CARSTEN HOPF ET AL: "Tyrosine phosphorylation of the muscle-specific kinase is exclusively induced by acetylcholine receptor-aggregating agrin fragments", EUROPEAN JOURNAL OF BIOCHEMISTRY, vol. 253, no. 2, 15 April 1998 (1998-04-15), pages 382 - 389, XP055173987, ISSN: 0014-2956, DOI: 10.1046/j.1432-1327.1998.2530382.x
LABOME: "Tags peptidiques/protéiques", 19 March 2012 (2012-03-19), XP055174038, Retrieved from the Internet
EUNICE L KWAK ET AL: "n Anaplastic Lymphoma Kinase Inhibition in Non-Small-Cell Lung Cancer From the Massachusetts General Hospi- tal Cancer Center ( Background", N ENGL J MED, 1 January 2010 (2010-01-01), pages 1693 - 703, XP055174180, Retrieved from the Internet
LIN W H ET AL: "A cell-based high-throughput screen for epidermal growth factor receptor pathway inhibitors", ANALYTICAL BIOCHEMISTRY, ACADEMIC PRESS INC, NEW YORK, vol. 377, no. 1, 1 June 2008 (2008-06-01), pages 89 - 94, XP022621095, ISSN: 0003-2697, [retrieved on 20080304], DOI: 10.1016/J.AB.2008.02.027
CHESI M; NARDINI E; BRENTS LA; SCHROCK E; RIED T; KUEHL WM; BERGSAGEL PL., NAT GENET, vol. 16, 1997, pages 260 - 4
DAILEY L; LAPLANTINE E; PRIORE R; BASILICO C., J CELL BIOL., vol. 161, 2003, pages 1053 - 66
GUAGNANO V; KAUFFMANN A; WÖHRLE S; STAMM C; ITO M ET AL., CANCER.DISCOV., vol. 2, 2012, pages 1118 - 33
KANG S; DONG S; GU TL; GUO A; COHEN MS ET AL., CANCER CELL., vol. 12, 2007, pages 201 - 14
KREJCI P; MASRI B; SALAZAR L; FARRINGTON-ROCK C; PRATS H ET AL., J BIOL CHEM., vol. 282, 2007, pages 2929 - 36
KREJCI P; SALAZAR L; GOODRIDGE HS; KAS IWADA TA; SCHIBLER MJ ET AL., J CELL SCI., vol. 121, 2008, pages 2 2 - 81
KREJCI P; PROCHAZKOVA J; SMUTNY J; CHLEBOVA K; LIN P. ET AL., BONE, vol. 47, 2010, pages 102 - 10
KREJCI P; AKLIAN A; KAUCKA M; SEVCIKOVA E; PROCHAZKOVA J ET AL., PLOS ONE., vol. 7, 2012, pages E35826
KIM HR; KIM DJ; KANG DR; LEE JG; LIM SM ET AL., J CLIN ONCOL, vol. 31, no. 6, 2013, pages 731 - 7
LAEDERICH MB; DEGNIN CR; LUNSTRUM GP; HOLDEN P; HORTON WA., J BIOL CHEM., vol. 286, 2011, pages 19597 - 604
LAX I; WONG A; LAMOTHE B; LEE A; FROST A; HAWES J; SCHLESSINGER J., MOL CELL, vol. 10, 2002, pages 709 - 19
LI C; SUN Y; FANG R; HAN X; LUO X ET AL., J THORAC ONCOL., vol. 7, 2012, pages 85 - 9
LIU Z; HOU P; JI MH; STUDEMAN K; JENSEN K ET AL., J CLIN. ENDOCR. METAB., vol. 93, 2008, pages 3106 - 16
OKABE T; OKAMOTO I; TSUKIOKA S; UCHIDA J; HATASHITA E ET AL., CLIN CANCER RES., vol. 15, 2009, pages 907 - 13
PASSOS-BUENO MR; WILCOX WR; JABS EW; SERTIÉ AL; ALONSO LG; KITOH H., HUM MUTAT., vol. 14, 1999, pages 115 - 25
RONCHETTI D; GRECO A; COMPASSO S; COLOMBO G; DELL'ERA P ET AL., ONCOGENE, vol. 20, 2001, pages 3553 - 62
CLAIMS 1. A method for in-cell activity profiling of human receptor tyrosine kinases (RTK), characterized in that it comprises the following steps: - transferring plasmid vectors carrying wild-type human C-terminaily V5/His-tagged RTKs or their disease-associated mutants into cell cultivation containers, - reconstituting the plasmid DNA and adding a transfection reagent, - adding cells to be transfected, - growing transfected cells to enable the RTK expression and spontaneous, ligand-independent dimerization and activation when in overexpressed state, - exposing the transfected cells to a compound to be tested and/or optionally to a cognate RTK ligand, - performing a detection method selected from: - fixing the cells and in situ immunocytochemical detection of active, phosphorylated forms of RTKs and their mutants with standardization of the expression of all RTKs using a single immunocytochemistry for V5 epitope, or, - cell lysis followed by WB with phosphorylation stage-dependent antibodies to detect the levels of RTK activation, followed by WB with V5 antibody, to determine and compare the levels of expression of all RTKs and their mutants using a single V5 epitope, or, - cell lysis followed by dual luciferase detection of firefly luciferase driven by Dusp6 gene promoter and Renilla luciferase driven by constant promoter. 2. The method according to claim 1, wherein for transfection, 293T, CHO or Cos7 cells are used. 3. The method according to any one of the preceding claims, wherein the reaction containers are 12, 24 or 96-weIl plates, tissue culture treated. 4. The method according to any one of the preceding claims, wherein the plasmids are bacterial plasmids containing pCMV promoter in front of their insert cDNA and an ampicillin resistance gene. 5. The method according to any one of the preceding claims, wherein the RTKs and/or mutated RTKs are selected from a group comprising the wild-type human RTKs and their disease-associated mutants shown in Table 1: 6. The method according to any one of the preceding claims, wherein the RTK and its disease- associated mutants are used which are listed in one line of the Table 1 shown in Claim 5. 7. The method according to any one of the claims 1 to 5, wherein all RTKs and their disease- associated mutants are used which are contained in the Table 1 shown in Claim 5. 8. The method according to any one of the preceding claims, wherein the transfection reagent is a cationic lipid-based transfection reagent. 9. The method according to any one of the preceding claims, wherein the RTK-carrying plasmids are produced by cloning of the RTK cDNAs into a plasmid backbone, N-terminally relative to the V5/6xHis tag. |
Field of Art
The present invention relates to a novel method of in-cell activity profiling of human receptor tyrosine kinases and their disease-associated mutants.
Background Art
Maintenance of tissue homeostasis depends on complex intercellular signalling networks that govern basic cell functions. Receptor tyrosine kinases (RTK) represent major molecular toolkit by which cells sense their extracellular environment and respond to communication signals. In humans, at least 54 different RTKs exist and belong to 18 distinct families. RTKs respond to more than hundred different signals delivered by growth factors, hormones, cytokines, components of extracellular matrix and other ligands.
Since discovery of the first RTK more than 25 years ago, intensive research illuminated many important aspects of the RTK function. We hold a detailed knowledge of tissue manifestations of both physiological and pathological RTK function. For most RTKs, we also have more or less complete understanding of molecular mechanism of ligand-mediated activation of the tyrosine kinase function. In contrast, the intracellular mechanisms of RTK signal transduction remain considerably less characterized to date, with major questions still awaiting their answers. To date, we have very limited knowledge of the intracellular signaling patterns specific to each of the 54 human RTKs, yet this area is critical to our understanding of RTK functions. It is estimated that only 40% of RTK signal transduction have been uncovered in past three decades of RTK research; 60% of knowledge still awaits the discovery. In order to accomplish the abovementioned, the research tools enabling systematic studies involving entire RTK component of the human genome must be developed first. This invention describes one such tool.
The importance of RTK signaling is further emphasiEed by evidence of their pathological functions. More than 80 human pathologies associate with mutations, gene amplifications or gene translocations involving RTK genes, including cancer, developmental defects, metabolic disorders, and bone and skin defects. Their pathological roles make RTKs attractive therapeutic targets; the development of drugs targeting RTK activity constitutes a major part of the current pharmacology. Many drugs target more than one RTK at the same time, necessitating the need for precise determination of their target specificity.
At present, direct tyrosine kinase inhibitors (TKI) targeting the ATP binding site on the kinases constitute the prevailing class of drugs under development for cancers driven by RTK mutations. Although a direct targeting of tyrosine kinase activity by TKIs generates impressive results in some cancers, many problems persist which compromise the successful therapy, such as poor TKI target specificity or development of secondary tumor resistance to TKI. It is now clear that novel avenues of RTK inhibition must be devised including targeting of downstream signaling components rather than the given RTK itself. Signaling proteins do not act in isolation but are organized in large, multiprotein complexes which dynamically change their spatio-temporal organization. Targeting the protein-protein interaction within the signaling complexes might represent one way to find novel opportunities to treat cancer caused by aberrant RTK signaling.
In order to find novel therapeutic strategies to target RTK signaling, we need methodologies allowing for comparative, in-cell profiling of human RTKs and their disease-associated mutants. Existing technologies allowing multiple (i.e. 50+) RTK activity profiling are invariantly based on in vitro (cell-free) kinase assays, do not contain many disease-associated mutants, and are expensive. Although many different RTKs might be profiled at the same time, the major disadvantage of such approach is that only direct RTK inhibitors may be evaluated. Compared to in vitro kinase assays, the major advantage of in-cell profiling is that it allows to determine RTK activity in intact cell environment, complete with regulatory networks and feedback control mechanisms. This is especially important when compounds targeting RTK interaction with its downstream signaling mediators are evaluated. Unfortunately, the existing technologies of in-cell RTK profiling are labor intensive, expensive and limited to profiling of only few RTKs (<20) at the same time.
This invention introduces a method that overcomes the abovementioned limitations of both existing in vitro and in-cell RTK profiling approaches; it allows for rapid, efficient and cost effective in-cell activity profiling, in one particular embodiment of 36 different human RTKs and 203 of their disease-associated mutants.
Disclosure of the Invention The present invention provides a method for in-cell activity profiling of human receptor tyrosine kinases (RTK), which comprises the following steps:
- transferring plasmid vectors carrying either wild-type C-terminally V5/His-tagged RTKs or their disease-associated mutants into cell cultivation containers,
- adding a transfection reagent,
- adding cells to be transfected,
- growing transfected cells to enable the RTK expression and spontaneous, ligand-independent dimerization and activation when in overexpressed state,
- exposing the transfected cells to a compound to be tested and/or optionally to a cognate RTK ligand (for RTKs that do not activate spontaneously when overexpressed),
- performing a detection method selected from:
- fixing the cells and in situ immunocytochemical detection of active, phosphorylated forms of RTKs and their mutants with standardization of the expression of all RTKs using a single immunocytochemistry for the V5 epitope,
or,
- cell lysis followed by western immunoblotting with phosphorylation stage-dependent antibodies to detect the levels of RTK activation, followed by western immunoblotting with V5 antibody, to determine and compare the levels of expression of all RTKs and their mutants using a single protein epitope, i.e. the V5 epitope,
or,
- transfection of A) reporter vector containing human Dusp6 gene promoter region driving expression of firefly luciferase responding to Erk pathway activation by most of the RTKs, and B) pTK-RL or similar internal standard vector containing Renilla luciferase under the control of constitutively-active promoter, followed by cell lysis and detection of firefly and Renilla luciferase activity via dual-luciferase activity assay. For transfection, any established or primary cell type permitting piasmid transfection via cationic lipid methods may be used. Preferred for the transfection are easily transfectable cells that produce a lot of transgenic protein from a given amount of transfected plasmid, such as 293T, CHO or Cos7 cells.
Reaction containers are preferably wells, e.g. standard 12, 24, or 96-weIl plates, tissue culture treated.
In a preferred embodiment, the plasmids are bacterial plasmids, more preferably containing pCMV promoter in front of their insert cDNA and an ampicillin resistance gene, such as pcDNA or pRK7 plasmids. The V5 epitope allows for quantification of the all RTKs expression via only one commercially available antibody, the 6xHis tag allows for easy purification of transgenic proteins from cell lysates using the nickel column. Preferably, one type of plasmid is used for all RTKs and their disease-associated mutants.
The RTKs and/or mutated RTKs are selected from a group comprising the wild-type human RTKs and their disease-associated mutants shown in Table 1. In one embodiment, the given RTK and its disease-associated mutants are used, which are selected from RTKs shown in Table 1, wherein each line of the Table 1 represents one RTK and its disease-associated mutants. In a preferred embodiment, one line in Table 1 represents one vector with cloned RTK and its mutants. Numbers indicate the mutation position relative to the reference protein sequence for the given RTK in Human Protein Reference Database (www.hprd.org) database, letters indicate the amino acid substitution. In another embodiment, a library of RTKs and disease-associated mutants containing all RTKs and disease-associated mutants of RTKs contained in Table 1 is used. Table 1 A list of RTKs and their disease associated mutants (numbers represent amino acid substitution in given RTK based on sequence numbering available for given RTK in the Human Protein Reference Database (www.hprd.org). Kinase-dead mutants serve as negative controls for RTK activation.
The RT -carrying plasmids can be produced by cloning of the RTK cDNAs into a plasmid backbone, N-terminally relative to the V5/6xHis tag. The mutations listed in Table 1 can be introduced into the RTK sequences via site-directed mutagenesis. The transfection reagent is any cationic lipid-based transfection reagent capable of transfecting the cells used for RTK profiling (established cell lines or primary cells), preferably FuGENE6 reagent (Promega) when the Cos7, 293T or CHO cells are used. Other transfection reagents which are non-toxic for the cell line used for transfection may also be used. This allows for high transfection efficiency due to prolonged time the reagent may be left with ceils.
This invention covers a novel method enabling a rapid and efficient in-cell activity profiling of different human RTKs and hundreds of their disease-associated mutants. This is a major advance over the current technologies, which typically measure activity of one or few RTKs at the time. Another advantage of the proposed technology lies in its cell-based nature. Compared to in vitro kinase assays which are a standard approach for determining RTK activity, the cell-based profiling allows to determine RTK activity in intact cell environment, complete with regulatory networks and feedback control mechanisms. This is especially important when compounds targeting RTK interaction with its downstream signaling mediators are evaluated for their anti-RTK activity. Additionally, the proposed method is much faster and less requiring as to the skill and equipment than the methods known in the art.
In one preferred embodiment, the method of the present invention is as follows: commercially available 6, 12, 24 or 96-well tissue plates containing each RTK-expressing bacterial vector deposited in multiple wells are used for in situ transfection of 293T cells via common methods utilizing cationic lipid transfection. 24 hours after transfection, the cells are ready for kinase activity profiling, accomplished by either WB, ICC or dual-Iuciferase assay. When the application is drug screening (Example 4), the cells are exposed to the tested compounds for 48-72 hours. The comparison of given RTK's activity in the presence or absence of the tested compound identifies which RTKs and their mutants are efficiently targeted. This may be confirmed in subsequent experiments utilizing only RTKs which were targeted in the initial activity profiling. In studies addressing RTK signaling complexes, RTK dimerization or RTK transact! vation, the RTK expressing cells are stimulated with commercially available ligands, cognate to the particular RTK, for desired periods of time before harvesting.
In another preferred embodiment, the method of the present invention is performed as follows: 1. Vectors (pcDNA3.1.) carrying C-terminally V5/His-tagged RTKs (Table 1) are first plated into the clear tissue culture-treated 96-well plates, using an automated dispenser at 0.5-1 g of plasmid DNA/well, 4 wells for each plasmid. Plates with plasmid DNA are used immediately, or the DNA is vacuum dried for long-term storage at -20 °C.
2. Plasmid DNA is mixed with 25μ1 of DMEM media without FBS or antibiotics for 30 minutes. Media (25μ1) containing the transfection reagent FuGENE6 are mixed with plasmid
DNA in 3.2/1 ratio (μΐ FuGENE6^g plasmid DNA, when FuGENE6 transfection reagent is used) in wells and incubated for 20 minutes at room temperature.
3. 293T cells, NIH3T3, RCS or other established cell models (in 50μ1 complete DMEM media containing 20% FBS and antibiotics) are added into the DNA/FuGENE6 mixture at 2-3xl0 4 cells/well. For proper transfection and transgenic RTK expression, cells are incubated for 24 hours without media change. After 24 hours, media are replaced with ΙΟΟμΙ of fresh media containing desired stimulants (growth factors, inhibitor compounds). Cells are left growing for 24-48 hours and analyzed.
4a. RTK activity profiling by B: Cultivation media are removed and cells are directly lysed with 2x Laemmli buffer (65.8 mM Tris-HCl, pH 6.8, 2.1% SDS, 26.3% (w/v) glycerol, 0.01% bromophenol blue). Plates are incubated on waterbath heated to 95°C for 10 minutes. The samples are then used for WB or stored frozen at -20°C. Standard protocol is used for WB detection of RTK expression and activation. Briefly, protein samples are resolved by SDS- PAGE, transferred onto a PVDF membrane and visualized by chemiluminiscence. The following commercially available antibodies can be used to detect active and total forms of profiled RTKs: V5, 4G10, pVEGFR2 Y1 175 , pALK Y1096 , pAxl Y702 , pcKit Y703 , pDDRl™ 2 , pEGFR™ 68 , pFGFR Y653 654 , pFLT3 Y842 , pHER2/ErbB 2 Y877 , pIGF-IR YU35 , pIGF- IRpY m5 1136 , InsRp Yl l50m isl , pIGF-IRp Y98 °, pM-CSFR Y699 , P M-CSFR Y723 , pMet Y1349 , pPDGFRa Y762 , pPDGFRp Y751 , P Ret Y905 , pTie2 Y992 , pTrkA Y674/675 , pTrkB Y706 ™ 7 .
4b. RTK activity profiling by ICC: Following the cultivation period, cells are fixed with 4% paraformaldehyde for 10 minutes, permeabilized with 0.25% Triton X-100, and blocked with 5% BSA for 1 hour at room temperature. ICC with V5 antibody will be used to localize RTKs within the cell, exploiting the fact that all cloned RTKs and their mutants (Table 1) carry C- terminal V5 epitope. Ceils are labeled with commercially available V5 monoclonal antibody in 1% BSA, followed by incubation with commercially available AlexaFluor594-conjugated secondary antibody. RTK activation is visualized by commercially available 4G10 antibody conjugated to FITC. The overlap of the green and red channel fluorescence indicate the presence of the active RTK in cells. ICC pictures are recorded and supervised high content image analysis is carried out to determine the effect of tested compound on RTK activity as well as other parameters, such as cell proliferation, apoptosis and changes in cellular morphology.
4c. RTK activity profiling using dual-luciferase assay: For dual -luciferase assay, RTKs are co- transfected to cells (step 1 above) with vector containing firefly luciferase driven by Dusp6 gene promoter and vector containing Renilla luciferase driven by constant promoter, in 3:1 (w/w) ratio. 24 hours after transfection, cells are harvested and the signal for both luciferases is measured by commercial dual-luciferase activity assay. Intensity of firefly luciferase monitors the level of RTK activation, Renilla luciferase serves as internal standard for transfection efficiency.
4a, 4b and 4c are alternatives.
The present invention provides numerous advantages over the prior art methods. The following table presents an overview of the advantages and drawbacks of the profiling methods known in the art and the method of the present invention:
Method Pros Cons
Existing in vitro RTK - rapid - only direct inhibitors may be tested profiling methods - hundreds of RTKs may be - high costs limiting the amounts of repeated profiled experiments
-limited amount of RTK disease-associated mutants
- not suitable for many diverse research applications Method Pros Cons
Existing in cell RTK - direct and indirect inhibitors may - only few RTKs (<20) may be profiled profiling methods be tested - no disease-associated RTK mutants may be
- suitable for many diverse research profiled
applications - relatively time consuming
- low cost compared to in vitro
profiling, many repeated
experiment may be carried-out
Our method - rapid
- hundreds of RT s may be
profiled
- disease-associated RTK mutants
may be profiled
- direct and indirect inhibitors may
be tested
- suitable for many diverse research
applications
- low cost compared to in vitro
profiling, many repeated
experiment may be carried-out
Brief description of Drawings
Figure 1 Activation of ERK MAP kinase by TRKA and EGFR RTKs: RCS cells were transfected with wild-type, V5-tagged TRKA or EGFR, treated with appropriate commercially available recombinant ligands (EGF, NGF) for one hour and analyzed for ERK MAP kinase activation by WB (P-Erk). WBs for total Erk and Actin serve as loading controls. TRKA and EGFR expression was monitored by V5 WB.
Figure 2 Activation of downstream signaling by FGFR2, MET, EGFR and their disease- associated mutants. Experiments were carried-out as described in Example 4: V5-tagged FGFR2, MET and EGFR were expressed in 293T cells and analyzed for activation of downstream signaling molecules by WB, using commercially available antibodies recognizing given molecules in their active, phosphorylated (P) state. FGFR2 autophosphorylation is also indicated (P-FGFR). Total amounts of given signaling intermediators are also shown. Actin serves as loading control. Control, untransfected cells; empty, cells transfected with empty plasmid. Note the differences in magnitude of pathway activation among different RTK mutants. For MET, two independent plasmid clones are shown to test the reproducibility of the results. Note the marked differences in levels of activation of given signaling intermediate mediated by different RTK mutants. Comparison of such differences among all RTKs and their activating mutants represent a first step in identification of signaling patterns specific to each of the RTK. These patterns are poorly known at present, and their identification hold a key to our understanding of the RTK role in physiology and disease. Also note the relative differences in downstream signaling activation.
Figure 3 Ligand-independent activation among FGFR family RTKs. Experiments were carried-out as described in Example 5: 293T cells were transfected with vectors carrying C- terminally V5-tagged wild-type (wt) FGFRl -4 or their activating mutants, and the levels of FGFR activation were determined by WB with antibody detecting FGFRs phosphorylated (p) at Y653/Y654. The levels of total FGFRs (WB with V5 antibody) and actin serve as loading controls. Cells transfected with empty vector or untransfected cells are also shown. Note the spontaneous, ligand-independent activation which take place in wt FGFRl and FGFR2 in contrast to FGFR3 and FGFR4 which are activated only when mutated. Determination of total FGFR 1-4 expression with one V5 antibody allows for relative estimation of differences in FGFR expression as well as for comparison of differences in FGFR activation. This is not achievable if four different FGFR-specific antibodies were used for detection of total levels of FGFR 1-4 expression.
Figure 4 Effect of small chemical targeting FGFR ATP-binding site on activity of FGFRl -4 and their activating mutants. The experiments were carried-out as described in Example 6: (A- F) FGFRs were expressed in 293T cells, treated with BGJ398 for 24 hours, and the levels of FGFR and ERK activating phosphorylation (p) were determined by WB with commercially available specific antibody. Total FGFR and ERK levels serve as loading control. Note the complete suppression of wt and mutant FGFRl and FGFR2 activation by low levels of BGJ398 (A-D). In contrast, suppression of FGFR3-K650E activation required high levels of BGJ398 (E); the activity of FGFR4-V550E was insensitive to even high levels of BGJ398 (F). A detailed analysis of the BGJ398 against all clinically relevant mutations in FGFRl-4 will take just two weeks to complete, at little cost. If the ICC is used to detect RTK activity, the whole analysis can be carried-out in few days, with minimal expenses compared to current methods of RTK profiling. Examples
Example 1 : Generation of RTK plasmid library
To obtain the uniform levels of expression and negate the effects of different plasmid backbones, all RTKs were cloned into one type of plasmid. Since quantitative measurements of signal transduction parameters are one of the main applications of the RTK library, all RTKs were equipped with single protein tag so that WB with single antibody may be used to achieve comparable levels of expression of all RTK. All RTKs were cloned into the commercially available Invitrogen's pcDNA3.1-V5/6xHis plasmid backbone with C-terminal V5-tag, sequenced and protein expression verified using Sanger sequencing methods and WB with commercial antibodies. To every RTK cloned so far we also prepared, via site-directed mutagenesis, major mutants associated with human inherited diseases and cancer, selected from Sanger Cosmic (http://cancer.sanger.ac.uk /cancergenome/projects/cosmic/) and OMIM databases (http://www.ncbi.nlm.nih.gov/pubmed) (Table 1). Because kinase-inactive RTK forms often associate more firmly with their substrates allowing for easier purification of the RTK-associated signaling complexes (as described in Example 1), we also prepared 'kinase- dead' mutants for some of the cloned RTKs.
Example 2: Method description
Commercially available 6, 12, 24 or 96-well tissue plates containing each RTK-expressing bacterial vector deposited in multiple wells are used for in situ transfection of 293T cells via common methods utilizing cationic lipid transfection. 24 hours after transfection, the cells are ready for kinase activity profiling, accomplished by either WB or ICC. When the application is drug screening (Example 4), the cells are exposed to the tested compounds for 48-72 hours. The comparison of given RTK's activity in the presence or absence of the tested compound identifies which RTKs and their mutants are efficiently targeted. This may be confirmed in subsequent experiments utilizing only RTKs which were targeted in the initial activity profiling. In studies addressing RTK signaling complexes, RTK dimerization or RTK transactivation, the RTK expressing cells are stimulated with commercially available ligands, cognate to the particular RTK, for desired periods of time before harvesting. Detailed experimental protocol is as following:
1. Vectors (pcDNA3.1.) carrying C-terminally V5/His-tagged RTKs (Table 1) are first plated into the clear tissue culture-treated 96-well plates, using an automated dispenser at 0.5-^g of plasmid DNA/well, 4 wells for each plasmid. Plates with plasmid DNA are used immediately, or the DNA is vacuum dried for long-term storage at -20 °C.
2. Plasmid DNA is in freshly added 25μ1 complete DMEM media without FBS or antibiotics for 30 minutes. Media (25μ1) containing the transfection reagent FuGENE6 are mixed with plasmid DNA in 3.2/1 ratio (μΐ plasmid DNA, when FuGENE6 transfection reagent is used) in wells and incubated for 20 minutes at room temperature.
3. 293T cells, NIH3T3, RCS or other established cell models (in 50μ1 complete DMEM media containing 20% FBS and antibiotics) are added into the DNA/FuGENE6 mixture at 2-3x10 4 cells/well. For proper transfection and transgenic RTK expression, cells are incubated for 24 hours without media change. After 24 hours, media are replaced with ΙΟΟμΙ of fresh media containing desired stimulants (growth factors, inhibitor compounds). Cells are left growing for 24-48 hours and analyzed.
4a. RTK activity profiling by WB: Cultivation media are removed and cells are directly lysed with 2x Laemmli buffer (65.8 mM Tris-HCl, pH 6.8, 2.1% SDS, 26.3% (w/v) glycerol, 0.01% bromophenol blue). Plates are incubated on waterbath heated to 95°C for 10 minutes. The samples are then used for WB or stored frozen at -20°C Standard protocol is used for WB detection of RTK expression and activation. Briefly, protein samples are resolved by SDS- PAGE, transferred onto a PVDF membrane and visualized by chemiluminiscence. The following commercially available antibodies can be used to detect active and total forms of profiled RTKs: V5 5 4G10, pVEGFR2 Y1175 , pALK YI0% , pAxI Y702 , pcKit Y703 , pDDRl Y792 5 pEGFR Y,068 J pFGFR Y653 654 , pFLT3 Y842 } pHER2 ErbB2 Y877 , pIGF-IRp Y1 135 , pIGF- ικβγ π35/. ΐ36 5 Ιη5Κβ ΥΠ50/ΥΠ5. ? p i G F-IRp Y98 °, P M-CSFR Y6 ", pM-CSFR Y723 , pMet Yf 349 , pPDGFRa Y762 , pPDGFRp Y751 , pRet Y90S , pTie2 Y992 3 pTrkA 4 ' 675 , pTrkB Y706 707 .
4b. RTK activity profiling by ICC: Following the cultivation period, cells are fixed with 4% paraformaldehyde for 10 minutes, permeabilized with 0.25% Triton X-100, and blocked with 5% BSA for 1 hour at room temperature. ICC with V5 antibody will be used to localize RTKs within the cell, exploiting the fact that all cloned RTKs and their mutants (Table 1) carry C- terminal V5 epitope. Cells are labeled with commercially available V5 monoclonal antibody in 1% BSA, followed by incubation with commercially available AlexaFluor594-conjugated secondary antibody. RTK activation is visualized by commercially available 4G10 antibody conjugated to FITC. The overlap of the green and red channel fluorescence indicate the presence of the active RTK in cells. ICC pictures are recorded and supervised high content image analysis is carried out to determine the effect of tested compound on RTK activity as well as other parameters, such as cell proliferation, apoptosis and changes in cellular morphology.
4c. RTK activity profiling using dual-luciferase assay: For dual-luciferase assay, RTKs are co- transfected to cells (step 1 above) with vector containing firefly luciferase driven by Dusp6 gene promoter and vector containing Renilla luciferase driven by constant promoter, in 3:1 (w/w) ratio. 24 hours after transfection, ceils are harvested and the signal for both Iuciferases is measured by commercial dual-luciferase activity assay. Intensity of firefly luciferase monitors the level of RTK activation, Renilla luciferase serves as internal standard for transfection efficiency.
Example 3. Identification of proteins interacting with RTKs via proteomics
Protein complexes interacting with active RTKs represent one key area of the signal transduction, yet this area remains very little characterized for most human RTKs. Eight large proteomic analyses were carried-out in 293T cells transfected with V5-tagged wild-type (wt)- FGFR3 RTK or its activating mutant K560E-FGFR3 (associated with skeletal dysplasia and cancer) (Ronchetti et al. s 2001; Passos-Bueno et al., 1999). FGFR3 protein complexes were isolated by V5 immunoprecipitation (IP). Samples were analyzed by mass spectrometry (MS) to identify proteins specifically associated with FGFR3 or its signaling complexes (those that were not present in negative controls for each experiment). We considered authentic only those interactors which were found in the majority of MS experiments carried-out. A total of 19 proteins were identified. Two of these proteins are known to interact with FGFR3 (RSK, HSP90) (Laederich et al., 2011; Kang et al., 2007). Interaction of 17 other proteins with FGFR3 has not yet been reported in the scientific literature, including SHOX2 transcriptional factor, serine/threonine kinases MAK and ICK, and tyrosine kinases YES, FYN, FGR, LYN, LCK, BLK and HCK)(Table 2). Next, phosphoprotemics approach was used to search for proteins that are tyrosine phosphorylated by FGFR3 but their interaction is transient in time, making them difficult to isolate by FGFR3 pull-down. Five phosphoproteomic experiments were carried-out in 293T cells expressing wt-FGFR3 activated by commercially available FGF ligand (FGF2), with tyrosine phosphorylated proteins purified by commercially available anti- phosphotyrosine (p-Y) antibody (4G10). With exception of SHOX2, we identified the same proteins as found in V5-FGFR3 immunocomplexes. In addition, two signaling adapters known to interact with FGFR3 (FRS2, GABl)(Lax et al. 2002) and one novel adapter (p OCAS) were found (Table 2). Next, we used RCS cells that represent the major cellular model to pathological FGFR signaling (Dailey et al., 2003). Endogenous FGFR3 was activated by addition of commercially available recombinant FGF (FGF2) ligand, tyrosine phosphorylated proteins were isolated by IP with commercially available 4G10 antibody, and subjected to MS analysis. This way, we again identified MAK, ICK, LCK, LYN, FYN, FGR, HCK and pl30CAS as being phosphorylated upon FGFR3 activation.
We also completed experiments aimed on confirmation of the interactions found in MS experiments via method independent of MS, i.e. via mutual co-immunoprecipitations (co-IP) of both partners expressed in 293T cells, using previously published protocols (Krejci et al., 2007). Using this method, we confirmed FGFR3 interactions with MAK, SHOX2, BLK, FYN, LCK, FGR and YES (Table 2). Altogether, the MS and co-IP experiments described above demonstrate the feasibility of the MS-based approach to identify proteins directly associated with active RTK. Two different methods, i.e. FGFR3 pull-down, and isolation of proteins tyrosine phosphorylated by either endogenous or ectopic FGFR3 activation, carried-out in 2 different cell types (293T and RCS cells) identified the same FGFR3-specific interactors. Our data validate the use of RTK library plasmids for identification of novel proteins associated with active RTKs, and suggests that the core proteins interacting with the given RTK might be conserved among different cell types (Table 2). Example 4. Determination of quantitative and qualitative differences in RTK-mediated activation of major signaling pathways Differences in the activation of downstream signal transduction by each RTK may represent a major determinant of the RTK-specific signaling pattern. Yet very little is known about magnitude and duration of activation of major signaling pathway by each RTK. The fact that all RTKs and their mutants are expressed from one plasmid backbone and tagged with the same V5 epitope (for precise quantification of expression), makes the RTK library an ideal research instrument for the studies aimed on differences among RTK signaling. RTKs will be expressed in cells and the effect of RTK expression on spontaneous or ligand-induced activation (Fig. 1) of major signal transduction pathways utilized by RTKs. Using WB with panel of commercial phosphorylation-dependent antibodies, members of ERK and p38 MAP kinase, PI3K/AKT, PLCg, STAT, NFkB and other pathways for their activation mediated by expressed RTKs (Fig. 2). WB data may be analyzed using standard densitometric software to determine the quantitative differences in activation of signaling pathways. Commercially available kinase assays and GST pull-down assays for activation of signaling intermediates such as Raf, Erk, SRC, p38 and Akt kinases, and small GTPases Ras, RhoA and Cdc42 may be employed to functionally test the pathways activation. For kinase assays, the given kinase may be immunopurifted from cells and subjected a cell-free kinase assay with appropriate recombinant substrate according to methods described before (Krejci et al., 2007; Krejci et al., 2010). In addition, transcriptional dual -lucif erase reporter assays may be used to test whether the pathway activation results in effect on gene transcription (Krejci et al., 2008; Krejci et al., 2012). The activation of STAT, NFkB, b-catenin, SMAD and SNAIL transcriptional modulators will be determined using commercially available dual-Iuciferase reporter technology.
The approaches described above may be applied also to disease-associated variants of RTKs, to determine the extent of the effect of each mutation on kinase activation and its downstream signaling (Fig. 3). Commercially available antibodies specific to phosphorylated form of the given RTK will be used to monitor the levels of autophosphorylation. The ability of each RTK to activate its downstream signaling Iigand-independently may be compared with other RTKs, as inclination to spontaneous, ligand-independent activation represents important aspect of the RTK signaling (described in detail in Example 5). Example 5. Research on spontaneous, ligand-independent RTK dimerization/activation
Some RTKs may undergo spontaneous, ligand independent dimerization and activation. This might have important implications for situations where the given RTK is ectopically overexpressed, as a result of gene amplification or upstream translocation of a strong promoter, as described for oncogenic signaling of FGFRl, FGFR3, ERBB2, EGFR, MET, VEGFR and other RTKs (Li et al., 2012; Chesi et ah, 1997; Kim et al., 2012; Okabe et al., 2009, Liu et al., 2008). We found that although FGFR, EGFR and TRK RTKs share many features of their signaling, they differ in the levels of ligand-independent activation (by spontaneous dimerization when overexpressed), which is much higher in FGFR and TRK RTKs compared to EGFR (P. Krejci, unpublished data).
When the ligand-independent activation is carefully compared among RTKs, even kinases belonging to one family reveal significant differences. We found that among FGFR family RTKs, which are generally considered prone to the ligand-independent activation when overexpressed, this feature is restricted only to wt FGFRl and FGFR2, in contrast to FGFR3 and FGFR4, which need to carry activating, disease-associated mutations to activate without the ligand (Fig. 3). This suggest that the propensity for ligand-independent activation varies significantly among the RTKs, and thus may contribute to specific pattern of the RTK signaling.
In addition to its importance for physiological RTK signaling, the propensity to ligand- independent activation represents a critical parameter in rational design of cancer treatment, namely in development of so called ,ligand trap' inhibitors (soluble proteins scavenging the RTK ligands). Yet the extent of ligand-independent activation was never systematically evaluated among human RTKs. The RTK library offers an ideal research tool to carry-on a comparative and detailed analysis of spontaneous versus ligand-induced activation in majority of human RTKs and their disease-associated mutants.
Example 6. In cell RTK activity profiling for drug screening Compared to cell-free kinase assays, a cell-based profiling allows to determine RTK activity in intact cell environment, complete with regulatory networks and feedback control mechanisms. This is especially important when compounds targeting RTK interaction with its downstream signaling mediators are evaluated. The RTK library allows for rapid activity profiling of 36 different human RTKs and 203 of their disease-associated mutants, using automated transfection of the RTK-carrying vectors into the cultured cells in a 96-well plate format, cell incubation for 48-72 hours, and determination of the RTK activity by WB or ICC or dual-luciferase assay.
A main advantage of the technology lies in its large scope, which involves profiling of majority of human RTKs and their disease-associated mutants at the same time. This is by far larger amount of activity profiled RTKs in one experiment, compared with technologies currently at the market. When a novel anti-cancer drug is, for instance, tested for its RTK target specificity, a comprehensive picture of the given inhibitor's activity against many human RTKs. Comparative screenings of many RTK disease-associated mutants is especially important in situations when newly developed inhibitors have no activity against particular RTK mutations, as we demonstrate for BJG398 (Fig. 4), an anti-FGFR inhibitor recently developed by Novartis (Guaqnano et al., 2012).
Table 2 Compilation of proteomics experiments aimed on identification of proteins interacting with FGFR3 RTK. The experiments were carried-out as described in Example 3: MS V5 IP, FGFR3 purified via V5 immunoprecipitation and the immunocomplexes analyzed by mass spectrometry (MS); pY IP, immunoprecipitation of proteins (via 4G10 anti-pY antibody) phosphorylated at tyrosine upon activation of transgenic FGFR3 (293T cells) or endogenous FGFR3 (RCS cells); Co-IP, confirmation of interaction via independent method, i.e. reciprocal co-immunoprecipitation of both partners expressed in 293T cells; N/A, not analyzed.
Fyn + + + +
Fgr + + + +
Lyn + + + +
Lck + + + +
Blk + + - +
Hck + + + N/A p!30CAS - + + -
References
Chesi M, Nardini E, Brents LA, Schrock E, Ried T, Kuehl WM, Bergsagel PL. Nat Genet 1997;16:260-4.
Dailey L, Laplantine E, Priore R, Basilico C. J Cell Biol. 2003; 161:1053-66.
Guagnano V, Kauffmann A, Wohrle S, Stamm C, Ito M et al. Cancer Discov. 2012;2:1118-33. Kang S, Dong S, Gu TL, Guo A, Cohen MS et al. Cancer Cell. 2007;12:201-14.
Krejci P, Masri B, Salazar L, Farrington-Rock C, Prats H et al. (2007) J Biol Chem. 282: 2929-36.
Krejci P, Salazar L, Goodridge HS, Kashiwada TA, Schibler MJ et al. (2008) J Cell Sci. 121: 272-81.
Krejci P, Prochazkova J, Smutny J, Chlebova K, Lin P et al. (2010) Bone 47:102-10.
Krejci P, Aklian A, Kaucka M 5 Sevcikova E, Prochazkova J et al. (2012) PLoS One. 7:e35826. Kim HR, Kim DJ, Kang DR, Lee JG, Lim SM et sA. JCtin Oncol. 2013;31(6):731-7.
Laederich MB, Degnin CR, Lunstrum GP, Holden P, Horton WA. J Biol Chem. 2011;286:19597-604.
Lax I, Wong A, Lamothe B, Lee A,. Frost A, Hawes J, Schlessinger J. Mol Cell. 2002;10:709- 19.
Li C, Sun Y, Fang R, Han X, Luo X et al. JThorac Oncol. 2012;7:85-9.
Liu Z, Hou P, Ji MH, Studeman K, Jensen K et al. J. Clin. Endocr. Metab. 2008;93:3106-16.
Okabe T, Okamoto I, Tsukioka S, Uchida J, Hatashita E et al. Clin Cancer Res. 2009;15:907-
13. Passos-Bueno MR, Wilcox WR, Jabs EW, Sertie AL, Alonso LG, Kitoh H. Hum Mutat. 1999;14: 1 15-25.
Ronchetti D, Greco A, Compasso S, Colombo G, Deli'Era P et al. Oncogene. 2001;20:3553- 62.