The increasing number of rare sequence variations identified in the CFTR gene substantially complicates test interpretation and consequently genetic counseling of patients and families. Diagnostic laboratories have to face a risk for misclassification of mutations and variations, especially of missense type. To allow an accurate interpretation of variant pathogenicity the French CFTR database collects molecular and clinical data for more than 4,400 subjects (CF and CFTR-RD patients, patients analysed in the context of newborn screening, fetal bowel anomalies and compound heterozygous unaffected parents). We have recorded so far 690 different variants, including 34% that remains unclassified (also called UVs). Near 75% of these UVs are missense variations.
In order to predict missense variants pathogenicity, we used bioinformatics softwares assessing their impact on splicing and on protein function. Nineteen missense variants predicted to affect splicing were selected to be tested using functional analysis at both transcript and protein level. We thus promote a collaborative network. Six French laboratories specialized with CFTR genetics and functional studies are willing to join their skills in a common project: Bordeaux (P Fergelot), Brest (MP Audrezet), Créteil (P Fanen), Paris-Cochin (E Girodon), Poitiers (A Kitzis) and Montpellier (C Raynal).
First, we will perform in vitro assays using a minigene approach to test splicing defects. Then, we will assess the impact on transcripts by RNA quantification and stability studies. Finally, effect on protein function will be assessed by analyzing protein maturation (by Western Blotting), intracellular trafficking (by immunofluorescence) and CFTR channel activity (by Cl- efflux measurement).
By implementing these combined approaches, we expect to (i) assess missense pathogenicity and help to classify mutations and variations for the purposes of diagnosis and genetic counseling, (ii) characterize the functional mechanisms involved, (iii) improve our knowledge on genotype/phenotype correlations and (iv) improve data on the reliability of bioinformatics softwares used to select candidate variants.
Despite the high complexity of CFTR disorders, improving our knowledge of their pathogenic mechanisms will help the classification of mutation according to the mechanisms by which they disrupt CFTR function, and will also help to target new molecules therapies for CF patients.