A truncated CFTR protein rescues endogenous DeltaF508-CFTR and corrects chloride transport in mice

Abstract

Cystic fibrosis (CF) is most frequently associated with deletion of phenylalanine at position 508 (DeltaF508) in the CF transmembrane conductance regulator (CFTR) protein. The DeltaF508-CFTR mutant protein exhibits a folding defect that affects its processing and impairs chloride-channel function. This study aimed to determine whether CFTR fragments approximately half the size of wild-type CFTR and complementary to the portion of CFTR bearing the mutation can specifically rescue the processing of endogenous DeltaF508-CFTR in vivo. cDNA encoding CFTR fragments were delivered to human airway epithelial cells and mice harboring endogenous DeltaF508-CFTR. Delivery of small CFTR fragments, which do not act as chloride channels by themselves, rescue DeltaF508-CFTR. Therefore, we can speculate that the presence of the CFTR fragment, which does not harbor a mutation, might facilitate intermolecular interactions. The rescue of CFTR was evident by the restoration of chloride transport in human CFBE41o- bronchial epithelial cells expressing DeltaF508-CFTR in vitro. More important, nasal administration of an adenovirus expressing a complementary CFTR fragment restored some degree of CFTR activity in the nasal airways of DeltaF508 homozygous mice in vivo. These findings identify complementary protein fragments as a viable in vivo approach for correcting disease-causing misfolding of plasma membrane proteins.

Figure 1.

A) Immunolocalization of WT-CFTR, CFTR1-633, and CFTR1-614KKAA. WT-CFTR is expressed at the plasma membrane in HEK293T cells. CFTR1-633 staining is mostly in the ER, and CFTR1-614KKAA colocalizes with the ER marker calreticulin. All results are representative of ≥3 experiments. B) Color profiling of the fluorescence of the CFTR constructs and calreticulin. Images were analyzed with QCapturePro image acquisition and analysis software (QImaging, Tucson, AZ, USA).

Figure 2.

Rescue of the trafficking of ΔF508-CFTR and Δ2-79-CFTR by a CFTR fragment containing an ER retention signal. A, B) Rescue of the processing of ΔF508-CFTR (A) and of CFTR mutant lacking the amino-terminal tail of CFTR (Δ2-79-CFTR) (B) by a CFTR fragment containing an ER retention signal (1-614KKAA). C) Time course of disappearance of rescued Δ2-79-CFTR by CFTR1-614KKAA. D) Time course of disappearance of rescued ΔF508-CFTR by CFTR1-614KKAA. CFTR1-614KKAA was coimmunoprecipitated with ΔF508-CFTR using a C-CFTR antibody. Cycloheximide was added at the beginning of the experiment to inhibit protein synthesis. Blots were analyzed by densitometry using Bio-Rad Quantity One program. In each case, the band C signal was normalized to the total band B plus band C signal and expressed as percentage of band C. E) Detection of a high-molecular-weight complex (HMWC) in HEK293T cells transfected with the indicated CFTR constructs. The HMWC was detected after immunoprecipitation using a C-CFTR monoclonal antibody and was probed with a C-CFTR antibody and ubiquitin antibody. All results are representative of 3 to 5 experiments.

Figure 3.

Detection of functional ΔF508-CFTR chloride channels after coexpression of a CFTR fragment. Microsomes, prepared from HEK293T cells transfected with CFTR constructs, were incorporated in planar lipid bilayers. A) CFTR channels could be detected only in cells transfected with ΔF508-CFTR and the CFTR fragment 1-633 using low PKA (25 U/ml). At least 2 channels were present in the lipid bilayer, and addition of 150 μM glybenclamide inhibited these channels. O, open state; C, closed state. Open probabilities (Po) were 0.41 ± 0.06 and 0.06 ± 0.02 after addition of forskolin and glybenclamide, respectively. B, C) No channel could be detected when the cells were transfected only with 1-633 (B) or ΔF508-CFTR alone (C). All results are representative of ≥3 experiments.

Figure 4.

Rescue of ΔF508-CFTR in CFBE41o-ΔF cells, using a CFTR fragment. The CFTR fragment 1–633 was delivered to the CFBE41o-ΔF cells using an adenovirus vector. A) CyclicAMP-activated currents in CFBE41o-ΔF cells transduced with Ad-1-633 (top panel) or Ad-Luc (bottom panel). B) Histogram representation of peak currents assessed by Ussing chambers. C) Expression of ΔF508-CFTR and CFTR1-633 fragment in CFBE41o-ΔF cells transduced with Ad-Luc or Ad-1–633. Same number of cells was used per immunoprecipitation. CFTR1-633 fragment coimmunoprecipitated with ΔF508-CFTR was detected using an N-CFTR monoclonal antibody (Chemicon). n = number of experiments. *P < 0.05.

Figure 5.

Restoration of chloride transport in CF mice after delivery of WT-CFTR. A) Measure of NPD in a mouse expressing WT-CFTR. Trace is representative of 10 WT mice. B) NPD measurements from a mouse expressing ΔF508-CFTR. Trace is representative of 19 CF mice (ΔF508/ΔF508). C) NPD measurements in ΔF508-CFTR mouse after delivery of WT-CFTR. Trace is representative of 3 CF mice (ΔF508/ΔF508). Brackets represent changes in NPD after stimulation of CFTR.

Figure 6.

Restoration of chloride transport in CF mice after expression of the CFTR fragment 1-614KKAA. A) Representative NPD measurements of CF mice expressing ΔF508-CFTR before (left panel) and after (right panel) delivery of the CFTR fragment 1-614KKAA. Traces represent 5 separate experiments. B) NPD changes on activation of the CFTR chloride channel in WT mice and in CF mice (ΔF508/ΔF508) before and after delivery of an adenovirus vector bearing CFTR1-614KKAA fragment. Brackets indicate changes in NPD after stimulation of CFTR. n = number of experiments. *P < 0.05.