HYBRID NUCLEIC ACID MOLECULES AND THEIR USE

Abstract

The invention relates to a nucleic acid molecule comprising: a. a first region comprising a nucleic acid sequence coding for the protein Cyclin D1, also called CCND1, said first region being controlled by means allowing the expression of said protein, and b. at least one second region, said second region comprising essentially a sequence from 14 to 59 nucleic acids, said second region corresponding to a transcribed region of a gene, said second region containing at least a genetic modification compared to the same region of the corresponding wild-type version of said gene, said second region being genetically isolated from the means allowing the expression of said protein such that said second region is not translated into a peptide.

Claims

1. A nucleic acid molecule comprising: a first region comprising a nucleic acid sequence coding for a Cyclin D1 protein, also called CCND1 protein, the CCND1 protein comprising the amino acid sequence as set forth in SEQ ID NO: 231 wherein said first region is controlled by means allowing the expression of said protein, and a second region comprising a nucleic acid sequence as set forth in SEQ ID NO: 232 ACGTGACACGTTCGGAGAATT, the second region being not translated into peptide.

2. The nucleic acid molecule according to claim 1, wherein the first region code for one of the CCND1 proteins as set forth in SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 233 and SEQ ID NO: 234.

3. The nucleic acid molecule according to claim 1, wherein said first region comprises, in its 5′-end a KOZAK sequence, and at its 3′-end a STOP codon, the STOP codon being in phase with the sequence coding for the CCND1 protein.

4. The nucleic acid molecule according to claim 3, wherein the KOZAK sequence consists of one of the sequences as set forth in SEQ ID NOs: 235 to 241.

5. The nucleic acid molecule according to claim 1, wherein the nucleic acid molecule comprises the first region which is followed, in the sense 5′->3′, by the second region.

6. The nucleic acid molecule according to claim 1, wherein the nucleic acid molecule comprises a third region comprising a restriction site of a restriction enzymes, wherein the third region is not translated into peptide.

7. The nucleic acid molecule according to claim 6, wherein the nucleic acid molecule comprises the first region which is followed, in the sense 5′->3′, by the third region, and wherein the third region, in the sense 5′->3′, is followed by the second region.

8. The nucleic acid molecule according to claim 6, wherein the third region comprises, between two restriction sites, a target sequence; wherein the target sequence has from 14 to 59 nucleic acids, said target sequence corresponding to a sequence of a transcribed region of a gene coding for a coded RNA, the coded RNA having a coded RNA sequence, the coded RNA coding, or not, for a coded protein, the coded protein having a coded protein sequence, wherein the sequence of the transcribed region of the gene coding for the coded RNA contains a genetic modification, the genetic modification being absent in the wild type transcribed region of the gene coding for the coded RNA, wherein said genetic modification when the coded RNA does not code for a coded protein, the genetic modification modifies an expression of the coded RNA, or a function of the coded RNA, and when the coded RNA codes for a coded protein, the genetic modification modifies an expression of the coded RNA, or a function of coded RNA, or an expression of the coded protein or a function of the coded protein.

9. The nucleic acid molecule according to claim 1, wherein said nucleic acid molecule comprises one of the sequences as set forth in SEQ ID NO: 242 and SEQ ID NO: 243.

10. The nucleic acid sequence according to claim 8, wherein the target sequence comprises one of the sequences as set forth in SEQ ID NOs: 244 to 246.

11. The nucleic acid molecule according to claim 8, wherein said nucleic acid molecule comprises one of the sequences as set forth in SEQ ID NOs: 247 to 249.

12. A cell comprising at least one copy of the nucleic acid molecule as defined in claim 1.

13. The cell according to claim 12, wherein said cell is a tumoral cell.

14. A genetically modified non-human animal, comprising at least one cell as defined in claim 12.

15. A subset of nucleic acid molecules, comprising a first nucleic acid molecule as defined in claim 1, and a second nucleic acid molecule, said second nucleic acid molecule comprising, i. the same first and second regions compared to the first and second regions of the first nucleic acid molecule.

16. The subset of nucleic acid molecules according to claim 15, wherein the second nucleic acid molecule comprise a third region, the third region comprising a target sequence, wherein the target sequence has from 14 to 59 nucleic acids, said target sequence corresponding to a sequence of a transcribed region of a wild type gene coding for a wild type coded protein, wherein the wild type gene coding for the wild type coded protein corresponds to the gene contained in the third region of the first nucleic acid molecule without the genetic modification.

17. A set of nucleic acid molecules, comprising: i. A subset according to claim 15, ii. A third nucleic acid molecule, said third nucleic acid molecule comprising, A first region comprising a nucleic acid sequence coding for reporter protein, said first region being controlled by means allowing translation of said reporter protein, and A second region corresponding to the second region found in the first nucleic acid molecule, and iii. A fourth nucleic acid molecule comprising, A first and a second region corresponding respectively to the first and the second region of said third nucleic acid molecule.

18. The set of nucleic acid molecules according to claim 17, wherein the fourth nucleic acid molecule comprises a third region, the third region comprising a target sequence, wherein the target sequence has from 14 to 59 nucleic acids, said target sequence corresponding to a sequence of a transcribed region of a wild type gene coding for a wild type coded protein, wherein the wild type gene coding for the wild type coded protein corresponds to the gene contained in the third region of the first nucleic acid molecule without the genetic modification.

19. A method for screening of small interfering nucleic acid molecules comprising a step of contacting a tumoral cell containing one of a nucleic acid molecule according to claim 1, or a subset of nucleic acid molecules comprising a first nucleic acid molecule as defined in claim 1, and a second nucleic acid molecule, said second nucleic acid molecule comprising, i. the same first and second regions compared to the first and second regions of the first nucleic acid molecule, and a set of nucleic acid molecules, comprising: i. a subset of nucleic acid molecules comprising a first nucleic acid molecule as defined in claim 1, and a second nucleic acid molecule, said second nucleic acid molecule comprising, i. the same first and second regions compared to the first and second regions of the first nucleic acid molecule. ii. a third nucleic acid molecule, said third nucleic acid molecule comprising, a first region comprising a nucleic acid sequence coding for reporter protein, said first region being controlled by means allowing translation of said reporter protein, and a second region corresponding to the second region found in the first nucleic acid molecule, and iii. a fourth nucleic acid molecule comprising, A first and a second region corresponding respectively to the first and the second region of said third nucleic acid molecule, with small interfering nucleic acid molecules, and a step of evaluating homeostasis of said tumoral cell.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0224] FIGS. 1-6—Functional hyper-specificity of TAG-RNAi

[0225] FIG. 1 represents the TAG-RNAi design to target tagged-Cyclin D1 mRNA (large arrow) but spare wildtype Cyclin D1. Flag tag is represented in black; HA tag in grey and Ccnd1 coding exons (numbered) or Untranslated Region (UTR) in white. WT mRNA is unaffected by TAG-RNAi but shares the similar off-target functional impact than Tagged responding cells.

[0226] FIG. 2 represents an immunoblot using anti-cyclin D1 (i.) or anti-actin (ii.) antibodies of RAS-G12V/DNP53 transformed MEFs of Ccnd1.sup.+/+ (1.), Ccnd1Ntag/Ntag (2.) or Ccnd1Ctag/Ctag (3.) genotype treated with scramble (A) Flag (B) or HA (C) TAG-siRNA.

[0227] FIG. 3 represents the Venn diagrams representing the genes differentially expressed and their degree of overlap within each other (expressed as % of similarity) after RNA interference using siRNAs specific to CycD1 in RAS-G12V/DNP53 transformed MEFs of Ccnd1Ctag/Ctag genotype. Nat corresponds to a previously described custom made siRNA24, Qia corresponds to a commercial siRNA sequence provided by the Qiagen company and Life corresponds to a commercial siRNA sequence provided by the Life Technologies company (see Table 1). Arrows highlight the three other G1-Cyclins (putative off-targets) that are affected by some of these siRNAs but not by TAG-siRNAs from FIG. 4.

[0228] FIG. 4 represents Venn diagrams representing the genes differentially expressed after RNA interference using siRNAs specific to Flag or HA Tags in RAS-G12V/DNP53 transformed MEFs of Ccnd1 Ctag/Ctag genotype. The arrow highlights Cyc-D2 which is the only other G1-Cyclin affected by the targeting of CycD1 using HA-RNAi.

[0229] FIG. 5 represents an immunoblot for G1-Cyclins by using anti-cyclin D1 (1), anti-cyclin D2 (2), anti-cyclin D3 (3b), anti-cyclin E1 (4) and anti CDK4 (5) antibodies (and controlled with anti-actin antibody (6) after RNA interference using three CycD1 “specific” siRNAs (Nat: D, Qui: E and Life: F), or Flag (B) and HA (C) siRNAs, in RAS-G12V/DNP53 transformed MEFs of Ccnd1 Ctag/Ctag genotype. Scramble siRNA treatment are shown in A

[0230] FIG. 6 represents a graph showing the in vivo RNAi functional impact on tumor burden dynamics of RAS-G12V/DNP53 transformed MEFs of Ccnd1.sup.−/− genotype rescued by Tagged-CycD1 (curve with squares) or Untagged-CycD1 (curve with diamonds) transgene. Note that TAG-RNAi has no significant impact in absence of Tagged-CycD1 transgene (black curve). HA-siRNA was used for TAG-RNAi in this experiment. TAG-RNAi treatment (illustrated by the bar) was initiated on the morning of Day 0 (see methods). Values are represented as average tumor size in mm.sup.3 of n=5 tumors+/−standard error of the mean. X-axis: days, y-axis: average tumor burden in mm.sup.3.

[0231] **p<0.01; pairwise comparison using two-tailed paired Student's t test.

[0232] FIGS. 7-11—Endogenous mutation-specific TAG-RNAi

[0233] FIG. 7 represents an immunoblot from lysates of cells expressing versions of CycD1 transgene depicted on the right schematic. The black box is the nucleotide sequence encoding for FLAG tag, the grey box is the nucleotide sequence encoding for HA tag and the box is the Kozak sequence (K). is Flag-CycD1-HA; HA-CycD1-Flag; Flag-K-CycD1-HA and iv: K-CycD1-HA-Stop-Flag. Cells are treated with scramble siRNA (A) or with Flag (B) or Ha (C) siRNAs. Proteins were labelled with anti-cyclin D1 (1) or actin (2) antibodies.

[0234] Note that the TAG-RNAi approach works equally when targeting the 5′ or the 3′ end of the mature messenger RNA and whether or not the genetic targeted TAG is translated as part of the coding sequence.

[0235] FIG. 8 is a schematic representing the generation of a TAG-RNAi strategy specific to the Kras mutation of the codon 12 (G12V-Endotag). The mutant G12V-Endotag (right dark grey) or the non-mutated WT-Endotag (right black) sequence spans from the −20 to the +20 nucleotides around the mutation and are fused to the non-coding part of the reporter gene encoding for Ntag (left black/grey)-CycD1.

[0236] FIG. 9 represents an histogram showing KRAS-G12V-Endotag specific knock down of the Ntag-CycD1 reporter constructs from FIG. 8 and measured by Tandem-HTRF (see methods), highlighting the major impact of the Ras-Endotag-siRNA #4 (C) on the expression of the mRNA carrying the mutation (grey bars) while only minor effect is observed on the mRNA carrying the non-mutated nucleotide sequence (black bars). Scramble-siRNA (Scr) is used as a negative control (A), HA-siRNA (B) is used as a positive control and Ras-Endotag-siRNA #12 (D) which targets equally both reporter constructs illustrates the specificity of the Ras-Endotag-siRNA #4 for the mutation. RAS-G12V/DNP53 transformed MEFs of Ccnd1−/− genotype were used for this experiment.

[0237] FIG. 10 represents a histogram showing the tumor burden dynamics of RAS-G12V/DNP53 transformed MEFs of Ccnd1−/− genotype rescued by the Tagged transgenes from FIG. 8. In vivo, TAG-RNAi illustrates the functional impact of the Ras-Endotag-siRNA #4 leading to the knock down of the CycD1 transgene fused to the G12V-Endotag (curve with circles) and to tumor growth arrest). No functional impact of Ras-Endotag-siRNA #4 is observed in tumors where the CycD1 transgene is fused to the KRAS-WT-Endotag (curve with triangles). Values are represented as average tumor size in mm.sup.3 of n=5 tumors+/−standard error of the mean. X-axis: days; y-axis: average tumor burden in mm.sup.3.

[0238] FIG. 11 represents an Immunoblot for KRAS (1.1 anti KRAS short exposure and 1.2 anti KRAS long exposure) using lysates from SW620 (KRAS-G12V mutated; i.) or HT29 (KRAS WildType; ii.) cell lines after treatment with Ras-Endotag-siRNA #4 (B and C) or irrelevant negative control HA-siRNA (A and D. Note the strong down-regulation of G12V-KRAS mutant in SW620 cell line by Ras-Endotag-siRNA #4, whereas only a marginal knock down is observed in wildtype KRAS HT29 cell line.

[0239] FIGS. 12-14—TAG-RNAi development using Cond1Ntag/Ntag and Cond1Ctag/Ctag MEFs

[0240] FIG. 12 represents a CycD1 Immunoblot (1.) of Ccnd1Ctag/Ctag MEFs lysates after TAG-RNAi titration using an increasing final concentration of 0.1, 1 or 10 nM HA-siRNA (B) compared to 10 nM of Scramble siRNA (A). Load charge is evaluated with an anti-actin antibody (2)

[0241] FIG. 13 represents Ntag-CycD1 (left graph) and Ctag-CycD1 (right graph) mRNA relative quantification by RT-qPCR after TAG-RNAi (1: scramble; 2: Flag siRNA and 3: HA siRNA) in two different clones of Ccnd1Ntag/Ntag or Ccnd1Ctag/Ctag MEFs transformed by HRAS-G12V and Dominant Negative P53 (DNP53). The black box is the nucleotide sequence encoding for FLAG tag, the grey box is the nucleotide sequence encoding for HA tag. Error bars=SD, n=3.

[0242] FIG. 14 represents the relative Ntag-CycD1 (left graph) and Ctag-CycD1 (right graph) protein abundance, after TAG-RNAi (1: scramble; 2: Flag siRNA and 3: HA siRNA) in respectively two clones (A, B) of MEFs of Ccnd1Ntag/Ntag (A, B left graph) or Ccnd1Ctag/Ctag (A, B right graph) genotype, measured by Tandem-HTRF using FLAG antibody as a donor and ha, sc, ab1 and ab3 antibodies as acceptors (see methods section)1. Error bars=SD, n=3.

[0243] FIGS. 15-16—G1-Cyclins off-targeting by CycD1 siRNAs revealed thanks to TAG-RNAi

[0244] FIG. 15 represents an immunoblot using anti-cyclin D1 (A) and actin (B) antibodies of lysates from RAS/DNP53 transformed MEFs of Ccnd1 Ctag/Ctag (i) and Ccnd1+/+(ii) genotype after TAG-RNAi (Flag: 2; HA; 3) or RNAi against CycD1 using three different siRNAs (Nat: 4, Qia: 5 and Life: 6) and as control scramble siRNA (1).

[0245] FIG. 16 represents an immunoblot for G1-Cyclins by using anti-cyclin D2 (1), anti-cyclin D3 (3b), anti-cyclin E1 (4), and anti CDK4 (5) antibodies (and controlled with anti-actin antibody (6)) of lysates from RAS/DNP53 transformed MEFs of Ccnd1−/− genotype after Flag-RNAi as a control (A) or RNAi against CycD1 using three different siRNAs (Nat: B, Qia: C, Life: D). Note the down-regulations of 1-CycD2 protein after CycD1 RNAi using the Nat siRNA, 2-CycD3 protein after CycD1 RNAi using both Life and Qia siRNAs and 3-CycE1 protein after all CycD1 RNAi compared to control FLAG-RNAi.

[0246] FIGS. 17A-J—Comparative analysis of transcriptome profiles after CycD1-RNAi or TAG-RNAi

[0247] Venn diagrams illustrating the degree of overlap (both in total number and %) of genes differentially expressed between two siRNAs applied on RAS/DNP53 transformed MEFs of Ccnd1 Ctag/Ctag genotype. The total number of genes differentially expressed after each siRNA treatment is 1—Flag-siRNA: 862; 2—HA-siRNA: 2670, 3—Nat-siRNA: 448, 4—Life-siRNA: 1700, 5—Qia-siRNA: 604. A represents the transcription profile of Flag-RNAi versus HA-RNAi, B represents HA versus Nature, C is FLAG vs. Life, D is Quia vs. Life, E is HA vs. Quia, F is HA vs. Life, G is Nature vs. Life, H is Flag vs. Nat, I is Flag vs. Quia and J is Quia vs. Nat,

[0248] FIGS. 18-20—In vivo Comparison of CycD1-RNAi or TAG-RNAi functional incidence

[0249] FIG. 18 represents an immunoblot after TAG-RNAi (Flag siRNA: 1, HA siRNA: 2, scramble siRNA: 3) on RAS/DNP53 transformed MEFs of Ccnd1−/− genotype rescued by FLAG-HA-Tagged (i) or Untagged-CycD1 (ii) transgene. A: anti-cyclin D1 antibody and B: anti actin antibodies

[0250] FIG. 19 is a schematic representation of the conventional RNA interference approach using siRNAs designed to target a specific gene of interest in wildtype cells while no impact is expected in cells where this target has been genetically ablated. In this setting, the transcriptome and functional impact should be unaltered in the genetic knock out cells. E: Off targets; A: Phenotype ? Transcriptome ?; B: Identical phenotype, Identical transcriptome and C: Good control for Gene A siRNA off-target functional incidence.

[0251] FIG. 20 is a graph representing the in vivo RNAi functional impact on tumor burden dynamics of RAS-G12V/DNP53 transformed MEFs of Ccnd1−/− genotype rescued by Tagged-CycD1 (curve with squares) or Untagged-CycD1 (curve with diamonds) transgene, or not (parental cells stably expressing GFP, curve with triangles). Note that CycD1-specific siRNAs induce a tumor progression arrest of CycD1-null tumors (curve with triangles) which is reversed after siRNA treatment arrest. In parallel, HA-siRNA was used in this experiment to demonstrate the specificity of TAG-RNAi (curves with squares for which tumor growth is inhibited or curve with diamonds where no effect is observed). RNAi treatment (illustrated by the bar) was initiated on the morning of Day 0 (see methods). Values are represented as average tumor size of n=6 tumors+/−standard error of the mean.

[0252] FIGS. 21-23—TAG-RNAi for the targeting of any gene in any cell type

[0253] FIG. 21 represents an immunoblot of mouse 3T3 (i) or human MCF7 cell lines (ii) expressing FLAG-HA-CycD1 transgene w/wo TAG-RNAi treatment (Flag (2) or HA (3)), and compared to the scramble siRNA treatment (1). Protein are revealed with anti-cyclin D1 (A) and actin (B) antibodies.

[0254] FIG. 22 represents the knock down efficiency of transgenic Flag-mCherry-HA (left panel) or Flag-CycD1-HA expression (right panel) in wildtype RAS/DNP53 transformed MEFs measured by RT-qPCR after TAG-RNAi using Flag-siRNA (2) or HA-siRNA (3) vs scramble siRNA (1). Error bars=SD, n=3.

[0255] FIG. 23 represents an immunoblot from lysates of Ccnd1 Ctag/+ MEFs expressing the CycD1 mRNA depicted on the right schematic (tagged and untagged). Note that only the tagged version of CycD1 (upper band) is affected by TAG-RNAi treatment and not the WT untagged version of CycD1 (lower band) when labelled with anti-cyclin D1 antibody (A). B: labelling with anti-HA antibody; C: labelling with anti-actin antibody.

[0256] FIGS. 24-28—In vivo TAG-RNAi versatility for the study of tumor growth dynamics after reversible gene knock down

[0257] FIG. 24 represents an immunoblot of tagged or untagged WT-CycD1 or T286A-CycD1 mutant (which is hyper-stable and oncogenic), expressed within the same Ccnd1−/− MEFs cell line. Note that only the tagged version (whether WT or mutated on T286) is sensitive to TAG-RNAi treatment (**, lane 6 to 9). 1: TAG-CycD1+T286A; 2: TAG-T286A+CycD1; 3: TAG-T286A; 4: TAG-CycD1; 5: CD1−/− (parental); 6: TAG-CycD1; 7: TAG-T286A; 8: TAG-T286A+CycD1 and 9: TAG-CycD1+T286A. #1: clone 1; #2 clone 2. *: treatment with siRNA. A1 and A2: band of cyclin D1. B: actin.

[0258] FIG. 25 represents a graph showing the tagged-T286A-CycD1 transgene driven tumor progression analysis w/wo HA-siRNA (curve with squares), FLAG-siRNA (curve with triangles) or Scramble-siRNA (curve with diamonds) (illustrated by the bars). Note the versatility of the approach with tumors relapsing after TAG-RNAi treatment pause, but remaining sensitive to the treatment when applied again later. Values represent the average tumor size of n=10 tumors+/−standard error of the mean. Ccnd1+/+3T3 cells were used for this experiment. Y-axis: Average tumor size (mm.sup.3) and x-axis: days.

[0259] FIG. 26 represents a graph showing T286A-CycD1 (curve with diamonds) or Tagged-T286A-CycD1 (curve with squares) driven tumor progression analysis w/wo in vivo TAG-RNAi. Note the specificity of the TAG-RNAi approach which specifically impinges on Tagged-T286A-CycD1-driven tumor progression on one flank of the mouse (squares) but has no off-target impact on untagged T286A-CycD1-driven tumor growth on the other flank of the same animal (diamonds). Values are represented as average tumor size of n=10 tumors+/−standard error of the mean. Ccnd1+/+3T3 cells and FLAG-siRNA were used in this experiment. Curve with triangle represents T286A-CycD1 treated with control scramble siRNA. Y-axis: Average tumor size (mm.sup.3) and x-axis: days.

[0260] FIG. 27 represents a histogram of 5 days Tumor growth index (size of the tumor/size of the tumor 5 days before) of Ccnd1+/+3T3 cells expressing Tagged-T286A-CycD1 (4-8) or untagged 1286-CycD1 (1-3) after in vivo TAG-RNAi. Note the equal efficiency of FLAG or HA-RNAi, but no significant additive effect of FLAG then HA RNA interference on tumor burden (see methods). Results are represented as average values of n=10 tumors+/−standard deviation, where the size of each tumor is measured and divided by its size 5 days before. 1: Flag-siRNA; 2: FLAG-siRNA+HA-siRNA; 3: HA-siRNA; 4: HA-siRNA; 5: FLAG-siRNA+HA-siRNA; 6: FLAG-siRNA; 7: SCR-siRNA and 8: vehicle. X-axis: days. *p<0.05, **p<0.01; pairwise comparison using two-tailed paired (brown bars versus blue bars) or unpaired (blue bars versus blue bars) Student's t test.

[0261] FIG. 28 represents a graph showing the tumor growth dynamics upon TAG-RNAi treatment, followed by treatment pause (represented by slashes preceding the relapse of the tumors on the graph), followed by TAG-RNAi treatment inversion, illustrating the various possibilities of the versatile TAG-RNAi approach. Measures of the tumor size are done in the morning just before TAG-RNAi treatment, and day 1 is the first day of treatment. Values are represented as average tumor size of n=10 tumors+/−standard error of the mean. Curve with diamonds: HA-siRNA/Vehicle; curve with squares: FLAG-siRNA/SCR-siRNA; curve with triangles: SCR-siRNA/FLAG-siRNA and curve with circles: Vehicle/HA-siRNA. Y-axis: Average tumor size in mm.sup.3 and x-axis: days.

[0262] FIGS. 29-33—In vivo TAG-RNAi applied to the murine HRAS-G12V oncogene

[0263] FIG. 29 is a schematic representation of the implantation of murine “Tagged-HRAS-G12V” expressing cancer cells on one flank of immune compromised mice (black circle) whereas murine “untagged-HRAS-G12V” control cells are implanted on the contralateral flank (grey circle). Cells from the black circle can be targeted by the siRNA specific to the genetic TAG whereas cells from the grey circle are insensitive to this siRNA. The cancer cells were generated using MEFs transformed by the SV40 Large T and the human HRAS-G12V transgenes.

[0264] FIG. 30 represents an immunoblot using lysates from murine RAS-G12V/Large T transformed MEFs that express RAS-G12V protein from a transgene that is fused (left immunoblot) or not (right immunoblot) to genetic sequences of Flag, schematized as a black box (in 5′ before the translation initiation Kozak sequence) and HA, schematized as a grey box, (in 3′ after the stop codon) localized in the untranslated region of the transgenic mRNA. Note that only the transgene carrying the Flag and HA sequences can be silenced by Flag or HA-specific siRNA. I: anti RAS antibody and ii: anti actin antibody. 1: untransformed cells; 2: cells treated with scramble siRNA, 3; cells treated with FLAG-siRNA and 4: cells treated with HA-siRNA. Grey box with K: schematic representation of Kozak sequence; white box with R: schematic representation of murine RAS-G12V cDNA. * represents a stop codon. +++: tumor

[0265] FIG. 31 is a graph showing Tagged-HRAS-G12V driven tumor progression w/wo in vivo TAG-RNAi using HA-siRNA or Scramble-siRNA. Note that the “Tagged” tumor progression is decreased after TAG-RNAi. Values are represented as average tumor size of n=10 tumors+/−standard error of the mean. Ccnd1+/+3T3 cells were used in this experiment. Curve with diamonds: HA-siRNA; curve with squares: scramble siRNA. Y-axis: tumor volume (mm.sup.3) and y-axis: days.

[0266] FIG. 32 is a graph showing the in vivo growth kinetics of Tagged-HRAS-G12V-driven tumors after TAG-RNAi. The size of each tumor is measured (Day 9 of graph FIG. 31) and divided by its size 5 days before (Day 5 of graph FIG. 31). Results are represented as average values+/−standard deviation from two independent experiments performed with two independent biological clones, each experiment comprising n=5 tumors per clone. A: scramble siRNA; B: HA-siRNA.

[0267] FIG. 33 is a graph showing the untagged-HRAS-G12V driven tumor progression w/wo in vivo RNAi using HA-siRNA or Scramble-siRNA. Note the absence of significant impact on tumor progression with both Scramble and HA-siRNA. Values are represented as average tumor size of n=10 tumors+/−standard error of the mean. Curve with squares: HA-siRNA; curve with diamonds: scramble siRNA. *p<0.05; ***p<0,001; pairwise comparison USING two-tailed unpaired Student's t test (FIGS. 32 and 33).

[0268] FIGS. 34-36—Rapid TAG-RNAi screening using 386-well plate Tandem-HTRF readouts

[0269] FIG. 34 represents the relative Ntag-CycD1 protein abundance measured by Tandem-HTRF using FLAG as a Förster Resonance Energy Transfer “donor” antibody and ha, ab1, ab3 and sc antibodies as “acceptors” for the screening of the V5 Tag-specific siRNAs (see Table 1). On the bottom schematic is represented the nucleotide sequence encoding for the V5 Tag (separated right grey box) and which corresponds to the peptide from 95 to 108 (GKPIPNPLLGLDST SEQ ID NO: 51) of RNA polymerase a subunit of simian parainfluenza virus type 5, but that has been fused to the non-coding region of the reporter transgene encoding for Ntag-CycD1. The asterisk illustrates the Stop codon of the transgene. Error bars=SD, n=3. 1: Scr-siRNA, 2: V5-siRNA1, 3: V5-siRNA2, 4: V5-siRNA4, 5: V5-siRNA5, 6: V5-siRNA6, 7: V5-siRNA7, 8: V5-siRNA8, 9: V5-siRNA9, 10: V5-siRNA10, 11: V5-siRNA11, 12: V5-siRNA12, 13: V5-siRNA13, 14: V5-siRNA14, 15: V5-siRNA15, 16: V5-siRNA16, 17: V5-siRNA17, 18: V5-siRNA18, 19: V5-siRNA19, 20: V5-siRNA20, 21: V5-siRNA21, 22: V5-siRNA22 and 23: FLAG-siRNA.

[0270] FIG. 35 represents the relative MYC-CDK4-V5 fusion protein abundance measured by Tandem-HTRF using MYC as a Förster Resonance Energy Transfer “donor” antibody and v5 antibody as an “acceptors” for the screening confirmation of V5 Tag-specific siRNAs compared to a (see Table 1). The nucleotide sequence encoding for the V5 Tag has been fused to the coding region of the reporter transgene encoding for CDK4 in this experiment. Error bars=SD, n=3. 1: Scr-siRNA, 2: V5-siRNA1, 3: V5-siRNA2, 4: V5-siRNA4, 5: V5-siRNA5, 6: V5-siRNA6, 7: V5-siRNA7, 8: V5-siRNA8, 9: V5-siRNA9, 10: V5-siRNA10, 11: V5-siRNA11, 12: V5-siRNA12, 13: V5-siRNA13, 14: V5-siRNA14, 15: V5-siRNA15, 16: V5-siRNA16, 17: V5-siRNA17, 18: V5-siRNA18, 19: V5-siRNA19, 20: V5-siRNA20, 21: V5-siRNA21 and 22: V5-siRNA22.

[0271] FIG. 36 represents the impact of mutations in the coding sequence of FLAG or HA peptides on the knock down efficiency by TAG-siRNAs (see table 1). The right schematic illustrates the mismatches (star for a match versus exclamation mark for a mismatch) that reside between the targeted sequence and the siRNA used. Note that due to the mismatches, FlagN-siRNA (i.) no longer inhibits Flag-CycD1-HA transgene and is less efficient than FlagC-siRNA (ii.) for the inhibition of Ctag-CycD1 transgene. Scramble-siRNA was used as a negative control for the basal level of each transgenic construct expression. Flag-siRNA and HA-siRNA were used as positive interfering RNAs working on all transgenic constructs. Y-axis: Relative protein abundance (%). 1: scramble siRNA; 2: Flag-siRNA; 3: HA-siRNA; 4: FlagC-siRNA and 5: FlagN-siRNA.

[0272] FIGS. 37-40—TAG-RNAi applied to the endogenous Ras-G12V genetic mutant tag

[0273] FIG. 37 represents the relative Ntag-CycD1 reporter protein abundance measured by Tandem-HTRF using FLAG as a Förster Resonance Energy Transfer “donor” antibody and ha, ab1, ab3 and sc antibodies as “acceptors” for the screening of Ras-G12V Endotag-specific siRNAs (see Table 1). Error bars=SD, n=3. 1: Scr-siRNA: 2: Flag-siRNA: 3: Ras-G12V-Endotag-siRNA1, 4: Ras-G12V-Endotag-siRNA2, 5: Ras-G12V-Endotag-siRNA 3, 6: Ras-G12V-Endotag-siRNA4, 7: Ras-G12V-Endotag-siRNA5, 8: Ras-G12V-Endotag-siRNA6, 9: Ras-G12V-Endotag-siRNA7, 10: Ras-G12V-Endotag-siRNA 8, 11: Ras-G12V-Endotag-siRNA9, 12: Ras-G12V-Endotag-siRNA10, 13: Ras-G12V-Endotag-siRNA11, 14: Ras-G12V-Endotag-siRNA12, 15: Ras-G12V-Endotag-siRNA 13, 16: Ras-G12V-Endotag-siRNA14, 17: Ras-G12V-Endotag-siRNA15, 18: Ras-G12V-Endotag-siRNA16, 19: Ras-G12V-Endotag-siRNA17, 20: Ras-G12V-Endotag-siRNA 18, 21: Ras-G12V-Endotag-siRNA19, 22: Ras-G12V-Endotag-siRNA20 and 23: Ras-G12V-Endotag-siRNA21.

[0274] Black columns: CycD1-STOP-RASWT Tag constructions, Dark grey columns: CycD1-STOP-RASG12V Tag constructions

[0275] FIG. 38 represents an immunoblot of HRAS-G12V/DNP53 transformed MEFs of Ccnd1−/− genotype rescued by the CycD1 Tagged transgenes (WT-Endotag (ii) or G12V-Endotag (i)) and targeted by TAG-RNAi using human Kras-G12V-Endotag-specific siRNA #4 (2), #5 (3), or #16 (4) and scramble (1), which were showing the most promising mutation specific knock down from the screening performed in FIG. 37. A: anti-cyclin D1; B: anti-actin.

[0276] FIG. 39 represents a histogram showing the in vivo tumor growth kinetics of HRAS-G12V/DNP53 transformed MEFs of Ccnd1−/− genotype rescued by the CycD1 Tagged transgenes (WT-Endotag or G12V-Endotag) and targeted by TAG-RNAi using human Kras-G12V-Endotag-specific siRNA #4 or HA-siRNA. The size of each tumor from FIG. 10 is measured before and after treatment and divided by its size 2 days before (Day 5/Day 3 before treatment in black bars and Day 7/Day 5 after treatment in grey bars). Results are represented as average values+/−standard deviation with n=5 tumors. *p<0.05; ***p<0,001; pairwise comparison USING two-tailed unpaired Student's t test. A: G12 V-Endotag/HA-siRNA; B: WT-Endotag/siRNA #4, C: G12V-Endotag/siRNA #4.

[0277] FIG. 40 represents an immunoblot for CycD1 (A) (and control actin (B)) using lysates from HT29 (i) and SW620 (ii) human cancer cell lines after treatment with CycD1-siRNA (1) or irrelevant negative control HA-siRNA (2). Note the strong down-regulation of CycD1 expression attesting for good siRNA transfection efficiency in both cell lines.

[0278] FIG. 41 is a schematic representing the generation of a TAG-RNAi strategy specific to the BRaf mutation (V600E-Endotag). The mutant V600E-Endotag (black) or the non-mutated WT-Endotag (grey) sequence spans from the −20 to the +20 nucleotides around the mutation and are fused to the non-coding part of the reporter gene encoding for CycD1.

[0279] FIG. 42 represents the relative CycD1 reporter protein abundance measured by Tandem-HTRF using SC450 as a Förster Resonance Energy Transfer “donor” antibody and ab1 and ab3 antibodies as “acceptors” for the screening of BRAF-V600E Endotag-specific siRNAs (see Table 1). Error bars=SD, n=3. 1: Scr-siRNA: 2: HA-siRNA: 3: Raf-V600E-Endotag-siRNA1, 4: Raf-V600E-Endotag-siRNA2, 5: Raf-V600E-Endotag-siRNA 3, 6: Raf-V600E-Endotag-siRNA4, 7: Raf-V600E-Endotag-siRNA5, 8: Raf-V600E-Endotag-siRNA6, 9: Raf-V600E-Endotag-siRNA7, 10: Raf-V600E-Endotag-siRNA 8, 11: Raf-V600E-Endotag-siRNA9, 12: Raf-V600E-Endotag-siRNA10, 13: Raf-V600E-Endotag-siRNA11, 14: Raf-V600E-Endotag-siRNA12, 15: Raf-V600E-Endotag-siRNA 13, 16: Raf-V600E-Endotag-siRNA14, 17: Raf-V600E-Endotag-siRNA15, 18: Raf-V600E-Endotag-siRNA16, 19: Raf-V600E-Endotag-siRNA17, 20: Raf-V600E-Endotag-siRNA 18, 21: Raf-V600E-Endotag-siRNA19, 22: Raf-V600E-Endotag-siRNA 20 and 23: Raf-V600E-Endotag-siRNA21.

[0280] Dark grey columns: CycD1-STOP-BRAFWT Tag constructions, Black columns: CycD1-STOP-BRAF-V600E Tag constructions

EXAMPLE

[0281] The inventors reasoned that ectopic RNAi could rely on a tag sequence linked to a specific locus of interest to be targeted. The idea behind using a tag complementary to the siRNA sequence, but absent from control cells, is that cells without the tag would correspond to scrambled controls in a classical siRNA experiment and also to rescued control cells. With the TAG-RNAi alternative, control cells are exposed to the exact same siRNA molecule than the responding cells to be challenged. This approach can theoretically unmask phenotypic alterations that arise with “off-target” interference. As a consequence, TAG-RNAi provides an accurate functional signature to fairly compare with gene ablation phenotypes.

[0282] To demonstrate our hypothesis, the inventors took advantage of genetically engineered mice expressing FLAG-HA tagged versions of Cyclin D1 (CycD1) at physiological levels. These strains produce functional N-terminal (Ntag-CycD1) or C-terminal (Ctag-CycD1) Flag-HA tagged-CycD1 protein. Hence, FLAG or HA RNA interference will be blind to wildtype Ccnd1 gene expression but interfere with Tagged-CycD1 mRNA translation (FIG. 1).

[0283] The inventors first isolated Flag or HA siRNAs specific to knock down Tagged-CycD1 in a dose-dependent manner, whether the target mRNA sequence is at the 5′ or at the 3′ end of the mature messenger RNA (FIG. 2, FIG. 12 and FIG. 14, Table 1).

[0284] Then, to test the specificity and the efficiency of TAG-RNAi compared to three independent siRNAs, the inventors performed RNA-Sequencing experiments on Mouse Embryonic Fibroblasts (MEFs) transformed by the HRAS oncogene and Dominant Negative P53 (DNP53). Following RNAi using either siFLAG, siHA, a published siRNA against CycD1, or two different commercial siRNA sequences against CycD1, the inventors collected the expression profiles of cells expressing Ctag-CycD1. As expected, Ctag-CycD1 mRNA was knocked down with all five siRNAs tested (FIG. 15). However, the global transcription profile deviates less between TAG-siRNAs than between CycD1-siRNAs (FIG. 4, FIG. 5, FIG. 17). More surprisingly, close inspection of genes only differentially expressed after CycD1 RNA interference but not TAG-RNAi, revealed the down-regulation of other G1-Cyclins, like Cyclin D3 and Cyclin E1 (FIG. 3-5). It is well established that all G1-Cyclins belong to the same functional group and promote cell cycle and tumor progression. Our results show unexpectedly that the targeting of CycD1-null cells by siRNAs supposed to be specific to CycD1, leads to the down-regulation of other G1-Cyclins too (FIG. 16). This suggests that a functional off-targeting by three different ectopic siRNAs against CycD1 could alter fundamental properties of these CycD1-null cancer cells. In contrast, TAG-RNAi technology provides functional observations that can confidently be attributed to the specific targeting of the tagged gene of interest.

[0285] Furthermore, TAG-RNAi offers in vivo an opportunity for the functional exploration intrinsic to the targeted cells. The strength of the approach relies on the biological response of “tagged tumors” on one flank of the recipient mouse, while no impact is expected on “untagged tumor” of the other flank of the same animal. Indeed the targeting of Tagged-CycD1 which induced tumor growth inhibition, as reported by conditional genetic ablation, demonstrated this (FIG. 6, FIG. 18). Surprisingly, whereas tumors expressing untagged-CycD1 remain unaffected by TAG-RNAi, a striking phenotype characterized by tumor progression arrest is induced in CycD1-null cancer cells treated with “CycD1-specific” siRNAs (FIG. 19 and FIG. 20). Such in vivo experimental artifact strongly suggest that “specific” CycD1 siRNAs exert an “off-target” functional pressure on tumors and should be used with caution to investigate the impact of CycD1 on cancer development. Besides, this off-target induced phenotype would not be revealed when performing parallel experiments using the usual Scramble siRNA control. TAG-RNAi on the other hand rules out the risk of biological misinterpretation following in vivo gene knock down by RNA interference.

[0286] Using the same siRNA molecule, TAG-RNAi is applicable in any cell type and for the targeting of any (single or multiple) tagged transgene(s) (FIG. 21 and FIG. 22). Thanks to heterozygous knock in strains, TAG-RNAi also allows the specific silencing of the product of one tagged allele while sparing the other wildtype (untagged) allele (FIG. 23). Thus, it is for example technically possible to co-express in the same cell, one mutant version that can be targeted by TAG-RNAi, and an additional wildtype untagged version for which the expression is unchanged, or vice versa (FIG. 24). Therefore, In vitro and in vivo, TAG-RNAi offers a wide range of opportunities to study the functional dynamics of transient knock down of any gene of interest (FIG. 25-27).

[0287] The targeting of any mRNA sequence can be achieved by TAG-RNAi conducted inside or outside of the translated region (FIG. 7). This useful alternative avoids peptide sequence modification of the candidate protein to be targeted and prevents the risk of subsequent loss-of-function. Tags added to the non-coding region of the Hras mRNA allows to target this untagged oncogene in vivo using TAG-RNAi (FIG. 31-33).

[0288] In a siRNA screening perspective, the inventors show using the V5 tag, that the isolation of specific siRNAs for any tag is relatively easy (FIG. 34). Additionally, while keeping a constant peptidic tag sequence, one can design mutations rendering Tagged-mRNA resistant to RNAi (FIG. 36).

[0289] For this reason, in the frame of human therapeutics, the inventors wanted to explore endogenous mutant genetic tags (Endotags) that could be targeted specifically by TAG-RNAi in native cells. Many diseases are linked to genetic mutations and silencing such alterations while sparing wildtype “healthy” version of the candidate target could be a specific means of targeting only mutated sick cells in a clinical assay. Consequently, the inventors focused on a known 35 G>T alteration of the oncogene Kras which occurs at the level of the codon 12 and changes the amino acid sequence from a Glycine to a Valine in human cancers. By extracting the 20 nucleotides upstream and downstream of this mutation the inventors generated the so-called G12V-Endotag (FIG. 8). As a control, the inventors used the same 40 nucleotides from the wildtype version of Kras and named this the WT-Endotag (FIG. 8). Moving along the G12V-Endotag sequence base by base, the inventors screened all 21 possible siRNAs that could potentially silence the reporter mRNA encoding for CycD1 and carrying the G12V-Endotag, to induce a minor effect on the reporter mRNA carrying the WT-Endotag (FIG. 37). From all the siRNA tested, G12V-Endotag-siRNA number 4 did knock down the reporter gene fused to G12V-Endotag but affected at the margin the reporter construct fused to the WT-Endotag (FIG. 9, FIG. 37 and FIG. 38). To probe for potential off-target side effects of this most promising “G12V-specific” siRNA #4 in vivo, the inventors investigated its impact on CycD1-driven tumor growth like the inventors did earlier (FIG. 6). #4 appeared to be efficient in targeting the CycD1 transgene carrying the G12V-Endotag to repress tumor growth, while having no significant biological impact on tumors carrying the WT-Endotag (FIG. 10, FIG. 39). Finally, by testing in parallel its efficiency in SW620 (KRAS-G12V mutated) and HT29 (KRAS wildtype) human colorectal cancer cell lines, the inventors confirmed that G12V-Endotag-siRNA #4 specifically knocks down the G12V mutation of Kras oncogene but presents a limited incidence on wildtype human KRAS (FIG. 11, FIG. 40).

[0290] Then the inventors performed the same kind of screening approach applied to another well-known genetic hit leading to the generation of the BRAF-V600E mutated protein. This strongly oncogenic 1799T>A genetic event on the gene coding for BRAF is associated with severe morbidity and resistance to modern anti-cancer therapies using monoclonal antibodies like Cetuximab. Like the inventors did for KRAS-G12V, they extracted the 20 nucleotides upstream and downstream of the BRAF-V600E mutation to generate the so-called BRAFV600E-Endotag (FIG. 41). Using Tandem-HTRF against CycD1 and following the targeting of BRAFV600E-Endotag by RNAi we isolated several siRNAs (#11 to #14) able to silence the CycD1 reporter gene containing this tag but not the reporter gene containing the BRAFWT-Endotag (FIG. 42). Since this mutation severely cripples the therapeutic response of colorectal cancer patients, the inventors tested these siRNAs in BRAF-V600E mutated HT-29 human colorectal cancer cell line in clonogenic assays. The inventors found that the siRNA #11 can strongly impair the viability of HT-29 cells even in absence of any chemotherapeutic stress (data not shown).

[0291] Thus, the screening of endogenous mutant genetic tags may provide a fruitful way of delivering novel and specific endogenous TAG-siRNAs to test their reliability in non-mutated cells in vitro and in vivo.

[0292] To date and despite sophisticated algorithm-based rationale design, siRNA selectivity remains difficult to evaluate8. However, the inventors have shown that TAG-RNA interference is an efficient way for acquiring high confidence functional genomics signatures. TAG-RNAi provides a novel elegant and robust approach to alter specific gene expression, without carrying over functional side-effects. The unique versatility of TAG-RNAi can be declined for any gene of interest, by using simple tagged-transgenic constructs, or by the specific editing of endogenous genomic locus using technologies like CRISPR-Cas9 for instance. Finally, the use of pathogenic mutations can provide a unique opportunity to search for disease-specific TAG-siRNAs and to rapidly test their pre-clinical safety. Ultimately, TAG-RNAi could perhaps be amenable to therapeutic perspectives by lowering off-target downsides for the safer use of RNAi in clinics.

[0293] Rationale for TAG-RNAi Development

[0294] RNA interference represents a strong potential therapeutic support to treat cancer. CycD1 is known to participate in cancer development. As a probe for potential therapeutic intervention by RNA interference and to ensure the unique knock down of CycD1, we ought to exclude “false-positive” phenotypic changes that could mislead our clinical strategy goal. By targeting Ctag-CycD1 and Ntag-CycD1 using FLAG or HA-siRNAs, we realized that no clear alteration of other G1-Cyclins was observed, contrary to the use of conventional CycD1-siRNAs. This led us to reconsider our view that CycD1 targeting by RNAi might be sufficient to prevent cell cycle in RAS-transformed cancer cells. It also alerted us on the dangerous scientific conclusion that could arise from such an experimental artifact related to the additional targeting of other G1-Cyclins by conventional siRNAs. That is why we decided to develop TAG-RNAi to gain confidence in the proper transcriptional and phenotypic signature of targeted cells. This way, any pharmacological intervention following RNAi screenings should have higher chances of success.

[0295] D-type Cyclins expression profile after conventional CycD1-RNAi or TAG-RNAi

[0296] Because G1-Cyclins levels of Ctag-CycD1 expressing cells seemed altered by CycD1-RNAi and not TAG-RNAi, we decided to explore the entire expression profile of these cells by RNA-Sequencing. It appears that CycD3 and CycE1 are only differentially expressed using one out of five siRNAs targeting CycD1. Considering that each of the five siRNAs tested (three raised against CycD1 and two raised against the FLAG-HA Tag), efficiently knocked down the expression of CycD1, we considered unlikely that this result would relate to the targeting efficacy of CycD1 between each siRNA. By testing these siRNAs in CycD1-null cells, we confirmed our prediction that they would alter other G1-Cyclins expression as an unspecific side-effect. We did not functionally explore whether this off-targeting is direct or indirect, but the simple alignment of the siRNAs tested with the cDNA of each G1-Cyclins reveals potential hybridization regions for the siRNAs we used. Concerning CycD2, we found that Nat-siRNA, life-siRNA and HA-siRNA lead to its down-regulation but not FLAG-siRNA. Hence, we remain skeptical about CycD2 expression being truly regulated by upstream CycD1 in Ras-transformed cells. In particular, Nat-siRNA also decreases CycD2 expression in CycD1-null cells.

[0297] Endotag-RNAi as a Clinical Intervention Perspective

[0298] Although TAG-RNAi appears as a reliable way for specifically targeting any gene of interest, its use for genetically unmodified primary human cells is still limited. To undertake functional studies in cellular models of human diseases, we believe that “natural” genetic mutations or SNPs may provide a powerful support for the development of base-specific Endotag-RNAi. Despite questionable mRNA off-targeting compared to TAG-RNAi based on 21 nucleotides rare genetic sequences, Endotag-RNAi provides at least a way to ensure the functional relevance of the targeted endogenous gene by using the same siRNA in cells that would not bare this mutation. Again, the benefit of this approach is to limit artificial phenotypes that would not only relate to the primary gene on-targeting but rather be the sum of multiple on and off-targeting consequences genome-wide. The subtraction of the off-targeting bias, at least at the functional level if not at the genome-wide transcription profile level, will certainly help to unmask true novel pharmacological targets and discard many false candidates for future therapeutics development. In addition the safety of such Endotag-siRNA can be easily tested using in cellulo viability models on healthy cells. For cancer therapeutic perspectives, any promising Endotag-siRNA can further be challenged in our model of Tagged-CycD1 reporter transgene, since we have shown that its targeting induces a tumor growth arrest but has no obvious incidence on the nude animal's health.

[0299] Material and Methods

[0300] Mice

[0301] Animal uses were performed in accordance with relevant guidelines and regulations. All experimental protocol were approved by the Regional Ethics committee (agreement number CEEA-LR-12070) and conducted according to approved procedures (Institute of Functional Genomics agreement number A 34-172-41, under F. Bienvenu agreement number A 34-513).

[0302] Ccnd1Ntag/Ntag and Ccnd1Ctag/Ctag mice have been described previously (Bienvenu et al. Nature 463, 374-378). Mice were bred at the Institute of Human Genetics animal care facility under standardized conditions with a 12 hours light/dark cycle, stable temperature (22±1° C.), controlled humidity (55±10%) and food and water ad libitum.

[0303] Genotyping of Cyclin D1-Tagged animals:

[0304] Genotyping of Ccnd1Ntag/Ntag and Ccnd1Ctag/Ctag animals was done as previously (Bienvenu et al. Nature 463, 374-378).

[0305] In Vivo siRNA Delivery and Tumor Growth Analysis

[0306] siRNAs (Genecust or Sigma) were dissolved in nuclease-free water and stored at −20° C. until use as described before (Lehmann et al. PLoS ONE 9(2): e88797). The soluble/lipid formulation was prepared extemporaneously to transport siRNAs across the animal body. At a well-defined ratio according to manufacturer's instructions, the siRNA lipid monophasic micro-emulsion was obtained by short vortex mixing of the lipid constituents with the siRNA solution. The formulation was kept at room temperature and protected from light until use.

[0307] In vivo the formulation at 1 mg/mL of siRNA was administered by rectal route using a micropipette (Eline lite dispenser 12026368, Biohit) and adapted conical tips (Dispenser tips 792028, Biohit). A constant dosage-volume of 20 μl of siRNA formulation per delivery was used (1 mg/kg).

[0308] siRNA treatment for tumor progression analysis was done every day twice by injection in the anal mucosa of the mice in the morning and in the evening. Tumor sizes were measured and calculated from the following formula: tumor size=L×W2/2, where L and W represent the length and the width of the tumor mass respectively.

[0309] Where indicated, FLAG and HA siRNA delivery was done alternately to assess a synergic or additive effect. Alternating the FLAG or HA siRNA delivery did not provoke any substantial difference in tumor growth response compared to FLAG only or HA only. However, the inventors tested decreasing the siRNA dose (i.e 0.5 mg/mL instead of 1 mg/mL) but it induced a less dramatic tumor growth arrest on RAS-G12V/DNP53-driven tumors.

[0310] Allograft Animal Models

[0311] For allografts in vivo experiments in nude mice, TAG responding cells and control cells pairs were prepared by one experimentator (J.C. or B.M) who gave them to a second experimentator (L.K.L) who was blind to the nature of each cell line. The second experimentator (L.K.L.) implanted the cells subcutaneously. Then, L.K.L. or J.C. performed the siRNA delivery as described above, and measured the tumor sizes. For each experimental design, TAG positive responding cells were implanted on one flank of the mouse, while control cells were implanted on the contralateral flank of the same mouse. A minimum of 5 mice were used per experiment. For each siRNA preparation to be tested, that is empty vehicle, Scramble, Nat, Qia, Life, Flag or HA as mentioned, all the mice were treated every day at 9 AM and then at 5 PM. In the FIG. 5e of Extended Data, CycD1-null mice were treated with Nat-siRNA at 9 AM, then with Qia-siRNA at 12 AM and finally with Life-siRNA at 5 PM of the same day and for three consecutive days.

[0312] Where mentioned in the figures, treatment pause was applied and restarted later on when tumors reached larger sizes for experimental purposes.

[0313] In TAG-RAS-G12V experiment (FIG. 31, 33), TAG-siRNA targeting settings (twice per day) slows down tumor progression of TAG-RAS-G12V-driven tumors, but the treatment is not sufficient to induce a steady-state or regression of the tumor size. However, increasing the TAG-RNAi delivery frequency improves the tumor growth inhibition of this aggressive cancer model (not shown).

[0314] T286A transformed 3T3 cells: 2.10.sup.6 cells were used per site of subcutaneous implantation.

[0315] RAS-G12V/DNP53 transformed MEFs: 0.5.10.sup.6 cells were used per site of subcutaneous implantation.

[0316] H-RAS-G12V TAGOUT and H-RAS-G12V NoTAG transformed/Large T immortalized MEFs: 0.5.10.sup.6 cells were used per site of subcutaneous implantation.

[0317] Each cell type was resuspended in 150 μl of RPMI 1640 and inoculated into the subcutaneous flanks of 6 weeks old female athymic nude mice (Harlan).

[0318] Cells

[0319] Mouse Embryonic Fibroblast Cells

[0320] MEF cells were prepared as previously described.

[0321] Ccnd1−/−, Ccnd+/+ MEFs and wildtype 3T3 cells were kindly provided by P. Sicinski.

[0322] Cell Culture

[0323] MEFs derived cells were cultured in Dulbecco's Minimal Essential Medium (41966-029, Gibco), supplemented with 5% fetal bovine serum (Life technology) and 1000 U/ml of Penicillin-Streptomycin (P/S) (Gibco). All cells lines were incubated in a 37° C. incubator in an atmosphere of 5% CO2 in air and maintained in sub-confluent culture conditions.

[0324] In Vitro siRNA Transfection

[0325] In-vitro siRNA delivery was done using Lipofectamine® RNAiMAX Transfection Reagent (Life Technologies) according to manufacturer's instructions. Cells to be transfected were seeded at 9 AM in the morning and transfected at 6 PM of the same day. The day after, cells were harvested at 9 AM or 6 PM for further biochemistry analysis.

[0326] Immunoblot

[0327] Immunoblots were performed as previously described (Bienvenu et al. Nature 463, 374-378). and with lysates obtained using HTRF lysis buffer (see below) supplemented with Protease Inhibitor Cocktail (S8830-20TAB). Antibodies used were HA (HA.11 Clone 16B12, Eurogentec, or Anti-HA EPITOPE TAG—600-401-384, Tebu-bio, or Hemagglutinin (HA) Rabbit Polyclonal Antibody, Life technologie), Cyclin D1 (sc-450, Santa cruz, or MS-210-PABX (AB1), Fisher scientific or RB-010-PABX (AB3), Fisher scientific), Actin (ab6276, Abcam), Tubulin (T9026, Sigma-Aldrich), Flag (F7425, Sigma-Aldrich), Ras (BD610002, BD Biosciences), Cyclin E1 (sc-481, Santa Cruz), CDK4 (sc-260, Santa Cruz), Cyclin D2 (MS-221-PABX (AB4), Fisher scientific), Cyclin D3 (MS-215-PABX (AB1), Fisher scientific). As secondary antibodies, peroxidase-conjugated IgG (Cell signaling) was used, followed by enhanced chemiluminescence detection (Millipore) and revealed with ChemiDoc™ XRS+ System (Biorad).

[0328] Tandem-HTRF

[0329] Cells in culture were washed with 1×PBS at 37° C. and then lysed in HTRF lysis buffer (Tris 10 mM, EDTA 1 mM, 0.05% NP-40). After centrifugation at 16000 g for 10 minutes, samples normalization were performed by adjusting total DNA content (nanodrop, Thermo Scientific) to 50 ng/μL. In each control experiment wild type cyclin D1 (or Cyclin D1-null) samples were used as negative control of noise signal (control 1). In addition, samples to be analyzed were incubated with donor antibody only in parallel (control 2). Comparison of both controls was performed for each Tandem-HTRF measure and gives identical background results19.

[0330] Tandem-HTRF detection of Cyclin D1 was performed with donor and acceptor antibody mixes according to manufacturer's instructions (Cisbio Bioassays—0.4 nM for the donor and 3 nM for the acceptor) within the linear range of HTRF signal (inside the linearity window of antibodies), to avoid high level saturation and keep low noise level. Donor antibodies were labeled with Europium (Eu) or Terbium (Tb) Cryptate fluorophore, and acceptor antibodies were labeled with XL665 fluorophore, or d2. List of antibodies can be found in Extended Data supplemental information.

[0331] Three independent samples were processed separately (biological triplicate) for Tandem-HTRF reaction. Each Tandem-HTRF sample being performed in technical triplicates as well.

[0332] The labeling of antibodies was made by the manufacturer Cisbio bioassays (to be contacted for more information).

[0333] For Tandem-HTRF measure, antibodies mix were diluted in q.s.p 5 μl of 0.2×PBS and added to 5 μl of sample per well of a Greiner black 384-well plate. After shaking and centrifugation (600 g for 1 minute), samples were kept at 4° C. overnight, protected from light.

[0334] HTRF was acquired by a PHERAstar FS microplate reader (BMG Labtech) as follows: after excitation with a laser at 337 nm (40 flashes per well), fluorescence emissions were monitored both at 620 nm (Lumi4-Tb emission) and at 665 nm (XL665 and d2 emission). A 400-μs integration time was used after a 60-μs delay to remove the short-lived fluorescence background from the specific signal.

[0335] The HTRF intensity was calculated using the following formula and is expressed as arbitrary units:

HTRF(intensity)={(ratio 665/620)sample}×10{circumflex over ( )}4−{(ratio 665/620)background}×10{circumflex over ( )}4

[0336] The background signal corresponds to cell lysates labeled with the Lumi4-Tb alone or control cell lysates devoid of the bait (wildtype cells). For each HTRF measure, the mean of technical replicates were used. Tandem-HTRF results outlined in the figures are the average of three biological independent experiments+/−standard deviation unless mentioned otherwise.

[0337] Retroviral Constructs

[0338] Plasmids:

[0339] All Cyclin D1 or RAS genetic constructs were inserted into BamH1-EcoR1 restriction sites of retro-viral vector pBABE-Puro or pBABE-hygro kindly provided by P. Sicinski, or MSCV retro-viral vector kindly provided by O. Ayrault. Large T encoding plasmid was kindly provided by L. Fajas, Ras-G12V/DNP53 plasmid (pL56-Ras) was kindly provided by L. LeCam. mCherry cDNA (CMV-mCherry) was kindly provided by V. Homburger and inserted into SnaB1-Nott restriction site of MSCV vector. Inserts sequences are listed in supplementary information.

[0340] Generation of Cyclin D1 rescue or H-RAS inserts

[0341] All retroviral constructs used were manipulated according to security measures and approved by the Institute of Functional Genomics.

[0342] cDNA Inserts of mouse origin (Cyclin D1 and Hras) were generated by RT-PCR using cDNA template from Ccnd1Ntag/Ctag E.13.5 embryonic head derived from C57BL/6j×129Sv mixed genetic background. The PCR products were inserted into retro-viral vectors and verified by sequencing after bacterial amplification.

[0343] Mutagenesis

[0344] T286A-CycD1 mutagenesis was performed using GeneArt® Site-Directed Mutagenesis System (LifeTechnologies) according to manufacturer's recommendations. Mutagenesis Primers are listed in Supplementary information.

[0345] G12V-Kras and WT-Kras Oligomers

[0346] Oligos were ordered at IDT-DNA and inserted into EcoR1-BgIII restriction sites of MSCV-Ntag-CycD1-Puro vector. An additional NdeI restriction site was used for cloning verification before sequencing of the resulting plasmid construct. Sequence of the oligos can be found in supplementary information below.

[0347] V600E-Braf and WT-Braf Oligomers

[0348] Oligos were ordered at IDT-DNA and inserted into BgIII-NotI restriction sites of MSCV-CycD1-Puro vector. An additional MfeI restriction site was used for cloning verification before sequencing of the resulting plasmid construct. Sequence of the oligos can be found in supplementary information below.

[0349] Stable Cell Lines Generation

[0350] Cells obtained by retroviral infection were done as described (Bienvenu et al. Nature 463, 374-378). Briefly, the day before transfection, Plat-E cells were seeded in 10 cm dishes at 50% confluence in DMEM (Gibco) supplemented with 10% Fetal Bovine Serum (Life technology).

[0351] Murine ecotrope retroviruses were produced by jetPEI transfection of Plat-E cells with 3 μg of pBabe-puro or MSCV-puro transfer vector or empty control vector (no resistance). 48 h after transfection, viral supernatant was harvested, filtered (0.45 um), supplemented with 8 μg/ml polybrene (H9268, Sigma) and used to infect recipient proliferating cells. 72 h after infection, medium of recipient cells was replaced and cells were selected for several days with 2 μg/ml of puromycin or 150 μg/ml of hygromycin, until all control cells exposed to empty virus are dead.

[0352] RT-q PCR

[0353] RNA Preparation

[0354] Total RNA was prepared using Trizol (Invitrogen) according to the manufacturer's instruction. Purified RNA was treated with the DNase I from the DNA-Free™ kit (Ambion) according to manufacturer's instructions.

[0355] Reverse Transcription

[0356] 1 μg of total RNA was reverse transcribed using 200 U M-MLV reverse transcriptase (Invitrogen) in the presence of 2.5 μM random hexamers, 0.5 mM dNTP, 10 mM DTT and 40 U of RNAse inhibitor (Invitrogen).

[0357] Real-Time PCR to semi-quantify Cyclin D1 mRNA

[0358] Four ng of the RT resulting cDNAs were used as template for real time PCR using LightCycler®480 Real-Time PCR System (Roche Applied Science) with the LightCycler® 480 SYBR Green I Master (Roche Applied Science). The sequences of all the primers used are listed in Extended Data. The PCR reaction was performed in 5 μl in the presence of 300 nM specific primers. Thermal cycling parameters were 10 min at 95° C., followed by 45 cycles of 95° C. for 15 s and 60° C. for 30 sec. At the end of the PCR, melting curve analyses of amplification products were carried out to confirm that only one product was amplified. The level of expression of each gene “X” was normalized to the geometric mean of the expression levels of the selected reference genes, R1 to R3, in the same PCR plate according to the formula:

[0359] Reference genes were selected among eight commonly used according to the GeNorm procedure (http://medgen.ugent.behjvdesomp/genorm/). Reference genes tested in this study were B2m (beta-2 microglobulin), Gapdh (glyceraldehyde-3-phosphate dehydrogenase), Mrpl32 (mitochondrial 39S ribosomal protein L32), Tbp (TFIID) (TATA box binding protein), Tubb2b (Tubulin beta2b), Trfr1 (transferrin receptor-1), all listed in Extended Data.

[0360] RNA-Sequencing

[0361] RNA Libraries Generation

[0362] RNA-Seq libraries were constructed with the Truseq stranded mRNA sample preparation (Low throughput protocol) kit from Illumina.

[0363] Poly-A Based mRNA Enrichment

[0364] One microgram of total RNA was used for the construction of the libraries,

[0365] The first step in the workflow involves purifying the poly-A containing mRNA molecules using poly-T oligo attached magnetic beads. Following purification, the mRNA is fragmented into small pieces using divalent cations under elevated temperature. The cleaved RNA fragments are copied into first strand cDNA using SuperScript II reverse transcriptase, Actinomycine D and random hexamer primers. The Second strand cDNA was synthesized by replacing dTTP with dUTP. These cDNA fragments then have the addition of a single ‘A’ base and subsequent ligation of the adapter. The products are then purified and enriched with 15 cycles of PCR. The final cDNA libraries were validated with a DNA 1000 Labchip on a Bioanalyzer (Agilent) and quantified with a KAPA qPCR kit.

[0366] For each sequencing lane of a flowcell V3, four libraries were pooled in equal proportions, denatured with NaOH and diluted to 7.5 pM before clustering. Cluster formation, primer hybridisation and single end-read 50 cycles sequencing were performed on cBot and HiSeq2000 (Illumina, San Diego, Calif.) respectively.

[0367] RNA-Sequencing Statistical Analysis

[0368] Image analysis and base calling were performed using the HiSeq Control Software and Real-Time Analysis component. Demultiplexing was performed using IIlumina's sequencing analysis software (CASAVA 1.8.2). The quality of the data was assessed using FastQC from the Babraham Institute and the Illumina software SAV (Sequence Analysis Viewer). Potential contaminants were investigated with the FastQ Screen software from the Babraham Institute.

[0369] RNA-seq reads were aligned to the mouse genome (UCSC mm10) with a set of gene model annotations (genes.gtf downloaded from UCSC on May 23 2014; GeneIDs come from the NCBI: gene2refseq.gz downloaded on Sep. 24 2015), using the splice junction mapper TopHat 2.0.1333 (with bowtie 2.2.334). Final read alignments having more than 3 mismatches were discarded. Gene counting was performed using HTSeq-count 0.6.1p1 (union mode)35. Since data come from a strand-specific assay, the read has to be mapped to the opposite strand of the gene. Before statistical analysis, genes with less than 20 reads (cumulating all the analysed samples) were filtered out.

[0370] DESeq2

[0371] Differentially expressed genes were identified using the Bioconductor36 package DESeq2 1.4.537. Data were normalized using the DESeq2 normalization method. Genes with adjusted p-value less than 5% (according to the FDR method from Benjamini-Hochberg) were declared differentially expressed. Generalized linear models was used to take into account paired samples.

[0372] Statistical Analysis

[0373] Data and statistical methods are expressed as outlined in figure legends. The means of two groups were compared using two-tailed paired or unpaired Student's t test.

[0374] Supplementary Information

[0375] Primers Used for T286A CycD1 Mutagenesis

TABLE-US-00005 SEQ ID NO: 52 Forward primer: GGTCTGGCCTGCGCGCCCACCGACGTG -  SEQ ID NO: 53 Reverse primer: CACGTCGGTGGGCGCGCAGGCCAGACC - 

[0376] RT-qPCR Primers

TABLE-US-00006 SEQ ID Gène SeqRef Forward Forward Sequence NO: B2μg  NM_00973 B2m-F TATGCTATCCAGAAAAC 54 (beta2 5 CCCTCAA microglo- bulin) GAPDH NM_00808 Gapdh- GGAGCGAGACCCCACTA 55 glyceral- 4 F ACA dehyde-3- phosphate dehydro- genase Trfr1 NM_01163 Trfr1- AGACCTTGCACTCTTTG 56 (trans- 8 F GACATG ferrin receptor- 1) Mrpl32 NM_02927 Mrpl32- AGGTGCTGGGAGCTGCT 57 (mitochon- 1 F ACA drial 39S  ribosomal protein  L32) Tbp (TFHD) NM_01368 Tbp2a- ATCGAGTCCGGTAGCCG 58 TATA box 4 F GTG binding  protein TUBULIN, NM_02371 Tubb2b- CTTAGTGAACTTCTGTT 59 BETA-2B 6 F GTCCTCCAGCA Cyclin D1 NM_00763 mCcnd1- AGGAGCAGAAGTGCGAA 60 [Mus 1 F GAG musculus] mCherry mCherry- CCTGTCCCCTCAGTTCA 61 F TGT Gène SeqRef Reverse Reverse Sequence B2ug  NM_00973 B2m-R GTATGTTCGGCTTCCCA 62 (beta2 5 TTCTC microglo- bulin) GAPDH NM_00808 Gapdh- ACATACTCAGCACCGGC 62 glyceral- 4 R CTC dehyde-3- phosphate dehydro- genase Gus (beta- NM_01036 Gus2-R GCCAACGGAGCAGGTTG 64 glucuro- 8 A nidase) Trfr1 NM_01163 Trfri-R GGTGTGTATGGATCACC 65 (trans- 8 AGTTCCTA ferrin receptor- 1) Mrpl32 NM_02927 Mrpl32- AAAGCGACTCCAGCTCT 66 (mitochon- 1 R GCT drial 39S  ribosomal protein  L32) Tbp (TFHD) NM_01368 Tbp2a-R GAAACCTAGCCAAACCG 67 TATA box 4 CC binding  protein TUBULIN, NM_02371 Tubb2b- AGGCAAACTGAGCACCA 68 BETA-2B 6 R TAATTTACAAA Cyclin D1 NM_00763 mCcnd1- CACAACTTCTCGGCAGT 69 [Mus 1 R CAA musculus] mCherry mCherry- CCCATGGTCTTCTTCTG 70 F CAT

TABLE-US-00007 TABLE 1 Active  SEQ    SEQ siRNA anti-sens ID Non-active Sense ID name sequence 5′-3′ NO: sequence 5′-3′ NO: Scram- AAUUCUCCGAAC  71 ACGUGACACGUUCGGAGA 121 ble GUGUCACGU Att HA UAGUCGGGCACG  72 CCUACGACGUGCCCGACU 122 UCGUAGGGG Att FLAG GUCAUCGUCGUC  73 CUACAAGGACGACGAUGA 123 CUUGUAGUC Ctt FLAGC CGACUUGUCAUC  74 GGACGACGAUGACAAGUC 124 GUCGUCCUU Gtt FLAGN GAGCUUGUCAUC  75 GGACGACGAUGACAAGCU 125 GUCGUCCUU Ctt Nat CCACAGAUGUGA  76 AAAUGAACUUCACAUCUG 126 AGUUCAUUU UGGtt Qia AACACCAGCUCC  77 CGCAGCACAGGAGCUGGU 127 UGUGCUGCG GUUtt Life CAGGAACAGAUU  78 AAGGGCUUCAAUCUGUUC 128 GAAGCCCUU CUGtt V5#1 cggguucggaaucggu  79 caaaccgauuccgaaccc 129 uugcc gTT V5#2 gcggguucggaaucgg  80 aaaccgauuccgaacccg 130 uuugc cTT V5#4 cagcggguucggaauc  81 accgauuccgaacccgcu 131 gguuu gTT V5#5 gcagcggguucggaau  82 ccgauuccgaacccgcug 132 caguu cTT V5#6 agcagcggguucggaa  83 cgauuccgaacccgcugc 133 ucggu uTT V5#7 cagcagcggguucgga  84 gauuccgaacccgcugcu 134 aucgg gTT V5#8 ccagcagcggguucgg  85 auuccgaacccgcugcug 135 aaucg gTT V5#9 cccagcagcggguucg  86 uuccgaacccgcugcugg 136 gaauc gTT V5#10 gcccagcagcggguuc  87 uccgaacccgcugcuggg 137 agaau cTT V5#11 ggcccagcagcggguu  88 ccgaacccgcugcugggc 138 cggaa cTT V5#12 aggcccagcagcgggu  89 cgaacccgcugcugggcc 139 ucgga uTT V5#13 caggcccagcagcggg  90 gaacccgcugcugggccu 140 uucgg gTT V5#14 ccaggcccagcagcgg  91 aacccgcugcugggccug 141 guucg gTT V5#15 uccaggcccagcagcg  92 acccgcugcugggccugg 142 gguuc aTT V5#16 auccaggcccagcagc  93 cccgcugcugggccugga 143 gaguu uTT V5#17 uauccaggcccagcag  94 ccgcugcugggccuggau 144 cgggu aTT V5#18 cuauccaggcccagca  95 cgcugcugggccuggaua 145 gcggg gTT V5#19 gcuauccaggcccagc  96 gcugcugggccuggauag 146 agcgg cTT V5#20 ugcuauccaggcccag  97 cugcugggccuggauagc 147 cagcg aTT V5#21 gugcuauccaggccca  98 ugcugggccuggauagca 148 gcagc cTT V5#22 ggugcuauccaggccc  99 gcugggccuggauagcac 149 agcag cTT G12V- ACAGCUCCAACU 100 UUGUGGUAGUUGGAGCUG 150 Ras- ACCACAAGC Utt Tag#1 G12V- AACAGCUCCAAC 101 UGUGGUAGUUGGAGCUGU 151 Ras- UACCACAAG Utt Tag#2 G12V- CAACAGCUCCAA 102 GUGGUAGUUGGAGCUGUU 152 Ras- CUACCACAA Gtt Tag#3 G12V- CCAACAGCUCCA 103 UGGUAGUUGGAGCUGUUG 153 Ras- ACUACCACA Gtt Tag#4 G12V- GCCAACAGCUCC 104 GGUAGUUGGAGCUGUUGG 154 Ras- AACUACCAC Ctt Tag#5 G12V- CGCCAACAGCUC 105 GUAGUUGGAGCUGUUGGC 155 Ras- CAACUACCA Gtt Tag#6 G12V- ACGCCAACAGCU 106 UAGUUGGAGCUGUUGGCG 156 Ras- CCAACUACC Utt Tag#7 G12V- UACGCCAACAGC 107 AGUUGGAGCUGUUGGCGU 157 Ras- UCCAACUAC Att Tag#8 G12V- CUACGCCAACAG 108 GUUGGAGCUGUUGGCGUA 158 Ras- CUCCAACUA Gtt Tag#9 G12V- CCUACGCCAACA 109 UUGGAGCUGUUGGCGUAG 159 Ras- GCUCCAACU Gtt Tag#10 G12V- GCCUACGCCAAC 110 UGGAGCUGUUGGCGUAGG 160 Ras- AGCUCCAAC Ctt Tag#11 G12V- UGCCUACGCCAA 111 GGAGCUGUUGGCGUAGGC 161 Ras- CAGCUCCAA Att Tag#12 G12V- UUGCCUACGCCA 112 GAGCUGUUGGCGUAGGCA 162 Ras- ACAGCUCCA Att Tag#13 G12V- CUUGCCUACGCC 113 AGCUGUUGGCGUAGGCAA 163 Ras- AACAGCUCC Gtt Tag#14 G12V- UCUUGCCUACGC 114 GCUGUUGGCGUAGGCAAG 164 Ras- CAACAGCUC Att Tag#15 G12V- CUCUUGCCUACG 115 CUGUUGGCGUAGGCAAGA 165 Ras- CCAACAGCU Gtt Tag#16 G12V- ACUCUUGCCUAC 116 UGUUGGCGUAGGCAAGAG 166 Ras- GCCAACAGC Utt Tag#17 G12V- CACUCUUGCCUA 117 GUUGGCGUAGGCAAGAGU 167 Ras- CGCCAACAG Gtt Tag#18 G12V- GCACUCUUGCCU 118 UUGGCGUAGGCAAGAGUG 168 Ras- AGGCCAACA Ctt Tag#19 G12V- GGCACUCUUGCC 119 UGGCGUAGGCAAGAGUGC 169 Ras- UACGCCAAC Ctt Tag#20 G12V- UGGCACUCUUGC 120 GGCGUAGGCAAGAGUGCC 170 Ras- CUACGCCAA Att Tag#21 V600E- UCUGUAGCUAGA 183 AUUUUGGUCUAGCUACAG 204 Raf- CCAAAAUCA att Tag#1 V600E- CUCUGUAGCUAG 184 UUUUGGUCUAGCUACAGa 205 Raf- ACCAAAAUC Gtt Tag#2 V600E- UCUCUGUAGCUA 185 UUUGGUCUAGCUACAGaG 206 Raf- GACCAAAAU Att Tag#3 V600E- UUCUCUGUAGCU 186 UUGGUCUAGCUACAGaGA 207 Raf- AGACCAAAA Att Tag#4 V600E- UUUCUCUGUAGC 187 UGGUCUAGCUACAGaGAA 208 Raf- UAGACCAAA Att Tag#5 V600E- AUUUCUCUGUAG 188 GGUCUAGCUACAGaGAAA 209 Raf- CUAGACCAA Utt Tag#6 V600E- GAUUUCUCUGUA 189 GUCUAGCUACAGaGAAAU 210 Raf- GCUAGACCA Ctt Tag#7 V600E- AGAUUUCUCUGU 190 UCUAGCUACAGaGAAAUC 211 Raf- AGCUAGACC Utt Tag#8 V600E- GAGAUUUCUCUG 191 CUAGCUACAGaGAAAUCU 212 Raf- UAGCUAGAC Ctt Tag#9 V600E- CGAGAUUUCUCU 192 UAGCUACAGaGAAAUCUC 213 Raf- GUAGCUAGA Gtt Tag#10 V600E- UCGAGAUUUCUC 193 AGCUACAGaGAAAUCUCG 214 Raf- UGUAGCUAG Att Tag#11 V600E- AUCGAGAUUUCU 194 GCUACAGaGAAAUCUCGA 215 Raf- CUGUAGCUA Utt Tag#12 V600E- CAUCGAGAUUUC 195 CUACAGaGAAAUCUCGAU 216 Raf- UCUGUAGCU Gtt Tag#13 V600E- CCAUCGAGAUUU 196 UACAGaGAAAUCUCGAUG 217 Raf- CUCUGUAGC Gtt Tag#14 V600E- UCCAUCGAGAUU 197 ACAGaGAAAUCUCGAUGG 218 Raf- UCUCUGUAG Att Tag#15 V600E- CUCCAUCGAGAU 198 CAGaGAAAUCUCGAUGGA 219 Raf- UUCUCUGUA Gtt Tag#16 V600E- ACUCCAUCGAGA 199 AGaGAAAUCUCGAUGGAG 220 Raf- UUUCUCUGU Utt Tag#17 V600E- CACUGCAUCGAG 200 GaGAAAUCUCGAUGGAGU 221 Raf- AUUUCUCUG Gtt Tag#18 V600E- CCACUCCAUCGA 201 aGAAAUCUCGAUGGAGUG 222 Raf- GAUUUCUCU Gtt Tag#19 V600E- CCCACUCCAUCG 202 GAAAUCUCGAUGGAGUGG 223 Raf- AGAUUUCUC Gtt Tag#20 V600E- ACCCACUCCAUC 203 AAAUCUCGAUGGAGUGGG 224 Raf- GAGAUUUCU Utt Tag#21

[0377] Oligos Used for G12V-TAG Specific siRNA Screening

TABLE-US-00008 EcoR1 G12V-Endotag BglII TOP,  (SEQ ID NO: 171) AATTCCATATGCTTGTGGTAGTTGGAGCTGtTGGCGTAGGCAAGAG TGCCA EcoR1 G12V-Endotag BglII Bottom,  (SEQ ID NO: 172) gatcTGGCACTCTTGCCTACGCCAaCAGCTCCAACTACCACAAGCA TATGg EcoR1 WT-Endotag BglII TOP,  (SEQ ID NO: 173) AATTCCATATGCTTGTGGTAGTTGGAGCTGgTGGCGTAGGCAAGAG TGCCA EcoR1 WT-Endotag BglII Bottom,  (SEQ ID NO: 174) gatcTGGCACTCTTGCCTACGCCAcCAGCTCCAACTACCACAAGCA TATGg

[0378] Oligos Used for V600E-TAG Specific siRNA Screening

TABLE-US-00009 Bam-BRAFwtTAG-Not-TOP  (SEQ ID NO: 225) GATCCCAATTGTAGTTAGTTTAGACCGGTTGATTTTGGTCTAGCTA CAGtGAAATCTCGATGGAGTGGGTACGCGTAGATCTTATTTGC Bam-BRAFwtTAG-Not-Bottom  (SEQ ID NO: 226) ggccGCAAATAAGATCTACGCGTACCCACTCCATCGAGATTTCaCT GTAGCTAGACCAAAATCAACCGGTCTAAACTAACTACAATTGG Bam-BRAFv600eTAG-Not-TOP  (SEQ ID NO: 227) GATCCCAATTGTAGTTAGTTTAGACCGGTTGATTTTGGTCTAGCTA CAGaGAAATCTCGATGGAGTGGGTACGCGTAGATCTTATTTGC Bam-BRAFv600eTAG-Not-Bottom  (SEQ ID NO: 228) ggccGCAAATAAGATCTACGCGTACCCACTCCATCGAGATTTCtCT GTAGCTAGACCAAAATCAACCGGTCTAAACTAACTACAATTGG

[0379] List of Tandem-HTRF Antibodies [0380] HA-XL, 610HAXLB, Flag-Tb, 61FG2TLB, and MYC-Eu, 61MYCKLA, Cisbio [0381] V5-d2, 64CUSDAYE, Cisbio (custom labelling of MA5-15253 (V5), Perbio) [0382] AB3-d2 64CUSDAZE, Cisbio (custom labelling of RB-010-PABX (AB3), Fisher scientific) [0383] AB1-d2 64CUSDAZE, Cisbio (custom labelling of MS-210-PABX (AB1), Fisher scientific) [0384] SC-450-d2 64CUSDAZE, Cisbio (custom labelling of SC-450, Santa Cruz)

[0385] Analysis of sequence homology between siRNAs and mouse CycD1, CycD2, CycD3 and CycE1 cDNA:

TABLE-US-00010 Flag siRNA sequence: (SEQ ID NO: 175) GACTACAAGGACGACGATGAC Potential off-target hybridization sites: 5. HA siRNA sequence: (SEQ ID NO: 176) CCCCTACGACGTGCCCGACTA Potential off-target hybridization sites: 7. Nat siRNA sequence: (SEQ ID NO: 177) CCACAGATGTGAAGTTCATTT Potential off-target hybridization sites: 23. Qiagen siRNA sequence: (SEQ ID NO: 178) AACACCAGCTCCTGTGCTGCG Potential off-target hybridization sites: 18 Life siRNA sequence: (SEQ ID NO: 179) CAGGAACAGATTGAAGCCCTT Potential off-target hybridization sites: 16.