Figure 1.
Differential EFEMP1 expression in normal endometrium and endometrial carcinoma.
A: Immunochemistry of EFEMP1 in endometrial tissues. (a) normal endometrium with strong staining, (b) atypical hyperplasia with moderate staining, (c) endometrioid carcinoma with no staining, (d) clear cell carcinoma with no staining, and (e) serous endometrial carcinoma with lymph node metastasis. Magnification:×200 (above) and×400 (below). B: Down-regulation of EFEMP1 mRNA expression in EC cell-lines as compared with EEC (endometrial epithelial cell). ESC is the abbreviation of endometrial interstitial cell. HeLa cells were used as a positive control. The experiment was repeated twice, and each experiment was done in triplicate. * P<0.05, #*P<0.01. Arithmetic means of the data (bars) and SD (error bars) are shown. C: EFEMP1 expression in various endometrial cancer cell-lines as determined by Western immunoblotting analysis.
Table 1.
Statistical difference between EFEMP1 expression in normal endometrium and endometrial carcinoma.
Table 2.
Immunohistochemical expression of EFEMP and its association with clinic-pathological variables.
Figure 2.
Methylation status of EFEMP1 in cell-lines and tissues.
A: MSP was used to determine the methylation status of EFEMP1 in EC cells and tissues, NC = H2O; N = normal endometrial tissue; T = endometrial carcinoma; M = PCR products amplified by MSP specific for methylated DNA; U = PCR products amplified by MSP specific for unmethylated DNA. B: Showing three examples of the different methylation sequences as revealed by bisulfite genomic sequencing. (a) Shows a fully methylated sequence with 15 methylated CpGs. The cytosine nucleotides were not converted into thymines. (b)Shows a partly methylated sequence and only CpGs 1, 4, 5, 6, 7, 8 and 11 were methylated. (c) Shows a fully unmethylated sequence since all cytosine nucleotides of the 15 CpGs were converted into thymines. C: Bisulfite genomic sequencing detected endogenous levels of methylation among the endometrial carcinoma cell-lines HEC-1B and RL95-2, normal endometrium (N1) and endometrial carcinoma tissues (T1, T2, T3). Columns indicate CpG dinucleotides 1–15, and horizontal lines analyzed clones 1–10. Black dots symbolize methylated CpGs, and white dots symbolize unmethylated CpGs. D: Showing immunohistochemical staining of tissues with known methylation status as above. Magnification x400.
Figure 3.
The effect of 5-aza-dC and TSA on EFEMP1 expression.
A: Endometrial carcinoma cell-lines HEC-1B, AN3CA and SPEC2, shown to be methylated in the DNA promoter regain of EFEMP1, and RL95-2 shown to be unmethylated. Cells were treated with 5-aza-dC (10 µM) and/or TSA (200 ng/ml) for 96 h, and the methylation status of EFEMP1 was assessed by MSP. B: Bisulfite genomic sequencing detected endogenous levels of methylation before and after treatment with 5-aza-dC (10 µM) and/or TSA (200 ng/ml). Columns indicate clones 1–10; horizontal lines show the analysis of CpG dinucleotides 1–15. Black dots symbolize methylated CpGs, and white dots symbolize unmethylated CpGs. C: Showing qRT-PCR analysis and evaluation of EFEMP1 mRNA expression before and after treatment with 5-aza-dC and/or TSA. Data representative of three independent experiments are shown. # P<0.05. D: Western immunoblot showing changes in EFEMP1 protein expression. Protein samples were extracted from cells that were treated with 5-aza-dC and/or TSA.
Figure 4.
The effect of EFEMP1 on tumor growth in vivo.
A: Stable transfection of HEC-1B cells and RL95-2 cells with pCMV6/GFP/Neo-EFEMP1 and shEFEMP1. The percentage of transfected cells with fluorescence was >95%. B: Levels of mRNA expression of EFEMP1 as evaluated by qRT-PCR after transfecting cells with these plasmids. The experiments here were repeated twice, and each was done in triplicate, ** P<0.01. C: Western immunoblot to demonstrate the efficacy of these plasmids. D: Showing the comparative growth rate of tumors formed from HEC-1B-NC and HEC-1B-exEFEMP1. Ten days after injection the size of the tumor was measured every 4 days. P<0.05. E: Four wks after injection of HEC-1B-NC and HEC-1B-exEFEMP1, the weight and volume of the tumors were measured. Arithmetic means (bars) and SD (error bars) are shown, * P<0.05, **P<0.01. F: Nude mice tumor tissues were paraffin embedded and tumor slides were stained with hematoxylin and eosin (magnification was ×400), antibody of EFEMP1 (× 400) and Ki67 (×400). The black arrow indicates positively-stained cells for Ki67, at least five regions were randomly selected to score positive cells (mean ± SD, P<0.01).
Figure 5.
Effect of EFEMP1 on endometrial carcinoma cell proliferation, migration and invasion.
A: Knockdown of EFEMP1 promoted cell growth in RL95-2. However cell proliferation was lower in HEC-1B-exEFEMP1 cells compared with negative control and HEC-1B. B: Plate colony formation assay was used to measure cell proliferation of EFEMP1 in HEC-1B-exEFEMP1 and RL95-2-shEFEMP1 and negative control. C: Representative images and statistical plots of Transwell assays in HEC-1B cells, which were stably transfected with exEFEMP1and vector control; RL95-2 cells and KLE cells transfected with shEFEMP1 and vector control (x200 magnification). The data were described as arithmetic mean ± SD of three independent experiments. D: Wound healing assays of cells expressing the vector control and exEFEMP1 in HEC-1B; and RL95-2 cells showing dampened expression of EFEMP1and vector control. Typical images were captured at the precise time of wounding at 0, 6, 12 and 24 h (original magnification×100). The percent of wound closure was measured in at least three randomly selected regions (mean ± SD). E: MMP-2 and MMP-9 secretion was decreased significantly in HEC-1B transfected with exEFEMP1 and increased in KLE and RL95-2 transfected with shEFEMP1cells. Data are expressed as mean concentrations ± SD of three independent experiments. * P<0.05, **P<0.01, ***P<0.001.
Figure 6.
The effect of EFEMP1 on E-cadherin and vimentin expression.
A: The nude mouse tumor tissues were paraffin embedded and tumor slides were stained with hematoxylin & eosin (×400), antibody of E-cadherin(× 400) and vimentin (×400). B: Western immunoblot analysis of the changes in expression of E-cadherin and vimentin in HEC-1B cells transfected with exEFEMP1 and vector. C: Cell spheroids cultured to mimic metastatic cancer cells. qRT-PCR analysis of the changes in EFEMP1 expression in various EC cells. Data are expressed as the mean; bars, ±SD, and normalized to adherent cells from three independent experiments. ***P<0.001.