Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Comparative Study
. 2014 Nov 15;5(21):10840-53.
doi: 10.18632/oncotarget.2535.

CDKL2 promotes epithelial-mesenchymal transition and breast cancer progression

Affiliations
Comparative Study

CDKL2 promotes epithelial-mesenchymal transition and breast cancer progression

Linna Li et al. Oncotarget. .

Abstract

The epithelial-mesenchymal transition (EMT) confers mesenchymal properties on epithelial cells and has been closely associated with the acquisition of aggressive traits by epithelial cancer cells. To identify novel regulators of EMT, we carried out cDNA screens that covered 500 human kinases. Subsequent characterization of candidate kinases led us to uncover cyclin-dependent kinase-like 2 (CDKL2) as a novel potent promoter for EMT and breast cancer progression. CDKL2-expressing human mammary gland epithelial cells displayed enhanced mesenchymal traits and stem cell-like phenotypes, which was acquired through activating a ZEB1/E-cadherin/β-catenin positive feedback loop and regulating CD44 mRNA alternative splicing to promote conversion of CD24(high) cells to CD44(high) cells. Furthermore, CDKL2 enhanced primary tumor formation and metastasis in a breast cancer xenograft model. Notably, CDKL2 is expressed significantly higher in mesenchymal human breast cancer cell lines than in epithelial lines, and its over-expression/amplification in human breast cancers is associated with shorter disease-free survival. Taken together, our study uncovered a major role for CDKL2 in promoting EMT and breast cancer progression.

PubMed Disclaimer

Conflict of interest statement

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Figures

Figure 1
Figure 1. Human kinase cDNA screen and validation for novel regulators of EMT
(A) human kinase cDNA screen was performed by measuring vimentin promoter luciferase reporter activities in firefly luciferase vector (VimPro-luc) that was transiently co-transfected with individual kinase cDNA vector and TK-renilla luciferase vector (pRL-TK, transfection control) into 293T cells. Shown on Y-axis is log2 transformation of the fold changes of vimentin promoter luciferase activity induced by individual kinases vs GFP control. X-axis represents 651 cDNA clones for 500 human kinases. (B) Western blotting analysis of mesenchymal markers (Vimentin, N-cadherin, and Fibronectin) and epithelial marker (Occludin) in stable HMLE cell lines expressing several candidate kinases, GFP negative control, and positive controls (MET and FYN). (C) mesenchymal-like morphological changes occurred in several stable HMLE cell lines expressing indicated kinases. Scale bars represent 50 μm. (D) FACS analysis of CD44 and CD24 in several stable cell lines. The percentage of the CD44high/CD24low mesenchymal subpopulation is indicated. (E) shRNA silencing of CDKL2 gene led to downregulation of mesenchymal markers and upregulation of epithelial marker in HMLE cells, as measured by RT-PCR and western blotting analysis (EV, empty vector). A high quality antibody specific for CDKL2 is unavailable.
Figure 2
Figure 2. CDKL2-transduced cells show both EMT and stem cell-like phenotypes
(A) CDKL2 induced migration and EMT marker expression in 3 epithelial cell lines, showing representative photos of migration (top), quantification of migration as the mean ± SD (middle) and vimentin expression (bottom). (B) HMLE-CDKL2 cells generated more mammospheres than HMLE-EV control cells. Phase-contrast images represent mammospheres formed by indicated cell lines. (C) HMLE cells expressing CDKL2 gained MSC-like capabilities for multilineage differentiation. Following culture in osteoblastic differentiation media, cells were tested for alkaline phosphatase (AP) activity, or analyzed by alizarin red S staining and silver nitrate (Von-Kossa) staining to determine calcium deposition and mineral deposition. Following culture in adipogenic differentiation media, cells were stained with oil red dye to detect oil droplets formation. (D) dose-response in survival and proliferation of HMLE-EV and HMLE-CDKL2 cells treated with different concentrations of paclitaxel or doxorubicin, or incubated with reducing concentrations of growth factors. IC50 values were obtained by using logistic nonlinear regression analyzing model of MicroCal Origin 7.0 software. Error bars denote SD from quadruplicate.
Figure 3
Figure 3. CDKL2 promotes direct transition of CD24high epithelial cells into CD44high cells and endows them with enhanced EMT and stem cell-like phenotypes
(A) CD44high (short for CD44high/CD24low) and CD24high (short for CD44low/CD24high) subpopulations of HMLE-EV and HMLE-CDKL2 parental cells were sorted out by FACS. The percentage of sorted cells is indicated. (B) representative photos of migration of above cells determined by Boyden Chamber assay. (C) osteogenic differentiation ability (Alkaline Phosphatase activity) of parental and sorted cells. (D) mammospheres formed by parental and sorted cells. (E) relative expression of epithelial markers (E-cadherin and CD44v8-9) and mesenchymal markers (Vimentin and CD44s) was measured by qRT-PCR. The data are reported as mean ± SD. (F) proliferation curves of CD44high sorted cells indicated that CD44high subpopulation from HMLE-CDKL2 cells did not proliferate faster than those from HMLE-EV cells. Bars denote standard error from quadruplicate. (G) FACS re-analysis after 7 passages of continuous culture of CD44high sorted cells indicated that most CD44high cells from HMLE-EV and -CDKL2 cells remained CD44high. (H) EV and CDKL2 were re-introduced into CD24high epithelial subpopulation sorted from HMLE parental cells. After 4 weeks of drug selection and in vitro passaging, cell migration ability, CD44/CD24 and EMT markers expression were re-analyzed. These results indicated that increased CD44high subpopulation in HMLE-CDKL2 cells was a result of true EMT by CDKL2 converting CD24high cells into CD44high cells.
Figure 4
Figure 4. ZEB1 is a key mediator in CDKL2-induced EMT
(A) Western blotting analysis of EMT markers (E-cadherin, Vimentin, N-cadherin, Fibronectin), EMT regulators (ZEB1 and AP1) in parental and CD44high subpopulation of HMLE-EV and HMLE-CDKL2 cells. (B) changes in the expression of ZEB1, ESRP1/2, CD44s and CD44v8-9 by CDKL2 knockdown in HMLE cells as determined by real time PCR. *P<0.05, **P<0.01, ***P<0.001. C and D, shRNA silencing of ZEB1 gene and resultant reversal of EMT in HMLE-CDKL2 cells. Showed are changes in the expression of ZEB1, EMT markers, ESRP1/2 and CD44s (C) as well as cellular morphology, migration ability and CD44/CD24 antigenic profile (D) by ZEB1 knockdown in HMLE-CDKL2 cells.
Figure 5
Figure 5. CDKL2 activates a positive feedback loop composed of ZEB1, E-cadherin and β-catenin
(A) immunofluorescence images of HMLE-EV and -CDKL2 cells stained for β-catenin and E-cadherin. Cell nuclei were stained with DRAQ5. Scale bars, 50 μm. (B) luciferase activity of TOPflash, ZEB1 and vimentin promoter luciferase-reporter constructs in HMLE-EV and -CDKL2 cells. (C-F) breaking the loop by β-catenin (CTNNB1) silencing in HMLE-CDKL2 cells resulted in reversal of CDKL2-induced EMT. Shown are alterations in the expression of ZEB1, EMT markers, ESRP1/2 and CD44s (C), immunofluorescence staining patterns of β-catenin and E-cadherin (D), luciferase reporter activities of TOPflash, ZEB1 and vimentin promoters (E), as well as morphology, CD44/CD24 antigenic profile and migration ability (F). Scale bar, 50 μm. (G-H) breaking the loop by ZEB1 silencing in HMLE-CDKL2 cells. Shown are alterations in luciferase reporter activities of TOPflash, ZEB1 and vimentin promoters (G), as well as the immunofluorescence staining patterns of β-catenin and vimentin (H). Scale bars, 25 μm. (I) a proposed model of CDKL2 regulation on CD44high and stem cell-like phenotypes through activation of a positive feedback loop composed of ZEB1, E-cadherin and β-catenin. The loop, on one hand, promotes CD44 transcription through activating Wnt/β-catenin signaling; on the other hand, represses ESRP1/2 expression by upregulating ZEB1, which promotes CD44 alternative splicing, resulting in a switch in expression from epithelial CD44v isoforms to mesenchymal CD44s isoform and CD44high subpopulation increases over the time not only in quantity but also with enhanced EMT and stem cell-like phenotypes.
Figure 6
Figure 6. CDKL2 promotes mammary gland tumor formation and lung metastasis in orthotopic xenograft mouse model
(A) tumor incidence and tumor weight of transformed HMLER cells expressing CDKL2 or EV. (B) in vivo and ex vivo bioluminescence imagining of NOD/SCID mouse hosts injected with HMLER cells expressing either EV or CDKL2 into the fourth inguinal mammary glands. The fourth inguinal mammary glands on either side of the same mice were injected with the same cell lines. Shown on left panel are representative in vivo images of orthotopic tumors and lymph node metastases 5 weeks after inoculation. Shown on right panel is representative ex vivo bioluminescence images of lung, liver and chest bones. (C) histological analysis of tumors developed from HMLER-EV and -CDKL2 cells. Shown are H&E and IHC staining for β-catenin, ZEB1, vimentin and E-cadherin. Scale bars, 200 μm.
Figure 7
Figure 7. CDKL2 expression in human breast cancer cell lines and human invasive breast cancers
(A) CDKL2, E-cadherin and Vimentin mRNA levels in a panel of epithelial (basal, luminal) and mesenchymal human breast cancer cell lines, as well as human mesenchymal stem cell (MSC) and fibroblast cell lines. Shown on Y axis is log2 transformation of fold changes for CDKL2, Vimentin and E-cadherin in the indicated cell lines relative to HCC1954 cell line. (B) scatter plots of log2 of fold changes for CDKL2 vs Vimentin (top, Spearman correlation coefficient rho= 0.73) and E-cadherin (bottom, rho= - 0.47). (C) CDKL2 alterations in human invasive breast cancers. A Kaplan-Meier plot of overall survival corresponding to 749 invasive breast cancers from the publicly available TCGA database is shown for two groups with and without CDKL2 amplifications and/or over-expression, as defined by cBio Portal. The P-value was calculated using the Log-rank test.

Similar articles

Cited by

References

    1. Kalluri R, Weinberg RA. The basics of epithelial-mesenchymal transition. J Clin Invest. 2009;119:1420–1428. - PMC - PubMed
    1. Thiery JP, Acloque H, Huang RY, Nieto MA. Epithelial-mesenchymal transitions in development and disease. Cell. 2009;139:871–890. - PubMed
    1. Polyak K, Weinberg RA. Transitions between epithelial and mesenchymal states: acquisition of malignant and stem cell traits. Nat Rev Cancer. 2009;9:265–273. - PubMed
    1. Yang J, Weinberg RA. Epithelial-mesenchymal transition: at the crossroads of development and tumor metastasis. Dev Cell. 2008;14:818–829. - PubMed
    1. Zhang J, Ma L. MicroRNA control of epithelial-mesenchymal transition and metastasis. Cancer Metastasis Rev. 2012;31:653–662. - PMC - PubMed

Publication types

MeSH terms