METHOD

Abstract

The use of Akt3 as a biomarker for detecting the occurrence of epithelial-to-mesenchymal transition (EMT) in a subject, and the use of Akt3 inhibitors to treat cancer is disclosed herein. Also disclosed are various methods for detecting the occurrence of epithelial-to-mesenchymal transition (EMT) in a subject by measuring Akt3 expression and/or activity.

Claims

1.-67. (canceled)

68. A method of diagnosing and treating cancer comprising: (a) obtaining a sample from a subject; (b) detecting a level of epithelial-to-mesenchymal transition (EMT) in the sample from the subject by contacting the sample with a reagent that specifically binds Akt3 protein or Akt3 mRNA and detecting the expression level of Akt3 in the sample; (c) diagnosing the subject with increased risk of cancer when the level of EMT is increased in the sample as compared to a reference level; and (d) administering an effective amount of a therapeutic agent or radiotherapeutic treatment to the subject diagnosed with increased risk of cancer.

69. The method of claim 68, wherein the cancer is a cancer selected from the group consisting of: acute myeloid leukemia (AML), myelodysplatic syndrome (MDS), non-small-cell lung carcinoma (NSCLC), melanoma, triple negative breast cancer, ovarian cancer, glioma, bladder cancer, gastric cancer, hepatocellular carcinoma (HCC), and prostate cancer.

70. The method of claim 68, wherein Akt3 mRNA is detected in step (b).

71. The method of claim 70, wherein Akt3 mRNA comprises an mRNA encoding the protein of SEQ ID NO: 1 or SEQ ID NO: 2.

72. The method of claim 68, wherein Akt3 protein is detected in step (b).

73. The method of claim 72, wherein Akt3 protein comprises SEQ ID NO: 1 or SEQ ID NO: 2.

74. The method of claim 68, wherein the therapeutic agent comprises a chemotherapeutic agent.

75. The method of claim 68, wherein the therapeutic agent comprises an Axl inhibitor and/or Akt3 inhibitor.

76. The method of claim 68, wherein the therapeutic agent comprises BGB324/R428.

77. A method of treating cancer in a subject comprising: contacting a sample from the subject with a reagent that specifically binds Akt3 protein or Akt3 mRNA; detecting the expression level of Akt3 to determine the level of epithelial-to-mesenchymal transition (EMT) as compared to a reference level; and administering a therapeutically effective amount of a therapeutic agent or radiotherapeutic treatment to the subject provided that a sample from the subject has an increased level of epithelial-to-mesenchymal transition (EMT) as compared to a reference level.

78. The method of claim 77, wherein the cancer is a cancer selected from the group consisting of: acute myeloid leukemia (AML), myelodysplatic syndrome (MDS), non-small-cell lung carcinoma (NSCLC), melanoma, triple negative breast cancer, ovarian cancer, glioma, bladder cancer, gastric cancer, hepatocellular carcinoma (HCC), and prostate cancer.

79. The method of claim 77, wherein Akt3 mRNA is contacted.

80. The method of claim 79, wherein Akt3 mRNA comprises an mRNA encoding the protein of SEQ ID NO: 1 or SEQ ID NO: 2.

81. The method of claim 77, wherein Akt3 protein is contacted.

82. The method of claim 81, wherein Akt3 protein comprises SEQ ID NO: 1 or SEQ ID NO: 2.

83. The method of claim 77, wherein the therapeutic agent comprises a chemotherapeutic agent.

84. The method of claim 77, wherein the therapeutic agent comprises an Axl inhibitor and/or Akt3 inhibitor.

85. The method of claim 77, wherein the therapeutic agent comprises BGB324/R428.

86. A cell line which is sensitive to inhibitors of epithelial-to-mesenchymal transition (EMT), the cell line having a level of Akt3 expression that is just insufficient to prevent EMT.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0228] FIG. 1 is a set of photographs of immunoblots depicting results of experiments on breast epithelial cells undergoing EMT;

[0229] FIG. 2 shows Akt3 is up-regulated when breast cancer cells undergo EMT, and these changes are Axl-dependent;

[0230] FIG. 3 shows Akt3, and not Akt1 is downregulated in response to Axl inhibition in triple negative breast cancer cells;

[0231] FIG. 4 is a set of photographs of immunoblots determining levels of Akt isoforms in EMT-induced breast cancer cells;

[0232] FIG. 5 shows Akt3, but not Akt1 and 2 mRNA correlate with EMT and stem genes in breast cancer cells and breast cancer biopsies;

[0233] FIG. 6 shows suppression of Akt3 expression is able to reverse EMT and CSC traits in breast epithelial cells;

[0234] FIG. 7 is a photograph of gel experiments on the activity of Akt isoforms;

[0235] FIG. 8 is a photograph of growth studies of breast cancer cells;

[0236] FIG. 9 is a set of photographs of mammosphere cultures of breast cancer cells and a graph;

[0237] FIG. 10 shows constitutively-active Akt3 (myr-Akt3), but not constitutively-active Akt1 is able to induce EMT;

[0238] FIG. 11 shows constitutively-active Akt3 (MyrAkt3), but not constitutively-active Akt1 is able to induce EMT and CSC traits in breast epithelial cells;

[0239] FIG. 12 shows cells expressing constitutively active Akt3, but not cells expressing constitutively active Akt1 show the ability to grow in mammospheres;

[0240] FIG. 13 shows cells expressing constitutively active Akt3 show a much higher ability to form tumors than cells expressing constitutively active Akt1

[0241] FIG. 14 is a photograph of growth studies of breast cancer cells;

[0242] FIG. 15 is an image of a gel from experiments in which breast cancer cells were treated with an Akt inhibitor; and

[0243] FIG. 16 is a photograph of gels from experiment to study activity of Akt isoforms.

[0244] FIG. 17 shows constitutively active Akt3, but not constitutively active Akt1 is localized to the cell nucleus

[0245] FIG. 18 shows Akt3 and Snail are found in the nuclear fraction

[0246] FIG. 19 shows Akt3 localizes to the nucleus in cultured triple negative breast cancer cells and primary human mammary epithelial cells

[0247] FIG. 20 shows suppression of Axl kinase expression reduces EMT induced nuclear localization of Akt3

[0248] FIG. 21 shows inhibition of Axl kinase activity inhibits nuclear localization of Akt3

[0249] FIG. 22 shows Akt1 and Akt3 kinases are able to directly phosphorylate Snail protein

[0250] FIG. 23 shows specific detection of phospho-Akt3 in MCF7, WM115 and LNCaP cells

EXAMPLES

Example 1 when Breast Epithelial Cells Undergo EMT, Akt3 is Up-Regulated while Akt2 is Down-Regulated, and these Changes are Ail-Dependent

[0251] MCF10A cells (American Type Culture Collection), a breast epithelial cell line used as a model for normal breast epithelial cells, were cultured in DMEM/F12 medium supplemented with 5% horse serum, 20 ng/mL EGF, 0.5 g/mL hydrocortisone, 100 ng/mL cholera toxin, 10 g/mL insulin, 100 U/mL penicillin and 100 g/mL streptomycin (Sigma-Aldrich). MCF10A cells were used in this experiment along with a retroviral vector (Slug) driving expression of the EMT inducer Slug and a retroviral vector (Axl2) driving expression of a shRNA that knocks down expression of Axl (vectors described in (Gjerdrum C et al. Axl is an essential epithelial-to-mesenchymal transition-induced regulator of breast cancer metastasis and patient survival. Proc Natl Acad Sci USA. 2010 Jan. 19; 107(3):1124-9). Briefly, Axl shRNA was expressed from a modified human U6 promoter in the LTR of the retroviral vectors RRI-Red/L087 (GenBank: EU424173), while the human Slug cDNA sequence from BC012910 (Open Biosystcms) was cloned into the CRUS-IRES-OFP retroviral vector (Lorens J B, Jang Y, Rossi A B, Payan D G, Bogenberger J M (2000) Optimization of regulated LTR-mediated expression. Virology 272:7-15). MCF10A cells were transduced with either retroviral vector Slug alone or with a combination of both the Slug and Axl2 retroviral vectors. Control cells were transduced with neither vector. Protein extracts were prepared from the control and transduced cell populations by lysis in RIPA buffer (PBS with 1% (vol/vol) Nonidet P-40 (Nonidet P-40), 0.5% (wt/vol) sodium deoxycholate, and 0.1% (wt/vol) SDS) supplemented with protease inhibitor (Ser. No. 13/457,200; Roche) and 0.2 mM PMSF. Protein concentration was determined by Bradford assay (BioRad), and 50 g of total protein was loaded in each well of SDS/PAGE. Running of gel and immunoblotting were carried out according to standard procedures. The antibodies used were mouse monoclonal anti-human Axl (MABL54; R&D Systems), AKT1 (Cell Signaling #2967), Akt2 (Cell Signaling #3063), Akt3 (Millipore #1586912), pAKT (Ser473, Cell Signaling 2971), -actin (Sigma-Aldrich), pERK (Cell Signaling #4695). The pAKT antibody reacts with Akt1, Akt2 and Akt3 when they are phosphorylated at the amino-acid corresponding to Ser473 in Akt1.

[0252] The results of these experiments are shown in FIG. 1. As expected, pAKT is increased in the cells undergoing EMT (compare Control, Slug lanes), and this increase is blocked when EMT is blocked by knocking down Axl (compare Control, Slug, Axl2-Slug lanes). Unexpectedly, Akt3 expression is strongly upregulated when these breast epithelial cells undergo EMT. In contrast, Akt1 expression remains constant and Akt2 is down-regulated. Blocking Axl in EMT induced cells (Axl2-Slug lane) also blocks the switch in Akt isoform indicating that the change in expression from Akt2 to Akt3 depends on Axl. Note that activated (phosphorylated) Akt (pAkt) follows Akt3 levels, and not Akt1 and 2 levels, suggesting that the major phospho-Akt isoform in these cells may be Akt3. This is confirmed in Example 4.

Example 2 when Transformed Breast Cancer Cells Undergo EMT, Akt3 is Up-Regulated, and these Changes are Axl-Dependent

[0253] HMLE and HMLER cells (Elenbaas B, Spirio L, and Weinberg R A. Human breast cancer cells generated by oncogenic transformation of primary mammary epithelial cells. Genes Dev. 2001 Jan. 1; 15(1):50-65)), experimentally transformed breast cancer cell lines, were cultured in MEGM (Lonza)/DMEMF12 (Sigma-Aldrich) medium supplemented with 5 ng/mL EGF, 10 g/mL insulin, 0.5 g/mL hydrocortisone, 100 U/mL penicillin and 100 g/mL streptomycin (Sigma-Aldrich). Cells were transduced with either retroviral vectors Slug, Axl2, control luciferase shRNA (ctr), or with a combination of both the Slug and Axl2 retroviral vectors as described in Example 1. Preparation of protein extracts, running of gel, immunoblotting and probing of membranes were performed as described in Example 1.

[0254] The results of these experiments are shown in FIG. 2. As found in Example 1, Akt3 expression is strongly upregulated when these breast epithelial cells undergo EMT (compare Control, Slug lanes). HMLE cells (left): Activated (phosphorylated) Akt (pAkt) follows Akt3 levels, and not Akt1 levels (compare Control and Slug lanes). HMLER cells (right): Blocking Axl in EMT induced cells (Slug-Axl2 lane) also blocks Akt3 expression, indicating that Akt3 expression levels depends on Axl.

Example 3 Akt3, and not Akt1 is Downregulated in Response to Axl Inhibition in Triple Negative Breast Cancer Cells

[0255] MDA-MB231 cells (American Type Culture Collection), a triple negative breast cancer cell line, were cultured in F12 media supplemented with 10% fetal bovine serum, 100 U/mL penicillin and 100 g/mL streptomycin (Sigma-Aldrich). The cells were transduced with retroviral constructs expressing shAxl (Axl2) or control luciferase shRNA (control) as described in Example 1 and Example 2. Preparation of protein extracts, running of gel, immunoblotting and probing of membranes were performed as described in Example 1 and 2.

[0256] Results are shown in FIG. 3. MDA-MB231 cells express Axl, Akt1 and Akt3 (see control lane). In line with data shown in Example 1 and 2, blocking of Axl (Axl2), also blocks Akt3 expression, but has no significant effect on Akt1 expression, indicating that Akt3, but not Akt1, expression levels depends on Axl.

Example 4 Akt3 Represents the Major P-Akt Isoform in MCF10A Cells Induced to Undergo EMT (by TGF Treatment)

[0257] MCF10A cells were treated with TGF (10 ng/ml) for 4 days. Cells were then lysed using NP40 Cell Lysis Buffer (40 mM HepesNAOH, 75 mM NaCl, 2 mM EDTA, 1% NP40, phosphatase inhibitor cocktail tablet, protease inhibitor cocktail tablet (Roche)), scraped off the plate, rotated at 4 C. for 30 min, centrifuged at 13000 rpm for 10 min, and supernatant harvested. For immunoprecipitation, Akt1 (2H10, Cell Signaling #2967), Akt2 (Cell Signaling #3063), Akt3 (Millipore #07-383) and control IgG (Abcam) antibodies (1 g/lysate) were added to lysates and incubated overnight at 4 C. Next day the pre-blocked protein-G beads (GE Healthcare) in lysis buffer were added and allowed to bind at 4 C. for 1 hour. Beads were then washed 3 times (20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% NP40), protein eluted by boiling in SDS-PAGE loading buffer. Running of SDS/PAGE and immunoblotting were carried out according to standard procedures. Membrane was probed using pAktS473 (Cell Signaling #9271) and PAN-Akt (Cell Signaling #9272) antibodies.

[0258] These data, shown in FIG. 4, demonstrate that phospho-Akt3 represents the major pAkt isoform in EMT-induced MCF10A cells. No phospho-Akt1 was detected. This was unexpected in view of the previous studies suggesting that Akt1 was responsible for the effects of Akt.

Example 5 Akt3, but not Akt1 and 2 mRNA Correlate with EMT and Stem Genes in Breast Cancer Cells and Breast Cancer Biopsies

[0259] The expression analysis of the breast cancer cell lines and human breast cancer biopsy samples (cancer, normal) was performed from published and GEO-submitted Affymetrix data as described (Kilpinen S, Autio R, Ojala K, Iljin K, Bucher E, Sara H, Pisto T, Saarela M, Skotheim R I, Bjrkman M, Mpindi J P, Haapa-Paananen S, Vainio P, Edgren H, Wolf M, Astola J, Nees M, Hautaniemi S, Kallioniemi O. Systematic bioinformatic analysis of expression levels of 17,330 human genes across 9,783 samples from 175 types of healthy and pathological tissues. Genome Biol. 2008; 9(9):R139.). Positive correlation is indicated with a plus (+) sign, darker gray indicate stronger positive correlation, and higher confidence is indicated with increasing number of asterisks (*). Similarly, negative correlation is indicated with a minus () sign, darker gray indicate stronger negative correlation, and higher confidence is indicated with increasing number of asterisks (*).

TABLE-US-00004 Epithelial Mesenchymal Cancer Stem Cell Gene marker marker EMT-mediator marker SOX2 X X SNAI1 X CD44 X X SEMA3C X? TWIST1 X ZEB1 X CDH2 X ID4 X VIM X ZEB1 X AXL X X SNAI2 X PLXNA2 X

[0260] Results are shown in FIG. 5. Akt3, not Akt1 and Akt2, show strong correlation with EMT and stem markers in breast cancer cell lines and breast cancer biopsies.

Example 6 Knocking Down Akt3 is Able to Reverse EMT and CSC Traits in Breast Epithelial Cells

[0261] MCF10A cells and MDA-MB 231 cells were cultured as described in Example 1 and 3. MCF10a were transduced with the EMT driver Slug (Slug/control) as described in Example 1. siRNA-mediated silencing of Akt3 was done using HiPerFect transfection reagent (Qiagen) according to the manufacturer's protocol, and the cells were cultured for 2-3 days. Annealed siRNAs against Akt3 (siAKT3; HsAkt3_2 HP), and GAPDH (control; Hs_GAPDH_3) were used at 60 nM final concentrations (all were from Qiagen). SDS/PAGE, Immunoblotting and antibodies as described in Example 1 except Rat anti-human Vimentin (MAB2105; R&D Systems). The 3D matrigel experiments were performed as described (Gjerdrum C, Tiron C, Heiby T, Stefansson I, Haugen H, Sandal T, Collett K, Li S, McCormack E, Gjertsen B T, Micklem D R, Akslen L A, Glackin C, Lorens J B. Axl is an essential epithelial-to-mesenchymal transition-induced regulator of breast cancer metastasis and patient survival. Proc Natl Acad Sci USA. 2010 Jan. 19; 107(3):1124-9.). The cells were visualized by fluorescence microscopy (DAPI nuclear stain) using a Nikon TE2000 microscope (Nikon).

[0262] These data, presented in FIG. 6, show that knocking down Akt3 is able to reverse EMT traits in two different breast epithelial cells as shown by down-regulation of the mesenchymal marker Vimentin, and inhibition of invasive, stellate growth in 3D Matrigel.

Example 7 Constitutively Active Akt3, but not Constitutively Active Akt1 is Able to Activate EMT and to Activate EMT Regulators

[0263] SDS/PAGE, Immunoblotting and antibodies as described in Example 1 and 6, except Rabbit anti-human N-cadherin (ab18203; Abcam), Rabbit anti-human Ecadherin (24E10; Cell Signaling), Rabbit anti-human 3-catenin (LS4E2; Cell Signaling), Mouse anti-human Twist (Twist2C1a; Abeam).

[0264] MCF10A cells, cultured as described in Example 1, were transduced with empty vector (CRUS-IRES-GFP described in Example 1), or the CRUS-IRES-GFP vector harboring a constitutively active myristylated form of Akt1 (myrAkt1) or a constitutively active myristylated form of Akt3 (myrAkt3). When directed to membranes by the addition of a src myristoylation sequence, Akt becomes constitutively active (Barthel A, Kohn A D, Luo Y, Roth R A. A constitutively active version of the Ser/Thr kinase Akt induces production of the ob gene product, leptin, in 3T3-L1 adipocytes. Endocrinology. 1997 August; 138(8):3559-62). MCF10A cells were transduced with either retroviral vector myr-AKT1 or with a retroviral vector myr-AKT3. Control cells (wt) were transduced with neither vector. Immunoblots of proteins extracted from these cell lines were probed with a panel of markers associated with epithelial and mesenchymal cell fates and EMT.

[0265] These data, shown in FIG. 7, unexpectedly show that constitutively active Akt3, but not constitutively active Akt1 is able to activate EMT as shown by up-regulation of mesenchymal markers (N-cadherin, Vimentin) and loss of epithelial markers (E-cadherin, b-catenin). The expression of myr-Akt3 also leads to activation of the EMT regulators Snail and Axl, and phosphorylation of Akt suggesting the existence of a positive feedback loop.

Example 8 Constitutively-Active Akt3 (MyrAkt3), but not Constitutively-Active Akt1 is Able to Induce EMT and CSC Traits in Breast Epithelial Cells

[0266] MCF10A cells were used in this experiment along with a retroviral vector (myr-AKT1) driving expression of constitutively active Akt1 and a retroviral vector (myr-AKT3) driving expression of constitutively active Akt3. The 3D matrigel experiments were performed as described in Example 6. The cells were visualized at indicated magnification by phase-contrast and fluorescence microscopy (DAPI nuclear stain) using a Nikon TE2000 microscope (Nikon).

[0267] Results shown in FIG. 8. These data show that constitutively-active Akt3, but not constitutively-active Akt1 is able to induce EMT and CSC traits in breast epithelial cells (fibmblastoid cell growth in 2D culture and invasive, stellate growth in 3D Matrigel).

Example 9 Cells Expressing Constitutively Active Akt3, but not Cells Expressing Constitutively Active Akt1 Show the Ability to Grow in Mammospheres

[0268] Cell lines used are as described in Example 1. Mammosphere culture of MCF10A cells was performed as described (Dontu G, Abdallah W M, Foley M, Jackson K W, Clarke M F, Kawamura M J, Wicha M S. In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes Dev. 2003 May 15; 17(10): 1253-70). Briefly, single cells were plated on 35 mm ultra-low attachment plates (Corning, USA, Cat. #3471), 20000 viable cells/ml at a total of 30000 cells per well. The mammospheres were cultured for 10 days, imaged and mammospheres quantified using ImageJ (http://rsbweb.nih.gov/ij/index.html). Statistical analyses were based on Students t-test. Results shown in FIG. 9.

[0269] The ability to form mammosphereslarge structures composed of many cellsis considered to be a trait of cancer stem cells. MCF10A cells expressing constitutively-active Akt3 were able to form as mammospheres. In contrast cells expressing constitutively active Akt1 and untreated MCF10A cells are not able to form mammospheres. The ability to form mammospheres is therefore triggered by signalling through Akt3 rather than through Akt1.

Example 10 Constitutively-Active Akt3 (Myr-Akt3), but not Constitutively-Active Akt1 is Able to Induce EMT, Leading to a Rise in EMT/Mesenchymal Markers (Axl, Vimentin, N-Cadherin) and Loss of Epithelial Markers (E-Cadherin)

[0270] HMLER cells were transduced with retroviral vectors that express myrAkt1 or myrAkt3 as described in Example 8 and analyzed for Axl receptor protein, epithelial (E-cadherin) and mesenchymal (vimentin, N-cadherin) marker expression, Akt1/3 and pAkt levels as described in Example 7.

[0271] The results are shown in FIG. 10. Constitutively active Akt3, but not constitutively active Akt1 is able to activate EMT and to activate EMT regulators.

Example 11 Constitutively-Active Akt3 (MyrAkt3), but not Constitutively-Active Akt1 is Able to Induce EMT and CSC Traits in Breast Epithelial Cells

[0272] HMLER cells expressing myrAkt1, myrAkt3 or empty vector were grown in 2D (Left) and 3D Matrigel (Right), and visualized by phase contrast microscopy as described in Example 8.

[0273] Results shown in FIG. 11.

[0274] HMLER cells typically have epithelial morphology (sheets of rounded cells) when grown on tissue culture plastic (Left), and are not invasive when grown embedded in matrigel (Right). These data show that constitutively-active Akt3, but not constitutively-active Akt1 is able to induce EMT and CSC traits in transformed breast epithelial HMLER cells (fibroblastoid cell growth in 2D culture and invasive, stellate growth in 3D Matrigel).

Example 12 Cells Expressing Constitutively Active Akt3, but not Cells Expressing Constitutively Active Akt1 Show the Ability to Grow in Mammospheres

[0275] HMLER cells expressing myrAkt1, myrAkt3 were grown in mammosphere culture and quantified as described in Example 9.

[0276] The results are shown in FIG. 12.

[0277] Tumorsphere formation is a characteristic of cancer stem cells. HMLER cells are normally not able to form mammospheres, indicating a lack of cancer stem cell characteristics. Transfection with a vector encoding activated Akt3 (myrAkt3) but not with a vector encoding activated Akt1 (myrAkt1) confers the ability to form mammospheres.

Example 13 Cells Expressing Constitutively Active Akt3 Show a Much Higher Ability to Form Tumors than Cells Expressing Constitutively Active Akt1

[0278] HMLER cells transduced with retroviral vectors HMLER/vector, HMLER/myrAkt1 and HMLER/myrAkt3 as described in Example 10, were injected into host mice at limiting dilutions as described (Mani S A, Guo W, Liao M J, Eaton E N, Ayyanan A, Zhou A Y, Brooks M, Reinhard F, Zhang C C, Shipitsin M, Campbell L L, Polyak K, Brisken C, Yang J, Weinberg R A. The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell. 2008 May 16; 133(4):704-15).

[0279] Results presented in FIG. 13. These data show that expressing constitutively active Akt3 (HMLER/myrAkt3) significantly increases the HMLER cell ability to form tumors, compared to control cells or cells expressing constitutively active Akt1 (see number of tumors formed at 1000 cells seeded).

Example 14 Constitutively Active Akt3 and Slug, but not Constitutively-Active Akt1 is Able to Induce the Mesenchymal Phenotype and Invasive Growth; Akt Inhibitors Inhibit the Mesenchymal Phenotype

[0280] MCF10A cells were cultured and transduced with retroviral constructs expressing myrAkt1, MyrAkt3 or Slug as described in FIGS. 1 and 3. The cells were treated with Akt inhibitors LY294002 (10 M, Cell Signaling Technology, Cat. #9901) and Akt VIII (10 M, Merck, Cat. #124018) for 12 hours as indicated. Cells were then either visualized at indicated magnification by phase-contrast microscopy, or seeded for invasive growth in 3D Matrigel as described in Example 3.

[0281] Results shown in FIG. 14.

[0282] These results show that constitutively-active Akt3 and Slug, but not constitutively-active Akt1 is able to induce the mesenchymal phenotype in 2D growth and invasive growth in 3D growth matrigel. Akt inhibitors LY-294002 and AKT Viii inhibit the mesenchymal phenotype induced by Akt3, but also the mesenchymal/invasive phenotype induced by Slug in 2D and 3D growth, suggesting that Akt3 is required for Slug signalling.

Example 15 Constitutively-Active Akt3 and Slug, but not Constitutively-Active Akt1 is Able to Activate EMT and Akt Inhibitors are Able to Partially Reverse the Mesenchymal Phenotype

[0283] MCF10A cells were cultured and transduced with retroviral constructs expressing myrAkt1, MyrAkt3 or Slug as described in FIGS. 1 and 3. The cells were treated with Akt inhibitor as described in FIG. 6. SDS/PAGE, Immunoblotting and antibodies as described in Examples 1 and 4.

[0284] Results are shown in FIG. 15.

[0285] These results show that constitutively-active Akt3 and Slug, but not constitutively-active Akt1 is able to activate EMT as shown by up-regulation of the markers N-cadherin, vimentin and loss of the epithelial marker E-cadherin. The Akt inhibitor AKT VIII (Left) is able to fully inhibit Akt activation (pAkt) and partially reverse the mesenchymal phenotype, as shown by significant reduction of the mesenchymal markers N-cadherin and vimentin (Left). The Akt inhibitor LY-294002 similarly shows partial inhibition of EMT, showing reduced vimentin expression. Neither inhibitor led to re-expression of the epithelial marker E-cadherin.

Example 16 RNAi Silencing of Akt3, but not Silencing of Akt1 in MCF10A Cells Induced to Undergo EMT (by Expression of H-RasV12 or Slug) Significantly Reduces P-Akt Levels

[0286] MCF10A cells were transduced with retroviral vectors encoding EMT inducers Slug, and H-RasV12. From these cells small interfering RNA-mediated silencing was done using HiPerFect transfection reagent (Qiagen) according to the manufacturer's protocol and the cells were cultured for 3 days. Annealed siRNAs against Akt1 (Hs_AKT1-_5 Flexitube siRNA), Akt3 (Hs_AKT3_2 HP siRNA) and GAPDH (Hs_GAPDH_3 HP validated siRNA) (all from Qiagen) as a negative control were used at 60 nM final concentrations. After silencing, cells were lysed with SDS-PAGE loading buffer, sonicated and boiled. Lysates were subjected to Western blot analysis and blots were probed using Akt1, Akt3, pAkt Ser473 antibodies as described in FIG. 2 and -tubulin 12g10 (Hybridoma Bank; http://dshb.biology.uiowa.edu/12G10-anti-alpha-tubulin).

[0287] These results are presented in FIG. 16. These results show that knocking down the level of Akt3, but not knocking down the level of Akt1 in cells induced to undergo EMT is able to significantly reduce the level of total P-Akt. This suggests that phospho-Akt3 represents the major pAkt isoform in EMT-induced MCF10A cells.

Example 17 Constitutively Active Akt3, but not Constitutively Active Akt1 is Localized to the Cell Nucleus

[0288] HMLER cells were cultured as described in Example 2, and SDS/PAGE, Immunoblotting and antibodies as described in Example 1 except anti-histone 3 (Cell Signaling, Histone H3 Antibody #9715). Nuclear extraction was done according to manufacturer' instructions (Universal Magnetic Co-IP Kit, Active Motif, 54002). Immunofluorescence staining of constitutively active Akt1 and Akt3 of fixated cells were done using antibodies as in Example 1, by the method described by the manufacturer (Cell Signaling, Immunofluorescence General Protocol).

[0289] These results are presented in FIG. 17. Activated Akt3 (MyrAkt3) localizes to the nuclear fraction in HMLER cells (top). Immunofluorescence staining reveals nuclear/peri-nuclear staining for myrAkt3 but exclusion from the nucleus for myrAkt1.

Example 18 Akt3 and the Transcription Factor Snail are Found Overwhelmingly in the Nuclear Fraction of Cultured Triple-Negative Breast Cancer Cells. Tubulin (Cytoplasmic) and Lamin (Nuclear) Markers Confirm Successful Fractionation

[0290] MDA-MB 231 cells were cultured as described in Example 3. SDS/PAGE, Immunoblotting and antibodies as described in Examples 1 and 16, exept Laminin A/C from Santa Cruz, sc-7292. Cytosolic and nuclear proteins were isolated as described in Example 17.

[0291] The results of these experiments are shown in FIG. 18. As expected, the transcription factor Snail is found in the nucleus. Unexpectedly, Akt3 was also almost exclusively nuclear.

Example 19 Akt3 Localizes to the Nucleus in Cultured Triple Negative Breast Cancer Cells and Primary Human Mammary Epithelial Cells

[0292] Culturing of MDA-MB-231 cells was as described in Example 3. Primary human mammary epithelial cells (HMEC) were isolated as described (Garbe J C, Pepin F, Pelissier F A, Sputova K, Fridriksdottir A J, Guo D E, Villadsen R, Park M, Petersen O W, Borowsky A D, Stampfer M R, Labarge M A. Accumulation of multipotent progenitors with a basal differentiation bias during aging of human mammary epithelia. Cancer Res. 2012 Jul. 15; 72(14):3687-701) Localization of Akt3 (anti-Akt3-FITC, top panels) in MDA-MB-231 and primary human mammary epithelial cells (HMEC) as in Example 17. Nucleus was stained by DAPI (lower panels). Bar: 50 micron

[0293] The results of these experiments are shown in FIG. 19. Akt3 protein (top panels) is localized to the nuclei (compare to nuclear stain, bottom panels) in both breast cancer cells and primary mammary epithelial cells.

Example 20 Knocking Down Axl Kinase Significantly Reduces EMT Induced Nuclear Localization of Akt3

[0294] HMLER and HMLER/Slug cells were transduced retroviral vectors that express Axl-targeting shRNA (shAxl2) or control luciferase shRNA (shLuc). Cytoplasmic and nuclear cell fractions were isolated as described in Example 17, and Akt3 protein level were measured by lmmunoblotting in nuclear and cytoplasmic cell fractions.

[0295] Immunofluorescence of transduced cells (GFP, cytoplasmic green) stained with anti-Akt3-FITC (nuclear green) or DAPI nuclear stain.

[0296] Results from this experiment in FIG. 20. Induction of EMT by Slug leads to nuclear localization of Akt3 in an Axl-dependent process.

Example 21 Inhibition of Axl Kinase Activity Inhibits Nuclear Localization of Akt3

[0297] Quantification of nuclear Akt3 immunofluorescence (anti-Akt3-FITC, top panels) in mammary epithelial cells (HMEC) treated with vehicle (DMSO), cKit TKI (1 uM imatinib) and Axl TKI (600 nM BGB324). Bar 50 micron. *P<0.05.

[0298] Results presented in FIG. 21 show that blocking Axl activity (BGB324) inhibit Akt3 nuclear localization, while inhibiting cKit (Imatinib) has no effect on Akt3 nuclear localization.

Example 22 Akt1 and Akt3 Kinases are Able to Directly Phosphorylate Snail Protein in a Biochemical Assay Containing No Other Proteins

[0299] SNAI1/Snail coding sequence were cloned into the pGEX-4T-1 vector (Promega) and sequence verified. GST fusion protein were expressed in Escherichia coli (Rosetta BL21DE3) and purified according to the manufacturer's instructions (BD Biosciences); the GST moiety was cleaved by using thrombin. In vitro kinase assays were performed using recombinant Akt1 and Akt3 (ProQinase GmbH), Snail and Slug substrate proteins, and 32P-ATP and detected by autoradiography as described (Tuomi et al., 2009).

[0300] Results presented in FIG. 22 show that Akt1 and Akt3 kinases are able to directly phosphorylate Snail protein in a biochemical assay containing no other proteins.

Example 23 the SureFire Assay Detects Phospho-Akt3 in Insulin-Stimulated Melanoma Cells (WM115) that Express Akt3. No Signal is Detected in Breast and Prostate Cell Lines that Express Akt1 and Akt2 but not Akt3 (MCF7 and LNCaP)

[0301] A SureFire assay was used to specifically detect activated (phosphorylated) Akt3 in WM115, MCF7 and LNCaP cells. Briefly cells were lysated, and antibodies recognizing phosphorylated Akt (p473) and Akt3 were coupled to acceptor and donor beads as described by the manufacturer (PerkinElmer). Cells were either unstimulated or stimulated with 10 nM insulin for 10 minutes to activate Akt.

[0302] The results of these experiments are shown in FIG. 23. As expected, insulin was able to significantly induce Akt3 activity (phosphorylation) in WM115 melanoma cells, while no signal was observed in MCF7 breast epithelial or LNCaP prostate cells.

Example 24 Reduction of Expression of Akt3

[0303] By identifying suitable shRNA sequences, for example as described in US2008014037, it is possible to knock down expression of Akt3 to different levels (e.g. 20%, 40%, 60%, 70%, 80%, 90%, 100%). Attempts to induce EMT in mammalian, preferably human, cells with different levels of Akt3 knockdown can be used to define the minimum degree of Akt3 knockdown required to prevent EMT, thus identifying the therapeutic window. Furthermore, by selecting a level of Akt3 knockdown that is just insufficient to prevent EMT, it is possible to generate a mammalian, preferably human, cell line that is particularly sensitive to inhibitors of EMT. In such a cell line there is only just enough Akt3 expression to allow EMT to occur, and compounds which inhibit Akt3 signalling even slightly will block EMT. Such cell lines are thus useful screening tools for inhibitors of EMT, and especially inhibitors of Akt3.

INDUSTRIAL APPLICATION

[0304] The invention is industrially applicable through operation of methods in accordance with the invention.