PMC

Elevated UGT8 expression highly correlates with BLBC. (A) Box plots indicated UGT8 mRNA expression in different subtypes of breast cancer from four gene expression datasets (NKI295, METABRIC, GSE25066, and TCGA). Comparisons were analyzed by one-way ANOVA. (B) Box plot indicated UGT8 protein expression in different subtypes of breast cancer from TCGA dataset. Comparisons were analyzed by one-way ANOVA. (C) Expression of UGT8 and sulfatide was examined in tumor samples from five cases of luminal and five cases of TNBC. (D) Box plots indicated UGT8 mRNA expression in luminal and BLBC cell lines from four gene expression datasets (GSE12777, GSE10890, E-TABM-157, and E-MTAB-181). Comparisons were made using the two-tailed Student’s t test. (E and F) Expression of UGT8 mRNA was analyzed by either semi-quantitative RT-PCR (E) or quantitative RT-PCR (F) in a representative panel of breast cancer cell lines. Data are shown as mean ± SD based on three independent experiments. *, P < 0.05 by Student’s t test. (G) Expression of UGT8 in cells from E was examined by Western blotting.

UGT8 activates αVβ5 signaling. (A–C) Expression of αVβ5 was measured by immunofluorescent staining in MDA-MB231 cells with stable empty vector or knockdown of UGT8 expression (A), MDA-MB231 cells treated with or without ZA (20 µM; B) as well as HCC1937 cells with stable empty vector or UGT8 expression (C). Nuclei were visualized with DAPI (blue). Bars, 20 µm. (D–F) The level of αVβ5 was analyzed by flow cytometry in MDA-MB231 cells with stable empty vector or knockdown of UGT8 expression (D), MDA-MB231 cells treated with or without ZA (10 µM; E) as well as HCC1937 cells with stable empty vector or UGT8 expression (F). Representative images were shown (left). Isotype controls are used to determine the staining specificity (unfilled). The level of αVβ5 in cells with knockdown of UGT8 expression or UGT8 expression as well as cells treated with ZA was shown as a percentage of the control (mean ± SD in three separate experiments). *, P < 0.01 by Student’s t test. (G) Expression of UGT8, Smad4, Smad5, p-Smad1/5/8, and ID4 was analyzed by Western blotting in MDA-MB231 cells with stable empty vector or knockdown of UGT8 expression (left), as well as MDA-MB231 cells treated with or without ZA (10 µM; right). (H) Expression of UGT8 and RelA was analyzed by Western blotting in MDA-MB231 cells with stable empty vector or knockdown of UGT8 expression (left) as well as MDA-MB231 cells treated with or without ZA (10 µM; right). (I) A proposed model to illustrate the transcription activation of UGT8 by Sox10, which activates sulfatide–αVβ5 signaling axis in BLBC (see Discussion). TFs, transcription factors.

UGT8 activates the sulfatide biosynthetic pathway and enhances breast cancer cell migration and invasion. (A) Sulfatide biosynthetic pathway. (B) Stable transfectants with empty vector or knockdown of UGT8 expression were established in MDA-MB231 and SUM159 cells, and stable clones with empty vector or UGT8 expression were also generated in BT549 and HCC1937 cells. (C) Expression of GalCer and sulfatide was examined by immunoblotting in MDA-MB231 cells with stable empty vector or knockdown of UGT8 expression, as well as HCC1937 cells with stable empty vector or UGT8 expression. (D) Expression of GalCer and sulfatide was measured by immunofluorescent staining in MDA-MB231 cells with stable empty vector or knockdown of UGT8 expression as well as BT549 cells with stable empty vector or UGT8 expression. Nuclei were visualized with DAPI (blue). Bars, 20 µm. (E and F) Migratory ability (E) and invasiveness (F) of BT549 cells with stable empty vector or UGT8 expression as well as BT549 cells treated with or without GalCer (2 µM) or sulfatide (2 µM) were analyzed. The percentage of migratory and invasive cells was shown in the bar graph (mean ± SD in three separate experiments). (G and H) Migratory ability (G) and invasiveness (H) of MDA-MB231 cells with stable empty vector or knockdown of UGT8 expression, as well as shUGT8-expressing MDA-MB231 cells treated with or without GalCer (2 µM) or sulfatide (2 µM) were analyzed. The percentage of migratory and invasive cells was analyzed as in E and F. *, P < 0.01 by Student’s t test. Data are shown as mean ± SD based on three independent experiments.

UGT8 knockdown and ZA treatment have similar gene expression profiles and UGT8-mediated sulfatide induces ITGAV expression. (A) KEGG pathway analysis of down-regulated genes after UGT8 knockdown (fold change less than −2; left panel) and ZA treatment (30 µM; fold change less than −1.5; right panel) in MDA-MB231 cells. The top 10 most significant KEGG pathway terms were listed. (B) Expression of ITGAV was measured by immunofluorescent staining in MDA-MB231 cells with stable empty vector or knockdown of UGT8 expression as well as HCC1937 cells with stable empty vector or UGT8 expression. Nuclei were visualized with DAPI (blue). Bars, 20 µm. (C) Expression of ITGAV mRNA was analyzed by quantitative RT-PCR in MDA-MB231 cells with stable empty vector or knockdown of UGT8 expression as well as HCC1937 cells with stable empty vector or UGT8 expression. (D) Expression of ITGAV was analyzed by Western blotting in MDA-MB231 cells with stable empty vector or knockdown of UGT8 expression as well as HCC1937 cells with stable empty vector or UGT8 expression. (E) Expression of ITGAV was analyzed by quantitative RT-PCR (left) or Western blotting (right) in MDA-MB231 cells treated with or without ZA (10 µM). (F) Expression of ITGAV was analyzed by quantitative RT-PCR (left) or by Western blotting (right) in HCC1937 cells treated with or without GalCer (2 µM) or sulfatide (2 µM). Data are shown as mean ± SD based on three independent experiments. *, P < 0.01 by Student’s t test.

ZA is a direct inhibitor of UGT8 and inhibits breast cancer cell migration and invasion. (A and B) Expression of GalCer and sulfatide was measured by immunofluorescent staining in MDA-MB231 and SUM159 cells treated with or without ZA (20 µM; A), HCC1937 cells with empty vector, or stable UGT8 expression as well as UGT8-expressing HCC1937 cells treated with or without ZA (20 µM; B). Nuclei were visualized with DAPI (blue). Bars, 20 µm. (C) In vitro activity assay of UGT8 was performed by mixing UDP-galactose, substrate, and lysate of MDA-MB231 cells. After treatment of the indicated concentration of ZA, UDP-galactose consumption and UDP production were tested by HPLC system. The percentage of UDP-galactose consumption and UDP production was shown in the bar graph (mean ± SD in three separate experiments). (D) Expression of GalCer and sulfatide was examined by immunoblotting in MDA-MB231 and SUM159 cells treated with the indicated concentration of ZA. (E and F) Migratory ability (E) and invasiveness (F) of MDA-MB231 and SUM159 cells treated with the indicated concentration of ZA. The percentage of migratory and invasive cells was shown in the bar graph (mean ± SD in three separate experiments). *, P < 0.05 by Student’s t test.

Inhibition of UGT8 suppresses metastasis in vivo and elevated UGT8 predicts poor survival. (A and B) MDA-MB231 cells (A) and MDA-MB231 cells with stable empty vector or knockdown of UGT8 expression (B) were injected into SCID mice via the tail vein. For evaluation of ZA, the mice received ZA (0.0186 mg/kg/d) or sterile PBS subcutaneously. After 4 wk, the development of lung metastases was monitored using bioluminescence imaging and quantified by measuring photon flux (mean of six animals + SEM; left). Three representative mice from each group were shown (middle). Lung metastatic nodules were examined in paraffin-embedded sections stained with hematoxylin and eosin. The arrowheads indicate lung metastases. Bar, 100 µm (A, right). (C–E) Kaplan-Meier survival analysis for overall survival (OS), RFS, and distant metastasis-free survival (DMFS) of patients in the NKI295 dataset according to UGT8 expression status. The p-value was determined using the log-rank test. (F and G) Kaplan-Meier survival analysis for RFS of patients with various subtypes (F) or BLBC (G) in an aggregate breast cancer dataset according to UGT8 expression status. The p-value was determined using the log-rank test.

Knockdown of UGT8 expression inhibits tumorigenicity in vitro and in vivo. (A) Soft-agar assay was performed using MDA-MB231, SUM159, MDA-MB435, and MDA-MB436 cells with stable empty vector or knockdown of UGT8 expression (A), as well as BT549, HCC1937, and BT20 cells with stable empty vector or UGT8 expression (B). Data are presented as a percentage of empty vector cell lines (mean ± SD in three separate experiments). (C and D) MDA-MB231 cells (C) and SUM159 cells (D) with stable empty vector or knockdown of UGT8 expression were injected into the mammary fat pad of SCID mice. The growth of tumors was examined every 2 d. Tumor size and weight were recorded. Data are shown as mean ± SEM from six mice. *, P < 0.001. (E) Box plots indicated UGT8 expression in different tumor size of breast cancer from NKI295 dataset. Comparisons were made using the two-tailed Student’s t test. (F) Box plots indicated UGT8 expression in different histological grades of breast cancer from multiple datasets (NKI295, GSE25066, and GSE1456). Comparisons between two groups were made using the two-tailed Student’s t test.

UGT8 positively correlates with Sox10 and is a direct transcriptional target of Sox10. (A) Analysis of GSE25066 and TCGA datasets for the expression of UGT8 and Sox10. The relative level of UGT8 was plotted against that of Sox10. Correlations were analyzed using Pearson’s correlation method and Spearman’s rank correlation test. (B) Box plots indicated Sox10 mRNA expression in different subtypes of breast cancer from GSE25066 and TCGA datasets. Comparisons were analyzed by one-way ANOVA. (C) Expression of UGT8 and Sox10 was analyzed by quantitative RT-PCR in SUM159 and MDA-MB436 cells infected with empty vector or Sox10-expressing vector. Data are shown as mean ± SD based on three independent experiments. *, P < 0.01 by Student's t test. (D) Expression of UGT8 and Sox10 was examined by Western blotting in SUM159 and MDA-MB436 cells infected with empty vector or Sox10-expressing vector. (E) Schematic diagram showing positions of potential Sox10-binding motifs in UGT8 promoter. UGT8 promoter luciferase construct and mutated derivatives were also shown. Sox10 consensus sequence: (A/T)(A/T)CAA(A/T)G. (F) UGT8 promoter luciferase construct (wtU) was coexpressed with empty vector or Sox10-expressing vector in HeLa, SUM159, and MDA-MB436 cells, respectively. After 48 h, luciferase activities were analyzed (mean ± SD in three separate experiments). (G) UGT8 promoter luciferase construct (wtU) as well as its mutants (mutU1, mutU2, and mutU3) were coexpressed with empty vector or Sox10-expressing vector in HEK-293T cells. Luciferase activities were analyzed as in F. Data are shown as mean ± SD based on three independent experiments. (H) ChIP analysis for binding of Sox10 to the UGT8 promoter in MDA-MB231 and HCC1428 cells.