Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2011 Feb 18;585(4):693-9.
doi: 10.1016/j.febslet.2011.01.033. Epub 2011 Jan 26.

The Evi1, microRNA-143, K-Ras Axis in Colon Cancer

Affiliations
Free PMC article

The Evi1, microRNA-143, K-Ras Axis in Colon Cancer

Jin-Song Gao et al. FEBS Lett. .
Free PMC article

Abstract

MicroRNA profiling of diseased/non-diseased tissue has identified expression signatures associated with a wide range of pathogenic conditions including malignancy. For example, colon cancer is associated with the under expression of miRNA-143 yet the molecular etiology of under expression is unknown. The K-Ras oncogene is a target of miRNA-143. Here, we show that the ecotropic viral integration site 1 oncoprotein (Evi1) is a transcriptional suppressor of the miRNA-143 gene. We find an indirect relationship between miRNA-143 and Evi1 expression. A complex molecular axis linking Evi1, miRNA-143 is operational in human colon cancer.

Figures

Figure 1
Figure 1
miRNA-143 targets K-Ras. a. The 3′UTR of K-Ras harbors five potential sites (A-E) of sequence homology to the seed sequence of miRNA-143. Blue letters denote the K-Ras mRNA target while red letters are used to denote miRNA-143 sequence. ( : ) indicates GU pair ( | ) indicates normal bond b. K-Ras 3′ UTR segments A-E were incorporated into a dual luciferase reporter plasmid. All constructs were introduced into HEK 293 cells with miRNA-143 or an empty control vector (EV) and luminescence was measured at 48 hours. Compared to experiments involving EV, luciferase activity was repressed in all experiments involving the 3′-UTR segments, albeit at variable levels. c. Point mutations were introduced into the seed sequence of miRNA-143 and four expression vectors were created (mutant 1-4). In the table, seed sequence is shown in bold lettering; sequence changes for each mutant are underlined. Wildtype (WT), mutant and control (EV) plasmids were transfected into HEK 293 cells along with the dual luciferase reporter plasmid and levels of luminescence were quantified at 48 hours.
Figure 1
Figure 1
miRNA-143 targets K-Ras. a. The 3′UTR of K-Ras harbors five potential sites (A-E) of sequence homology to the seed sequence of miRNA-143. Blue letters denote the K-Ras mRNA target while red letters are used to denote miRNA-143 sequence. ( : ) indicates GU pair ( | ) indicates normal bond b. K-Ras 3′ UTR segments A-E were incorporated into a dual luciferase reporter plasmid. All constructs were introduced into HEK 293 cells with miRNA-143 or an empty control vector (EV) and luminescence was measured at 48 hours. Compared to experiments involving EV, luciferase activity was repressed in all experiments involving the 3′-UTR segments, albeit at variable levels. c. Point mutations were introduced into the seed sequence of miRNA-143 and four expression vectors were created (mutant 1-4). In the table, seed sequence is shown in bold lettering; sequence changes for each mutant are underlined. Wildtype (WT), mutant and control (EV) plasmids were transfected into HEK 293 cells along with the dual luciferase reporter plasmid and levels of luminescence were quantified at 48 hours.
Figure 1
Figure 1
miRNA-143 targets K-Ras. a. The 3′UTR of K-Ras harbors five potential sites (A-E) of sequence homology to the seed sequence of miRNA-143. Blue letters denote the K-Ras mRNA target while red letters are used to denote miRNA-143 sequence. ( : ) indicates GU pair ( | ) indicates normal bond b. K-Ras 3′ UTR segments A-E were incorporated into a dual luciferase reporter plasmid. All constructs were introduced into HEK 293 cells with miRNA-143 or an empty control vector (EV) and luminescence was measured at 48 hours. Compared to experiments involving EV, luciferase activity was repressed in all experiments involving the 3′-UTR segments, albeit at variable levels. c. Point mutations were introduced into the seed sequence of miRNA-143 and four expression vectors were created (mutant 1-4). In the table, seed sequence is shown in bold lettering; sequence changes for each mutant are underlined. Wildtype (WT), mutant and control (EV) plasmids were transfected into HEK 293 cells along with the dual luciferase reporter plasmid and levels of luminescence were quantified at 48 hours.
Figure 2
Figure 2
Differential expression of miRNA-143, let-7a and K-Ras in colon cancer a. Real-time RT-PCR was used to quantify levels of mature miR-143 and let-7a in twenty samples of human colon cancer in both frank cancerous tissue and surrounding normal adjacent tissue (NAT). Patterns of expression are presented as fold comparisons between cancerous and NAT for each miRNA (miRNA-143: dark blue bars and let-7a: light blue bars). Data from 18 samples are presented; two outliers (cases # 4 and #13) with a >50 fold difference in miRNA-143 expression are omitted for graphical clarity. We observed a statistically significant under-expression of miRNA-143 in cancerous tissue vs. NAT (p=0.001). No such relationship was found for let-7 (p=0.263). b. In agreement with Real Time RT-PCR data for mature miRNA expression level, northern blots indicated a decreased level of miRNA-143 in cancerous tissue compared to NAT. No such difference was observed for let-7a in the samples studied. A strong inverse relationship was observed between levels of miRNA-143 and K-Ras, the latter quantified by western blot. GAPDH protein levels served as a loading control for western blot.
Figure 2
Figure 2
Differential expression of miRNA-143, let-7a and K-Ras in colon cancer a. Real-time RT-PCR was used to quantify levels of mature miR-143 and let-7a in twenty samples of human colon cancer in both frank cancerous tissue and surrounding normal adjacent tissue (NAT). Patterns of expression are presented as fold comparisons between cancerous and NAT for each miRNA (miRNA-143: dark blue bars and let-7a: light blue bars). Data from 18 samples are presented; two outliers (cases # 4 and #13) with a >50 fold difference in miRNA-143 expression are omitted for graphical clarity. We observed a statistically significant under-expression of miRNA-143 in cancerous tissue vs. NAT (p=0.001). No such relationship was found for let-7 (p=0.263). b. In agreement with Real Time RT-PCR data for mature miRNA expression level, northern blots indicated a decreased level of miRNA-143 in cancerous tissue compared to NAT. No such difference was observed for let-7a in the samples studied. A strong inverse relationship was observed between levels of miRNA-143 and K-Ras, the latter quantified by western blot. GAPDH protein levels served as a loading control for western blot.
Figure 3
Figure 3
EVi1 is a transcriptional suppressor of miRNA-143 a. Schematic diagram of miRNA-143 genomic locus on chromosome 5. Four putative Evi1 binding sites exist in a 2kb region upstream of the start of pre-miRNA-143, denoted by black boxes. Consensus binding sequence is annotated below each box. Arrows ‘a’ and ‘b’ correspond to primer locations for amplification of recovered DNA for use in the ChIP assay. b. Chromatin was immunoprecipitated with anti-Evi1 antibody and oligonucleotides ‘a’ and ‘b’ were used to PCR amplify the recovered DNA. Control amplifications were carried out using primers specific for the promoter region of the PLZF gene, a known transcriptional target of EVi1. A gene with no known regulation by EVI1 (GAPDH) served as a negative control. c. We generated a series of miRNA-143 promoter region luciferase reporter constructs containing 0 (TF1), 2 (TF2) or all 4 (TF3) putative Evi1 binding sites. Lower panel: The constructs were transfected into HEK 293 cells and luminescence was measured at 48 hours. Data are shown as average reduction (+/-S.D.) in promoter activity relative to the basal endogenous pri-miRNA-143 promoter activity (TF1). Promoter activity was reduced by ~80% in experiments involving TF3.
Figure 3
Figure 3
EVi1 is a transcriptional suppressor of miRNA-143 a. Schematic diagram of miRNA-143 genomic locus on chromosome 5. Four putative Evi1 binding sites exist in a 2kb region upstream of the start of pre-miRNA-143, denoted by black boxes. Consensus binding sequence is annotated below each box. Arrows ‘a’ and ‘b’ correspond to primer locations for amplification of recovered DNA for use in the ChIP assay. b. Chromatin was immunoprecipitated with anti-Evi1 antibody and oligonucleotides ‘a’ and ‘b’ were used to PCR amplify the recovered DNA. Control amplifications were carried out using primers specific for the promoter region of the PLZF gene, a known transcriptional target of EVi1. A gene with no known regulation by EVI1 (GAPDH) served as a negative control. c. We generated a series of miRNA-143 promoter region luciferase reporter constructs containing 0 (TF1), 2 (TF2) or all 4 (TF3) putative Evi1 binding sites. Lower panel: The constructs were transfected into HEK 293 cells and luminescence was measured at 48 hours. Data are shown as average reduction (+/-S.D.) in promoter activity relative to the basal endogenous pri-miRNA-143 promoter activity (TF1). Promoter activity was reduced by ~80% in experiments involving TF3.
Figure 3
Figure 3
EVi1 is a transcriptional suppressor of miRNA-143 a. Schematic diagram of miRNA-143 genomic locus on chromosome 5. Four putative Evi1 binding sites exist in a 2kb region upstream of the start of pre-miRNA-143, denoted by black boxes. Consensus binding sequence is annotated below each box. Arrows ‘a’ and ‘b’ correspond to primer locations for amplification of recovered DNA for use in the ChIP assay. b. Chromatin was immunoprecipitated with anti-Evi1 antibody and oligonucleotides ‘a’ and ‘b’ were used to PCR amplify the recovered DNA. Control amplifications were carried out using primers specific for the promoter region of the PLZF gene, a known transcriptional target of EVi1. A gene with no known regulation by EVI1 (GAPDH) served as a negative control. c. We generated a series of miRNA-143 promoter region luciferase reporter constructs containing 0 (TF1), 2 (TF2) or all 4 (TF3) putative Evi1 binding sites. Lower panel: The constructs were transfected into HEK 293 cells and luminescence was measured at 48 hours. Data are shown as average reduction (+/-S.D.) in promoter activity relative to the basal endogenous pri-miRNA-143 promoter activity (TF1). Promoter activity was reduced by ~80% in experiments involving TF3.
Figure 4
Figure 4
We profiled levels of pri-miRNA-143 and Evi1 in three colon cancer cell lines (CaCO2, HT29 and HCT116). (a) We observed an inverse relationship between levels of Evi1 and pri-miRNA-143, as measured by Real Time PCR. (b). Western blotting revealed a direct relationship between levels of Evi1 and K-Ras in HCT116 and CaCO2 cells. Beta actin and GAPDH were used to normalize RNA and protein input for RT-PCR and western blot, respectively. (c). Overexpression of Evi1 in HCT116 cells by transient transfection of an expression plasmid led to a ~50% reduction in pri-miRNA-143 levels. In contrast, depletion of Evi1 in CaCO2 cells by specific siRNA or arsenic trioxide (ATO) both for 24h led to a ~2-fold increase in pri-miRNA-143 levels (d). Western blot of cells treated with either anti-Evi1 siRNA (48h) or ATO (24h) revealed a decrease in levels of Evi1 and K-Ras proteins in CaCO2 cells (e). We measured the effect of miRNA-143 over expression in Caco2 cells by MTT assay (f) and motility assay (g). Over expression was associated with decreased proliferation and motility, respectively.
Figure 4
Figure 4
We profiled levels of pri-miRNA-143 and Evi1 in three colon cancer cell lines (CaCO2, HT29 and HCT116). (a) We observed an inverse relationship between levels of Evi1 and pri-miRNA-143, as measured by Real Time PCR. (b). Western blotting revealed a direct relationship between levels of Evi1 and K-Ras in HCT116 and CaCO2 cells. Beta actin and GAPDH were used to normalize RNA and protein input for RT-PCR and western blot, respectively. (c). Overexpression of Evi1 in HCT116 cells by transient transfection of an expression plasmid led to a ~50% reduction in pri-miRNA-143 levels. In contrast, depletion of Evi1 in CaCO2 cells by specific siRNA or arsenic trioxide (ATO) both for 24h led to a ~2-fold increase in pri-miRNA-143 levels (d). Western blot of cells treated with either anti-Evi1 siRNA (48h) or ATO (24h) revealed a decrease in levels of Evi1 and K-Ras proteins in CaCO2 cells (e). We measured the effect of miRNA-143 over expression in Caco2 cells by MTT assay (f) and motility assay (g). Over expression was associated with decreased proliferation and motility, respectively.
Figure 4
Figure 4
We profiled levels of pri-miRNA-143 and Evi1 in three colon cancer cell lines (CaCO2, HT29 and HCT116). (a) We observed an inverse relationship between levels of Evi1 and pri-miRNA-143, as measured by Real Time PCR. (b). Western blotting revealed a direct relationship between levels of Evi1 and K-Ras in HCT116 and CaCO2 cells. Beta actin and GAPDH were used to normalize RNA and protein input for RT-PCR and western blot, respectively. (c). Overexpression of Evi1 in HCT116 cells by transient transfection of an expression plasmid led to a ~50% reduction in pri-miRNA-143 levels. In contrast, depletion of Evi1 in CaCO2 cells by specific siRNA or arsenic trioxide (ATO) both for 24h led to a ~2-fold increase in pri-miRNA-143 levels (d). Western blot of cells treated with either anti-Evi1 siRNA (48h) or ATO (24h) revealed a decrease in levels of Evi1 and K-Ras proteins in CaCO2 cells (e). We measured the effect of miRNA-143 over expression in Caco2 cells by MTT assay (f) and motility assay (g). Over expression was associated with decreased proliferation and motility, respectively.
Figure 4
Figure 4
We profiled levels of pri-miRNA-143 and Evi1 in three colon cancer cell lines (CaCO2, HT29 and HCT116). (a) We observed an inverse relationship between levels of Evi1 and pri-miRNA-143, as measured by Real Time PCR. (b). Western blotting revealed a direct relationship between levels of Evi1 and K-Ras in HCT116 and CaCO2 cells. Beta actin and GAPDH were used to normalize RNA and protein input for RT-PCR and western blot, respectively. (c). Overexpression of Evi1 in HCT116 cells by transient transfection of an expression plasmid led to a ~50% reduction in pri-miRNA-143 levels. In contrast, depletion of Evi1 in CaCO2 cells by specific siRNA or arsenic trioxide (ATO) both for 24h led to a ~2-fold increase in pri-miRNA-143 levels (d). Western blot of cells treated with either anti-Evi1 siRNA (48h) or ATO (24h) revealed a decrease in levels of Evi1 and K-Ras proteins in CaCO2 cells (e). We measured the effect of miRNA-143 over expression in Caco2 cells by MTT assay (f) and motility assay (g). Over expression was associated with decreased proliferation and motility, respectively.
Figure 4
Figure 4
We profiled levels of pri-miRNA-143 and Evi1 in three colon cancer cell lines (CaCO2, HT29 and HCT116). (a) We observed an inverse relationship between levels of Evi1 and pri-miRNA-143, as measured by Real Time PCR. (b). Western blotting revealed a direct relationship between levels of Evi1 and K-Ras in HCT116 and CaCO2 cells. Beta actin and GAPDH were used to normalize RNA and protein input for RT-PCR and western blot, respectively. (c). Overexpression of Evi1 in HCT116 cells by transient transfection of an expression plasmid led to a ~50% reduction in pri-miRNA-143 levels. In contrast, depletion of Evi1 in CaCO2 cells by specific siRNA or arsenic trioxide (ATO) both for 24h led to a ~2-fold increase in pri-miRNA-143 levels (d). Western blot of cells treated with either anti-Evi1 siRNA (48h) or ATO (24h) revealed a decrease in levels of Evi1 and K-Ras proteins in CaCO2 cells (e). We measured the effect of miRNA-143 over expression in Caco2 cells by MTT assay (f) and motility assay (g). Over expression was associated with decreased proliferation and motility, respectively.
Figure 4
Figure 4
We profiled levels of pri-miRNA-143 and Evi1 in three colon cancer cell lines (CaCO2, HT29 and HCT116). (a) We observed an inverse relationship between levels of Evi1 and pri-miRNA-143, as measured by Real Time PCR. (b). Western blotting revealed a direct relationship between levels of Evi1 and K-Ras in HCT116 and CaCO2 cells. Beta actin and GAPDH were used to normalize RNA and protein input for RT-PCR and western blot, respectively. (c). Overexpression of Evi1 in HCT116 cells by transient transfection of an expression plasmid led to a ~50% reduction in pri-miRNA-143 levels. In contrast, depletion of Evi1 in CaCO2 cells by specific siRNA or arsenic trioxide (ATO) both for 24h led to a ~2-fold increase in pri-miRNA-143 levels (d). Western blot of cells treated with either anti-Evi1 siRNA (48h) or ATO (24h) revealed a decrease in levels of Evi1 and K-Ras proteins in CaCO2 cells (e). We measured the effect of miRNA-143 over expression in Caco2 cells by MTT assay (f) and motility assay (g). Over expression was associated with decreased proliferation and motility, respectively.
Figure 4
Figure 4
We profiled levels of pri-miRNA-143 and Evi1 in three colon cancer cell lines (CaCO2, HT29 and HCT116). (a) We observed an inverse relationship between levels of Evi1 and pri-miRNA-143, as measured by Real Time PCR. (b). Western blotting revealed a direct relationship between levels of Evi1 and K-Ras in HCT116 and CaCO2 cells. Beta actin and GAPDH were used to normalize RNA and protein input for RT-PCR and western blot, respectively. (c). Overexpression of Evi1 in HCT116 cells by transient transfection of an expression plasmid led to a ~50% reduction in pri-miRNA-143 levels. In contrast, depletion of Evi1 in CaCO2 cells by specific siRNA or arsenic trioxide (ATO) both for 24h led to a ~2-fold increase in pri-miRNA-143 levels (d). Western blot of cells treated with either anti-Evi1 siRNA (48h) or ATO (24h) revealed a decrease in levels of Evi1 and K-Ras proteins in CaCO2 cells (e). We measured the effect of miRNA-143 over expression in Caco2 cells by MTT assay (f) and motility assay (g). Over expression was associated with decreased proliferation and motility, respectively.
Figure 5
Figure 5
Multiple isoforms of EVi1 exist due to alternative splicing and usage of 5′-ends. Intergenic splicing leads to the formation of a fusion protein with MDS1. Western blotting of colonic tissue revealed over expression of Evi1,MDS1/Evi1 and a truncated version of Evi1 (ΔEvi1) in tumor tissue (T) compared to surrounding normal tissue (N). Similar expression patterns of MDS1/Evi1 and Evi1 have been recently described in ovarian cancer tissue and cell lines (17).

Similar articles

See all similar articles

Cited by 14 articles

See all "Cited by" articles

Publication types

MeSH terms

Feedback