not prevent cAMP from opening the channel. How- 
ever, when all four serines were mutated, the chan- 
nel failed to respond to cAMP agonists. Nevertheless 
a triple mutant containing only one of the four ser- 
ines still responded to cAMP. These results led to the 
conclusion that phosphorylation-dependent regula- 
tion of the CFTR chloride channel is degenerate. 
That is, normally more than one site is involved, but 
no single site is essential. Furthermore a single site 
alone may be sufficient. This is a novel mechanism 
of regulation not described for any other channel. 
In CFTR, the function of the nucleotide-binding 
domains (NBDs) has been an enigma. On the one 
hand, the NBDs are the conserved feature of the 
CFTR amino acid sequence that suggests that it be- 
longs to a family of proteins, many of which serve as 
pumps. On the other hand, why should an ion chan- 
nel contain a domain that might hydrolyze ATP? Add- 
ing to the mystery, the NBDs are the site of a majority 
of naturally occurring CF mutations. To discover the 
function of the NBDs, the Welsh laboratory tested 
the hypothesis that ATP would regulate the CFTR 
chloride channel. They found that once the channel 
was phosphorylated by cAMP-dependent protein ki- 
nase, ATP is required to keep it in the open state. 
Moreover, because only hydro lyzable analogues of 
ATP were effective, the data suggested that ATP hy- 
drolysis may be required for the channel to open. 
That result suggested that some CF-associated muta- 
tions in the NBDs may cause dysfunction of CFTR by 
disrupting its normal regulation. 
Some further clues about the function of CFTR 
and abnormalities in disease came from the studies 
of its cellular location. Antibodies to CFTR were 
used to localize CFTR. By using confocal laser scan- 
ning microscopy, CFTR was identified in the apical 
region of several chloride-secreting intestinal epi- 
thelial cell lines. More direct evidence that CFTR is 
located within the apical membrane came from stud- 
ies showing that an antibody directed against an ex- 
tracellular epitope labeled the apical membrane of 
unpermeabilized intestinal epithelial cell lines. Lo- 
calization of CFTR in the apical membrane places it 
in a position where it can directly mediate chloride 
transport; for a chloride channel to mediate chlo- 
ride transport directly it must be located in the api- 
cal membrane. 
Numerous mutations are reported to cause CF, 
but deletion of phenylalanine at position 508 
(AF508) is the most common. Previous studies from 
the Welsh and other laboratories suggested that 
CFTR AF508 is not completely processed. Those 
observations suggested that the mutant proteins are 
retained in the endoplasmic reticulum rather than 
being transported to the plasma membrane. As a re- 
sult the apical membrane would not contain func- 
tional chloride channels. To test the hypothesis that 
such mutants are not delivered to the plasmid mem- 
brane. Dr. Welsh and his colleagues used primary 
cultures of airway epithelia grown on permeable 
supports; under these conditions they polarized and 
expressed the CF defect in apical chloride perme- 
ability. They developed a semiquantitative assay us- 
ing nonpermeabilized epithelia, an antibody di- 
rected against an extracellular epitope of CFTR, and 
large fluorescent beads bound to secondary antibod- 
ies. They observed specific binding to airway epithe- 
lia from non-CF subjects, indicating that CFTR is lo- 
cated in the apical membrane of airway epithelia. In 
contrast, there was no specific binding to the apical 
membrane of CF airway epithelia. These data were 
supported by qualitative studies using confocal mi- 
croscopy: the most prominent immunostaining was 
in the apical region of non-CF cells and in the cyto- 
plasmic regions of CF cells. The results indicate that 
CFTR is either missing from the apical membrane of 
the CF cells or is present at a much reduced level. 
Thus those results explain the lack of chloride per- 
meability in most CF airway epithelia. 
Dr. Welsh is also Professor of Internal Medicine 
and of Physiology and Biophysics at the Univer- 
sity of Iowa College of Medicine, Iowa City. 
Books and Chapters of Books 
Anderson, M.P., Rich, D.P., Gregory, R.J., Cheng, S., 
Smith, A.E., and Welsh, M.J. 1992. Function and 
regulation of the cystic fibrosis transmembrane 
conductance regulator. In Adenine Nucleotides 
in Cellular Energy Transfer and Signal Trans- 
duction (Papa, S., Azzi, A., and Tager, J.M., Eds.). 
Basel: Birkhauser Verlag, pp 399-413. 
Anderson, M.P., and Welsh, M.J. 1991. Regulation 
of apical membrane chloride channels by phos- 
phorylation and fatty acids in normal and cystic 
fibrosis airway epithelium. In Signaling Mecha- 
nisms in Secretory and Immune Cells (Martinez, 
J.R., Edwards, B.S., and Seagrave, J.C., Eds.). San 
Francisco, CA: San Francisco Press, pp 1-5. 
Krause, K.-H., Lew, D P., and Welsh, M.J. 1991. 
Electrophysiological properties of human neutro- 
phils. In New Aspects of Human Polymorphonu- 
clear Leukocytes (Horl, W.H., and SchoUmeyer, 
P.J., Eds.). New York: Plenum, pp 1-11. 
Welsh, M.J. 1992. Abnormal chloride and sodium 
channel function in cystic fibrosis airway epithe- 
lia. In Lung Injury (Crystal, R.G., and West, J. B., 
Eds.). New York: Raven, pp 313-321. 
118 
