Limits...
DOT1L cooperates with the c-Myc-p300 complex to epigenetically derepress CDH1 transcription factors in breast cancer progression.

Cho MH, Park JH, Choi HJ, Park MK, Won HY, Park YJ, Lee CH, Oh SH, Song YS, Kim HS, Oh YH, Lee JY, Kong G - Nat Commun (2015)

Bottom Line: DOT1L has emerged as an anticancer target for MLL-associated leukaemias; however, its functional role in solid tumours is largely unknown.DOT1L recognizes SNAIL, ZEB1 and ZEB2 promoters via interacting with the c-Myc-p300 complex and facilitates lysine-79 methylation and acetylation towards histone H3, leading to the dissociation of HDAC1 and DNMT1 in the regions.Collectively, we suggest that cooperative effect of DOT1L and c-Myc-p300 is critical for acquisition of aggressive phenotype of breast cancer by promoting EMT/CSC.

View Article: PubMed Central - PubMed

Affiliation: Department of Pathology, College of Medicine, Hanyang University, Seoul 133-791, Korea.

ABSTRACT
DOT1L has emerged as an anticancer target for MLL-associated leukaemias; however, its functional role in solid tumours is largely unknown. Here we identify that DOT1L cooperates with c-Myc and p300 acetyltransferase to epigenetically activate epithelial-mesenchymal transition (EMT) regulators in breast cancer progression. DOT1L recognizes SNAIL, ZEB1 and ZEB2 promoters via interacting with the c-Myc-p300 complex and facilitates lysine-79 methylation and acetylation towards histone H3, leading to the dissociation of HDAC1 and DNMT1 in the regions. The upregulation of these EMT regulators by the DOT1L-c-Myc-p300 complex enhances EMT-induced breast cancer stem cell (CSC)-like properties. Furthermore, in vivo orthotopic xenograft models show that DOT1L is required for malignant transformation of breast epithelial cells and breast tumour initiation and metastasis. Clinically, DOT1L expression is associated with poorer survival and aggressiveness of breast cancers. Collectively, we suggest that cooperative effect of DOT1L and c-Myc-p300 is critical for acquisition of aggressive phenotype of breast cancer by promoting EMT/CSC.

No MeSH data available.


Related in: MedlinePlus

DOT1L is associated with aggressive breast cancer by promoting EMT-inducedbreast tumour metastasis.(a) Analysis of overall and disease-free survival according to DOT1Lexpression in 91 ER-negative human breast cancers usingKaplan–Meier method with log-rank test. (b) Thequantification of the association with DOT1L expression and invasion andmetastasis in 182 human breast cancers. The P values were calculatedby χ2-test. (c) Bright-field images ofthe indicated cell types for observation of changes in cell morphology afterDOT1L overexpression and knockdown using lentiviruses encoding DOT1L cDNA orTet-inducible DOT1L shRNA (shDOT1L), respectively. CON (control), emptyvector; shCON, shRNA control. (d) The expression levels of epithelialand mesenchymal markers in the indicated cells were analysed usingimmunoblotting. (e,f) DOT1L-overexpressing MCF10A and itscontrol cells were treated with 100 pM TGF-β1 for48 h, and subjected to immunoblotting for analysis of DOT1Lexpression and activity (e). The invasion (upper) and migration(lower) by the indicated cells were analysed and quantified (f).* and †P<0.05 compared withCON/vehicle and CON/TGF-β, respectively (Student'st-test). (g) Analysis of invasion (left) and migration(right) by Tet-inducible DOTL-knockdown MDA-MB-231 cells.*P<0.05 versus shCON (Student'st-test). The data in f,g represent the means±s.d.of triplicate assays. (h) Effect of DOT1L on lung metastasis ofbreast cancer in xenografted mice. The in vivo experimental procedurewas shown as a timeline (top). A representative images of histologicalanalysis of lungs isolated from xenograft mice bearing Control (shCON) orTet-inducible DOT1L-knockdown (shDOT1L)-MDA-MB-231 tumours (left).Arrowheads in the haematoxylin and eosin (H&E) image indicate lungmetastatic nodules. To confirm DOT1L knockdown, immunohistochemical analysisof DOT1L expression was performed using lung tissues from the mice (bottomleft). Data were quantified by counting the number of surface lung nodules(middle right) or assessing the incidence of lung metastasis in each groupof mice (n=5; bottom right). Error bars indicate themeans±s.e.m. *P<0.05 versus shCON(Student's t-test). Poisson distribution analysis was usedfor calculation of statistical significance of tumour incidence difference.Scale bars in c,f–h,100 μm.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4525167&req=5

f1: DOT1L is associated with aggressive breast cancer by promoting EMT-inducedbreast tumour metastasis.(a) Analysis of overall and disease-free survival according to DOT1Lexpression in 91 ER-negative human breast cancers usingKaplan–Meier method with log-rank test. (b) Thequantification of the association with DOT1L expression and invasion andmetastasis in 182 human breast cancers. The P values were calculatedby χ2-test. (c) Bright-field images ofthe indicated cell types for observation of changes in cell morphology afterDOT1L overexpression and knockdown using lentiviruses encoding DOT1L cDNA orTet-inducible DOT1L shRNA (shDOT1L), respectively. CON (control), emptyvector; shCON, shRNA control. (d) The expression levels of epithelialand mesenchymal markers in the indicated cells were analysed usingimmunoblotting. (e,f) DOT1L-overexpressing MCF10A and itscontrol cells were treated with 100 pM TGF-β1 for48 h, and subjected to immunoblotting for analysis of DOT1Lexpression and activity (e). The invasion (upper) and migration(lower) by the indicated cells were analysed and quantified (f).* and †P<0.05 compared withCON/vehicle and CON/TGF-β, respectively (Student'st-test). (g) Analysis of invasion (left) and migration(right) by Tet-inducible DOTL-knockdown MDA-MB-231 cells.*P<0.05 versus shCON (Student'st-test). The data in f,g represent the means±s.d.of triplicate assays. (h) Effect of DOT1L on lung metastasis ofbreast cancer in xenografted mice. The in vivo experimental procedurewas shown as a timeline (top). A representative images of histologicalanalysis of lungs isolated from xenograft mice bearing Control (shCON) orTet-inducible DOT1L-knockdown (shDOT1L)-MDA-MB-231 tumours (left).Arrowheads in the haematoxylin and eosin (H&E) image indicate lungmetastatic nodules. To confirm DOT1L knockdown, immunohistochemical analysisof DOT1L expression was performed using lung tissues from the mice (bottomleft). Data were quantified by counting the number of surface lung nodules(middle right) or assessing the incidence of lung metastasis in each groupof mice (n=5; bottom right). Error bars indicate themeans±s.e.m. *P<0.05 versus shCON(Student's t-test). Poisson distribution analysis was usedfor calculation of statistical significance of tumour incidence difference.Scale bars in c,f–h,100 μm.

Mentions: A recent study analysing a breast cancer genomic database implies that theDOT1L mRNA level is highly expressed in breast cancer and isespecially associated with oestrogen receptor (ER)-negative breast cancer10. However, clinical evidence for role of DOT1L in breast cancerprogression is still unclear. By using immunohistochemical analysis of DOT1L, weinvestigated the relationship of DOT1L with clinicopathological features in 182human breast cancers. In our cohort, DOT1L was also associated withER-α negativity (P=0.046,χ2-test), as well as progesterone receptornegativity (P=0.014, respectively,χ2-test) and triple-negative breast cancer(P=0.027, χ2-test), whichare aggressive breast cancer subtypes (Supplementary Table 1). Furthermore, the high DOT1L expression levelin the ER-negative invasive ductal carcinoma subtype was associated withsignificantly worse overall survival (P=0.042,Gehan–Breslow–Wilcoxon tests; hazard ratio, 2.624) but nostatistically significant difference in recurrence (disease-free survival;P=0.184; Fig. 1a and Supplementary Table 2). DOT1L expression wasnot associated with survival in 182 patients and a subset of ER-positive breastcancers (Supplementary Table 2). Wefurther found that higher DOT1L expression was significantly correlated withlymph node metastasis (P=0.003,χ2-test) and lymphatic invasion(P<0.001, χ2-test; Fig. 1b and SupplementaryTable 1). Since Zhang et al.10 also showed thatDOT1L inhibitors suppress breast cancer invasion in vitro, we nextexamined whether DOT1L is involved in regulation of EMT, an initiation step fortumour invasion and metastasis. In DOT1L-overexpressing MCF10A cells, amorphological change of epithelial cells into mesenchymal-like cells, loss ofepithelial marker E-cadherin expression and gain of the expression ofmesenchymal markers, N-cadherin and Vimentin were induced (Fig.1c,d and Supplementary Fig.1a,b). Moreover, treatment with DOT1L siRNA reversed the changes inthe EMT marker expression in the DOT1L-overexpressing MCF10A cells, but not inthe siRNA-resistant DOT1L mutant-expressing cells (Supplementary Fig. 1c). Consistently,tetracycline (Tet)-inducible DOT1L short hairpin RNA (shRNA) expression forconditional DOT1L knockdown in mesenchymal-like MDA-MB-231 cells, which havemoderate DOT1L expression, led to a reversible morphological change, whichsuggested that the MET was accompanied by induction of E-cadherin anddownregulation of mesenchymal markers (Fig. 1c,d).Transient DOT1L knockdown in MDA-MB-231 cells using multiple siRNAs also led toupregulation of E-cadherin (SupplementaryFig. 1d). Furthermore, DOT1L overexpression alone was sufficient forincreasing the migration and invasion abilities of non-invasive MCF10A cellsregardless of the treatment with transforming growth factor (TGF)-β, anEMT inducer, or TGF-β inhibitor SB 431542 (Fig.1e,f and Supplementary Fig.2a,b). This effect was reversed by inhibition of DOT1L enzymaticactivity with EPZ004777 treatment (Supplementary Fig. 2a,b). In highly invasive and metastaticMDA-MB-231 cells, Tet-inducible DOT1L depletion inhibited these abilities (Fig. 1g). Similar results were shown in DOT1L-overexpressingT47D cells and DOT1L knockdown SK-BR-3 cells (Fig. 1c andSupplementary Fig.3a–c). Furthermore, orthotopic xenograft mice withTet-inducible DOT1L-knockdown MDA-MB-231 cells displayed significantly inhibitedtumour metastasis in terms of the incidence of lung metastasis (four of five incontrols versus one of five in DOT1L-knockdown cells) and number of nodules onthe lung surface (Fig. 1h). The morphological analysisindicated that the lungs of control mice had numerous variable-sizedinfiltrative nodules composed of atypical cells, while the lungs of xenograftmice with DOT1L-knockdown cells showed well-preserved bronchiolar and alveolarstructures without evidence of metastatic carcinoma (Fig.1h, left). Collectively, our clinical, in vitro and invivo evidence indicated that DOT1L is a marker of aggressive phenotypein human breast cancer that is associated with a worse clinical outcome inER-negative breast cancer and advanced tumour progression related with EMT,invasion and metastasis.


DOT1L cooperates with the c-Myc-p300 complex to epigenetically derepress CDH1 transcription factors in breast cancer progression.

Cho MH, Park JH, Choi HJ, Park MK, Won HY, Park YJ, Lee CH, Oh SH, Song YS, Kim HS, Oh YH, Lee JY, Kong G - Nat Commun (2015)

DOT1L is associated with aggressive breast cancer by promoting EMT-inducedbreast tumour metastasis.(a) Analysis of overall and disease-free survival according to DOT1Lexpression in 91 ER-negative human breast cancers usingKaplan–Meier method with log-rank test. (b) Thequantification of the association with DOT1L expression and invasion andmetastasis in 182 human breast cancers. The P values were calculatedby χ2-test. (c) Bright-field images ofthe indicated cell types for observation of changes in cell morphology afterDOT1L overexpression and knockdown using lentiviruses encoding DOT1L cDNA orTet-inducible DOT1L shRNA (shDOT1L), respectively. CON (control), emptyvector; shCON, shRNA control. (d) The expression levels of epithelialand mesenchymal markers in the indicated cells were analysed usingimmunoblotting. (e,f) DOT1L-overexpressing MCF10A and itscontrol cells were treated with 100 pM TGF-β1 for48 h, and subjected to immunoblotting for analysis of DOT1Lexpression and activity (e). The invasion (upper) and migration(lower) by the indicated cells were analysed and quantified (f).* and †P<0.05 compared withCON/vehicle and CON/TGF-β, respectively (Student'st-test). (g) Analysis of invasion (left) and migration(right) by Tet-inducible DOTL-knockdown MDA-MB-231 cells.*P<0.05 versus shCON (Student'st-test). The data in f,g represent the means±s.d.of triplicate assays. (h) Effect of DOT1L on lung metastasis ofbreast cancer in xenografted mice. The in vivo experimental procedurewas shown as a timeline (top). A representative images of histologicalanalysis of lungs isolated from xenograft mice bearing Control (shCON) orTet-inducible DOT1L-knockdown (shDOT1L)-MDA-MB-231 tumours (left).Arrowheads in the haematoxylin and eosin (H&E) image indicate lungmetastatic nodules. To confirm DOT1L knockdown, immunohistochemical analysisof DOT1L expression was performed using lung tissues from the mice (bottomleft). Data were quantified by counting the number of surface lung nodules(middle right) or assessing the incidence of lung metastasis in each groupof mice (n=5; bottom right). Error bars indicate themeans±s.e.m. *P<0.05 versus shCON(Student's t-test). Poisson distribution analysis was usedfor calculation of statistical significance of tumour incidence difference.Scale bars in c,f–h,100 μm.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4525167&req=5

f1: DOT1L is associated with aggressive breast cancer by promoting EMT-inducedbreast tumour metastasis.(a) Analysis of overall and disease-free survival according to DOT1Lexpression in 91 ER-negative human breast cancers usingKaplan–Meier method with log-rank test. (b) Thequantification of the association with DOT1L expression and invasion andmetastasis in 182 human breast cancers. The P values were calculatedby χ2-test. (c) Bright-field images ofthe indicated cell types for observation of changes in cell morphology afterDOT1L overexpression and knockdown using lentiviruses encoding DOT1L cDNA orTet-inducible DOT1L shRNA (shDOT1L), respectively. CON (control), emptyvector; shCON, shRNA control. (d) The expression levels of epithelialand mesenchymal markers in the indicated cells were analysed usingimmunoblotting. (e,f) DOT1L-overexpressing MCF10A and itscontrol cells were treated with 100 pM TGF-β1 for48 h, and subjected to immunoblotting for analysis of DOT1Lexpression and activity (e). The invasion (upper) and migration(lower) by the indicated cells were analysed and quantified (f).* and †P<0.05 compared withCON/vehicle and CON/TGF-β, respectively (Student'st-test). (g) Analysis of invasion (left) and migration(right) by Tet-inducible DOTL-knockdown MDA-MB-231 cells.*P<0.05 versus shCON (Student'st-test). The data in f,g represent the means±s.d.of triplicate assays. (h) Effect of DOT1L on lung metastasis ofbreast cancer in xenografted mice. The in vivo experimental procedurewas shown as a timeline (top). A representative images of histologicalanalysis of lungs isolated from xenograft mice bearing Control (shCON) orTet-inducible DOT1L-knockdown (shDOT1L)-MDA-MB-231 tumours (left).Arrowheads in the haematoxylin and eosin (H&E) image indicate lungmetastatic nodules. To confirm DOT1L knockdown, immunohistochemical analysisof DOT1L expression was performed using lung tissues from the mice (bottomleft). Data were quantified by counting the number of surface lung nodules(middle right) or assessing the incidence of lung metastasis in each groupof mice (n=5; bottom right). Error bars indicate themeans±s.e.m. *P<0.05 versus shCON(Student's t-test). Poisson distribution analysis was usedfor calculation of statistical significance of tumour incidence difference.Scale bars in c,f–h,100 μm.
Mentions: A recent study analysing a breast cancer genomic database implies that theDOT1L mRNA level is highly expressed in breast cancer and isespecially associated with oestrogen receptor (ER)-negative breast cancer10. However, clinical evidence for role of DOT1L in breast cancerprogression is still unclear. By using immunohistochemical analysis of DOT1L, weinvestigated the relationship of DOT1L with clinicopathological features in 182human breast cancers. In our cohort, DOT1L was also associated withER-α negativity (P=0.046,χ2-test), as well as progesterone receptornegativity (P=0.014, respectively,χ2-test) and triple-negative breast cancer(P=0.027, χ2-test), whichare aggressive breast cancer subtypes (Supplementary Table 1). Furthermore, the high DOT1L expression levelin the ER-negative invasive ductal carcinoma subtype was associated withsignificantly worse overall survival (P=0.042,Gehan–Breslow–Wilcoxon tests; hazard ratio, 2.624) but nostatistically significant difference in recurrence (disease-free survival;P=0.184; Fig. 1a and Supplementary Table 2). DOT1L expression wasnot associated with survival in 182 patients and a subset of ER-positive breastcancers (Supplementary Table 2). Wefurther found that higher DOT1L expression was significantly correlated withlymph node metastasis (P=0.003,χ2-test) and lymphatic invasion(P<0.001, χ2-test; Fig. 1b and SupplementaryTable 1). Since Zhang et al.10 also showed thatDOT1L inhibitors suppress breast cancer invasion in vitro, we nextexamined whether DOT1L is involved in regulation of EMT, an initiation step fortumour invasion and metastasis. In DOT1L-overexpressing MCF10A cells, amorphological change of epithelial cells into mesenchymal-like cells, loss ofepithelial marker E-cadherin expression and gain of the expression ofmesenchymal markers, N-cadherin and Vimentin were induced (Fig.1c,d and Supplementary Fig.1a,b). Moreover, treatment with DOT1L siRNA reversed the changes inthe EMT marker expression in the DOT1L-overexpressing MCF10A cells, but not inthe siRNA-resistant DOT1L mutant-expressing cells (Supplementary Fig. 1c). Consistently,tetracycline (Tet)-inducible DOT1L short hairpin RNA (shRNA) expression forconditional DOT1L knockdown in mesenchymal-like MDA-MB-231 cells, which havemoderate DOT1L expression, led to a reversible morphological change, whichsuggested that the MET was accompanied by induction of E-cadherin anddownregulation of mesenchymal markers (Fig. 1c,d).Transient DOT1L knockdown in MDA-MB-231 cells using multiple siRNAs also led toupregulation of E-cadherin (SupplementaryFig. 1d). Furthermore, DOT1L overexpression alone was sufficient forincreasing the migration and invasion abilities of non-invasive MCF10A cellsregardless of the treatment with transforming growth factor (TGF)-β, anEMT inducer, or TGF-β inhibitor SB 431542 (Fig.1e,f and Supplementary Fig.2a,b). This effect was reversed by inhibition of DOT1L enzymaticactivity with EPZ004777 treatment (Supplementary Fig. 2a,b). In highly invasive and metastaticMDA-MB-231 cells, Tet-inducible DOT1L depletion inhibited these abilities (Fig. 1g). Similar results were shown in DOT1L-overexpressingT47D cells and DOT1L knockdown SK-BR-3 cells (Fig. 1c andSupplementary Fig.3a–c). Furthermore, orthotopic xenograft mice withTet-inducible DOT1L-knockdown MDA-MB-231 cells displayed significantly inhibitedtumour metastasis in terms of the incidence of lung metastasis (four of five incontrols versus one of five in DOT1L-knockdown cells) and number of nodules onthe lung surface (Fig. 1h). The morphological analysisindicated that the lungs of control mice had numerous variable-sizedinfiltrative nodules composed of atypical cells, while the lungs of xenograftmice with DOT1L-knockdown cells showed well-preserved bronchiolar and alveolarstructures without evidence of metastatic carcinoma (Fig.1h, left). Collectively, our clinical, in vitro and invivo evidence indicated that DOT1L is a marker of aggressive phenotypein human breast cancer that is associated with a worse clinical outcome inER-negative breast cancer and advanced tumour progression related with EMT,invasion and metastasis.

Bottom Line: DOT1L has emerged as an anticancer target for MLL-associated leukaemias; however, its functional role in solid tumours is largely unknown.DOT1L recognizes SNAIL, ZEB1 and ZEB2 promoters via interacting with the c-Myc-p300 complex and facilitates lysine-79 methylation and acetylation towards histone H3, leading to the dissociation of HDAC1 and DNMT1 in the regions.Collectively, we suggest that cooperative effect of DOT1L and c-Myc-p300 is critical for acquisition of aggressive phenotype of breast cancer by promoting EMT/CSC.

View Article: PubMed Central - PubMed

Affiliation: Department of Pathology, College of Medicine, Hanyang University, Seoul 133-791, Korea.

ABSTRACT
DOT1L has emerged as an anticancer target for MLL-associated leukaemias; however, its functional role in solid tumours is largely unknown. Here we identify that DOT1L cooperates with c-Myc and p300 acetyltransferase to epigenetically activate epithelial-mesenchymal transition (EMT) regulators in breast cancer progression. DOT1L recognizes SNAIL, ZEB1 and ZEB2 promoters via interacting with the c-Myc-p300 complex and facilitates lysine-79 methylation and acetylation towards histone H3, leading to the dissociation of HDAC1 and DNMT1 in the regions. The upregulation of these EMT regulators by the DOT1L-c-Myc-p300 complex enhances EMT-induced breast cancer stem cell (CSC)-like properties. Furthermore, in vivo orthotopic xenograft models show that DOT1L is required for malignant transformation of breast epithelial cells and breast tumour initiation and metastasis. Clinically, DOT1L expression is associated with poorer survival and aggressiveness of breast cancers. Collectively, we suggest that cooperative effect of DOT1L and c-Myc-p300 is critical for acquisition of aggressive phenotype of breast cancer by promoting EMT/CSC.

No MeSH data available.


Related in: MedlinePlus