M.J. Welsh and A.E. Smith, RAC Application 
experiment established the feasibility of gene therapy to reverse the effects of the abnormal 
gene in CF patients. 
Several antibodies to CFTR have been developed and these have allowed identification of 
the protein (22,24-31). Moreover, we have shown that in normal human airway epithelia, 
the protein is located in the apical membrane (32), where it can mediate chloride transport 
Although complementation of chloride transport in CF cells did not identify the function of 
CFTR, data obtained since then provide compelling evidence that CFTR is a chloride 
channel with novel regulation (for a review see 7). The protein has been expressed in a 
variety of recombinant cells (25,29,33-39); in each case, c AMP-regulated chloride channels 
were generated. Channels produced by recombinant CFTR and those endogenous to the 
apical membrane of normal secretory epithelia (40-43) have the same regulatory and 
biophysical properties. In addition, we have shown that alteration of specific amino acid 
residues in CFTR alters the anion selectivity of chloride channels (35). This observation 
indicates that the protein itself interacts with permeating anions, thereby indicating that 
CFTR forms the channel pore. Most recently, CFTR has been purified, reconstituted into 
proteoliposomes, and fused with planar lipid bilayers (44); in a bilayer, CFTR had 
regulatory and biophysical properties similar to those observed in the native cell membrane. 
Other experiments on the expression of CFTR in recombinant cells indicate that the CFTR 
chloride channel is regulated by phosphorylation by cAMP-dependent protein kinase 
(34,38,45,46). We showed by site-directed mutagenesis of CFTR, followed by 
electrophysiological studies, that the R domain acts to regulate passage of chloride through 
the channel. ATP is also required on the cytosolic surface to keep CFTR chloride channels 
open (47). Recent work indicates that ATP interacts with the nucleotide-binding domains 
of CFTR to control chloride channel opening and closing (48). 
These data indicate that CFTR is a regulated chloride channel. They do not, however, 
exclude the possibility that CFTR also has other functions (49-52). Further, although 
CFTR is active in the apical surface of epithelial cells, it may also function at other 
locations within the cell (53). The possibility that CFTR might function other than as a 
chloride channel or at other cellular locations make gene therapy a particularly attractive 
therapeutic strategy. 
2.3.3 Cystic fibrosis-associated mutations in CFTR. 
The most common CF-associated mutation, accounting for about 70% of CF chromosomes, 
is deletion of phenylalanine at position 508 (AF508) (6,54). Phenylalanine 508 is located in 
the middle of the first nucleotide-binding domain. Over 100 other mutations have been 
discovered on CF chromosomes (for reviews, see 3,55,56). 
Mutations in CFTR appear to result in a loss of chloride channel activity in one of three 
general ways. 
a. The mutated protein does not traffic to the apical membrane. Our recent studies (57,58) 
indicate that several CF-associated mutations, including the most common, AF508, lead to 
incomplete protein processing. This inference was based on the finding that wild-type 
CFTR expressed in heterologous cells underwent two stages of glycosylation, a core 
glycosylation (endoglycosidase H sensitive) characteristic of processing in the endoplasmic 
reticulum, and more extensive glycosylation, characteristic of processing in the Golgi 
complex. In contrast, CFTRAF508 only underwent core glycosylation. This result 
suggested that the mutant protein was retained in the endoplasmic reticulum, did not reach 
the Golgi complex, and was not delivered to the plasma membrane. We proposed that the 
Recombinant DNA Research, Volume 16 
[859] 
