peptide N-glycosidase F (PNGaseF) and neuraminidase (ND) were used to treat extracted proteins prior to WB; swainsonine (Sw) was used to treat tumor cells prior to protein extraction. (D) CD44 protein quantification in lectin precipitates (LPs) from protein extracts of tumor cells treated ± swainsonine. (E) Generation of control and MGAT5 knockdown derivatives of indicated cell lines and CD44 protein quantification in LPs pulled down from extracts of the respective derivative. (F) s.c. xenograft primary tumor growth periods and tumor weights as well as spontaneous lung metastasis numbers at necropsy of shControl versus shCD44 derivatives of the indicated cell lines (F, ^#p < 0.05 considering tumor weight as a covariate; ^§p < 0.05 considering tumor growth period as a covariate, ANCOVA). (G) s.c. xenograft primary tumor growth periods and tumor weights as well as spontaneous lung metastasis numbers at necropsy of shCD44 derivatives of the indicated cell lines treated with PBS (control) or BZM. This treatment was carried out as in Figures 2A and 2B. The effect on the tumor weight (HOS-shCD44) was considered as a covariate in the statistical analysis of lung metastasis numbers (ANCOVA). Bar charts represent mean ± SD of three replicates; black lines in histograms represent isotype controls; black lines in scatterplots represent means; ∗p < 0.05.