English version

Équipe KIN

Kinases essentielles dans les carcinomes rénaux




Équipe labellisée La Ligue

Leader
 
Claude Cochet
iRTSV/BCI
17 rue des Martyrs
CEA Grenoble
38 054 Grenoble cedex 09
Tel. : 04 38 78 42 04
Fax : 04 38 78 50 58
 
Members of the team
 
 
Jean Boutonnat
MCU-PH Université Joseph Fourier
Odile Filhol-Cochet
Inserm Researcher
   
Florin Anne
Assistante Hospitalo-Universitaire
Florence Teillet
CEA
Post-doctoral fellow
   
Eve Duchemin-Pelletier
CNRS
Technician
Elodie Spreux
CEA
Technician
   
Caroline Roelants
Master 2
 
 
Presentation
 
Our team has a long-standing interest in protein kinase-mediated signaling pathways whose dysregulation participates in the development of all forms of cancers.
Ongoing studies are aimed at:

1) correlating the expression level of Protein-kinase CK2 in different cancers (prostate, breast, kidney) with unfavourable prognostic factors;
2) identifying new chemical CK2 inhibitors and evaluating their potential in human tumor xenograft models;
3) investigating the role of CK2 in the maintenance of epithelial cell morphology and polarity and intercellular adhesion;
4) probing the human kinome to identify kinases as potential clinical bio-markers and/or new drug targets in renal cell carcinomas.
 
Research activities
 
CK2 as a prognostic marker
 
There is a growing consensus that the outcome of patient diagnosed with cancer may be significantly improved by personalized medicine. New molecular definition of cancer is rapidly replacing traditional pathology-based disease descriptions. Growth and survival of cancer cells rely on the aberrant activation of several signaling pathways in which protein-kinases play a key role. Therefore, the implication of these enzymes in deregulated signaling pathways together with their susceptibility to small molecule inhibitors make them the first choice for a « Signal Transduction Therapy ».
CK2 is a multi-subunit protein kinase composed of two catalytic subunits (αα, α'α' or αα') and a dimer of regulatory ß subunits. (
Figure 1).



Figure 1. Three-dimensional structure of protein kinase CK2.
Production and purification of the recombinant CK2 holoenzyme.

In mice a functional loss of CK2ß is lethal, demonstrating that this subunit is essential for cell viability (Buchou, et al., 2003). In addition, we showed that targeted CK2ß-knockdown in embryonic neural stem cells compromises proliferation and oligodendrogenesis in the mouse telencephalon (Huillard et al., 2010). Live-cell fluorescence imaging studies provided evidence of independent and rapid movements of CK2α and CK2ß (Filhol et al., 2003) and this kinase appears as a mobile enzyme, which can be rapidly recruited to target specific proteins in response to different stimuli (Filhol and Cochet, 2009).
Both CK2α/α' catalytic subunits display constitutive activity, but CK2ß was shown to operate as a targeting subunit affecting the accessibility of binding substrates to the catalytic site.

Increasing evidence indicates that CK2 is involved in protein kinase networks controlling several aspects of transformation and cancer (
Figure 2).



Figure 2. Overexpression of CK2 creates a favorable environment for tumor development.

For a long time, evaluation of CK2 expression level by immunohistochemistry (IHC) in human cancers has been hampered due to the lack of suitable CK2 antibodies. Therefore, we developped a highly specific polyclonal antibody against CK2α that was for the first time used to determine the global levels of CK2 in a retrospective cohort of 131 prostate adenocarcinomas. This study
demonstrated a strong association between CK2α staining pattern and poor pronostic factors (Laramas et al., 2007).
It was reported that enforced CK2α expression in mouse mammary gland leads to tumor formation. Thus, we have measured by IHC relative protein levels of CK2 together with 32 other biomarkers using digitized microscopic analysis of 905 node-negative breast carcinoma specimens (Figure 3). Both univariate and multivariable models indicated that CK2α overexpression correlated with metastatic risk of breast carcinoma (Giusiano et al., 2011). Thus, CK2α can be included in a signature with prognostic value for predicting the risk of developing metastatic breast cancer that may complement the currently available methods to improve the evidence-based management of patients. As new drugs targeting CK2 are entering the clinic, these findings have important therapeutic implications.




Figure 3.
A - TMA cores of node negative breast carcinomas.
B - CK2α staining.
C - Disease-free survival.
D - CK2α is included in a signature of metastatic risk of breast carcinoma.
 
CK2 as a drug target
 
A new era of molecularly targeted agents begun when several chemical kinase inhibitors demonstrated powerful clinical activity in tumors in which the target kinase was deregulated ensuring a prominent position for chemical inhibition of protein kinases in pharmacology.
Several CK2 inhibitors have been developed showing variable efficiency, emphasizing the need to expand the chemical diversity of those inhibitors. To this end, we screened several chemical libraries on human recombinant CK2α catalytic subunit using an automated luminescence-based kinase assay. Several families of molecules with different scaffolds were identified as potent CK2 inhibitors. Among them, ellipticine derivatives were characterized as new ATP-competitive CK2 inhibitors (
Figure 4). These compounds can decrease tumorigenesis of U373 glioblastoma cells in mice. A rationale basis was provided for their mode of action based on the crystal structure of CK2α-inhibitor complexes (Prudent et al., 2010).



Figure 4.
This crystal structure shows how a derivative of ellipticine inhibits CK2. The planar structure of this chemical molecule allows it to engage in the catalytic site taking the place of the substrate of the enzyme. Amino acids of CK2 are in blue, and the derivative of ellipticine in green.

We also reported the identification of 2,8-difurandicarboxylic acid derivatives as a new class of nanomolar ATP-competitive inhibitors. Selectivity profiling pointed out Pim kinases as the only other kinases that were significantly inhibited. By combining structure-activity relationship analysis with structural determination, we were able to determine the binding mode of these inhibitors for both kinases and to explain their strong inhibitory potency. Given the similar but not redundant pathophysiological functions of CK2 and Pim family members, such inhibitors would provide new attractive leads for targeted cancer therapy.
Because most small-molecule kinase inhibitors target the highly conserved ATP-binding pocket of this enzyme family, resulting in unforeseen crossreactivities, there are compelling reasons to identify new classes of inhibitors that have substantially different mechanisms of action. Towards this goal, we took advantage of the peculiar molecular architecture of CK2 to discover small molecules targeting different surface areas of CK2 subunits (Prudent and Cochet, 2009). Using high-throughput and structure-based virtual screening approaches, we have identified different classes of small molecules targeting different surface area such as exosites on CK2α or at the CK2α/CK2ß interface (
Figure 5). Polyoxometalates which are inorganic complexes of early transition metal ions and oxo ligands represent nonclassical kinase inhibitors showing a remarkable inhibitory activity (Prudent et al., 2008). Some of the identified compounds showed good pharmacologic profile, causing inhibition of CK2 activity in cancer cells and demonstrating antitumor activity in mouse xenograft models. As drugs targeting CK2 are entering the clinic, our work lays the foundation for development of new clinically useful CK2 inhibitors.



Figure 5.
Polyoxometalates are inorganic compounds consisting of metals (molybdenum and / or tungsten) complexed to oxo ligands. They specifically inhibit CK2 in a unique mode of action. Amino acids whose mutation affects the inhibition of CK2 by POMs are in red.
 
Role of CK2 in the control of epithelial cell plasticity
 
Dissociation of cell/cell contacts is one of the first characteristics of Epithelial to Mesenchymal Transition (EMT). EMT programs constitute recognized mechanisms for dispersing cells in embryos, forming mesenchymal cells in injured tissues, and initiating the invasive and metastatic behaviour of epithelial cancers. Thus, EMT is an extreme form of cellular plasticity defined by loss of epithelial cell morphology, dissociation of cell-cell contacts, reduction in proteins mediating cell-cell contacts, remodelling of the actin. cytoskeleton, and acquisition of a mesenchymal cell shape. Multiple signalling pathways are involved in EMT triggered by different members of the TGFß superfamily. Several protein kinases are key regulators of the signalling pathways involved in TGFß-induced EMT.
Beside its recognized central role in the control of several physiological and pathological processes, CK2 may be involved in the maintenance of mammalian cell morphology and polarity and intercellular adhesion (Deshiere et al., 2008). Along this line, we showed recently that CK2ß silencing in human mammary epithelial cells triggered changes in their morphology correlated with the acquisition of mesenchymal phenotype that were reminiscent to TGFß-induced EMT. However, we showed that alternative signaling pathways were involved in the acquisition of this phenotype. Given the importance of CK2 in tumorigenesis, a dysregulation of CK2ß expression might contribute to EMT induction during cancer progression.



Figure 6. Epithelial cell plasticity.
A) In response to extracellular signals (eg TGFß), epithelial cells undergo epithelial-mesenchymal transition (EMT) characterized by loss of expression of membrane proteins and cytoskeleton and the induction of transcription factors and acquisition of mesenchymal specific markers.

B) A disturbance in the expression of CK2 subunits in human breast epithelial cells leads to phenotypic changes which are characteristic of EMT.
 
Predictive models for the identification of new therapeutic targets in renal carcinomas
 
Renal cell carcinoma accounts for 2-3% of all malignant diseases in adults. In Europe, more than 63,000 news cases of renal cell carcinoma (RCC) and 26,000 deaths were reported in 2006. It has been realized that this disease is not one entity but rather a collection of different types of tumours. Most common is the clear cell renal cell carcinoma that represent 70-75 % of all renal cancer cases. RCC show a deletion of the von Hippel-Lindau (VHL) tumor suppressor gene that is considered to be one of the primary events in the carcinogenesis. The fact that small tumors may have metastatic potential and that the overall RCC mortality has not yet dropped (Hollingsworth et al., 2006), highlights the need to identify markers for the individual tumor aggressiveness. Expression profiling with microarrays have shown that many of the renal tumour subtypes are genetically distinct entities. This genetic heterogeneity associated with the lack of efficacy of hormonal and chemotherapy in treating renal tumours highlights the need:
1) to develop new predictive models of the response or non-response of patients to treatments,
2) to identify new therapeutic targets.

Consequently, our goals are to set-up and compare two 3D culture systems that preserve the original microenvironment thereby mimicking the tumour heterogeneity.

Multicellular tumour spheroids
Mixed spheroids of renal cancer cells and endothelial cells are routinely generated by centrifugation methods. While growing, spheroids display a gradient of proliferating cells located in the outer cell-layers whilst the quiescent cells are located more centrally. This type of heterogeneity is typically observed in neo-vascularised renal carcinomas. When the central cells become deprived of oxygen and glucose, important cellular effects take place such as: HIF-1α activation and acidification, increased glycolysis, autophagie, cell death or necrosis. Classical IHC on frozen sections are performed to assess the activation of proliferation and survival signalling pathways.

Tissue slice technique
The special advantage of ex vivo tissue slices is the ability to both maintain organ and cellular architecture, while preserving the integrity of the tumor-stroma interaction. This method represents a means by which pathways activation and pharmacological inhibition can be rapidly studied with respect to the native heterogeneity of a patient's tumor. It is therefore well suited to investigate resistance to therapies. Malignant human renal tissue are obtained from patients undergoing surgery at the Grenoble hospital (Urology Department). After harvesting from surgery, the tumors are rapidly sliced (300-500 mm) on a Vibratome VT1200 (Leica) and tissue slices are then cultured on organotypic inserts (Millicell-CM membrane; Millipore) for 4-5 days. Thereafter, slices are formalin-fixed and paraffin-embedded for morphologic and IHC evaluation of proliferation and survival signalling pathways.


Identifying essential protein kinases by RNAi screens
Commercial panels of synthetic short duplex RNA (siRNAi) and vector-encoded short hairpin RNA (shRNA) directed to all human kinases are available in the laboratory. Initially, our study will be focused on the identification of kinase requirements for renal carcinoma cells grown as spheroids. The tissue slice technique will be particularly well suited to compare the effects of kinase inhibitors.
It is expected that kinases identified in renal carcinomas will represent bona fide targets for the development of new anti-cancer drugs.



Figure 7. Strategies for identifying drug targets and new bioamarkers in renal carcinomas.