Cystic fibrosis: rescue of the function and of the processing of cftr mutants by pharmacological agents and by interacting proteins (CF-PRONET)

A mouse breeding programme was started to obtain congenic F508del-CFTR mice in different genetic backgrounds (FVB; C57BL/6; 129/Sv; BALB/C). Attempts to include DBA as the fifth background were stopped because of the poor breeding performance of the Cftr+/- heterozygotes. The FVB and C57BL/6 strains reached their congenic state (13th backcross) in 2003, whereas congenic 129/Sv and BALB/C strains will become available in November 2004 and March 2005, respectively. Both the FVB and C57BL/6 strains were analyzed for possible phenotypic differences with respect to - The level of F508del-Cftr processing in intestinal and airway epithelium, using a combination of immunological and functional assays (Western blots, immunocytochemistry, ion transport in mini-Using chambers); - The occurrence of distal intestinal obstruction syndrome (DIOS) in response to a shift in diet from Hope farms SRM-A to SDS-RM3 (obstruction, perforation and necrosis of the terminal ileum followed by death within 3-4 days after the diet shift); - The occurrence of early symptoms of CF lung disease (goblet cell hyperplasia, interstitial collagen deposits, hypercellularity of the bronchioli, increase of inflammatory markers). Whereas the percentage of F508del-Cftr protein escaping from the processing defect and the residual, Cftr-mediated transepithelial chloride secretion in intestinal and nasal mucosa did not differ significantly between both mouse strains, there was a remarkable strain difference in their sensitivity towards intestinal and lung disease: 100% of the homozygous F508del CF FVB mice, but 0% of the C57BL/6 mice died within 4 days after shifting to the SDS-RM3 diet, whereas all homozygous F508del CF C57BL/6 mice, but none of the FVB mice (kept on Hope Farms diet), showed clear symptoms of "spontaneous" CF lung disease. Awaiting the mapping of the genetic loci determining these strain differences in diet tolerance and susceptibility to CF lung disease, the congenic F508del CF mouse models created in this programme can already be used advantageously to test the efficacy of novel F508del Cftr rescue therapies (both genetic and pharmacologic) in ameliorating CF intestinal disease (FVB strain) and lung disease (C57BL/6 strain), respectively.

Gene expression profiles from various cell lines were generated. Gene expression profiling for pharmacological testing was optimised. For this, experiments were carried out that should lead to an adaptation of the Clondiag AT-platform (a tool developed for genotyping) for use in gene expression profiling. The AT-platform is limited in the number of genes that can be analysed, but first experiments indicate, that a much higher sensitivity may be possible. We optimised the experimental conditions for the analysis of lower abundant expressed genes (low amounts of gene specific mRNA) and optimised the proprietry array tube system for gene expression profiling. For this optimisation we focused on 50 human genes which are intensively studied at the Klinik für Innere Medizin (University Jena) and for which reference clones are available. A major focus of the work was to optimise the algorithm for the calculation of oligonucleotide probes that are most promising for gene expression analysis. This includes secondary structure prodictation, information about heterolougous sequence hybridization and exclusion of repeat and repeat like sequences. We also assayed the quality of oligonucleotide synthesis, which has a significant impact on the specificity, and sensitivity of the system. The most important improvement for the quality of the assay is our new optimised labelling protocol. Unfortunately, until now we could not fully confirm, that the quantitative results obtained give a true quantitative reflection on the expression of the respective genes. This problem however is to date not solved with any array based method. (Only RNAse protection assays and with slightly lower precision real time PCR allow to calculate absolute number of nucleic acid targets.) Reference experiments with rectal biopsies from healthy donors allowed us to establish an optimal homogenisation protocol of unfrozen tissue that can be combined with the Qiagen RNA extraction method. Gene expression profiling from biopsies with and without electrophysiological measurements (in cooperation with Partner in Hannover), showed that no significant change could be observed when samples are analysed in an Using Chamber before RNA extraction. This does not exclude that "sensitive" genes exhibit changes in gene expression upon preanalysis treatment.

A high throughput system to search for small molecules that would affect the function of CFTR was developed. The screening process is performed routinely using an automate for rapid measurement of radiotracer flux allowing multiple tests of molecules on living cells cultured in multiwell plates. The rate of activation is determined after mathematical fitting, the EC50 (for an activator) and/or IC50 (for an inhibitor) are calculated using the software GraphPad Prism version 3.0 for Windows (Graphpad Software, San Diego, California, USA). Several protocols have been used. An adaptation to the robotic system was done in preliminary experiments during the first three month of the project. Then, four different protocols were designed. Protocol #1 was designed to screen molecules on wild type CFTR activity expressed in CHO and Calu-3 cells. In the protocol 1a, drugs are tested alone in the absence of cAMP-promoting agent. In the protocol 1b, drugs are tested in the presence of low concentration of the cAMP-promoting agent forskolin used at 1 microM. The first protocol will tell us whether a drug is able to stimulate directly CFTR. The second one will determine whether the drug potentiates the response obtained with low concentration of forskolin. Protocol #2 was designed to screen molecule on the class III CF mutation G551D-CFTR. Despite the fact that this mutant does not respond to forskolin at concentration as high as 10 microM, this mutant need high concentration of forskolin before further activation. In other words, the R-domain of CFTR need to be phosphorylated by protein kinase A (PKA). Drugs are tested on this mutant in the presence of 10 microM of forskolin. After incubation of cells with the selected molecule the activity of defective CFTR can be addressed and compared to that of wild-type CFTR. Protocol #3 was designed to screen molecules on F508del-CFTR activity. The F508del homozygous human airway epithelial cell line CF15 are used for this protocol. Cells are cultured at 27°C for 24 to 48 hours before functional analysis. Protocol #4 was designed to screen molecules on F508del-CFTR trafficking. The ability of molecules to activate F508del-CFTR is determined after incubation (2-4h, 37°C) to allow the localization of proteins toward the apical compartment. The rescue of F508del-CFTR protein at the cell surface is determined using a cAMP-promoting cocktail containing forskolin (10 microM) and genistein (30 microM). Protocol #3 is now used extensively not only to screen new agents able to rescue F508del from the ER to the plasma membrane but also to evaluate the effect of modulators of the biosynthetic pathway added in association with these agents. For example using the brefeldin A allow to demonstrate that F508del move from the ER to the golgi. With proteasome inhibitors or Ca-ATPase inhibitors or other agent we can potentiate the effect of the first agents. These items are now used in routine to further study the mechanism of action of any new agent that rescue F508del.

Aim: to raise polyclonal antibodies in rabbits against human CFTR and CF-related (mouse) proteins for use on antibody chips, in immunocytochemistry and in Western blots. - Rabitt poyclonal antibodies against human CFTR - CFTR: Extracellular loops - EnaC: Extracellular loops of a,b, and g subunits - Rabitt polyclonal against Cytokeratins - CK8 (Cytokeratin 8) - CK18 (Cytokeratin 18) - Rabitt polyclonal against human beta defensins 1,2,3,4 - Chicken polyclonal against human beta defensins 1,2,3,4 - Mouse monoclonals against huamn beta defensins 2,3,4 Mouse proteins: - Mouse chloride channels CIC-Ka and CIC-Kb (see for selection discriminating peptides: option 3 from e-mail correspondence with Ann Andries, 18/09/03) - Mouse barttin (=regulator of CIC-K channels, identified in mouse nasal epithelium) - Mouse a-, b-, and g-subunits of the epithelial sodium channel ENaC - Mouse b2-adrenergic receptor (b2-AR) - The CFTR-regulated chloride/bicarbona te exchangers; -- 0 mouse SLC 26 A3 (=DRA, CLD) -- 0 mouse SLC 26 A6 (=PAT-1) -- 0 mouse SLC 26 A9 (lung-specific) - Mouse PDZ-adapter protein CAP-70 (binds to C-terminus of CFTR and CLD). - Mouse CFTR chloride channel -- 0 first extracellular loop (cf. sequence hCFTR, Chemicon Ab). -- 0 C-terminal13 aa or cloned C-terminal fragment (80aa). - Mouse ROMK (CFTR-regulated epithelial potassium channel in kidney and lung). - Mouse chloride channel CIC-3. - Mouse SGK-1 (serum-and glucocorticoid-activated protein kinase, regulates ENaC). - Mouse chloride channel CIC-2. - Mouse YAP-65 (=Yes-associated protein, involved in CFTR-ENaC coupling). - Mouse D-AKAP2 (AKAP and putative CFTR regulator).

A new set of molecules from various chemical families were tested in the screening process. This included novel MPB derivatives as part of a Structure Activity Relationship (SAR) study. Our criteria to exclude drugs for further analysis is 20% or less of stimulation at 250 microM concentration. Among the different derivatives tested, some have promising activity on CFTR. The best candidates as activator of wild-type CFTR and G551D-CFTR are MPB-91, MPB-97, MPB-77, MPB-104, MPB-107, MPB-96, and MPB-95. A similar analysis is concerning the effect of these derivatives on F508del-expressing cells, still needs to be performed and is under way. A structure-activity study demonstrated the importance and the nature of chemical groups within the MPB skeleton. For example MPB-80, and MPB-70 are not active on CFTR. More recently, we found that by adding an alkyl group in C5 we obtained MPB-91 a derivative that is more potent than MPB-07. Now we discovered MPB-104 and MPB-107 that are 100-fold more potent than MPB-07. These two agents are subject of extensive study to determine the mechanism of action to rescue F508del.

Risultati finali

Condividi questa pagina

Scarica