M.J. Welsh and A.E. Smith, RAC Application 
a. CF is an autosomal recessive disease and heterozygotes have no lung disease (1). Thus, 
50% of wild-type CFTR would appear sufficient for normal function. 
b. This issue was tested in mixing experiments using CF cells and recombinant CF cells 
expressing wild-type CFTR (66). The data obtained showed that when an epithelium is 
reconstituted with as few as 6-10% of corrected cells, chloride secretion is comparable to 
that observed with an epithelium containing 100% corrected cells. Although CFTR 
expression in the recombinant cells is probably higher than in normal cells, this result 
suggests that in vivo correction of all CF airway cells may not be required. 
c. Recent observations show that CFTR containing some CF-associated mutations retains 
residual chloride channel activity (67,68). These mutations are associated with mild lung 
disease. Thus, even a very low level of CFTR activity may at least partly amilorate the 
electrolyte transport abnormalities. 
d. As indicated in our experiments described below (8.2. b. (3). 1), complementation of CF 
epithelia, under conditions that probably would not cause expression of CFTR in every cell, 
restored cAMP-stimulated chloride secretion. 
e. Levels of CFTR in normal human airway epithelia are very low and are barely 
detectable. We have not been able to detect it by routine biochemical techniques such as 
immunoprecipitation or immunoblotting. It is also exceedingly difficult to detect with 
immunocytochemical techniques (32). Although we have been able to detect CFTR using 
laser-scanning confocal microscopy in some cases, the signal is at the limits of detection 
and cannot be detected above background in every case. Despite the minimal levels of 
CFTR, this small amount is sufficient to generate substantial cAMP-stimulated chloride 
secretion. The reason that a very small number of CFTR chloride channels can support a 
large chloride secretory rate is that a large number of ions can pass through a single channel 
(10 6 - 10 7 ions/sec) (69). 
f. Previous studies using quantitative PCR have reported that the airway epithelial cells 
contain at most one to two transcripts per cell (70). 
3.4 Effect of Overexpressing CFTR 
Fortunately, gene therapy for CF would appear to have a wide therapeutic index. Just as 
partial expression may be of therapeutic value, overexpression of wild-type CFTR appears 
unlikely to cause significant problems. This conclusion is based on both theoretical 
considerations and experimental results. 
Because CFTR is a regulated channel, and because it has a specific function in epithelia, it 
is unlikely that overexpression of CFTR will lead to uncontrolled chloride secretion. First, 
secretion would require activation of CFTR by cAMP-dependent phosphorylation. 
Activation of this kinase is a highly regulated process. Second, even if CFTR chloride 
channels open in the apical membrane, secretion will not ensue without regulation of the 
basolateral membrane transporters that are required for chloride to enter the cell from the 
interstitial space. At the basolateral membrane, the sodium-potassium-chloride 
cotransporter and potassium channels serve as important regulators of transepithelial 
secretion (62). 
[862] 
Recombinant DNA Research, Volume 16 
