KINASE AND UBIQUITIN LIGASE INHIBITORS AND USES THEREOF

Abstract

A suppressor or inhibitor of expression and/or function of 4 a gene, a kinase or ubiquitin ligase, for use in the treatment of a protein conformational disorder is provided.

Claims

1. A method of treating and/or preventing a protein conformational disorder comprising administering to a patient in need thereof a therapeutically effective amount of & molecule which suppresses or inhibits the expression and/or function of a gene selected from the group consisting of: JNK2/MAPK9, CAMK1, CDC42, HPK1/MAP4K1, PRKAA1(AMPK), PRKAA2(AMPK), RAC2, TGFBR-2, MAPK11, MAPK14, MAPK8/JNK1, CALML5, ITPR2, RNF215, UBOX5, SART1, PDGFRB, CD2BP2, CKII/CSNK2A1, ASB8, STAG2, FBXO7, PIK3CB, MLK3/MAP3K11, CTDSP1, VEGFR2/KDR, GTSE1, PRPF8, MED1, OSMR, DSN1, NFKB2, SENP6, PDGFRA, MKK7/MAP2K7, PIK3CG, MAPK15, NUP50, CAMKK2, MIS18BP1/C14orf106, YWHAH, VEGFR1/FLT1, TEP1, MED13 and PROKR1 with the proviso that said molecule is not oxozeanol, SU5402 and SU6668.

2. The method molecule for use according to claim 1, wherein said molecule does not suppress or inhibit the expression and/or function of at a gene selected from the group consisting of: FGFBP1, DCLK1, DNAJC2, S100A7, MKK1/MAP2K1, BIN2, RBM7, ERBB4, MKI67, MKK2/MAP2K2, PIK3CD, MKK3/MAP2K3, MKK4/MAP2K4, AKAP8 and CYC1.

3. The method according to claim 1, wherein the molecule: a) selectively suppresses or inhibits the expression and/or function of at least one: i) of the kinases or of the kinase regulators selected from the group consisting of: JNK2/MAPK9, CAMK1, CAMKK2, CDC42, CKII/CSNK2A1, HPK1/MAP4K1, MAPK15, MKK7/MAP2K7, MLK3/MAP3K11, PDGFRA, PDGFRB, PIK3CB, PIK3CG, PRKAA1(AMPK), PRKAA2(AMPK), RAC2, TGFBR-2, VEGFR1/FLT1, VEGFR2/KDR, MAPK11, MAPK14, MAPK8/JNK1, CALML5, ITPR2 or ii) of ubiquitin ligases selected from the group consisting of: RNF215, UBXO5, ASB8, FBXO7 and b) does not suppress or inhibit the expression and/or function of a kinase selected from the group consisting of: ERBB4, MKK1/MAP2K1, MKK2/MAP2K2, MKK3/MAP2K3, MKK4/MAP2K4 and PIK3CD.

4. The method according to claim 1, wherein the protein conformational disorder is selected from cystic fibrosis or Wilson disease.

5. The method according to claim 1, wherein the molecule selectively suppresses or inhibits the expression and/or function of one of the following combinations of kinases selected from the group consisting of MLK3/MAP3K11 and CAMKK2, MLK3/MAP3K11 and CKII/CSNK2A1, MLK3/MAP3K11 and RNF215, CAMKK2 and CKII/CSNK2A1.

6. The method according to claim 1, wherein the molecule selectively suppresses or inhibits the expression and/or function selected from the group consisting of: JNK2/MAPK9, CAMK1, CAMKK2, CDC42, CKII/CSNK2A1, HPK1/MAP4K1, MAPK15, MKK7/MAP2K7, MLK3/MAP3K11, PDGFRA, PDGFRB, PIK3CB, PIK3CG, PRKAA1(AMPK), PRKAA2(AMPK), RAC2, TGFBR-2, VEGFR1/FLT1 and VEGFR2/KDR, or any combination thereof and wherein the protein conformational disorder is cystic fibrosis.

7. The method according to claim 1, wherein the molecule selectively suppresses or inhibits the expression and/or function selected from the group consisting of: MLK3/MAP3K11, MAPK8 (JNK1), MAPK11 (p38) and MAPK14 (p38), or any combination thereof and wherein the protein conformational disorder is Wilson disease.

8. The method according to claim 1, wherein the molecule is selected from the group consisting of: a) a polypeptide; b) a polynucleotide coding for said polypeptide; c) a polynucleotide able to inhibit the expression of said gene; d) a vector comprising or expressing the polynucleotide as defined in b-c); e) a host cell genetically engineered expressing said polypeptide or said polynucleotide; and f) a small molecule.

9. The method according to claim 1 wherein the molecule is selected from the group consisting of: JNKi IX, SP600125/JNKi II, BIRB-796, VX-745, JNKi XI, SB202190, Pazopanib, Dovitinib lactate, Bexarotene, Flunarizine, Cannabidiol, CPI-1189 and ENMD-2076.

10. The method according to claim 1 wherein the molecule is selected from the group consisting of: JNKi IX, SP600125/JNKi II, JNKi XI, Pazopanib, Dovitinib lactate, Bexarotene, and wherein the protein conformational disorder is cystic fibrosis.

11. The method according to claim 8, wherein the molecule is selected from the group consisting of: VX-745, BIRB-796, JNKi II, SB202190, Bexarotene, Cannabidiol, CPI-1189 and ENMD-2076 wherein the protein conformational disorder is Wilson disease.

12. The method according to claim 1, wherein said polynucleotide able to inhibit the expression of said gene is an RNAi agent targeting said gene.

13. The method according to claim 1, in combination with a therapeutic agent.

14. The method according to claim 13, wherein the therapeutic agent is the pharmacochaperone VX-809 and the protein conformational disorder is cystic fibrosis.

15. (canceled)

16. A method of treating and/or preventing a protein conformational disorder comprising administering to a patient in need thereof a therapeutically effective amount of a molecule as defined in claim 1.

Description

[0044] FIG. 1: Corrector drugs modulate a set of CORE genes.

[0045] A. Schema of the FIT method. The upregulated (top 20%) and downregulated genes (bottom 20%) were fuzzy intersected to identify CORE genes. B. To obtain optimal fuzzy cut-off for the analysis, the corrector drug profiles (MANTRA dataset) as well as random profiles from MANTRA database were intersected with variable fuzzy cut-offs (represented as number of drugs out of 11). The enlargement (inset) shows that at the optimal fuzzy cut-off (0.7; 8 out of 11 drugs), the signal-to-noise ratio was close to 3 (108 probe-sets in the corrector drug intersection vs 32 in the random). C. Next, with a fuzzy cut-off of 0.7, the number of random drug profiles used was varied, and the number of probe-sets present in the intersection is shown. D. Using the optimal parameters (see B, C) the FIT analysis resulted in 402 upregulated and 219 downregulated CORE genes. E. The number of CORE genes associated with the enriched GO terms is shown. Those genes that did not associate with enriched GO terms were excluded from the chart. F. Protein-protein interactions between the CORE and the proteostasis genes (restricted to those that occur between the two groups) are shown.

[0046] FIG. 2: Validation of the selected CORE genes.

[0047] A-D. CFBE cells were treated with siRNAs targeting CORE genes and changes in F508del-CFTR proteostasis monitored by western blotting. The fold change in the levels of band C obtained by downregulating negative correction (A) and positive correction (D) genes and the fold change in levels of band B (B) and band C/band B ratio (C) after downregulation of the negative correction genes are shown. The effects of negative control siRNAs (dashed line) and VX-809 (dark grey) are indicated. E. The validated CORE genes were assembled into coherent networks based on information from databases. Non-directional interactions denote protein-protein interaction, directional interactions represent phosphorylation cascades and dashed arrows indicate indirect connections through intermediaries. F. Treatment of CFBE cells with mitoxantrone (2.5 to 20 M for 48 h), a potential corrector identified using downregulation of anti-corrector genes as selection criteria, increased the levels of both band C and band B. G. Treatment of CFBE cells with the indicated combinations of siRNAs targeting CORE genes led to a synergistic increase in the band C levels. A representative blot is shown in the insert.

[0048] FIG. 3: Downregulation of CORE genes rescues F508del-CFTR more efficiently than the corrector drugs used originally, without altering the F508del-CFTR mRNA levels (related to FIG. 2). A. CFBE cells were treated with indicated corrector drugs for 48 h and then lysed and prepared for western blotting, to assay the rescue of F508del-CFTR from ERQC. The changes in the levels of band C after drug treatment are shown as meanSEM (n>3).

[0049] B. CFBE cells were treated with indicated siRNAs (targeting the anti-correction genes) for 72 h, and then total RNA from the cells was purified. The levels of CFTR mRNA were then quantitated by RT-PCR. The data is presented as mRNA levels relative to the negative control siRNAs. The values are expressed as meanSEM (n=4). C. Representative blot used for quantitation's represented in FIG. 2A-C. D. Representative blot used for quantitation's represented in FIG. 2D

[0050] FIG. 4: Delineation of the MLK3 pathway branch that controls F508del-CFTR proteostasis.

[0051] A. CFBE cells were treated with indicated siRNAs targeting the upstream activators of MLK3 and their effect on F508del-CFTR proteostasis monitored by western blotting. The fold change in band C levels is shown. Reduction in TGF receptor, HPK, CDC42 and RAC2 levels rescued F508del-CFTR from ERQC. The rescue obtained with TNFR2 siRNA was quite variable and so was not considered further. B. JNK isoforms were tested for their effect on F508del-CFTR proteostasis after siRNA-mediated downregulation of their levels. Downregulation of JNK2 leads to efficient rescue of F508del-CFTR that is comparable to that obtained with MLK3. C. CFBE cells were transfected with activators of the MLK3 pathway to study their effect on F508del-CFTR proteostasis. All of them reduced the levels of both band C (not shown) and band B of F508del-CFTR. The corresponding increase in the levels of phospho-c-jun indicates an increase activation of the MLK3 pathway activity. D-E. Schematic representation of the proposed MLK3 (D) and CAMKK2 (E) pathways that regulate F508del-CFTR proteostasis. The directional interactions proposed between the components of the pathways are based on published literature.

[0052] FIG. 5: Delineation of the MLK3 and CAMKK2 pathway branches that regulate F508del-CFTR proteostasis (related to FIG. 4).

[0053] A. HeLa cells [HeLa cells stably expressing HA-tagged F508del-CFTR] were treated with indicated siRNAs targeting MLK3 pathway components including p38 MAPK (mix of siRNAs targeting all 4 isoforms) and JNK (mix of siRNAs targeting all 3 JNKs). The effect on F508del-CFTR proteostasis monitored by western blotting. Fold Change in the levels of band C was quantitated and represented as meanSEM (n>3), with a representative blot shown in the insert. The downregulation of the MLK3 pathway components (including p38 MAPK) leads to the rescue of F508del-CFTR in HeLa cells. SiRNAs targeting Rma1 and Aha1 used as positive controls for rescue of F508del-CFTR.

[0054] B. Screening for F508del-CFTR proteostasis regulators among the CORE genes led to the identification of CAMKK2 as an anti-correction hit. Three downstream components and 9 upstream components of the CAMKK2 signaling pathway (as derived from literature mining) were tested, by siRNA-mediated downregulation, for their role in regulation of F508del-CFTR proteostasis. CFBE cells were treated with the indicated siRNAs for 72 h and their effect on F508del-CFTR proteostasis monitored by western blotting. Four of them (CALML5, ITPR2, CAMK1 and AMPK [by a mix of siRNAs targeting PRKAA1 and PRKAA2]) rescued F508del-CFTR from ERQC as seen by an increase in band C levels. C. The changes in the levels of band C from (B) were quantitated and are represented as meanSEM (n>3). See FIG. 4 for a representation of the derived CAMKK2 pathway that regulates F508del-CFTR proteostasis.

[0055] FIG. 6: MLK3 pathway regulates the degradation of F508del-CFTR.

[0056] A-B. CFBE cells pretreated with siRNAs were treated with CHX (50 g/mL) for indicated times and the levels of band B of F508del-CFTR was monitored (A). The levels were quantitated and represented in (B). Downregulation of MLK3 or JNK2 reduced the kinetics of reduction of band B of F508del-CFTR. C-D. CHX chase assay (see above) after overexpression of the activators of MLK3 pathway. The activation of MLK3 pathway increases the rate of degradation of band B (C). Quantitation of the blot is shown in (D). The results are representative of 3 independent experiments. E-F. CFBE cells were treated with indicated siRNAs followed by incubation at 26 C. for 6 h followed by shift to 37 C. for the indicated time periods. The changes in band C levels were monitored as measure of PQC (C). See (F) for quantitation of band C levels. G-H. PQC assay (see above) after overexpression of CDC42 or JNK2 shows an increased rate of degradation of band C (G) upon CDC42 overexpression. JNK2 overexpression has no effect on the PQC of F508del-CFTR. The blots were quantified and presented in (H).

[0057] FIG. 7: Characterization of the mode of action of the MLK3 pathway on F508del-CFTR proteostasis (related to FIG. 6). A-B. CFBE cells treated with MLK3 siRNA were pulsed with radioactive [35S]-cysteine and methionine for 15 min, and then chased for the indicated times. CFTR was immunoprecipitated and processed for autoradiography (A). The signals corresponding to band B from (A) were quantitated and presented in (B). The data are representative of 2 independent experiments. Note the reduced degradation of F508del-CFTR upon downregulation of MLK3. C. Down-regulation of the MLK3-JNK pathway does not affect the activity of proteasomes. CFBE cells treated with MLK3 or JNK2 siRNA for 72 h were transfected with Proteasome ZsProsensor-1 for the final 24 h, and the levels Proteasome ZsProsensor-1 monitored by fluorescence microscopy. Treatment with MG132 (20 g/ml for 3 h), a proteasomal inhibitor, was used as a positive control. While treatment with MG132 increases the fluorescence levels of ZsProsensor-1 compared to untreated cells indicating a reduced proteasome activity, downregulation of MLK3 or JNK2 did not change the levels of fluorescence, suggesting that proteasome activity is not changed under these conditions. D. CFBE cells were treated with MLK3 or JNK2 siRNA and processed for western blotting to monitor the accumulation of poly ubiquitinated proteins. There was no change in the levels of poly ubiquitinated proteins suggesting that these treatments do not affect proteasome activity. E. Down-regulation of MLK3 does not affect the folding of F508del-CFTR. CFBE cells expressing wild type CFTR or F508del-CFTR were treated with MLK3 siRNA as indicated. Untreated CFBE cells incubated at 26 C. for 24 h were used as a positive control for the promotion of folding. Membrane fractions from the cells were isolated and subjected to trypsin digestion for 10 min on ice, followed by western blotting with M3A7 antibody that recognizes the NBD2 domain of F508del-CFTR, or with 3G11 antibody that recognizes NBD1. The wild-type CFTR and its NBD domains show more resistance to trypsin digestion compared to F508del-CFTR. There was no change in the stability of F508del-CFTR or its NBD domains upon down-regulation of MLK3, while the low temperature treatment enhanced the stability of F508del-CFTR and its NBD1 domain. The Western blots are representative of at least 3 different experiments.

[0058] FIG. 8: Inhibitors of the MLK3 pathway rescue F508de1-CFTR.

[0059] (A) CFBE cells were treated with the indicated inhibitors of the MLK3 pathway or VX-809 for 48 h, and the rescue of F508del-CFTR from was monitored by increase in band C western blotting.

[0060] (B) Fold changes in the levels of band C, normalized concentration refers to concentration [VX-809, JNKi IX and Oxozeaenol (1.25, 2.5, 5, 10 M), JNKi II (6.25, 12.5, 25, 50 M), JNKi XI, Pazopanib, Dovitinib lactate and Bexarotene (3.12, 6.25, 12.5, 25 M)] values that were normalized to the maximum used concentrations of the respective drugs. Also refer panel A for concentrations (M) [SEM (n>3)]. C. CFBE cells were treated with inhibitors of the MLK3 pathway and/or VX-809 (5 M) for 48 h and changes in band C levels monitored. The concentrations of the MLK3 pathway inhibitors used were: JNKi II (12.5 M), JNKi IX (50 M), JNKi XI (25 M) and oxozeaenol (5 M). Wild type CFTR (wt-CFTR) was used as a control. D. Quantitation of band C levels from (C), normalized to the levels of band C after VX-809 treatment are shown. The results show that synergy obtained between the MLK3 pathway inhibitors and VX-809 brings the levels of band C to about 40% of the wild type levels.

[0061] FIG. 9: Small-molecule inhibitors of the MLK3 pathway rescue F508del-CFTR and other structurally related mutant proteins from degradation (related to FIG. 8 and Table 5).

[0062] A. CFBE cells were treated with indicated JNK inhibitors for 24 h and processed for western blotting. The levels of phospho-c-jun as a measure of JNK inhibition was monitored. MLK3 pathway inhibitors reduce phospho-c-jun levels efficiently indicating a strong reduction in the activity of JNK and hence presumably of the MLK3 pathway. B. CFBE cells were treated with TAK1 or MLK3 siRNA as indicated and changes in F508del-CFTR proteostasis were monitored by western blotting. TAK1 does not regulate F508del-CFTR proteostasis, as evidenced by the absence of change in the levels of bands C or B. The fold change in the band C levels were quantitated and plotted as meanSD (n=2).

[0063] C. CFBE cells were treated with 5 M oxozeaenol for 48 h, or with MLK3 siRNA, or with both, and the correction of the F508del-CFTR folding/trafficking defect was monitored by changes in the levels of band C. There was no additive effect observed with the combination of MLK3 downregulation and oxozeaenol treatment. The quantitated band C levels are expressed as meanSD (n>3). D. CFBE cells were treated with 5 M oxozeaenol for 24 h, and the activity of the JNK pathway was measured by western blotting for phospho c-jun levels and F508del-CFTR. The levels of phospho c-jun were reduced suggesting that oxozeaenol leads to a reduction in the activity of JNK. The increase in band C levels of F508del-CFTR show that the reduction in the activity of JNK is accompanied by a rescue of F508del-CFTR from ERQC. E. CFBE cells were treated with flunarizine (at concentrations 6.25-50 M) targeting the CAMKK2 pathway for 48 h and the effect on F508del-CFTR proteostasis measured by western blotting. Treatment with flunarizine increased the levels of band C of F508del-CFTR. Other small molecules known to inhibit the CAMKK2 pathway components (verapamil and STO-609) did not show any effect on correction of F508del-CFTR. F. CFBE cells transiently transfected with the P-glycoprotein mutant (P-gp DY490), the NCC mutant (R948X), or the hERG mutant (G601S) were treated with JNKi II for 24 h, and the effect of the drug on their proteostasis monitored by western blotting. While the trafficking of P-gp DY490 out of the ER was enhanced by this treatment (seen as an increase in the Golgi-associated band C, indicated by arrows), other mutants are subjected to enhanced degradation upon drug treatment, as shown by a decrease in the levels of both bands B and C.

[0064] FIG. 10: Small molecule inhibitors of MLK3 pathway rescue the channel function of F508del-CFTR. A. F508del-CFTR and Halide sensitive YFP (Galietta et al., 2001) expressing CFBE (CFBE-YFP) cells were treated with MLK3 pathway inhibitors and/or VX-809 for 48 h, and the anion transport measured as described in the Materials and methods (Halide sensitive YFP assay for CFTR activity). The rate constants of the decrease in YFP fluorescence (K), a measure of anion conductance, after inhibitor treatments are shown. The data are expressed as meanSEM (n>3). B. Anion transport was measured in CFBE-YFP cells after downregulating the MLK3 pathway activity by siRNA-mediated knockdown MLK3 or JNK2. The rate constants of the decrease in YFP fluorescence (K), a measure of anion conductance, after downregulation of the indicated MLK3pathway components are shown. The data are expressed as meanSEM (n>3). Treatment with VX-809 was used as a positive control for the rescue. C. CFBE41o-cells were grown under polarising conditions before addition of oxozeaenol at indicated concentrations for 48 h, followed by the measurement of short circuit currents using Ussing chamber assays. The columns show the measured values of the short circuit current after oxozeaenol treatment at the indicated concentrations. The values are meanSEM (n>3).

[0065] FIG. 11 Silencing of several MLK3 pathways genes corrects localization and trafficking of the ATP7BH1069Q mutant.

[0066] (A) HeLa cells were incubated with siRNA, which target specific genes (indicated in graph) belonging to p38 and JNK pathways, then infected with Ad-ATP7BH1069Q-GFP (Chesi et al., 2016) and incubated for 2 h with 100 M CuSO4. Fixed cells were then labeled for TGN46 and visualized under confocal microscope. Silencing of MAPK8, MAPK11, MAPK14 or MAP3K11 results in rescue of ATP7BH1069Q from the ER and its movement to post-Golgi vesicles (arrows) and PM. (B) Cells were treated as in panel B. The percentage of the cells (averageSD, n=10 fields) with ATP7BH1069Q signal in the ER was calculated. RNAi of MAPK8, MAPK11, MAPK14 and MAP3K11 reduced the percentage of the cells exhibiting ATP7BH1069Q in the ER. Scale bar: 4.7 m.

[0067] FIG. 12 Inhibitors of MLK3 pathway inhibitors correct localization and trafficking of ATP7BH1069Q mutant.

[0068] (A) HeLa cells were infected with Ad-ATP7BWT-GFP (Chesi et al., 2016) or Ad-ATP7BH1069Q-GFP, incubated overnight with 200 M BCS, and incubated for an additional 2 h with 100 M CuSO4. In response to Cu, ATP7BWT traffics from the Golgi to PM and vesicle, while ATP7BH1069Q are retained within the ER under high Cu conditions. Addition of p38 inhibitor SB202190 (5 M), VX-745(1 M), JNK inhibitor SP600125 (2 M) and Oxozeaenol (5 M) (as indicated in corresponding panels) to the cells 24 h corrects ATP7BH1069Q from the ER to PM and vesicles (arrows) (B) Cells were treated as in panel A. The percentage of the cells (averageSD, n=50 fields) with an ATP7B signal in the ER, were calculated. The p38 inhibitors SB202190 (5 M), VX-745(1 M), JNK inhibitor SP600125 (2 M) and Oxozeaenol (5 M) reduced the percentage of the cells exhibiting ATP7BH1069Q in the ER and increases the number of cells in which ATP7B was corrected to PM and vesicles.

[0069] FIG. 13: Small-molecule inhibitors of MLK3-JNK pathway rescue ATP7B H1069Q localization to the Golgi apparatus.

[0070] Transiently ATP7B H1069Q-GFP expressing HeLa cells (A), HEPG2 cells and (C) human primary hepatocytes (E) treated with the inhibitors for 24 hours, cells are processed for immunofluorescence assay to measure the arrival of the ATP7B wt and H1069Q mutant from the ER to Golgi compartment. A, C, E) Normalized Golgi fluorescence of ATP7B is measured and plotted (n >50 cells). B, D, F) EC50 and recue effect (%) compared to the level ATP7B WT Golgi fluorescence calculated from (A), (C) and (E) respectively. Inhibitor SB202190 and VX-745 (or VX745) was used as positive control in our rescue assay and it has been shown to rescue the transport and function of ATP7B H1069Q (Chesi, Hegde et al. 2016). All the inhibitor drugs except SB202190 are in clinical trial for treatment of various other diseases.

[0071] FIG. 14: The VX-745 and BIRB-796 correctors reduce Copper levels in cells expressing ATP7BH1069Q mutant. HepG2 cells overexpressing indicated ATP7B constructs were incubated with 1 M each of VX-745, BIRB-796 for 24 hours, for the last 2 hours of incubation with the inhibitors 200 M copper sulphate is added to the cells with or without 100 M BCS (copper chelator) as indicated. Cells were incubated with HBSS for 3 hours before lysed and used for copper estimation. Fold change in copper normalized to control is plotted (n=4). Both VX-745 and BIRB-796 reduced Copper levels in ATP7BH1069Q expressing cells (BCS is used in the assay as a control for the sensitivity of the assay).

EXAMPLES

[0072] Materials and Methods

[0073] Cell Culture, Antibodies, Plasmids and Transfection

[0074] CFBE cells stably expressing wild type CFTR or F508del-CFTR (Bebok et al. 2005) and stably expressing halide sensitive YFP (Pedemonte et al. 2005) and HeLa cells stably expressing HA-tagged F508del-CFTR (Okiyoneda et al. 2010) were used. CFBE cells were cultured in Minimal Essential Medium supplemented with 10% foetal bovine serum, non-essential amino acids, glutamine, penicillin/streptomycin and 2 g/ml puromycin. This media additionally supplemented with 50 g/ml G418 was used for the CFBE-YFP cells. HeLa cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% foetal bovine serum, glutamine, penicillin/streptomycin and 1 g/ml puromycin. The antibodies used were: anti-phospho-c-jun (Cell Signaling Technology), monoclonal anti-HA, anti-actin and anti-tubulin (Sigma), rat anti-CFTR (3G11; CFTR Folding Consortium), mouse monoclonal anti-CFTR (M3A7), HRP-conjugated anti mouse, rabbit and rat IgG (Merckmillipore) and Anti-Na/K+ATPase 1 (Thermoscientific). The plasmids used were: JNK2 (pCDNA3 Flag MKK7B2Jnk2a2; Addgene plasmid #19727) and MKK7 (pCDNA3 Flag MKK7b1; Addgene plasmid #14622,) from Roger Davis (University of Massachusetts Medical School, Worcester, USA), ZsProSensor-1 proteasome sensor (Clontech), VSVG tagged with GFP (Jennifer Lippincott-Schwartz, NICHD, NIH, Bethesda, USA), Cdc42 (A. Hall, Sloan-Kettering Institute, New York, N.Y., USA), P-glycoprotein wild type, G268V and DY490 mutants (David M. Clarke, University of Toronto, Canada) and hERG wild type and G601S mutant (Alvin Shrier, McGill University, Montreal, Canada).

[0075] The reagents used include: VX-809 (Selleckchem), JNKi II (SP600125), JNKi IX and JNKi XI (Merck Millipore), oxozeaenol (Tocris Bioscience), siRNAs (Table 3), lipofectamine 2000 (Invitrogen) and ECL (Luminata crescendo from Merck Millipore), BIRB-796 (Sigma), VX-745 (Sigma), SB202190 (Sigma), pazopanib, dovitinib lactate (Sigma), bexarotene (Sigma), flunarizine (Sigma), cannabidiol (Sigma), CPI-1189 (Sigma) and ENMD-2076 (Sigma).

[0076] Analysis of Corrector-Induced Gene Expression Changes by Microarray

[0077] Polarised CFBE410-cells (cystic fibrosis bronchial epithelial cells) cultured at the air-liquid interface were treated with the corrector drugs of interest (CFBE dataset, Table 4) for 24 h. Total RNA was extracted and hybridization was carried out on to Whole Human Genome 44 K arrays (Agilent Technologies, product G4112A) following the manufacturer's protocol. See (Zhang et al. 2012) for experimental details. The microarray data for ouabain and low temperature treatments have been published (Zhang et al. 2012).

[0078] FIT Analysis of Microarray Profiles

[0079] The microarrays from the connectivity map database (https://www.broadinstitute.org/cmap/) were processed to produce prototype ranked lists (PRLs) (Iorio et al. 2010). In these PRLs, cell line specific responses are diluted, thus summarising consensual transcriptional responses to drug treatment. In each PRL, microarray probe-sets are ordered from the most upregulated to most downregulated one. Inventors downloaded PRLs for the whole panel of small molecules in the connectivity map (www.connectivitymap.org) from which the MANTRA database is derived (http://mantra.tigem.it/). Inventors used these in conjunction with ranked lists of probe sets based on fold-changes (and assembled following the guidelines provided in (Iorio et al., 2010)) from microarray profiles that inventors generated in house (CFBE dataset). The FIT analysis identifies microarray probe-sets that tend to respond consistently to a group of drugs (see also (Iorio et al. 2010) for description of a similar method). The top and bottom 20% of the probe-sets (corresponding to the up- and downregulated probe-sets respectively) were used for the analysis. The 20% cut-off was used since the merging of individual gene expression profiles into PRLs precludes the application of other thresholds based on fold-change (or p-value) to identify significantly differentially expressed genes. To build a null model against which the significance of the final genes sets can be tested (as detailed below), a fixed number of PRLs (N) from the MANTRA dataset were randomly selected and the upregulated or downregulated probe-sets from this selection were intersected by varying the fuzzy cut-off threshold (i.e. the ratio of drugs that a given probe-set should transcriptionally respond to, in order to be considered consistently regulated, hence to be included in the fuzzy intersection). After 1000 of these iterations, inventors derived an empirical null distribution of the number of probes included in the resulting fuzzy intersections and used it for p-value assignments (FIG. 1B). For the CFBE dataset, (generated on an Agilent platform, which is different from that used for the connectivity map and MANTRA database) inventors derived this null distribution by randomly permuting all the individual probes. Finally, inventors determined the optimal fuzzy cut-off values for the transcriptional profiles elicited by the corrector drugs (11 contained in MANTRA and 13 in the CFBE dataset). Briefly, inventors selected the value such that the number of probes present in the final fuzzy intersection was at least 3 fold higher than that expected by random chance and its p-value <0.05 (according to the computed null models). By using this method, no significantly upregulated probes from the MANTRA dataset were identified across all of the range of tested fuzzy cut-offs. For the downregulated probe-sets a fuzzy cut-off of 8 (out of 11 corrector drugs) or above produced significant fuzzy intersection of probe-sets. For the CFBE dataset, a significant cut-off of 6 drugs (out of 13) and above was identified. To further optimise the selection of these cut-offs, inventors chose the maximal cut-off yielding a fuzzy intersection of probe-sets enriched in one or more Gene Ontology terms. With this criterion, inventors obtained a final cut-off value of 8 for the MANTRA down-regulated probe set and cut-off of 9 for the CFBE dataset. Intersecting the corrector-induced gene expression profiles using this optimal fuzzy cut-off resulted in 541 upregulated probe-sets (mapping 402 unique genes) and 191 downregulated probe-sets (mapping to 117 unique genes) for the CFBE dataset, and 108 downregulated probe-sets (mapping 102 genes) for the MANTRA dataset. Note, that most of the CORE genes (about 500 out of the 600) are derived from the CFBE dataset. This inventors suppose is due to the use of PRLs in the case of cMAP dataset and use of data derived from a single cell line in the case of CFBE dataset. The use of single cell line derived data can potentially lead to high number of false positives since perturbation-independent response of cell lines to treatments is usually stronger than the perturbation-dependent response (Iorio et al. 2010). Inventors finally validated the optimal number of drugs that need to be considered for a fuzzy cut-off of 70% (corresponding to 8 out 11 drugs cut-off from the MANTRA dataset), providing a minimum number of false positives in the intersection (i.e. genes expected to be contained in the resulting intersections by random chance). This was performed by a permutation test where, in a series of iterations, the fuzzy cut-off is kept constant and the number of randomly selected drugs varied within a given range (specifically from 1 to 20). At each of these iterations inventors computed the cardinality 1 of the resulting fuzzy intersections, observing that this value reached a plateau at 10 drugs (FIG. 1C), which suggests that the number of drugs that was used in the analysis (i.e. 11 drugs in the cMAP dataset) was fairly close to the optimal level.

[0080] Protein-Protein Interaction

[0081] The protein-protein interactions were downloaded from the STRING database (http://string-db.org/) (Franceschini et al. 2013), and those with a confidence level of >0.7 were used for the analysis. To build the proteostasis gene (PG) dataset, inventors included known proteostatic regulators of CFTR i.e., proteins where their expression/activity level changes have been shown to affect CFTR proteostasis. Inventors also included the interactors of CFTR and CF pathology related genes/proteins present in GeneGO Metaminer Cystic Fibrosis database. The number of interactions observed among the CORE gene dataset and the proteostasis gene dataset as well as among the CORE gene dataset were more than expected on a random basis and were statistically significant. For details on the statistical test used see (Franceschini et al. 2013).

[0082] Ingenuity Pathway Analysis (IPA)

[0083] The gene sets were analyzed using the CORE analysis application of the Ingenuity pathway analysis, a web-based software application. The default settings of the analysis were used. Each network had an assigned significance score based on the p-value (calculated using Fischer's exact test) for the probability of finding the focus genes in a set of genes randomly selected from the global molecular network. The upregulated and downregulated genes of the CFBE dataset and the downregulated genes of the cMAP dataset were analyzed separately and also together, to infer common pathways or networks embedded among them.

[0084] Cell Lysis, Western Blotting and Analysis

[0085] Cells were washed three times in ice-cold Dulbecco's phosphate-buffered saline, and lysed in RIPA buffer (150 mM NaCl, 1% Triton X-100, 0.5% deoxycholic acid, 0.1% SDS, 20 mM Tris-HCl, pH 7.4), supplemented with protease inhibitor cocktail and phosphatase inhibitors. The lysates were clarified by centrifugation at 15000g for 15 min, and the supernatants were resolved by SDS-PAGE. BCA Protein Assay kit (Pierce) was used to quantitate protein levels before loading. The western blots were decorated with appropriate antibodies and developed using ECL. The blots were then exposed to x-ray films and exposure time was varied to obtain optimal signal. The x-ray films were then scanned and the bands were quantitated using ImageJ gel-analysis tool. The protein concentration and the exposures used for quantitation of the blots were optimized to be in a linear range of detection.

[0086] Biochemical Screening Assay:

[0087] Each gene was targeted by 3 siRNAs and as control non-targeting siRNAs provided by the manufacturer were used (Table 3). A gene was considered as active if: (1) at least two different siRNAs targeting a gene gave concordant changes in the levels of band C that was >2 SD from the mean value of the control siRNAs and (2) the change in band C levels was 20% of the level of band C obtained with the control siRNAs. Those genes that increased band C levels significantly upon their downregulation were termed anti-correction genes and those that decreased band C levels were termed pro-correction genes.

[0088] Immunoprecipitation

[0089] HeLa cells cultured in 10-cm plates (80% confluence) were treated with appropriate corrector drugs for 24 h. The cells then were washed three times in ice-cold Dulbecco's phosphate-buffered saline, and lysed in immunoprecipitation buffer (150 mM NaCl, 1% Triton X-100, 20 mM Tris-HCl, pH 7.4) on ice for 30 min. The lysates were clarified by centrifugation at 15000g for 15 min, and the protein content of the supernatants BCA quantitated by BCA Protein Assay kit (Pierce). Equal amounts of proteins from control and treated cell lysates were incubated with Protein-G sepharose beads conjugated with anti-HA antibody (Sigma) overnight at 4 C. The beads were then washed in the immunoprecipitation buffer 5 times and the bound proteins eluted with HA-peptide (Sigma) at a concentration of 100 g/ml. The eluted proteins were then resolved by SDS-PAGE and then immunoblotted.

[0090] Partial Trypsin Digestion of CFTR

[0091] The trypsin digestion assay was similar to that described previously (Zhang, Kartner, and Lukacs 1998). Cells were grown in a 10-cm plate and post-treatment they were washed three times with 10 mL phosphate-buffered saline (PBS). They were then scraped in 5 ml PBS, and pelleted at 500g for 5 min in 4 C. The cell pellet was resuspended in 1 mL of hypertonic buffer (250 mM sucrose, 10 mM Hepes, pH 7.2) and the cells were then homogenized using a ball bearing homogenizer. The nuclei and unbroken cells were removed by centrifugation at 600g for 15 min. The membranes were then pelleted by centrifugation at 100,000g for 30 min, and then resuspended in digestion buffer (40 mM Tris pH 7.4, 2 mM MgCl2, 0.1 mM EDTA). Then membranes corresponding to 50 g of protein were incubated with different concentrations of trypsin (1 to 50 g/ml) on ice for 15 min. The reactions were stopped with the addition of soya bean trypsin inhibitor (Sigma) to a final concentration of 1 mM, and the samples were immediately denatured in sample buffer (62.5 mM Tris-1 HCL, pH 6.8, 2% SDS, 10% glycerol, 0.001% bromophenol, 125 mM dithiothreitol) at 37 C. for 30 min. The samples were resolved on 4% to 16% gradient SDS-PAGE (Tris-glycine) and transferred onto nitrocellulose membranes. These membranes were developed with the 3G11 anti-CFTR antibodies (that recognize nucleotide binding domain 1NBD1) or the M3A7 clone (that recognizes nucleotide binding domain 2NBD2).

[0092] Plasma Membrane Quality Control Assay

[0093] The PQC assay was essentially as described previously (Okiyoneda et al. 2010). CFBE cells were untreated or treated with siRNAs for 72 h and for the final 31 h they were kept at low temperature (26 C.) and for an additional 5 h at 26 C. with CHX (100 g/ml). Then the cells were shifted to 37 C. for 1.5 h with 100 g/ml CHX before the turnover measurements started at 37 C. The cells were lysed at 0, 1, 3 and 5 h and the kinetics of degradation of band C was examined by immunoblotting.

[0094] Halide Sensitive YFP Assay for CFTR Activity

[0095] Twenty-four hours after plating, the CFBE cells that stably expressed halide sensitive YFP were incubated with the test compounds at 37 C. for 48 h. At the time of the assay, the cells were washed with PBS (containing 137 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, 1.5 mM KH2PO4, 1 mM CaCl2, 0.5 mM MgCl2) and stimulated for 30 min with 20 m forskolin and 50 m genistein. The cells were then transferred to a Zeiss LSM700 confocal microscope, where the images were acquired with a 20 objective (0.50 NA) and with an open pinhole (459 m) at a rate of 330 ms/frame (each frame corresponding to 159.42 m159.42 m), at ambient temperature. The excitation laser line 488nm was used at 2% efficiency coupled to a dual beam splitter (621 nm) for detection. The images (8-bit) were acquired in a 512512 format with no averaging to maximize the speed of acquisition. Each assay consisted of a continuous 300-s fluorescence reading with 30 s before and the rest after injection of an iodide-containing solution (PBS with Cl replaced by I; final I concentration in the well, 100 mM). To determine the fluorescence-quenching rate associated with I influx, the final 200 s of the data for each well were fitted with a mono-exponential decay, and the decay constant K was calculated using GraphPad Prism software.

[0096] Ussing Chamber Assay for Short Circuit Current Recordings

[0097] Short-circuit current (Isc) was measured across monolayers in modified Ussing chambers. CFBE41o-cells (1106) were seeded onto 12-mm fibronectin-coated Snapwell inserts (Corning Incorporated) and the apical medium was removed after 24 h to establish an air-liquid interface. Transepithelial resistance was monitored using an EVOM epithelial volt-ohmmeter and cells were used when the transepithelial resistance was 300-400 .Math.cm2. CFBE41o-monolayers were treated on both sides with optiMEM medium containing 2% (v/v) FBS and one of the following compound: 0.1% DMSO (negative control), or compounds at the stated dosage for 48 h before being mounted in EasyMount chambers and voltage clamped using a VCCMC6 multichannel current-voltage clamp (Physiologic Instruments). The apical membrane conductance was functionally isolated by permeabilising the basolateral membrane with 200 g/ml nystatin and imposing an apical-to-basolateral Cl gradient. The basolateral bathing solution contained 1.2 mM NaCl, 115 mM Na-gluconate, 25 mM NaHCO3, 1.2 mM MgCl2, 4 mM CaCl2, 2.4 mM KH2PO4, 1.24 mM K2HPO4 and 10 mM glucose (pH 7.4). The CaCl2 concentration was increased to 4mM to compensate for the chelation of calcium by gluconate. The apical bathing solution contained 115 mM NaCl, 25 mM NaHCO3, 1.2 mM MgCl2, 1.2 mM CaCl2, 2.4 mM KH2PO4, 1.24 mM K2HPO4 and 10 mM mannitol (pH 7.4). The apical solution contained mannitol instead of glucose to eliminate currents mediated by Na+-glucose co-transport. Successful permeabilization of the basolateral membrane was obvious from the reversal of Isc under these conditions. Solutions were continuously gassed and stirred with 95% O2-5% CO2 and maintained at 37 C. Ag/AgCl reference electrodes were used to measure transepithelial voltage and pass current. Pulses (1 mV amplitude, is duration) were delivered every 90 s to monitor resistance. The voltage clamps were connected to a PowerLab/8SP interface for data collection. CFTR was activated by adding 10 M forskolin to the apical bathing solution.]).

[0098] Immunofluorescence Assay for Correction of ATP7B

[0099] Cells were fixed with 4% paraformaldehyde in 0.2 M HEPES for 10 mins, permeabilized, labeled with primary and secondary antibodies, and examined with a ZEISS LSM 700 confocal microscope equipped with a 631.4 numerical aperture oil objective. The cells were scored based on the disappearance of ATP7B from the ER.

[0100] Morphological Assay to Estimate the Exit of ATP7B Exit from ER to Golgi:

[0101] Cells were transfected with ATP7B-WT-GFP or ATP7B-H1069Q-GFP, incubated overnight with 200 M BCS and/or drugs. Fixed cells were further labeled for TGN46 to mark and visualize the Golgi area under a confocal microscope. Under low copper conditions ATP7B-WT traffics to the Golgi from the ER, while ATP7B-H1069Q is retained within the ER. If the drug treatments induce the rescue of trafficking from the ER to the Golgi, the ATP7B-H1069Q-GFP fluorescence in the Golgi area increases. This is measured by quantifying (in 10 different microscopy fields in 100 cells) the increased in fluorescence of ATP7BWT-GFP or ATP7BH1069Q-GFP in the Golgi area (marked by TGN46) and normalizing this value to total cell fluorescence

[0102] Copper Detection by Inductively Coupled Plasma-Mass Spectrometry (ICP-MS)

[0103] To determine intracellular Cu concentrations, cell pellets were lysed. The protein concentration in each sample was evaluated using Bradford Protein Assay (BioRad, Segrate, Italy). Cu concentration in the cell lysates was analyzed by ICP-MS. An aliquot of each sample was diluted 1:10 v/v with 5% HNO3 and analyzed with an Agilent 7700 ICP-MS (Agilent Technologies, Santa Clara, Calif., USA) all values of Cu concentration were normalized for protein content in corresponding cell lysates.

[0104] Copper Detection Coppersensor 3 (CS3):

[0105] Coppersensor 3 (CS3), which becomes fluorescent in the presence of bioavailable Cu (Dodani, Domaille et al. 2011). For fluorescent Cu detection, cells were incubated with 5 M CS3 solution for 15 min at 37 C. CS3 was excited with 561 nm laser of LSM710, and its emission was collected from 565 to 650 nm. The signals were measured using ZEISS ZEN 2008 software and reported in arbitrary units.

[0106] Copper Estimation:

[0107] Cells are lysed in RIPA buffer (150 mM NaCl, 1% Triton X-100, 0.5% deoxycholic acid, 0.1% SDS, 20 mM Tris-HCl, pH 7.4). 500 g of total protein lysate in 100 l is taken for copper estimation using Copper assay kit (MAK127 sigma-aldrich) according to the manufacturer's protocol.

[0108] Results

[0109] Proteostasis Correctors have a Shared Transcriptional Signature

[0110] As noted, the proteostasis regulators share the ability to correct (albeit weakly) the F508del-CFTR folding-trafficking defect but have principal pharmacological effects not related to F508del-CFTR correction. Since the correction-related MOAs of these drugs are transcription-dependent, the gene signatures of the correctors should comprise genes related to F508del-CFTR correction in addition to those related to the principal actions of these drugs. If the correctors act through common mechanisms, the former genes, but not the latter, should be shared by all or most of the corrector gene signatures. To uncover this potential correction-related (CORE) gene pool, inventors developed a method based on the fuzzy intersection of transcriptional profiles (FIT) (FIG. 1A), by which corrector gene signatures are intersected to identify their commonalities (FIG. 1A). The intersections among the majority of the signatures should include the CORE genes but exclude genes related to the heterogeneous principal effects of the drugs. The main parameters of the FIT analysis (number of correctors; number of genes to be analyzed in each signature, and cut-off threshold for inclusion in the correction relevant gene pool (FIG. 1B-C)) were selected to identify a sufficiently large CORE gene pool for pathway analysis, and also to minimize the number of false CORE genes. Inventors included in our analysis two sets of correctors (24 drugs/conditions altogether) with different chemical structures and pharmacological activities (Table 4) while excluding known pharmacochaperones. The gene signatures of 13 correctors were obtained in our laboratories using immortalised CF bronchial epithelial (CFBE410-) cells (Kunzelmann et al. 1993) and an Agilent microarray platform (CFBE dataset; and GEO accession number GSE67698 for the expression profiles). Another 11 signatures were extracted from the MANTRA (Mode of action by network analysis; www.mantra.tigem.it; MANTRA dataset) (Iorio et al. 2010). The transcriptional profiles of glafenine and ouabain obtained from CFBE dataset were similar to those present in the MANTRA database (Zhang et al. 2012) suggesting that profiles obtained from CFBE and those downloaded from MANTRA database are similar enough to be treated together.

[0111] The FIT analysis of the gene signatures resulted in 219 downregulated and 402 upregulated CORE genes (FIG. 1D,E). Each of these CORE genes was shared by 70% of the corrector signatures. The number of CORE genes was 3-fold higher than that expected on a random basis. This indicates that common transcriptional programmes are indeed embedded in the signatures of proteostasis correctors.

[0112] Identification of CORE Genes/Pathways Involved in F508del-CFTR Correction

[0113] To understand the relation of CORE genes to CFTR proteostasis, inventors built a dataset of known F508del-CFTR proteostasis-relevant genes by assembling literature data and mapped their interactions with the CORE pool using STRING (Franceschini et al. 2013). Inventors found extensive and statistically significant protein-protein interactions among the nodes of the union of these two datasets (FIG. 1F), indicating that (at least a fraction of) the CORE genes are related to CFTR proteostasis. Significant interactions were also found between the CORE genes from CFBE and the MANTRA datasets, suggesting they were related and thus can be analyzed together. Inventors next applied standard bioinformatic tools to the CORE gene pool to identify functionally coherent pathways/networks/groups. Gene Ontology (GO)-based searches for proteostasis components among CORE genes retrieved 48 folding/degradation and 24 transport-machinery components, some of which are known to be involved in F508del-CFTR proteostasis. A search for signaling molecules yielded 24 kinases and 6 phosphatases. STRING and the Ingenuity pathway analysis (IPA) tool identified several statistically significant networks. The IPA networks comprised also (predicted) interactors of CORE genes, some of which were network hubs. Such hubs were often constituents of signaling pathways such as growth-factor-mediated pathways (e.g., receptors for vascular endothelial growth factor [VEGF] and platelet-derived growth factor [PDGF], phosphatidylinositol 3-kinase [PI3K], and mitogen-activated protein kinases [MAPKs]), inflammation-associated pathways (NF-B subunits, Toll-like receptor 4 [TLR4]), stress activated protein kinase (SAPK) pathways (MKK3/6, MKK4/7), and casein-kinase pathway (CSNK2A1/CKII). These hubs might control the CORE genes. Of note, many of the hubs were frequently present in the gene signatures of the individual correctors, although below the fuzzy cut-off threshold of 0.7 required for inclusion in the CORE gene pool.

[0114] Analysis of the promoters of CORE genes aimed at the identification of upstream transcription factors did not generate interpretable results.

[0115] Inventors then turned to experimental validation of the role of the CORE genes in the regulation of F508del-CFTR proteostasis. Experiments were carried out using a characterised biochemical assay that detects both the amount of core-glycosylated CFTR trapped in the ER (band B with Western blotting) and the amount of CFTR fully glycosylated in the Golgi (most of which presumably resides at the PM; band C with Western blotting). As a model system, inventors used non-polarised CFBE41o-cells stably expressing F508del-CFTR (Bebok et al. 2005) (hereafter referred to as CFBE); but many experiments were carried out also in HeLa, BHK and polarized CFBE cells, with results that were in good qualitative agreement with the CFBE data (unless specified otherwise).

[0116] While this assay is not suitable for large-scale screening, it provides quantitative information on the main proteostasis parameters including CFTR accumulation in the ER, ER-associated CFTR degradation, and transport and processing in the Golgi complex. Moreover, this assay is specific for proteostasis as it separates the effects on the F508del-CFTR protein from the effects on conductance as revealed by faster chloride-permeability assays (Pedemonte et al. 2005). Experimental validation was restricted to a limited set of genes: downregulated CORE genes (to exploit the availability of siRNA based downregulation and of small-molecule inhibitors) that showed functional coherence, i.e., were found in protein-protein interaction networks or in enriched GO groups; or were network hubs from Ingenuity analysis, or ubiquitin ligases and signaling molecules. In total, this resulted in a group of 108 genes. Notably, these genes had no previously reported role in the regulation of F508del-CFTR proteostasis.

[0117] CFBE cells were treated with siRNAs against these genes and the effects on both bands B and C were monitored. As a reference for correction, inventors used the investigational drug VX-809 (Van Goor et al. 2006), a robust corrector that acts as a pharmacochaperone. VX-809 treatment increased band C levels by 4-5-fold over control in most experiments. In all, 47 (Table 2) out of the 108 genes tested were found to be active in regulating F508del-CFTR proteostasis (FIG. 2A-D). Of these, 32 genes (when depleted) enhanced the levels of bands B and C by 1.5-fold to more than 10-fold over controls (importantly, they also increased the ratio between bands C and B; see below), while 15 decreased bands B and C by 20 to 80% of the control levels. Inventors refer to these as negative correction and positive correction genes, respectively. Among these genes 30 were CORE genes and 17 were hubs in IPA networks. Notably, the correction that was induced by depletion of many negative correction genes was greater than that achieved by VX-809 (FIG. 2A), or by the corrector drugs originally included in the study (FIG. 3A). This was in particular the case for a group of four poorly characterised ubiquitin ligases (RNF215, UBXOS, ASB8, FBXO7) that were not known to regulate F508del-CFTR proteostasis. RNF215 depletion increased the levels of bands C to over 10-fold the control levels (FIG. 2A). Given these strong effects, RNF215 is a worthy candidate for further studies as a potential ERAD machinery component. Also notably, the depletion of many negative correction genes not only enhanced the bands B and C but also markedly increased (to different extents) the band C/band B ratio (FIG. 2C), suggesting that these genes affect the efficiency of export of F508del-CFTR protein from the ER and/or the stability of this protein after export. It is to be noted that the downregulation of negative correction genes did not change the levels of CFTR mRNA (FIG. 3B) and thus the observed effect is not due to increased synthesis. It might appear surprising to find both positive correction and negative correction genes within the downregulated CORE gene pool. However, these genes are, presumably, components of complex transcriptional modules whose role is to control cellular functions in a balanced manner. To this end, the concomitant operation of regulatory systems of opposite signs is probably necessary (Hart and Alon 2013). These observations are therefore a reflection of the organization of the transcriptional programs that regulate proteostasis. Based on these results, inventors sought to identify putative pathways/networks/groups (collectively, networks) within the 47 active CORE gene pool, using literature data and pathway building tools (FIG. 2E). This resulted in several small potential networks (each comprising 2 to 6 connected elements), 4 of which were composed of signaling molecules and will be referred to by the name of their central components: MAP3K11 (MLK3), CAMKK2 PI3K- and , and CKII (with predominantly negative correction activity) and ERBB4 (with positive correction activity). A recent kinome-wide screening identified several kinases that regulate the rescue of F508del-CFTR (Trzcinska-Daneluti et al. 2015), with no overlap with the hits identified by the present inventors (possibly due to the different functional assays and cell types used in this versus our study). Other 3 of the networks shown in FIG. 2 comprised spliceosome, centromere and mediator complex components; and 2 were groups of ubiquitin-ligases and kinases. Of note, to minimise the possibility of false positives due to off-target effects of siRNA, in addition to the 3 siRNAs used in the screening, inventors also additionally tested up to 5 siRNAs for selected genes (MLK3, CAMKK2, RNF215, NUP50 and CD2BP2; Table 3) and found concordant effects on correction. Moreover, the presence of networks and pathways among the active CORE genes provides further evidence that the observed effect on correction was not due to off-target or other non-specific effects of the siRNAs used.

[0118] Proteostasis Corrector Drugs Act in Part by Modulating the Expression of CORE Genes

[0119] Inventors next sought to verify whether the effects of correctors on the CORE genes might explain the action of these drugs. Inventors first analyzed the frequency of the active CORE genes among the genes downregulated by the corrector drugs. The CORE genes were about 3-fold more enriched in the signatures of correctors compared to those of other 200 drugs taken at random from the MANTRA database. Inventors next searched for MANTRA drugs that significantly downregulate the CORE genes (anti-correctors) using Gene set enrichment analysis (GSEA; specifically two-tailed symmetric GSEA as implemented in MANTRA; www.mantra.tigem.it). The top 25 hits included 3 of the correctors that inventors had used for the FIT analysis. From the remaining 22 inventors selected 8 drugs (based on availability) for testing in the correction assay. Among these, mitoxantrone was found to potently increase both band C and band B. (FIG. 2F); in addition, among the top 5 hits was Vorinostat, an HDAC inhibitor that was shown to act as a corrector (Hutt et al. 2010). Thus, at least 20% of the short-listed drugs were correctors, while among a large number (>20) of randomly selected drugs none showed correction activity. These data suggest that the downregulation of CORE genes is a useful criterion to identify correctors. Inventors then extended our analysis of the top hits by comparing the 5 active drugs with those that failed to correct and also by examining upregulated by their gene expression profiles. The correctors showed a high frequency (2 to 3 fold more than non-corrector drugs) of up-regulation of the potent pro-corrector genes MKK1, MKK3 and FGFBP1 (FIG. 2D) while the non-corrector drugs up-regulated more frequently (2 to 3 fold more than corrector drugs) the anti corrector genes NF-B2 and MKK7. These results thus suggest that considering also the upregulation of CORE genes will help in further defining the search space for new correctors. Altogether, the above data indicate that the drug-induced modulation of CORE genes is a significant component of the MOAs of corrector drugs. Thousands of gene signatures of drugs and perturbagens are being deposited in specialized databases (http://www.lincscloud.org/). These and a more extensive search for CORE genes will provide useful tools for a more refined bioinformatic identification of new correctors.

[0120] Epistatic Interactions between CORE Pathways

[0121] As described earlier, the advantage of using this approach (deconvolution of drug MOA) to identify regulatory pathways is the possibility of discovering synergistic pathways. So in order, to explore the possible epistatic interactions between the CORE networks/pathways, siRNAs against selected targets were combined and tested on F508del-CFTR rescue. These candidates were chosen for their potential druggability and/or strong effects on correction. Strong synergistic interactions were observed between various combinations of siRNAs against CKII, CAMKK2, MLK3 and NUP50 (a spliceosomal network component) (FIG. 2G), thus validating our choice of the method. As a note of caution here, the efficacy of the combined siRNA treatments was more variable than that observed with single siRNAs. In our experience, this is because siRNAs in combinations are less effective than the individual siRNAs in depleting their target proteins, and a depletion threshold must be reached to achieve synergy. Inventors conclude that, using the FIT technique and a series of bioinformatic and experimental filters, inventors have identified a set of synergistic molecular networks that show strong control over F508del-CFTR proteostasis.

[0122] Delineation of the MLK3 and CAMKK2 Signaling Pathways Regulating F508del-CFTR Proteostasis

[0123] Next, inventors sought to define the composition and the role in correction of two representative CORE-networks, namely, the MLK3 and the CAMKK2 pathways. MLK3 (or MAP3K11) is part of a group of 14 MAP3 kinases that act through cascades of MAP2K and MAPK enzymes. MLK3 can be activated by various PM receptors, which include the TNF-, TGF-, VEGF and PDGF receptors, through at least two MAP4Ks (haematopoietic progenitor kinase [HPK] 1 and germinal centre kinase [GCK]) and glycogen synthase kinase (GSK)3, or via the CDC42/Rac family [summarised in (Karen Schachter 2006)]. MLK3 can also be activated by stress, e.g., oxidative stress (Lee et al. 2014) (i.e., it is a Stress Activated Protein Kinase, or SAPK). It can, in turn, trigger three main kinases: p38 MAPK, c-Jun N-terminal kinase (JNK), and extracellular signal regulated kinase (ERK), depending on cell type and conditions, through the intermediate kinases MAP2K3/6, MAP2K4/7 and MAP2K1/2, respectively (Karen Schachter 2006). MLK3 is also known to be an upstream activator of NF-kB (Hehner et al. 2000). Inventors thus sought to determine which components of the MLK3 pathway have roles in F508del-CFTR correction. The VEGF and PDGF receptors, MAP2K7 (MKK7), and NF-B2, like MLK3, appear to be components of the correction-relevant branch of the MLK3 pathway, as indicated by the screening data in FIG. 2A. Among the components upstream of MLK3, inventors found TGF receptors, CDC42, Rac2, and HPK1 to be active in correction (i.e. their depletion induced correction) (FIG. 4A). Within the cascade downstream of MLK3, MKK7 (FIG. 2A) and further downstream, JNK2 (FIG. 4B) were active components (JNK2, is highly expressed in bronchial epithelial cells (http://biogps.org). The p38 MAPK, also downstream of MLK3 (through MAP2K3 and 6) was inactive in CFBE but moderately active in HeLa cells indicating some cell-type-dependent specificity in the effects of these kinases (FIG. 5A). Thus, altogether, suppression of the MKK7-JNK branch of the MLK3 pathway induced F508del-CFTR correction. Conversely, when the activity of the MLK3 pathway was enhanced by transfection of the MLK3 activator CDC42 or MKK7 or JNK2 into the CFBE cells, the levels of both bands B and C dropped markedly (FIG. 4C), confirming that the MLK3 pathway has a tonic negative effect on the proteostasis of F508del-CFTR. In sum, as shown in FIG. 4D, a signal regulating F508del-CFTR proteostasis flows from the ligands and receptors upstream of MLK3, through HPK1 and CDC42/Rac2, to impinge on MLK3 and is then passed on through the JNK2 arm. NF-B2 is also a probable downstream component of this proteostasis regulatory pathway. Inventors also tested (again by siRNA silencing) seven other MAP3Ks (including TAK1/MAP3K7, see below) that can activate JNK or p38, for their effect on F508del-CFTR proteostasis. They had no effect. This highlights the remarkable specificity of MLK3 in the regulation of proteostasis, possibly due to spatial/temporal compartmentalization of the MAPK networks (Engstrom, Ward, and Moorwood 2010). A similar series of experiments was performed to characterise the CAMKK2 pathway in F508del-CFTR correction. The results are reported in detail (FIG. 5 B-C, FIG. 4E), and indicate that the CAMKK2 pathway has negative effects on F508del-CFTR proteostasis similar to those found for the MLK3 pathway.

[0124] The MLK3 Pathway Exerts Complex Regulatory Effects on F508del-CFTR Proteostasis.

[0125] The increase in band B induced by inhibition of the MLK3 pathway might be due to increased synthesis or to decreased degradation of F508del-CFTR. Downregulation of MLK3 did not increase the CFTR mRNA levels (FIG. 3 B), speaking against the former possibility. Inventors then examined the degradation of band B using both a cycloheximide (CHX) chase and a radioactive pulse-chase assays. Downregulation of the MLK3 pathway markedly slowed the degradation of band B when measured by CHX chase (FIG. 6A,B) and similar effects were obtained with the radioactive pulse-chase method (FIG. 7A,B). Inventors also examined the effects of enhancing the activity of MLK3 pathway by overexpressing CDC42 or MKK7 or JNK2: under these conditions the rate of degradation of band B increased 2-fold [FIG. 6C,D; see also (Ferru-Clement et al. 2015)]. The ubiquitin-proteasome system itself was not detectably affected by modulation of the MLK3 pathway activity, as judged by the lack of effects on both the proteasome sensor-ZsProsensor-1 (FIG. 7C) and the accumulation of poly-ubiquitinated proteins (FIG. 7D). Thus, the MLK3 pathway appears to regulate the ERQC/ERAD of F508del-CFTR at a step prior to proteasomal digestion.

[0126] In addition, silencing of the MLK3 pathway (and of several CORE genes) increased also the band C/band B ratio (see FIG. 2C). This is not explained by reduced ERAD alone and suggested that the MLK3 pathway might have additional effects on the folding/export of F508del-CFTR, or on the stability of band C at the PM (or both). The trypsin susceptibility assay to assess the folding status of F508del-CFTR and an assay for protein transport out of the ER using vesicular stomatitis virus G protein (VSVG), a classical probe to study secretory trafficking, ruled out large effects of the MLK3 pathway on F508del-CFTR folding or on the general ER-export machinery (FIG. 7E,F). Inventors next tested the effect of MLK3 on the stability of F508del-CFTR at the PM. Inventors depleted MLK3 and exposed the cells to low temperature (26 C.), to accumulate F508del-CFTR at the cell surface, and then shifted the cells back to 37 C., a temperature at which the F508del-CFTR at the PM is subjected to accelerated ubiquitination and degradation (Okiyoneda et al. 2010). Under these conditions, the depletion of MLK3 slowed the degradation rate of band C, increasing the t1/2 from 2 to 4 h (FIG. 6E,F), whereas overexpression of CDC42 to activate MLK3 enhanced band C degradation rate (FIG. 6G, H). These data suggest that also the peripheral QC of F508del-CFTR is regulated by MLK3. In contrast, the knockdown of JNK2 (or its overexpression) did not change the degradation kinetics of band C, although it increased the band C/band B ratio, suggesting that JNK2 may have additional effects on the folding and/or export of F508del-CFTR. Similar effects on F508del-CFTR folding seem likely to be induced also by the CORE genes whose depletion greatly increases the band C/band B ratio, in some cases up to 4-fold over control levels (FIG. 2C). In conclusion here, the receptor and stress-activated MLK3 signaling pathway markedly activates both the ER-associated and peripheral degradation processes of F508del-CFTR while possibly at the same time reducing the efficiency of F508del-CFTR folding/ER export. As a consequence, inhibition of the MLK3 pathway results in large increases in the levels of the Golgi-processed mature form of F508del-CFTR. 22 ROS and the CF modifiers TNF-, TGF- enhance F508del-CFTR degradation in an MLK3-dependent fashion.

[0127] Inventors next examined the effects on F508del-CFTR proteostasis of agents known to activate MLK3 such as TNF-, TGF- (Karen Schachter 2006) and reactive oxygen species (ROS) (Lee et al. 2014). TNF- and TGF- have been proposed to be genetic modifiers of CF (Cutting 2010) and ROS have been reported to 1 be enhanced in CF cells (Luciani et al. 2010) and to be massively produced by neutrophils during the inflammatory reactions that are common in CF patients (Witko-Sarsat et al. 1995). Inventors treated CFBE cells with TNF-, TGF- or H2O2 (to increase ROS), and monitored the effects on F508del-CFTR. The effects of H2O2 at non-toxic concentrations were dramatic, with a marked drop of the F508del-CFTR levels within a few minutes. Also TNF- and TGF- induced rapid, though less complete (50%) decreases in levels of F508del-CFTR. Under these conditions, the reduction in F508del-CFTR levels was completely abolished by MLK3 downregulation, confirming the crucial role of MLK3 pathway in F508del-CFTR QC/degradation. These results, and in particular the effects of H2O2, provide evidence for extremely rapid and potent mechanisms of protein degradation that involve the MLK3 pathway and act on F508del-CFTR (and presumably on other misfolded mutant proteins). These regulatory mechanisms might have pathological relevance, as discussed below.

[0128] Chemical inhibitors of the MLK3 pathway act as CFTR correctors and potently synergize with the pharmacochaperone VX-809 Inventors next tested the effect of selected kinase inhibitors on F508del-CFTR proteostasis in CFBE cells. A well-known characteristic of the kinase inhibitors is their promiscuity. In our experience, inhibitors that nominally target the same kinases can cause divergent effects on correction (see below), most likely because they target other kinases with different or competing effects. Inventors sought to overcome this difficulty by selecting kinase inhibitors with different structures and modes of action, and by using information from the KINOMEscan library (http://lines.hms.harvard.edu/data/kinomescan/). For JNK, inventors tested a set of 10 reported JNK inhibitors (JNKi), three of which led to robust increases in the levels of band B and band C (FIG. 8A,B; JNKi II, JNKi IX, JNKi XI) at concentrations that were required for JNK inhibition in CFBE cells (FIG. 9A). These JNK inhibitors have different chemical structures; moreover, while JNKi II and JNKi IX are ATP-competitive inhibitors of JNK, JNKi XI is an inhibitor of substrate/scaffold binding to JNK. These JNK inhibitors therefore appear to be reliable tools to correct F508del-CFTR by targeting the MLK3-JNK pathway. For MLK3, a previously proposed MLK3 inhibitor (K252a) had no clear effects on correction, perhaps because of its weak effect on MLK3 itself and diverging effects on other kinases (see http://www.kinase-screen.mrc.ac.uk/screening-compounds/345892). Inventors thus searched the KINOMEscan library for a molecule that had a suitable inhibitory pattern on the MLK3 pathway. (5Z)-7-oxozeaenol (herein referred as oxozeaenol) (Ninomiya-Tsuji et al. 2003) potently inhibits the MLK3 pathway members VEGF and PDGF receptor kinases and (less potently) MLK3 itself and MKK7, as well as, more weakly, a few kinases with antagonistic effects on correction (http://lincs.hms.harvard.edu/db/datasets/20211/). Oxozeaenol treatment markedly increased the bands B and C of F508del-CFTR (FIG. 8A). This drug had earlier been identified as a corrector in a screening study, and had been proposed to have F508del-CFTR corrector properties as an inhibitor of TAK1 (MAP3K7) (Trzcinska-Daneluti et al. 2012). However, the downregulation of TAK1 itself had no effect on correction (FIG. 9B). The data thus indicate that oxozeaenol acts by inhibiting the kinases of the MLK3 pathway. In line with this notion, the corrective effects of oxozeaenol were not additive with MLK3 knockdown (FIG. 9 C) and were accompanied by a reduction in phospho c-jun levels (c-jun phosphorylation is diagnostic of JNK activation) (FIG. 9D). Further inventors also tested 25 drugs that are either FDA-approved or under clinical trial kinase inhibitors that target MLK3-JNK pathway (FIG. 4D) anti-correction kinases but not pro-correction kinases. Among them Pazopanib, Dovitinib lactate and Bexarotene led to robust increases in the levels of band B and band C (FIG. 8A,B).

[0129] Thus, selected chemical blockers of the MLK3 (and CAMKK2; FIG. 9E) pathway potently increase band C (FIG. 8A,B). The level of correction obtained with these inhibitors is higher than the effects of the corrector compounds from which the pathways were deduced (FIG. 3A), and similar to, or higher than, the effects of VX-809. Inventors also noted that, while inhibitors of MLK3 pathway lead to large increases in the ER-localised band B, the pharmacochaperone VX-809 increases F508del-CFTR exit from the ER with a limited increase in band B (FIG. 8A), presumably because it primarily enhances F508del-CFTR folding. Inventors reasoned that if the band B protein accumulating in the ER following the inhibition of MLK3 pathway is in a foldable state, VX-809 might enhance its folding, and greatly increase the generation of the band C mature protein. Indeed when inventors added both MLK3 pathway inhibitors and VX-809, there was potent synergy between them (FIG. 8C,D; FIG. 10) with increases in levels of band C that were over 20-fold the basal band C level and 4-fold over those obtained with VX-809. Though VX-809 alone had only limited benefits in clinical trials (Clancy et al. 2012), recent laboratory-based research and additional clinical trials have shown promising results with combination therapies of VX-809 and other pharmacochaperones or potentiators (Jones and Barry 2015, Okiyoneda et al. 2013, Phuan et al. 2014). Given these observations, the observed additive/synergistic effects with MLK3 pathway inhibitors provide a potential therapeutic opportunity. The MLK3 pathway exerts selective effects on the proteostasis of F508del-CFTR and of structurally related mutant proteins. Inventors next examined the effects of the MLK3 pathway inhibition on the proteostasis of other conformational disease mutants. Inventors transfected CFBE (and HeLa) cells with different conformational mutants (i.e., Sodium-chloride symporter [NCC, R948X mutant]; P-glycoprotein, [P-gp, G268V and DY490 mutants]; human Ether-a-go-go-Related Gene [hERG, G601S mutant]; Wilson's disease associated protein [ATP7B, H1069Q and R778L mutants]) and then treated cells with JNKi II and monitored these proteins by assessing changes in their glycosylation patterns (NCC, P-gp, hERG mutants) or in their intracellular movement from the ER to the Golgi complex (ATP7B mutants). JNKi II rescued some of these mutants (P-gp DY490 and ATP7B mutants), while it had no effects or had negative effects, on others (FIG. 9F and Table 5). The effects on ATP7B were large and led to almost complete correction.

[0130] Both P-glycoprotein and ATP7B, like CFTR, have two groups of transmembrane domains with an interconnecting nucleotide-binding domain. Moreover, the mutations (DY490 and H1069Q) are located in the nucleotide binding domains of these proteins, and result from either a loss or substitution of aromatic amino acids, as for F508del-CFTR. These similarities suggest that common proteostatic machinery might be involved in the detection of these defects and might be targeted by the MLK3 pathway in a selective fashion. Prompted by the effects of the MLK3 kinase cascade on the CFTR-D-508 mutant, inventors examined the effects of the MLK3 pathway inhibition on the Wilson's disease (WD) associated protein mutants (ATP7B, H1069Q and R778L mutants, the main mutations found in Wilson patients). This is because CFTR and ATP7B are structurally similar, and the above mutations (DY490 and H1069Q) are located in the nucleotide-binding domains of the protein, and result from either a loss or substitution of aromatic amino acids, as for F508del-CFTR. These similarities suggest that the same proteostatic machinery acting on CFTR-D-508 might be involved in the detection of these defects and might be targeted by the MLK3 pathway in a selective fashion. This led us to test the relevance of the MLK3 pathway components and inhibitors on Wilson's disease ATP7B, H1069Q and R778L mutants.

[0131] MLK3, p38 MAPK and JNK as New Targets for Correction of Wilson Disease-Causing ATP7B Mutants.

[0132] Inventors silenced MAP3K11, the upstream activator of both p38 and JNK, and isoforms of p38 (MAPK11-MAPK14) and JNK (MAPK8-MAPK10), in HeLa cells expressing the ATP7BH1069Q (FIG. 11A, B) to evaluate whether this treatment improves ATP7BH1069Q recovery from the ER. Both a reduction in ER retention and a recovery of Golgi and vesicle targeting of the mutant was detected after depletion of MAP3K11, MAPK8 (JNK1), MAPK11 (p3813) and MAPK14 (p38) (FIG. 11). Thus, siRNA depletion of these kinases strongly corrected the trafficking defect of mutant ATP7B.

[0133] Inventors then tested the chemical inhibitors of p38 and JNK VX-745, SB202190 (SB90), Oxozeaenol and SP600125 (SP125) respectively (FIG. 12). Both p38 and JNK inhibitors (added for 24 h) did not affect ATP7BWT but strongly corrected the ATP7BH1069Q defect.

[0134] Altogether, the finding in this study is that MAP3K11, MAPK8 (JNK1), MAPK11 (p3813) and MAPK14 (p38) p38 and JNK kinases play an important role in WD by promoting retention and degradation of the ATP7BH1069Q mutant in the ER. Thus, suppression of these kinases allows ATP7BH1069Q to reach the post-Golgi vesicles and the apical surface in hepatocytes, from where it can contribute to the removal of excess Cu from the cell. As a consequence, treatments with the appropriate kinase inhibitors restore normal trafficking dynamics of the ATP7B mutants and reduce Cu accumulation in cells expressing them. Thus, MAP3K11, MAPK8 (JNK1), MAPK11 (p3813) and MAPK14 (p38) represent attractive targets for correction of the ATP7B mutant localization and function and could be considered for development of new therapeutic strategies.

[0135] Screening Assays

[0136] About 70 repositionable clinical phase drugs were acquired and tested in screening assays in cells expressing ATP7B H1069Q-GFP.

[0137] 1) Traffic-Based Screening in Hela Cells

[0138] This screening was based on a morphological assays that reveals the ability of the H1069Q to exit the ER and reach the Golgi complex.

[0139] Inventors found that 5 inhibitors (BIRB-796, Bexarotene, Cannabidiol, CPI-1189, ENMD-2076) potently rescue the mutant protein localization. A large fraction of the cellular of the mutant protein exits the ER and reaches the Golgi compartment upon inhibitor treatments (FIG. 13A,B).

[0140] 2) Traffic-Based Screening in Hepatocytes

[0141] Liver hepatocytes are the main cells that express ATP7B and the Wilson disease affects primarily liver cells. HEPG2 cells (hepatocytes from human liver carcinoma) and human primary hepatocytes are therefore a disease-relevant models to study the efficacy of the rescue by drugs. Inventors have therefore used the assay developed in HeLa cells to test drugs that rescue the ATP7B-H1069Q also in HEPG2 cells and human primary hepatocytes expressing ATP7B H1069Q-GFP. Inventors found that BIRB-796 and VX-745 rescue H1069Q potently in these cells (FIG. 13C-F).

[0142] 3) Test of Copper Excretion in Hepatocytes

[0143] ATP7B protein functions in the excretion of copper out of cells and tissue. As ATP7B H1069Q trafficking to the plasma membrane is impaired, the cells cannot excrete the copper, which leads to higher level of intracellular copper. If the corrector drugs promote the correct localization of the mutant, then the copper should be excreted, leading to lower intracellular levels. Inventors have tested the two best correctors of the localization defect of the ATP7B H1069Q mutant by estimating their intracellular copper levels upon treatment with BIRB-796 and VX-745 Inventors found that cells treated with VX-745 and BIRB-796 show low intracellular copper levels, indicating that the copper excretion function is recovered up on drugs treatments (FIG. 14).

[0144] Discussion

[0145] In this study, inventors have developed a bioinformatic method based on the fuzzy intersection of drug transcriptomes (FIT) that reveals the transcriptional components of the MOAs of proteostasis correctors. Using this method, inventors have uncovered a set of correction relevant genes (CORE genes), some of which belong to signaling networks that potently and selectively regulate the proteostasis of F508del-CFTR and of structurally related protein mutants. These are the first example of signaling cascades that specifically control the proteostasis machinery acting on AF508-CFTR. Physio-pathological significance of the CORE signaling networks. Based on literature data, interaction databases and our own experimental findings, the correction-relevant components inventors identified can be organised into five signaling cascades, which, for brevity, inventors refer to here by the names of their central components: namely, MLK3, CAMKK2, PI3K, CKII, and ERBB4. Other networks are made up of constituents of the spliceosome, centromere and mediator (transcriptional) complexes, or are groups of ubiquitin ligases.

[0146] The physiological role of the CORE signaling systems might be to regulate the stringency of the QC and degradation processes according to cellular needs. Most of the CORE pathways enhance the efficiency of QC and degradation. This is the case of the MLK3 pathway, which is activated by selected cytokines and by cellular stresses. The ERBB4 pathway, in contrast, is activated under growth conditions, and appears to have the effect of suppressing the QC and degradation processes. It may be speculated that cells under stress need to reduce the toxic burden of unfolded proteins to survive, while growing cells might need to tolerate higher levels of folding/unfolded proteins to proliferate, and that the CORE pathways regulate the proteostasis machinery according to needs. In addition, the CORE pathways might function as part of an internal control system (Cancino et al. 2014, Luini et al. 2014) that senses, and reacts to the presence of misfolded proteins. Interestingly in this regard, MLK3 interacts directly with (and might be activated by) HSP90 (Zhang et al. 2004), a component of the F508del-CFTR folding and QC machinery. More in general, the function of the CORE networks, considering that they exert selective effects on the degradation of different protein classes (FIG. 9F), might be to sculpt the proteome according to functional requirements. With respect to cystic fibrosis and similar diseases, the MLK3 and other CORE networks can be deleterious in that they enhance the degradation of mutants that retain the potential to function, such as F508del-CFTR. Also important, they can be hyper activated under pathological conditions, leading to vicious circles. For example, large amounts of ROS are produced by neutrophils in the inflamed lungs of CF patients (Witko-Sarsat et al. 1995); and elevated serum VEGF are detected in some CF patients (McColley et al. 2000). Both of these molecules act via the MLK3 pathway to enhance the degradation of F508del-CFTR, and in particular the ROS do so with striking efficacy and speed (FIG. 5A,B). These effects most probably results in lowering F508del-CFTR below the levels determined by the primary folding defect, which might be harmful because even low residual levels of F508del-CFTR may help to improve the CF phenotype in the long term (Amaral 2005). Blocking the MLK3 pathway is thus probably necessary to stop maladaptive processes that can adversely affect therapeutic efforts. Similar considerations apply to the CAMKK2 and other CORE-derived pathways.

[0147] Mechanism of Action of the MLK3 Signaling Network

[0148] The ER quality control relies on chaperones such as HSP90 and HSC70 that are also involved in folding and can switch between folding and quality control /degradation roles depending on their dwell-time on the folding client proteins (Zhang, Bonifacino, and Hegde 2013). The simplest interpretation of the data is therefore that inhibition of the MLK3 pathway regulates this folding/degradation switch by impairing the entry of F508del-CFTR into the degradation pathway and giving the mutant more time to fold and exit the ER. It cannot be excluded, however, although MLK3 does not measurably affect the folding of F508del-CFTR as detected by trypsin assay, that MLK3 (and other CORE genes) might exert subtle direct actions on the folding/ER export mechanisms. This is supported by the strong effects of some of the CORE pathways on the band C/band B ratio, and by the observation that the inhibition of MLK3 stimulates a mutant of ATP7B (similar in structure to CFTR) to leave the ER in a functional form (see Table 5).

[0149] At the molecular level, the mechanisms underlying these rescue effects remain unclear. Some initial insight might come from our observation that the phosphoprotein HOP co-precipitates much less efficiently with F508del-CFTR in cells treated with JNK inhibitors that in control cells. HOP serves as a link between HSC70 and HSP90, and its depletion induces rescue of F508del-CFTR (Marozkina et al. 2010), possibly by acting on the folding/ERQC switch discussed above. It is thus possible that a reduced interaction of HOP with the F508del-CFTR-associated QC/folding complex might be one of the modes of action of MLK3 on F508del-CFTR rescue. However, a complete analysis of the effects of the MLK3 pathway on the interactions and posttranslational modifications of the ERQC/ERAD machinery components remains a task for future work. Relevance of the CORE signaling networks for the pharmacological correction of F508del-CFTR.

[0150] Signaling cascades are eminently druggable (the majority of the known drug targets are signaling components (Imming, Sinning, and Meyer 2006)), and an enormous repertoire of drugs directed at kinases and other related molecules has been developed by the pharmaceutical industry for the therapy of major diseases. For instance, over 120 inhibitors against the correction-related kinases identified in this study are currently in clinical trial. Moreover, as shown for the case of oxozeaenol (FIG. 8A-B), suitable kinase inhibitors can be selected in a rational fashion by matching the list of CORE kinases with the kinase inhibitory patterns of the many available drugs of this class, according to polypharmacology principles (Aggarwal et al. 2007). It is thus quite possible that some of these drugs may be repositioned for CF therapy. In addition, the group of CORE ubiquitin ligases, particularly RNF215, are also attractive targets in view of their potent effect on F508del-CFTR correction (FIG. 2A). Although the technology for developing ubiquitin ligase inhibitors is still in its early stages, robust progress is being made in this direction (Goldenberg et al. 2010).

[0151] A further consideration is that the inhibitors of the CORE pathways show corrective effects that are (partially) selective for F508del-CFTR (and structurally related mutants) (see FIG. 9 F); and that these effects are complementary and synergic with those of the pharmacochaperone VX-809. Since these synergies lead to levels of correction that are several-fold higher than those achieved by VX-809 alone, it is possible that they result in combination therapies of clinical interest. Also of note is that the MOA-based approach used here can be exploited further in the future to identify more CORE pathways as well as more effective and specific correctors. In addition to the above, a key requirement for translating our findings towards clinical treatments is the conservation of the CORE pathways in epithelial bronchial cells in situ. Inventors have observed fundamentally similar role of the CORE networks across several human and mammalian cell lines, both under polarized and non-polarized conditions, suggesting that these networks are well conserved. Moreover, JNK has been reported to be hyperactive in the lungs of a mice model of CF (Grassme et al. 2014), as is p38 MAPK (also activated by MLK3) in the lungs of CF patients (Berube et al. 2010) indicating that a SAPK pathway is stimulated under these conditions. Also notably, the MLK3 pathway inhibitor oxozeaenol has been shown to be effective in correcting the F508del-CFTR proteostasis defect in the primary human bronchial epithelial cells (Trzcinska-Daneluti et al. 2012). These observations, together with the fact that the CF genetic modifiers TNF- and TGF- potently affect F508del-CFTR proteostasis, support the notion that a regulatory network similar to that uncovered in CFBE cells operates on the proteostasis machinery in bronchial cells in CF patients. In sum, this study builds on previous screening studies and on the accumulated knowledge on the F508del-CFTR proteostasis machinery (Balch, Roth, and Hutt 2011, Farinha, Matos, and Amaral 2013, Lukacs and Verkman 2012, Turnbull, Rosser, and Cyr 2007) to identify the first signaling pathways acting on F508del-CFTR proteostasis, and thereby opens new exciting possibilities to pharmacologically correct the folding and trafficking defect of this mutant protein. To establish the efficacy of these interventions in human bronchial epithelia and relevant animal models (Yan et al. 2015) will be the next stage towards the rational development of effective F508del-CFTR proteostasis regulators for CF patients.

[0152] Tables

TABLE-US-00001 TABLE 1 Kinases active on correction: Anti-corrector kinase Gen bank Accession SEQ ID NO: CAMK1 NM_003656.4 42 CAMKK2 NM_006549.3 43 NM_172215.2 44 CDC42 NM_001039802.1 45 NM_044472.2 46 CSNK2A1/CKII NM_177559.2 47 FLT1/VEGFR1 NM_002019.4 48 NM_001159920.1 49 KDR/VEGFR2 NM_002253.2 50 MAP2K7/MKK7 NM_145185.3 51 BC005365.1 52 MAP3K11/MLK3 NM_002419.3 53 MAP4K1/HPK1 NM_001042600.2 54 MAPK11 NM_002751.6 55 MAPK14 NM_001315.2 56 NM_139013.2 57 MAPK15 NM_139021.2 58 MAPK8/JNK1 NM_001278547.1 59 AB451271.1 60 MAPK9/JNK2 NM_002752.4 61 NM_139068.2 62 PDGFRA NM_006206.4 63 BC015186.1 64 PDGFRB NM_002609.3 65 PIK3CB NM_006219.2 66 PIK3CG NM_002649.3 67 PRKAA1 (AMPK) NM_206907.3 68 PRKAA2 (AMPK) NM_006252.3 69 RAC2 NM_002872.4 70 TGFBR2 NM_001024847.2 71 pro-corrector kinase Gen bank Accession ERBB4 NM_005235.2 72 MKK1/MAP2K1 NM_002755.3 73 MKK2/MAP2K2 NM_030662.3 74 MKK3/MAP2K3 NM_145109.2 75 MKK4/MAP2K4 NM_003010.3 76 PIK3CD NM_005026.3 77

[0153] The anti-corrector kinases when depleted by siRNA rescue F508del-CFTR from degradation and increase band C levels which can function at PM. The pro-corrector kinases when depleted by siRNA increase degradation of F508del-CFTR and band C levels reduce.

TABLE-US-00002 TABLE 2 CORE genes regulating the F508del-CFTR. anti- Corrector F508del-CFTR Gen bank Accession SEQ ID NO: ASB8 NM_024095.3 78 CAMKK2 NM_006549.3 43 NM_172215.2 44 CD2BP2 NM_006110.2 79 CSNK2A1 NM_177559.2 47 CTDSP1 NM_021198.2 80 NM_182642.2 81 DSN1 NM_024918.3 82 FBXO7 NM_012179.3 83 FLT1 NM_002019.4 48 NM_001159920.1 49 GTSE1 NM_016426.6 84 KDR NM_002253.2 50 MAP2K7/MKK7 NM_145185.3 51 BC005365.1 52 MAP3K11/MLK3 NM_002419.3 53 MAPK15 NM_139021.2 58 MED1 NM_004774.3 85 MED13 NM_005121.2 86 NFKB2 NM_001288724.1 87 NM_002502.5 88 NM_001077494.3 89 NUP50 NM_007172.3 90 NM_153645.2 91 OSMR NM_003999.2 92 NM_001168355.1 93 PDGFRA NM_006206.4 63 BC015186.1 64 PDGFRB NM_002609.3 65 PIK3CB NM_006219.2 66 PIK3CG NM_002649.3 67 PROKR1 NM_138964.2 94 PRPF8 NM_006445.3 95 RNF215 NM_001017981.1 96 SART1 NM_005146.4 97 SENP6 NM_015571.3 98 STAG2 NM_001042749.2 99 TEP1 NM_007110.4 100 UBOX5 NM_014948.3 101 YWHAH NM_003405.3 102 ITPR2 NM_002223.3 103 CALML5 NM_017422.4 104 MIS18BP1/C14orf106 NM_018353.4 105 Pro-corrector- F508del-CFTR Gen bank Accession AKAP8 NM_005858.3 106 BIN2 NM_016293.3 107 CYC1 NM_001916.4 108 DCLK1 NM_004734.4 109 DNAJC2 NM_014377.1 110 ERBB4 NM_005235.2 72 FGFBP1 NM_005130.4 111 MAP2K1 NM_002755.3 73 MAP2K2 NM_030662.3 74 MAP2K3 NM_145109.2 75 MAP2K4 NM_003010.3 76 MKI67 NM_002417.4 112 PIK3CD NM_005026.3 77 RBM7 NM_016090.3 113 S100A7 NM_002963.3 114

[0154] The anti-corrector when depleted by siRNA rescue F508del-CFTR from degradation and increase band C levels which can function at PM. The pro-corrector when depleted by siRNA increase degradation of F508del-CFTR and band C levels reduce.

TABLE-US-00003 TABLE3 ThesiRNAsusedinthestudy. Gene siRNAID/SensesiRNASequence(5'-3')/catalogueno. Symbol siRNA1 siRNA2 siRNA3 siRNA4 siRNA5 AKT1 s659 s660 s661 AKT2 s1215 s1216 s1217 CALM1 s2340 s2341 s2342 CALM2 s2343 s2344 s2345 CALM3 s2346 s2347 s2348 CALML3 s2349 s2350 s2351 CENPA s2906 s2907 s2908 CENPE s2917 s2915 s2916 CSNK2A2 s3639 s3640 s3641 CSNK2B s3642 s3643 s3644 CYC1 s3790 s3791 s3792 ELAVL1 s4608 s4609 s4610 ERBB4 s4781 s4782 s4783 FARSA s5027 s5028 s5029 FLNB s5278 s5279 s5280 HNF4A s6696 s6697 s6698 ONECUT1 s6702 s6703 s6704 ITPR1 s7631 s7632 s7633 ITPR2 s7634 s7635 s7636 ITPR3 s265 s266 s267 IVL s7640 s7641 s7642 KDR s7822 s7823 s7824 KRT34 s8011 s8012 s8013 LMNB1 s8224 s8225 s8226 MAL s8472 s8473 s8474 MITF s8790 s8791 s8792 MKI67 s8796 s8797 s8798 MAP3K11 s8814 s8815 s8814 NFKB1 s9504 s9505 s9506 PDE3A s10183 s10184 s10185 PDGFRA s10234 s10235 s10236 PDGFRB s10242 s10240 s10241 PIK3CA s10520 s10521 s10522 PIK3CB s10524 s10525 s10526 PIK3CD s10529 s10530 s10531 PIK3CG s10532 s10533 s10534 MED1 s10889 s10890 s10891 MAPK1 s11137 s11138 s11139 MAPK3 s11141 s230179 s230180 MAPK6 s11146 s11147 s11148 MAPK7 s11149 s11150 s11151 MAP2K1 s11167 s11168 s11169 MAP2K2 s11170 s11171 s11172 MAP2K3 s11173 s11175 s11176 MAP2K5 s11176 s11177 s11178 MAP2K6 s11180 s11181 s11182 MAP2K7 s11182 s11183 s11184 PXN s11627 s11628 s11629 RELB s11917 s11918 s11919 S100A7 s12419 s12420 s12421 MAP2K4 s12703 s12701 s12702 SPRR1A s13381 s13382 s13383 SPRR1B s13383 s13384 s13385 SPRR3 s13397 s13398 s13399 TEP1 s13985 s13986 s13987 TLR4 s14194 s14195 s14196 TOP3A s14310 s14311 s14312 TP53 s605 s606 s607 VHL s14789 s14790 s14791 YWHAH s14967 s14968 s14961 ZAP70 s14973 s14974 s14975 AKAP1 s15665 s15666 s15667 PPAP2B s16384 s16385 s16386 PRPF4B s17018 s17017 s17018 NOL3 s300 s301 s302 SART1 s17343 s17344 s17345 OSMR s17542 s17543 s17544 DCLK1 s17584 s17585 s17586 WTAP s18431 s18432 s18433 DHX38 s18906 s18907 s18908 MED13 s19365 s19366 s19367 FGFBP1 s19392 s19393 s19394 SCO2 s19424 s19425 s19426 AKT3 s19427 s19428 s19429 TROAP s229661 s229662 s229663 KIF20A s19676 s19677 s19678 RBM7 s19835 s19836 s19837 AKAP8 s20070 s20068 s20069 RGS19 s20107 s20108 s20109 CD2BP2 s20381 s20382 s20383 PRPF8 s20796 s20797 s20798 CAMKK2 s20925 s20926 s20927 STAG2 s21089 s21090 s21091 NUP50 s21138 s21139 s21140 EHD1 s21513 s21514 s21515 WDR6 s22068 s22069 s22070 UBOX5 s22595 s22596 s22597 ZC3H3 s23133 s23134 s23135 DICER1 s23754 s23755 s23756 PATZ1 s24176 s24177 s24178 FBXO7 s24491 s24492 s24493 SENP6 s25023 s25024 s25025 DNAJC2 s25685 s25686 s25687 GEMIN4 s27064 s27065 s27066 PDE11A s27187 s27188 s27189 BIN2 s28102 s28103 s28104 GTSE1 s28240 s28241 s28242 CALML5 s28669 s195236 s195237 SHC3 s28721 s28722 s28723 CYCS s28896 s28897 s28898 DGCR8 s29061 s29062 s29063 EXOSC4 s29112 s29113 s29114 C14orf106 s30720 s30721 s30722 CTDSP1 s33804 s33805 s33806 DSN1 s36760 s36761 s36762 ALPK1 s37074 s37072 s37073 ASB8 s44282 s44283 s44284 CALML6 s46468 s46469 s46470 CSNK2A1 s3636 s3637 s3638 MAPK15 s48270 s48271 s48272 NFKB2 s9507 s9508 s9509 PROKR1 s21384 s21385 s21386 PROKR2 s43338 s43339 s43340 RELA s11914 s11915 s11916 RNF215 s47218 s47217 s47218 FLT1 s5287 s5288 s5289 FLT4 s5294 s5295 s5296 Non CCGCACUC GCACCGUCCUAA GCUGGGUGG targeting- CUGAACUU UCGUCGAtt(SEQ CGGAUAAGU CHGG_05424 GAAtt(SEQ IDNO:2) Att(SEQID IDNO:1) NO:3) Non CAGUCGAA CGAGUCCGUGGA ACCGCACUC targeting- GAAGAUGG UAUCGUUtt(SEQ CUGAACUUG CHGG_05426 UUAtt(SEQ IDNO:5) Att(SEQID IDNO:4) NO:6) MAPK8 GUGGAAAGAA UUGAUAUAU AA(SEQID NO:7) MAPK9 AAGAGAGCUU AUCGUGAACU U(SEQID NO:8) MAPK10 CCGCAUGUGU CUGUAUUCAU A(SEQID NO:9) MAPK11 CAGGAUGGAG CUGAUCCAGU A(SEQID NO:10) MAPK12 CUGGACGUAU UCACUCCUGA U (SEQID NO:11) MAPK13 CCGGAGUGGC AUGAAGCUGU A(SEQID NO:12) MAPK14 AACUGCGGUU ACUUAAACAU A(SEQID NO:13) MKK7/MAP2K7 AGACUGCCUU ACUAAAGAU (SEQID NO:14) CAMK1 CAGGUGCUGG AUGCUGUGAA A(SEQID NO:15) PRKAA1 CCCACGAUAU (AMPK) UCUGUACACA A(SEQID NO:16) PRKAA2 CCGAAGUCAG (AMPK) AGCAAACCGU A(SEQID NO:17) CAMKK2 GGAUCUGAUC GCAUCGAGUACUUA AAAGGCAUC CACUA(SEQID (SEQID NO:19) NO:18) MAP3K11 GCAGCGACGU GCAGUGACGUCUGG GGGCAGUGACG CUGGAGGA GGAGGAGUCAC GUCGAGCUU* AGUUU(SEQID UCUGGAGUUU CUCAAGCA AGCAUACATT (SEQID NO:21) (SEQID AUG(SEQ (SEQID NO:20) NO:22) IDNO:23) NO:24) RAC1 UUUACCUACA GCUCCGUCUU U(SEQID NO:25) RAC2 AACUACUCAG CCAAUGUGAU G(SEQID NO:26) RAC3 CGCGCCCAUG CAGGCCAUCA A(SEQID NO:27) GSK3B GUAAUCCACC UCUGGCUAC (SEQID NO:28) MAP4K1 CUGACUAAGA GUCCCAAGA (SEQID NO:29) BRAF AAGUGGCAUG GUGAUGUGG CA(SEQID NO:30) TGFBR1 GCCUUAUUAU GCAAUGGGCUUAGU GAUCUUGUA AUUCU(SEQID (SEQID NO:32) NO:31) TGFBR2 GGAGAAAGAA CCAGCAAUCCUGAC UGACGAGAA UUGUU(SEQID (SEQID NO:34) NO:33) TGFBR3 GACAAUGACC 5GGAGUCAGGUGAU AAAUCAAUA AAUGGA(SEQID (SEQID NO:36) NO:35) TNFRSF1A CGGUGACUGU GAACCUACUUGUAC CCCAACUUU AAUGA(SEQID (SEQID NO:38) NO:37) TNFRSF1B AGAAUACUAU GCCUUGGGUCUACU GACCAGACA AAUAA(SEQID (SEQID NO:40) NO:39) MAP4K2 SASI_Hs01_ 00059138 CDC42 SASI_Hs01_ 00113094 CSNK2A1 SASI_Hs01_ SASI_Hs01_ 00110178 00110179 NUP50 SASI_Hs01_ SASI_Hs01_ 00193418 00193419 siControl-1 SI03650318 (AllStars Negative ControlsiRNA) siControl-2 UAGCGACUAA ACACAUCAA (SEQID NO:41) *indicates the siRNA for MLK3 that was used for all the experiments except for the original screening study (FIG. 2A-D). Of note, all the different siRNAs to MLK3 mentioned here led to qualitatively similar rescue of F508del-CFTR. indicates the siRNA used for epistatic interactions (FIG. 2G).

[0155] The Supplier for the siRNAs corresponding to from AKT1 to non targeting-CHGG_05426 is Life Technologies. The Supplier for the siRNAs corresponding to from MAPK8 to NUP50 and to siControl-2 is Sigma-Aldrich (USA). The Supplier for the siRNAs corresponding to siControl-1 (AllStars Negative Control siRNA) is Qiagen (Germany).

TABLE-US-00004 TABLE 4 The list of corrector drugs used in the study with their corresponding known primary MOAs. Drugs of the CFBE dataset (Reference for correction activity) Primary Use/Class 4-AN, PARP1 inhibitor PARP1 inhibitor (Anjos et al., 2012) ABT888 (Anjos et al., A poly(ADP-ribose) polymerase (PARP) -1 2012) and -2 inhibitor with chemosensitizing and antitumor activities. ABT-888 inhibits PARPs, thereby inhibiting DNA repair and potentiating the cytotoxicity of DNA- damaging agents. Glafenine (Robert et An anthranilic acid derivative with al., 2010) analgesic properties used for the relief of all types of pain (1) GSK339 Androgen receptor ligand (Norris et al., 2009). Ibuprofen (Carlile et Ibuprofen is a nonsteroidal anti- al., 2015) inflammatory drug. It is a non-selective inhibitor of cyclooxygenase. JFD03094 PARP inhibitor KM11060 (Robert et PDE5 inhibitor (an analog of sildenafil). al., 2008) Latonduine (Carlile PARP3 inhibitor et al., 2012) Minocycline H (D Y A tetracycline analog that inhibits Thomas lab protein synthesis in bacteria. Also known unpublished) to inhibit 5-lipooxygenase in the brain (2). Ouabagenin (Zhang et A cardiaoactive glycoside obtained from al., 2012) the seeds of Strophanthus gratus. Acts by inhibiting Na+/K+_ATPase, resulting in an increase in intracellular sodium and calcium concentrations (2). Ouabain (Zhang et A cardiaoactive glycoside obtained from al., 2012) the seeds of Strophanthus gratus. Acts by inhibiting Na+/K+ ATPase, resulting in an increase in intracellular sodium and calcium concentrations (2). PJ34 (Anjos et PARP1 inhibitor al., 2012) Low temperature (Denning et al., 1992) Drugs of the MANTRA dataset (Reference for correction activity) Primary Use/Class Chloramphenicol Inhibitor bacterial protein synthesis by (Carlile et al., 2007) binding to 23S rRNA and preventing peptidyl transferase activity (2). Chlorzoxazone (Carlile Muscle relaxant. Acts by inhibiting et al., 2007) degranulation of mast cells and preventing the release of histamine and slow-reacting substance of anaphylaxis. It acts at the level of the spinal cord and subcortical areas of the brain where it inhibits multi- synaptic reflex arcs involved in producing and maintaining skeletal muscle spasm (2). Dexamethasone (Caohuy Is a synthetic glucocorticoid agonist. Its et al., 2009) anti-inflammatory properties are thought to involve phospholipase A.sub.2 inhibitory proteins, lipocortins (2). Doxorubicin (Maitra DNA intercalator that inhibits et al., 2001) topoisomerase II activity by stabilizing the DNA-topoisomerase II complex (2). Glafenine (Robert et An anthranilic acid derivative with al., 2010) analgesic properties used for the relief of all types of pain (1). Liothyronine (Carlile L-triiodothyronine (T3, liothyronine) et al., 2007) thyroid hormone is normally synthesized and secreted by the thyroid gland. Most T3 is derived from peripheral monodeiodination of T4 (L- tetraiodothyronine, levothyroxine, L- thyroxine). The hormone finally delivered and used by the tissues is mainly T3. Liothyronine acts on the body to increase the basal metabolic rate, affect protein synthesis and increase the body's sensitivity to catecholamines (such as adrenaline). It is used to treat hypothyroidism (2). MS-275 (Hutt et al., Also known as Entinostat. An inhibitor of 2010) Class Ihistone deacetylases (preferentially HDAC 1, also HDAC 3) (Hu et al., 2003). Scriptaid (Hutt et al., An inhibitor of Class I histone 2010) deacetylases (HDAC1, HDAC3 and HDAC8) (Hu et al., 2003). Strophanthidin (Carlile A cardioactive glycoside that inhibits et al., 2007) Na+/K+_ATPase. Also known to inhibit the interaction of MDM2 and MDMX (1). Thapsigargin (Egan et A sesquiterpene lactone found in roots of al., 2002) Thapsia garganica. A non-competitive inhibitor of sarco/endoplasmic Ca.sup.2+ ATPase (SERCA) (1). Trichostatin-A (Hutt et An inhibitor of histone deacetylases al., 2010) (HDAC1, HDAC3, HDAC8 and HDAC7) (Hu et al., 2003). (1) http: //pubchem.ncbi.nlm.nih.gov/ (2) www.drugbank.ca

TABLE-US-00005 TABLE 5 MLK3 pathway regulates the proteostasis of mutant proteins that are structurally related to CFTR. Correction (% of wild type) * Mutant Proteins Control (DMSO) JNKi II (5 M) P-Glycoprotein DY490 24 44 hERG R948X 44 24 NCC G601S 10 9 ATP7B H1069Q 32 80 ATP7B R778L 12 40 Note: * in case of ATP7B mutants denotes fraction of protein in Golgi as calculated by fluorescence microscopy, and in other cases the protein that was processed by the Golgi are calculated by a biochemical assay similar to the one used for CFTR.

[0156] CFBE or HeLa cells (in case of ATP7B) were transfected with constructs encoding the indicated mutant proteins and treated with JNKi II for 48 h. The effect of JNKiII on proteostasis of these mutants was monitored by western blotting (to measure the change in Golgi processed band C or ER localized band B; or in the case of ATP7B using fluorescence microscopy to monitor the efficiency of translocation of the ER-localized mutant proteins to the Golgi. Treatment with JNKi II corrects the folding-trafficking defects of mutant proteins that have similar structure to F508del-CFTR (P-gp and ATP7B) while it does not have any effect or has an opposite effect on other multi-transmembrane proteins. ATP7B mutants displayed efficient correction after downregulation of the MLK3 pathway, where the localization of the mutant proteins to the Golgi reached almost the WT levels.

TABLE-US-00006 TABLE 6 Chemical structures of tested molecules Name of the drug Chemical structure JNKi XI [00003] SP600125/JNKi II [00004] JNKi IX [00005] SB202190 [00006] VX-745 [00007] Pazopanib [00008] Dovitinib lactate [00009] Bexarotene [00010] Flunarizine/Flunarizine dihydrochloride [00011] Cannabidiol [00012] CPI-1189 [00013] ENMD-2076 [00014] BIRB-796 [00015]

REFERENCES

[0157] Aggarwal, B. B., G. Sethi, V. Baladandayuthapani, S. Krishnan, and S. Shishodia. 2007. Targeting cell signaling pathways for drug discovery: an old lock needs a new key. J Cell Biochem no. 102 (3):580-92. doi: 10.1002/jcb.21500.

[0158] Amaral, M. D. 2005. Processing of CFTR: traversing the cellular mazehow much CFTR needs to go through to avoid cystic fibrosis? Pediatr Pulmonol no. 39 (6):479-91. doi: 10.1002/ppul.20168.

[0159] Anjos, S. M., R. Robert, D. Waller, D. L. Zhang, H. Balghi, H. M. Sampson, F. Ciciriello, P. Lesimple, G. W. Carlile, J. Goepp, J. Liao, P. Ferraro, R. Phillipe, F. Dantzer, J. W. Hanrahan, and D. Y. Thomas. 2012. Decreasing Poly(ADP-Ribose) Polymerase Activity Restores DeltaF508 CFTR Trafficking. Frontiers in pharmacology no. 3:165. doi: 10.3389/fphar.2012.00165.

[0160] Balch, W. E., D. M. Roth, and D. M. Hutt. 2011. Emergent properties of proteostasis in managing cystic fibrosis. Cold Spring Harbor perspectives in biology no. 3 (2). doi: 10.1101/cshperspect.a004499.

[0161] Bebok, Z., J. F. Collawn, J. Wakefield, W. Parker, Y. Li, K. Varga, E. J. Sorscher, and J. P. Clancy. 2005. Failure of cAMP agonists to activate rescued deltaF508 CFTR in CFBE41o-airway epithelial monolayers. The Journal of physiology no. 569 (Pt 2):601-15. doi: 10.1113/jphysiol.2005.096669.

[0162] Berube, J., L. Roussel, L. Nattagh, and S. Rousseau. 2010. Loss of cystic fibrosis transmembrane conductance regulator function enhances activation of p38 and ERK MAPKs, increasing interleukin-6 synthesis in airway epithelial cells exposed to Pseudomonas aeruginosa. J Biol Chem no. 285 (29):22299-307. doi: 10.1074/jbc.M109.098566.

[0163] Calamini, B., M. C. Silva, F. Madoux, D. M. Hutt, S. Khanna, M. A. Chalfant, S. A. Saldanha, P. Hodder, B. D. Tait, D. Garza, W. E. Balch, and R. I. Morimoto. 2012. Small-molecule proteostasis regulators for protein conformational diseases. Nature chemical biology no. 8 (2):185-96. doi: 10.1038/nchembio.763.

[0164] Cancino, J., A. Capalbo, A. Di Campli, M. Giannotta, R. Rizzo, J. E. Jung, R. Di Martino, M. Persico, P. Heinklein, M. Sallese, and A. Luini. 2014. Control systems of membrane transport at the interface between the endoplasmic reticulum and the Golgi. Dev Cell no. 30 (3):280-94. doi: 10.1016/j.devce1.2014.06.018.

[0165] Caohuy, H., C. Jozwik, and H. B. Pollard. 2009. Rescue of DeltaF508-CFTR by the SGK1/Nedd4-2 signaling pathway. The Journal of biological chemistry no. 284 (37):25241-53. doi: 10.1074/jbc.M109.035345.

[0166] Carlile, G. W., R. A. Keyzers, K. A. Teske, R. Robert, D. E. Williams, R. G. Linington, C. A. Gray, R. M. Centko, L. Yan, S. M. Anjos, H. M. Sampson, D. Zhang, J. Liao, J. W. Hanrahan, R. J. Andersen, and D. Y. Thomas. 2012. Correction of F508del-CFTR trafficking by the sponge alkaloid latonduine is modulated by interaction with PARP. Chemistry & biology no. 19 (10):1288-99. doi:10.1016/j.chembiol.2012.08.014.

[0167] Clancy, J. P., S. M. Rowe, F. J. Accurso, M. L. Aitken, R. S. Amin, M. A. Ashlock, M. Ballmann, M. P. Boyle, I. Bronsveld, P. W. Campbell, K. De Boeck, S. H. Donaldson, H. L. Dorkin, J. M. Dunitz, P. R. Dune, M. Jain, A. Leonard, K. S. McCoy, R. B. Moss, J. M. Pilewski, 1 D. B. Rosenbluth, R. C. Rubenstein, M. S. Schechter, M. Botfield, C. L. Ordonez, G. T. Spencer-Green, L. Vernillet, S. Wisseh, K. Yen, and M. W. Konstan. 2012. Results of a phase IIa study of VX-809, an investigational CFTR corrector compound, in subjects with cystic fibrosis homozygous for the F508del-CFTR mutation. Thorax no. 67 (1):12-8. doi: 10.1136/thoraxjnl-2011-200393.

[0168] Cutting, G. R. 2010. Modifier genes in Mendelian disorders: the example of cystic fibrosis. Annals of the New York Academy of Sciences no. 1214:57-69. doi: 10.1111/j.1749-6632.2010.05879.x.

[0169] Engstrom, W., A. Ward, and K. Moorwood. 2010. The role of scaffold proteins in JNK signaling. Cell Prolif no. 43 (1):56-66. doi: 10.1111/j.1365-2184.2009.00654.x.

[0170] Farinha, C. M., P. Matos, and M. D. Amaral. 2013. Control of cystic fibrosis transmembrane conductance regulator membrane trafficking: not just from the endoplasmic reticulum to the Golgi. The FEBS journal no. 280 (18):4396-406. doi: 10.1111/febs.12392.

[0171] Ferru-Clement, R., F. Fresquet, C. Norez, T. Metaye, F. Becq, A. Kitzis, and V. Thoreau. 2015. Involvement of the Cdc42 Pathway in CFTR Post-Translational Turnover and in Its Plasma Membrane Stability in Airway Epithelial Cells. PLoS One no. 10 (3):e0118943. doi: 10.1371/journal.pone.0118943.

[0172] Franceschini, A., D. Szklarczyk, S. Frankild, M. Kuhn, M. Simonovic, A. Roth, J. Lin, P. Minguez, P. Bork, C. von Mering, and L. J. Jensen. 2013. STRING v9.1: protein-protein interaction networks, with increased coverage and integration. Nucleic Acids Res no. 41 (Database issue):D808-15. doi: 10.1093/nar/gks1094.

[0173] Gentzsch, M., X. B. Chang, L. Cui, Y. Wu, V. V. Ozols, A. Choudhury, R. E. Pagano, and J. R. Riordan. 2004. Endocytic trafficking routes of wild type and DeltaF508 cystic fibrosis transmembrane conductance regulator. Molecular biology of the cell no. 15 (6):2684-96. doi: 10.1091/mbc.E04-03-0176.

[0174] Giannotta, M., C. Ruggiero, M. Grossi, J. Cancino, M. Capitani, T. Pulvirenti, G. M. Consoli, C. Geraci, F. Fanelli, A. Luini, and M. Sallese. 2012. The KDEL receptor couples to Galphaq/11 to activate Src kinases and regulate transport through the Golgi. EMBO J no. 31 (13):2869-81. doi: 10.1038/emboj.2012.134.

[0175] Goldenberg, S. J., J. G. Marblestone, M. R. Mattern, and B. Nicholson. 2010. Strategies for the identification of ubiquitin ligase inhibitors. Biochem Soc Trans no. 38 (Pt 1):132-6. doi: 10.1042/BST0380132.

[0176] Grassme, H., A. Carpinteiro, M. J. Edwards, E. Gulbins, and K. A. Becker. 2014. Regulation of the inflammasome by ceramide in cystic fibrosis lungs. Cellular physiology and biochemistry: international journal of experimental cellular physiology, biochemistry, and pharmacology no. 34 (1):45-55. doi: 10.1159/000362983.

[0177] Hart, Y., and U. Alon. 2013. The utility of paradoxical components in biological circuits. Mol Cell no. 49 (2):213-21. doi: 10.1016/j.molce1.2013.01.004.

[0178] Hehner, S. P., T. G. Hofmann, A. Ushmorov, O. Dienz, I. Wing-Lan Leung, N. Lassam, C. Scheidereit, W. Droge, and M. L. Schmitz. 2000. Mixed-lineage kinase 3 delivers CD3/CD28-derived signals into the IkappaB kinase complex. Mol Cell Biol no. 20 (7):2556-68.

[0179] Hutt, D. M., D. Herman, A. P. Rodrigues, S. Noel, J. M. Pilewski, J. Matteson, B. Hoch, W. Kellner, J. W. Kelly, A. Schmidt, P. J. Thomas, Y. Matsumura, W. R. Skach, M. Gentzsch, J. R. Riordan, E. J. Sorscher, T. Okiyoneda, J. R. Yates, 3rd, G. L. Lukacs, R. A. Frizzell, G. Manning, J. M. Gottesfeld, and W. E. Balch. 2010. Reduced histone deacetylase 7 activity restores function to misfolded CFTR in cystic fibrosis. Nature chemical biology no. 6 (1):25-33. doi: 10.1038/nchembio.275.

[0180] Imming, P., C. Sinning, and A. Meyer. 2006. Drugs, their targets and the nature and number of drug targets. Nat Rev Drug Discov no. 5 (10):821-34. doi: 10.1038/nrd2132.

[0181] Iorio, F., R. Bosotti, E. Scacheri, V. Belcastro, P. Mithbaokar, R. Ferriero, L. Murino, R. Tagliaferri, N. Brunetti-Pierri, A. Isacchi, and D. di Bernardo. 2010. Discovery of drug mode of action and drug repositioning from transcriptional responses. Proceedings of the National Academy of Sciences of the United States of America no. 107 (33):14621-6. doi: 10.1073/pnas.1000138107.

[0182] Iskar, M., G. Zeller, P. Blattmann, M. Campillos, M. Kuhn, K. H. Kaminska, H. Runz, A. C. Gavin, R. Pepperkok, V. van Noort, and P. Bork. 2013. Characterization of drug-induced transcriptional modules: towards drug repositioning and functional understanding. Mol Syst Biol no. 9:662. doi: 10.1038/msb.2013.20.

[0183] Jensen, T. J., M. A. Loo, S. Pind, D. B. Williams, A. L. Goldberg, and J. R. Riordan. 1995. Multiple proteolytic systems, including the proteasome, contribute to CFTR processing. Cell no. 83 (1):129-35.

[0184] Jones, A. M., and P. J. Barry. 2015. Lumacaftor/ivacaftor for patients homozygous for Phe508del-CFTR: should we curb our enthusiasm? Thorax no. 70 (7):615-6. doi: 10.1136/thoraxjnl-2015-207369.

[0185] Kalid, O., M. Mense, S. Fischman, A. Shitrit, H. Bihler, E. Ben-Zeev, N. Schutz, N. Pedemonte, P. J. Thomas, R. J. Bridges, D. R. Wetmore, Y. Marantz, and H. Senderowitz. 2010. Small molecule correctors of F508del-CFTR discovered by structure-based virtual screening. Journal of computer-aided molecular design no. 24 (12):971-91. doi: 10.1007/s10822-010-9390-0.

[0186] Karen Schachter, Geou-Yarh Liou, Yan Du, Kathleen A Gallo. 2006. MLK3. UCSD Nature Molecule Pages. doi: doi:10.1038/mp.a001551.01.

[0187] Kunzelmann, K., E. M. Schwiebert, P. L. Zeitlin, W. L. Kuo, B. A. Stanton, and D. C. Gruenert. 1993. An immortalized cystic fibrosis tracheal epithelial cell line homozygous for the delta F508 CFTR mutation. Am J Respir Cell Mol Biol no. 8 (5):522-9. doi: 10.1165/ajrcmb/8.5.522.

[0188] Lee, H. S., C. Y. Hwang, S. Y. Shin, K. S. Kwon, and K. H. Cho. 2014. MLK3 is part of a feedback mechanism that regulates different cellular responses to reactive oxygen species. Sci Signal no. 7 (328):ra52. doi: 10.1126/scisignal.2005260.

[0189] Loo, M. A., T. J. Jensen, L. Cui, Y. Hou, X. B. Chang, and J. R. Riordan. 1998. Perturbation of Hsp90 interaction with nascent CFTR prevents its maturation and accelerates its degradation by the proteasome. EMBO J no. 17 (23):6879-87. doi: 10.1093/emboj/17.23.6879.

[0190] Lu, J. J., W. Pan, Y. J. Hu, and Y. T. Wang. 2012. Multi-target drugs: the trend of drug research and development. PLoS One no. 7 (6):e40262. doi: 10.1371/journal.pone.0040262.

[0191] Luciani, A., V. R. Villella, S. Esposito, N. Brunetti-1 Pierri, D. Medina, C. Settembre, M. Gavina, L. Pulze, I. Giardino, M. Pettoello-Mantovani, M. D'Apolito, S. Guido, E. Masliah, B. Spencer, S. Quaratino, V. Raia, A. Ballabio, and L. Maiuri. 2010. Defective CFTR induces aggresome formation and lung inflammation in cystic fibrosis through ROS-mediated autophagy inhibition. Nature cell biology no. 12 (9):863-75. doi: 10.1038/ncb2090.

[0192] Luini, A., G. Mavelli, J. Jung, and J. Cancino. 2014. Control systems and coordination protocols of the secretory pathway. F1000Prime Rep no. 6:88. doi: 10.12703/P6-88.

[0193] Lukacs, G. L., and A. S. Verkman. 2012. CFTR: folding, misfolding and correcting the DeltaF508 conformational defect. Trends in molecular medicine no. 18 (2):81-91. doi: 1016/j.molmed.2011.10.003.

[0194] Marozkina, N. V., S. Yemen, M. Borowitz, L. Liu, M. Plapp, F. Sun, R. Islam, P. Erdmann-Gilmore, R. R. Townsend, C. F. Lichti, S. Mantri, P. W. Clapp, S. H. Randell, B. Gaston, and K. Zaman. 2010. Hsp 70/Hsp 90 organizing protein as a nitrosylation target in cystic fibrosis therapy. Proc Natl Acad Sci USA no. 107 (25):11393-8. doi: 10.1073/pnas.0909128107.

[0195] McColley, S. A., V. Stellmach, S. R. Boas, M. Jain, and S. E. Crawford. 2000. Serum vascular endothelial growth factor is elevated in cystic fibrosis and decreases with treatment of acute pulmonary exacerbation. Am J Respir Crit Care Med no. 161 (6):1877-80. doi: 10.1164/ajrccm.161.6.9905022.

[0196] Meacham, G. C., Z. Lu, S. King, E. Sorscher, A. Tousson, and D. M. Cyr. 1999. The Hdj-2/Hsc70 chaperone pair facilitates early steps in CFTR biogenesis. EMBO J no. 18 (6):1492-505. doi: 10.1093/emboj/18.6.1492.

[0197] Ninomiya-Tsuji, J., T. Kajino, K. Ono, T. Ohtomo, M. Matsumoto, M. Shiina, M. Mihara, M. Tsuchiya, and K. Matsumoto. 2003. A resorcylic acid lactone, 5Z-7-oxozeaenol, prevents inflammation by inhibiting the catalytic activity of TAK1 MAPK kinase kinase. The Journal of biological chemistry no. 278 (20):18485-90. doi: 10.1074/jbc.M207453200.

[0198] Odolczyk, N., J. Fritsch, C. Norez, N. Servel, M. F. da Cunha, S. Bitam, A. Kupniewska, L. Wiszniewski, J. Colas, K. Tarnowski, D. Tondelier, A. Roldan, E. L. Saussereau, P. Melin-Heschel, G. Wieczorek, G. L. Lukacs, M. Dadlez, G. Faure, H. Herrmann, M. Ollero, F. Becq, P. Zielenkiewicz, and A. Edelman. 2013. Discovery of novel potent DeltaF508-CFTR correctors that target the nucleotide binding domain. EMBO Mol Med no. 5 (10):1484-501. doi: 10.1002/emmm.201302699.

[0199] Okiyoneda, T., H. Barriere, M. Bagdany, W. M. Rabeh, K. Du, J. Hohfeld, J. C. Young, and G. L. Lukacs. 2010. Peripheral protein quality control removes unfolded CFTR from the plasma membrane. Science no. 329 (5993):805-10. doi: 10.1126/science.1191542.

[0200] Okiyoneda, T., G. Veit, J. F. Dekkers, M. Bagdany, N. Soya, H. Xu, A. Roldan, A. S. Verkman, M. Kurth, A. Simon, T. Hegedus, J. M. Beekman, and G. L. Lukacs. 2013. Mechanism-based corrector combination restores DeltaF508-CFTR folding and function. Nature chemical biology no. 9 (7):444-54. doi: 10.1038/nchembio.1253.

[0201] Pedemonte, N., G. L. Lukacs, K. Du, E. Caci, O. Zegarra-1 Moran, L. J. Galietta, and A. S. Verkman. 2005. Small-molecule correctors of defective DeltaF508-CFTR cellular processing identified by high-throughput screening. The Journal of clinical investigation no. 115 (9):2564-71. doi: 10.1172/JCI24898.

[0202] Phuan, P. W., G. Veit, J. Tan, A. Roldan, W. E. Finkbeiner, G. L. Lukacs, and A. S. Verkman. 2014. Synergy-based Small-Molecule Screen Using a Human Lung Epithelial Cell Line Yields DeltaF508-CFTR Correctors that Augment VX-809 Maximal Efficacy. Mol Pharmacol. doi: 10.1124/mol.114.092478.

[0203] Popescu, F. D. 2003. New asthma drugs acting on gene expression. Journal of cellular and molecular medicine no. 7 (4):475-86.

[0204] Pulvirenti, T., M. Giannotta, M. Capestrano, M. Capitani, A. Pisanu, R. S. Polishchuk, E. San Pietro, G. V. Beznoussenko, A. A. Mironov, G. Turacchio, V. W. Hsu, M. Sallese, and A. Luini. 2008. A traffic-activated Golgi-based signaling circuit coordinates the secretory pathway. Nat Cell Biol no. 10 (8):912-22. doi: 10.1038/ncb1751.

[0205] Riordan, J. R. 2008. CFTR function and prospects for therapy. Annual review of biochemistry no. 77:701-26. doi: 10.1146/annurev.biochem.75.103004.142532.

[0206] Rosser, M. F., D. E. Grove, L. Chen, and D. M. Cyr. 2008. Assembly and misassembly of cystic fibrosis transmembrane conductance regulator: folding defects caused by deletion of F508 occur before and after the calnexin-dependent association of membrane spanning domain (MSD) 1 and MSD2. Molecular biology of the cell no. 19 (11):4570-9. doi: 10.1091/mbc.E08-04-0357.

[0207] Sampson, H. M., R. Robert, J. Liao, E. Matthes, G. W. Carlile, J. W. Hanrahan, and D. Y. Thomas. 2011. Identification of a NBD1-binding pharmacological chaperone 30 that corrects the trafficking defect of F508del-CFTR. Chemistry & biology no. 18 (2):231-42. doi: 10.1016/j.chembiol.2010.11.016. Santagata, S., M. L. Mendillo, Y. C. Tang, A. Subramanian, C. C. Perley, S. P. Roche, B. Wong, R. Narayan, H. Kwon, M. Koeva, A. Amon, T. R. Golub, J. A. Porco, Jr., L. Whitesell, and S. Lindquist. 2013. Tight coordination of protein translation and HSF1 activation supports the anabolic malignant state. Science no. 341 (6143):1238303. doi: 10.1126/science.1238303.

[0208] Trzcinska-Daneluti, A. M., A. Chen, L. Nguyen, R. Murchie, C. Jiang, J. Moffat, L. Pelletier, and D. Rotin. 2015. RNA interference screen to identify kinases that suppress rescue of deltaF508-CFTR. Mol Cell Proteomics. doi:10.1074/mcp.M114.046375.

[0209] Trzcinska-Daneluti, A. M., L. Nguyen, C. Jiang, C. Fladd, D. Uehling, M. Prakesch, R. Al-awar, and D. Rotin. 2012. Use of kinase inhibitors to correct DeltaF508-CFTR function. Molecular & cellular proteomics: MCP no. 11 (9):745-57. doi: 10.1074/mcp.M111.016626.

[0210] Turnbull, E. L., M. F. Rosser, and D. M. Cyr. 2007. The role of the UPS in cystic fibrosis. BMC Biochem no. 8 Suppl 1:S11. doi: 10.1186/1471-2091-8-S1-S11.

[0211] Van Goor, F., K. S. Straley, D. Cao, J. Gonzalez, 1 S. Hadida, A. Hazlewood, J. Joubran, T. Knapp, L. R. Makings, M. Miller, T. Neuberger, E. Olson, V. Panchenko, J. Rader, A. Singh, J. H. Stack, R. Tung, P. D. Grootenhuis, and P. Negulescu. 2006. Rescue of DeltaF508-CFTR trafficking and gating in human cystic fibrosis airway primary cultures by small molecules. American journal of physiology. Lung cellular and molecular physiology no. 290 (6):L1117-30. doi: 10.1152/ajplung.00169.2005.

[0212] Wang, Y., T. W. Loo, M. C. Bartlett, and D. M. Clarke. 2007. Correctors promote maturation of cystic fibrosis transmembrane conductance regulator (CFTR)-processing mutants by binding to the protein. The Journal of biological chemistry no. 282 (46):33247-51. doi: 10.1074/jbc.C700175200.

[0213] Ward, C. L., S. Omura, and R. R. Kopito. 1995. Degradation of CFTR by the ubiquitin13 proteasome pathway. Cell no. 83 (1):121-7.

[0214] Witko-Sarsat, V., C. Delacourt, D. Rabier, J. Bardet, A. T. Nguyen, and B. Descamps-Latscha. 1995. Neutrophil-derived long-lived oxidants in cystic fibrosis sputum. American journal of respiratory and critical care medicine no. 152 (6 Pt 1):1910-6. doi: 10.1164/ajrccm.152.6.8520754.

[0215] Yan, Z., Z. A. Stewart, P. L. Sinn, J. C. Olsen, J. Hu, P. B. McCray, Jr., and J. F. Engelhardt. 2015. Ferret and pig models of cystic fibrosis: prospects and promise for gene therapy. Hum Gene Ther Clin Dev no. 26 (1):38-49. doi: 10.1089/humc.2014.154.

[0216] Zhang, D., F. Ciciriello, S. M. Anjos, A. Carissimo, J. Liao, G. W. Carlile, H. Balghi, R. Robert, A. Luini, J. W. Hanrahan, and D. Y. Thomas. 2012. Ouabain Mimics Low Temperature Rescue of F508del-CFTR in Cystic Fibrosis Epithelial Cells. Frontiers in pharmacology no. 3:176. doi: 10.3389/fphar.2012.00176.

[0217] Zhang, F., N. Kartner, and G. L. Lukacs. 1998. Limited proteolysis as a probe for arrested conformational maturation of delta F508 CFTR. Nature structural biology no. 5 (3):180-3.

[0218] Zhang, H., W. Wu, Y. Du, S. J. Santos, S. E. Conrad, J. T. Watson, N. Grammatikakis, and K. A. Gallo. 2004. Hsp90/p50cdc37 is required for mixed-lineage kinase (MLK) 3 signaling. J Biol Chem no. 279 (19):19457-63. doi: 10.1074/jbc.M311377200.

[0219] Zhang, Z. R., J. S. Bonifacino, and R. S. Hegde. 2013. Deubiquitinases sharpen substrate discrimination during membrane protein degradation from the ER. Cell no. 154 (3):609-22. doi: 10.1016/j.cell.2013.06.038.

[0220] Farinha, C. M., D. Penque, M. Roxo-Rosa, G. Lukacs, R. Dormer, M. McPherson, M. Pereira, A. G. Bot, H. Jorna, R. Willemsen, H. Dejonge, G. D. Heda, C. R. Marino, P. Fanen, A. Hinzpeter, J. Lipecka, J. Fritsch, M. Gentzsch, A. Edelman, and M. D. Amaral. 2004. Biochemical methods to assess CFTR expression and membrane localization. J Cyst Fibros. 3 Suppl 2:73-77.

[0221] Anjos, S. M., R. Robert, D. Waller, D. L. Zhang, H. Balghi, H. M. Sampson, F. Ciciriello, P. Lesimple, G. W. Carlile, J. Goepp, J. Liao, P. Ferraro, R. Phillipe, F. Dantzer, J. W. Hanrahan, and D. Y. Thomas. 2012. Decreasing Poly(ADP-Ribose) Polymerase Activity Restores DeltaF508 CFTR Trafficking. Front Pharmacol. 3:165.

[0222] Balch, W. E., R. I. Morimoto, A. Dillin, and J. W. Kelly. 2008. Adapting proteostasis for disease intervention. Science. 319:916-919.

[0223] Calamini, B., and R. I. Morimoto. 2012. Protein homeostasis as a therapeutic target for diseases of protein conformation. Curr Top Med Chem. 12:2623-2640.

[0224] Caohuy, H., C. Jozwik, and H. B. Pollard. 2009. Rescue of DeltaF508-CFTR by the SGK1/Nedd4-2 signaling pathway. J Biol Chem. 284:25241-25253.

[0225] Carlile, G. W., R. A. Keyzers, K. A. Teske, R. Robert, D. E. Williams, R. G. Linington, C. A. Gray, R. M. Centko, L. Yan, S. M. Anjos, H. M. Sampson, D. Zhang, J. Liao, J. W. Hanrahan, R. J. Andersen, and D. Y. Thomas. 2012. Correction of F508del-CFTR trafficking by the sponge alkaloid latonduine is modulated by interaction with PARP. Chem Biol. 19:1288-1299.

[0226] Carlile, G. W., R. Robert, J. Goepp, E. Matthes, J. Liao, B. Kus, S. D. Macknight, D. Rotin, J. W. Hanrahan, and D. Y. Thomas. 2015. Ibuprofen rescues mutant cystic fibrosis transmembrane conductance regulator trafficking. J Cyst Fibros. 14:16-25.

[0227] Carlile, G. W., R. Robert, D. Zhang, K. A. Teske, Y. Luo, J. W. Hanrahan, and D. Y. Thomas. 2007. Correctors of protein trafficking defects identified by a novel high-throughput screening assay. Chembiochem. 8:1012-1020.

[0228] Chesi, G., R. N. Hegde, S. lacobacci, M. Concilli, S. Parashuraman, B. P. Festa, E. V. Polishchuk, G. Di Tullio, A. Carissimo, S. Montefusco, D. Canetti, M. Monti, A. Amoresano, P. Pucci, B. van de Sluis, S. Lutsenko, A. Luini, and R. S. Polishchuk. 2016. Identification of p38 MAPK and JNK as new targets for correction of Wilson disease-causing ATP7B mutants. Hepatology. 63:1842-1859.

[0229] Denning, G. M., M. P. Anderson, J. F. Amara, J. Marshall, A. E. Smith, and M. J. Welsh. 1992. Processing of mutant cystic fibrosis transmembrane conductance regulator is temperature-sensitive. Nature. 358:761-764.

[0230] Egan, M. E., J. Glockner-Pagel, C. Ambrose, P. A. Cahill, L. Pappoe, N. Balamuth, E. Cho, S. Canny, C. A. Wagner, J. Geibel, and M. J. Caplan. 2002. Calcium-pump inhibitors induce functional surface expression of Delta F508-CFTR protein in cystic fibrosis epithelial cells. Nat Med. 8:485-492.

[0231] Galietta, L. J., P. M. Haggie, and A. S. Verkman. 2001. Green fluorescent protein-based halide indicators with improved chloride and iodide affinities. FEBS Lett. 499:220-224.

[0232] Gregersen, N., P. Bross, S. Vang, and J. H. Christensen. 2006. Protein misfolding and human disease. Annual review of genomics and human genetics. 7:103-124.

[0233] Hu, E., E. Dul, C. M. Sung, Z. Chen, R. Kirkpatrick, G. F. Zhang, K. Johanson, R. Liu, A. Lago, G. Hofmann, R. Macarron, M. de los Frailes, P. Perez, J. Krawiec, J. Winkler, and M. Jaye. 2003. Identification of novel isoform-selective inhibitors within class I histone deacetylases. J Pharmacol Exp Ther. 307:720-728.

[0234] Hutt, D. M., D. Herman, A. P. Rodrigues, S. Noel, J. M. Pilewski, J. Matteson, B. Hoch, W. Kellner, J. W. Kelly, A. Schmidt, P. J. Thomas, Y. Matsumura, W. R. Skach, M. Gentzsch, J. R. Riordan, E. J. Sorscher, T. Okiyoneda, J. R. Yates, 3rd, G. L. Lukacs, R. A. Frizzell, G. Manning, J. M. Gottesfeld, and W. E. Balch. 2010. Reduced histone deacetylase 7 activity restores function to misfolded CFTR in cystic fibrosis. Nat Chem Biol. 6:25-33.

[0235] Maitra, R., C. M. Shaw, B. A. Stanton, and J. W. Hamilton. 2001. Increased functional cell surface expression of CFTR and DeltaF508-CFTR by the anthracycline doxorubicin. Am J Physiol Cell Physiol. 280:C1031-1037.

[0236] Norris, J. D., J. D. Joseph, A. B. Sherk, D. Juzumiene, P. S. Turnbull, S. W. Rafferty, H. Cui, E. Anderson, D. Fan, D. A. Dye, X. Deng, D. Kazmin, C. Y. Chang, T. M. Willson, and D. P. McDonnell. 2009. Differential presentation of protein interaction surfaces on the androgen receptor defines the pharmacological actions of bound ligands. Chem Biol. 16:452-460.

[0237] Robert, R., G. W. Carlile, J. Liao, H. Balghi, P. Lesimple, N. Liu, B. Kus, D. Rotin, M. Wilke, H. R. de Jonge, B. J. Scholte, D. Y. Thomas, and J. W. Hanrahan. 2010. Correction of the Delta phe508 cystic fibrosis transmembrane conductance regulator trafficking defect by the bioavailable compound glafenine. Mol Pharmacol. 77:922-930.

[0238] Robert, R., G. W. Carlile, C. Pavel, N. Liu, S. M. Anjos, J. Liao, Y. Luo, D. Zhang, D. Y. Thomas, and J. W. Hanrahan. 2008. Structural analog of sildenafil identified as a novel corrector of the F508del-CFTR trafficking defect. Mol Pharmacol. 73:478-489.

[0239] Zhang, D., F. Ciciriello, S. M. Anjos, A. Carissimo, J. Liao, G. W. Carlile, H. Balghi, R. Robert, A. Luini, J. W. Hanrahan, and D. Y. Thomas. 2012. Ouabain Mimics Low Temperature Rescue of F508del-CFTR in Cystic Fibrosis Epithelial Cells. Front Pharmacol. 3:176.