Cystic fibrosis is manifestly a disease of exocrine organs, if we recall that the respiratory, digestive and reproductive tracts all rely extensively on exocrine secretions. In cystic fibrosis, the lungs lose their ability to maintain a sterile surface and are gradually destroyed by colonies of bacteria. The intestinal lining secretes less fluid than normal, and is therefore susceptible to blockade from improperly hydrated stools. Stools have excessive amounts of fat, caused by loss of pancreatic enzymes. Additional symptoms include the blockage and eventual degeneration of the vas deferens in males, dehydrated cervical mucus and a failure of the mucus to show appropriate hydration during ovulation in females, and greatly elevated concentrations of salt in the sweat. Symptoms variably present include focal biliary cirrhosis and an unusually small gallbladder.

All of these defects arise because of mutations in CFTR. Can they all be explained by the consequent loss of CFTR-mediated Cl- conductance, or does the complicated CFTR protein play other roles as well? If Cl- conductance is the critical physiological defect, how can the pleiotropic effects be explained? If CFTR plays other roles, what are they, and what are their consequences? These questions are a pervasive theme in modern CF research. One way to approach them is by examining the role of CFTR and the consequences of its dysfunction organ-by-organ.

The sweat gland, as an almost pure serous gland, represents the consequences of the CF defect uncomplicated by obstruction or infection. Sweat secretion is achieved by two parallel systems. The physiologically relevant system that mediates thermal sweating operates via a cholinergically-stimulated increase in Ca2+. This system is intact in CF; but unfortunately the mechanisms responsible for Ca2+-mediated sweat secretion are unknown. Small amounts of sweating can also be stimulated by elevating cAMP.

Heat ® cholinergic stimulus ® ® increase [Ca2+ ]i ® [unknown] ® sweat secretion

? ® adrenergic stimulus ® ® increase [cAMP]i ® ­ CFTR Po ® sweat secretion

The physiological role of the cAMP pathway is unknown, since 100% of thermally-induced sweating is blocked by the cholinergic blocker atropine. In any event, the cAMP-mediated secretory response is entirely absent in CF individuals and is reduced to half normal values in heterozygotes, indicating that CFTR Cl- channels are rate-limiting for cAMP-mediated sweat secretion.

The secretory defect in CF sweating is only detectable in the laboratory, but reduced salt reabsorption from sweat is apparent to anyone with CF. Normally, as primary sweat moves along the reabsorptive duct, most of the salt is reabsorbed. Reabsorption is driven by the large electrochemical gradient for Na+ which flows into the cell through amiloride-blockable Na+ channels in the apical membrane. The basolateral Na+, K+-ATPase then transports Na+ out of the cell and into the blood. The Na+ movement instantaneously creates a negative voltage in the lumen, which provides an electrochemical gradient that forces Cl- out of the lumen and into the ductal cells via apically located CFTR Cl- channels. (The paracellular pathway in duct cells has a high resistance to ion flow.) The apical membrane of sweat ducts contains perhaps the most dense concentration of CFTR Cl- channels known in normal tissues. Accordingly, the conductance of a normal sweat duct is among the highest known for epithelial tissues (about 110 mS/cm2 ).

In CF sweat ducts, the Cl- conductance is virtually abolished, and the duct behaves as though it were permeable only to Na+. Thus when Na+ attempts to flow out of a CF duct unaccompanied by Cl- , it creates a large excess of negative charge in the duct which sets up an opposing electrical gradient for Na+ and so greatly retards its movement. The net result is that both Na+ and Cl- are poorly reabsorbed by the CF duct, leading to the high salt content of CF sweat.

· The epididymus and vas deferens

The thin duct that conveys sperm from the testes is also a Cl--based fluid secreting organ. It is perhaps more vulnerable than any other organ to the destructive effects of CF lesions. A classic study by Oppenheimer and Esterly established that the loss of the vas deferens in CF males is not the result of failure to develop, but is instead caused by degeneration secondary to obstruction. The graph shows the percentage of normal observations for the head and tail of the epididymus and for the vas deferens, for four age groups. Although the graph is based on incomplete information, it makes the point that normal genital tract structures are present in many young CF males, but that after age 10 the vas is absent, with variable sparing of the epididymus.

Subjects with congenital bilateral absence of the vas deferens (CBAVD) but no other obvious signs of CF have an exceptionally high incidence of mutations in CFTR, adding support to a suggestion made long ago by Holsclaw that CBAVD is a variant form of CF. We recently examined a subject who is homozygous for R117H, an allele that results in a low conductance CFTR Cl- channel. This subject has recurrent bronchitis and CBAVD, but all other indications are normal, including clear sinuses as assessed with CT. However, careful sweat tests revealed that the subject had consistently higher sweat Cl- values than a set of age-matched controls, although still well within the values considered normal in clinical tests.

· The pancreas

The pancreatic acinar cells secretes digestive enzymes and some fluid that then mixes with a bicarbonate-rich fluid secreted by the duct cells in response to secretin, which elevates cAMP in the usual way. The pancreatic juice flows along the pancreatic duct into the duodenum, where it participates in digestion, especially of fats. Because of significant species differences in ion transport mechanisms in the pancreas, and the inaccessibility of human pancreatic tissue, many uncertainties remain about the details of CFTR's role in the pancreas. However, a likely model is that CFTR is the only significant Cl- channel in the apical membrane of ductal cells, and it functions there in conjunction with an anion exchanger to effect bicarbonate secretion. When defective there is a reduction in fluid secretion, which eventually leads to blockage of smaller ducts because of enzyme precipitation or mucus accumulation. The pancreas is unusual in that the acini entirely lack myoepithelial cells as supporting structures, and hence are extremely sensitive to damage by even minor increases in intraductal pressure. Once damage and inflammation start, a rapid loss of pancreatic function ensues.

Of the hundreds of CF mutations that have been identified, the great majority cause pancreatic insufficiency. These include all mutations that cause biosynthetic arrest of CFTR, or that are missense or nonsense mutations that result in no mature protein, and also includes normally processed proteins that are incapable of opening. However, a small set of CFTR mutations that allow some residual CFTR channel conductance lead to milder forms of CF in which pancreatic function remains sufficient for digestion. It is now widely appreciated that slight residual function can have dramatic consequences at the organ system level. Thus, if as little as 5% of pancreatic secretions are spared, digestion is nearly normal, no exogenous pancreatic enzymes are needed, and patients are said to be pancreatic sufficient. The finding that point mutations in the Cl- conducting pore region of CFTR cause milder forms of CF strongly supports the Cl- conductance hypothesis.

· The intestine

CFTR expression in the intestine is most dense in the crypts and stops abruptly as the crypt cells migrate across the crypt-villus border. Because the crypts are the site of fluid secretion by the intestine, loss of CFTR function would be predicted to lead to underhydration of the intestines, and that is exactly what is found. In direct measurements of Cl- secretion from intestinal biopsies, tissues from CF individuals were shown to lack Cl- secretion to any of the secretagogues tested, regardless of whether they activated cAMP, cGMP, or Ca2+. That finding is consistent with other evidence that intestinal crypt cells lack a Ca2+-activated Cl- channel. In a larger study that took genotype into consideration, 30 of 51 (59 %) of CF patients entirely lacked Cl- secretion and in fact showed a reversed Isc another 11 (22 %) showed a small amount of residual secretion and a third set of 10 (20%) showed a still larger residual secretion, amounting to about 1/4 of the control response. (None of the groups differed from controls with regard to the amiloride sensitivity of the intestinal tissue.)

As predicted by the Cl- channel hypothesis of CF, those patients with residual Cl- secretion had milder disease in general, as assessed by various measures such as age at diagnosis and degree of pancreatic sufficiency. Almost all D F508 homozygotes were in the group having no residual secretion, and mutations associated with pancreatic sufficiency had residual intestinal secretion too. However, some subjects that should lack CFTR entirely, (i.e. G542X homozygotes) also had some residual secretion, which might be attributable to a small contribution of Ca2+-activated Cl- channels in these subjects, either because of compensatory expression, or because of this mutation occurs on a different genetic background.

Mice also lack a significant Ca2+-activated Cl- secretion in their intestines, and intestinal blockage is the major cause of mortality in cftr -/- mice.

· The lung

  The above examples illustrate how the reduction of fluid secretion and salt absorption across epithelia, caused by loss of CFTR-mediated Cl- conductance, can directly explain most CF disease. However, persistent lung infection is the most life-threatening symptom of CF, and the role of CFTR in preventing such infections is poorly understood. CFTR protein is not abundant in lung, and other epithelial Cl- channels exist in lung. Therefore, it seems plausible that the effects of CFTR in the lung must somehow be amplified to enable it to have such a significant effect. This suggestion is supported by evidence, so far found only for lung cells, that CFTR is able to influence other lung transport proteins. Loss of CFTR leads to increased Na+ reabsorption, probably because of increased activity of sodium channels. CFTR also influences another Cl- channel of unknown function, and it may do other things as well. It is not yet known how CFTR mediates these effects on other channels nor if they are important for lung disease.

Another way to explain the powerful role played by small amounts of CFTR in the lung, without going beyond the Cl- conductance hypothesis, is to propose that CFTR Cl- channels are strategically located in the lung to play a key role in fighting infections and in keeping the airway ducts free of mucus accumulation. This idea is given credence by the finding of a highly heterogeneous distribution of CFTR within the airway mucosa, with the most dense staining occurring in serous cells of the submucosal glands.

Submucosal glands are complex glands that have, via their many tubules, a greatly expanded surface area. In humans they provide most of the mucin secretion in the upper airways. Most of the gland cell volume is made up of serous cells, which form the secretory endpieces of the glands, while mucus cells line the tubules. CFTR is not found in the mucous cells. Serous cells are the primary source of fluid secretion in the glands and are essential to the formation of properly hydrated mucus. In addition, serous cells secrete a host of antibiotic compounds and tissue-protecting protease inhibitors, leading Basbaum and her colleagues to designate them as "the primary defensive cell of the mucosa".

It seems apparent that if the antibiotic rich fluid secretion of airway submucosal glands were lost, the lungs should be more vulnerable to infections. Nevertheless, a role for gland malfunction in CF was ignored until recently because the glands normally secrete to agents they raise cellular Ca2+ levels, and, as we have seen, Ca2-mediated Cl- secretion is typically spared in CF. However, intestinal crypt cells are an exception: they normally secrete to both cAMP and Ca2+, and as we saw both forms of secretion are lost from CF subjects. The same finding applies to submucosal gland serous cells.

The mechanism for this has been explored in a human cell line, called Calu-3, that shares a great many features with native serous cells. In Calu-3 cells, CFTR channels in the apical membrane are constitutively active and sustain a low level of constitutive Cl- secretion. Elevation of cystosolic Ca2+ causes a large increase in Cl- secretion by opening basolateral K+ channels to increase the driving force for Cl- exit. Thus, airway serous cells are another example of an epithelium in which Ca2+ -mediated fluid secretion should be abolished by CFTR mutations, and that has been observed.

· The chloride channel hypothesis of cystic fibrosis

  We saw how fluid secretion in many epithelia depends upon Cl- flow through CFTR. Because CFTR is a channel, the direction of ion flow through it depends only on the electrochemical driving force. So CFTR also plays a key role in salt absorption in some tissues, such as the sweat duct. Thus, both hyposecretion of fluid and poor reabsorption are fundamental defects of CFTR mutations. Problems caused by hyposecretion may be most serious for health. Most dramatically, the poor hydration of luminal contents can lead to complete obstruction of the lumen in organs like pancreas and even the intestine. Intestinal block once killed 10-15% of children with CF; and presently kills a much higher proportion of cystic fibrosis mice.

The hypothesis that loss of Cl- conductance can explain the majority of CF symptoms can account for many observations that initially appear to be at odds with such a simple idea. At this point, it might be useful to restate the hypothesis and some of its corollaries more formally. The Cl- conductance hypothesis of CF states that the critical factors determining the consequences of CFTR mutations are:

Some additional factors that determine organ patency are:

This hypothesis can in principle account for a good portion of the wide variations in symptom severity among individuals, among organs, and among species. To summarize some specific supporting data:

The Cl- conductance hypothesis obviously can't account for all inter individual variation, including the substantial variation due to environmental factors as diverse as nutrition, exposure to pathogens, exercise, and medical care. In fact, genetic factors will become more apparent when lifestyle and treatment for CF are optimized.

· The link to secretory diarrheas

  It was pointed out in the first talk that intestinal CFTR Cl- channels are essential both for normal fluid elaboration and for the profound excess encountered in secretory diarrheas. It was hypothesized in the early eighties that CF carriers should lose less fluid during bouts of secretory diarrhea and so should be partially protected. This could help account for the prevalence of CF mutations. It may also help explain some other curious features that have been noted about CFTR.

One way to make biological sense of all of this is to postulate that CFTR is under dual selective pressures, and that more recent pressures caused by an increasing prevalence of secretory diarrheas that exploit CFTR-mediated secretion may have begun to favor a variety of mutations that reduce the efficiency of processing, trafficking, splicing and operation of CFTR to down-regulate such secretion.

· Other hypotheses

The pleiotropic effects of CFTR mutations are considerable, as is the complexity of CFTR's molecular structure. Because of that, many investigators hypothesize that CFTR has roles beyond those of a plasma membrane Cl- channel.

A modest extension of the hypothesis is that CFTR might function as an important Cl- channel within internal membranes as well as the plasma membrane. The Cl- conductance of some internal compartments in the cell is known to be a key component of pH regulation within those compartments--pH is in turn known to affect enzymes differentially, so this is one plausible scenario by which mutations in CFTR could lead to altered biosynthesis of mucins.

Another hypothesis is that CFTR-mediated Cl- conductance, or some other aspect of CFTR function, is critical for vesicle trafficking to and from the plasma membrane. Depending on what is in the vesicle membrane and lumen, such a mechanism could profoundly effect the composition of the apical membrane and might also influence the secretion of macromolecules.

A more profound extension of the hypothesis is based on the similarity in domain structure to P-glycoprotein, a known transport ATPase. This has led to the hypothesis that CFTR may transport some substance from the cell, which might then act to influence other channels or transporters. A twist on this hypothesis is that CFTR actually conducts larger molecules from the cells, such as ATP.

Finally, CFTR might have some influence on the functioning of white blood cells.

None of these hypothesis denies the critical importance of CFTR-mediated Cl- conductance across the apical membrane of epithelia, and its sometimes critical role in fluid secretion and salt absorption. There is evidence for each of these hypotheses, and they are not mutually exclusive.

· Summary

Since first formulated by Quinton in 1983, the hypothesis that Cl- conductance is central to CF pathophysiology has gradually emerged as the most powerful unifying hypothesis to explain cystic fibrosis. The cloning of the CFTR and the analysis of its function have strengthened the hypothesis considerably. However, some of the pleiotropic effects of CFTR mutations, some of the complex structure of CFTR, and some of the still mysterious properties of CF lung disease remain to be explained, and may yet require a considerable elaboration of CFTR's basic functions.

· Suggestions for further reading

BOOKS

A comprehensive review of modern findings in basic and clinical cystic fibrosis research by some of the leading workers in the field.

Cystic Fibrosis. Taussig, L.M. (ed.) New York: Thieme-Stratton, 1984

Now dated in parts, but still excellent descriptions of some of the classic work in the field, especially clinical observations.

REVIEWS

Chloride channels in the apical membrane of normal and cystic fibrosis airway and intestinal epithelia.

American Journal of Physiology, 1992 Jul, 263(1 Pt 1):L1-14.

Anderson MP; Sheppard DN; Berger HA; Welsh MJ.

Chloride channels in the apical membrane of normal and cystic fibrosis airway and intestinal epithelia.

American Journal of Physiology, 1992 Jul, 263(1 Pt 1):L1-14.

McIntosh I; Cutting GR.

Cystic fibrosis transmembrane conductance regulator and the etiology and pathogenesis of cystic fibrosis.

FASEB Journal, 1992, 6: 2775-2782.

Fuller CM; Benos DJ.

CFTR!.

American Journal of Physiology, 1992 Aug, 263(2 Pt 1):C267-86.

Riordan JR.

The cystic fibrosis transmembrane conductance regulator.

Annual Review of Physiology, 1993, 55:609-30.

Riordan JR; Chang XB.

CFTR, a channel with the structure of a transporter.

Biochimica et Biophysica Acta, 1992 Jul 17, 1101(2):221-2.

Welsh MJ; Smith AE.

Molecular mechanisms of CFTR chloride channel dysfunction in cystic fibrosis.

Cell, 1993 Jul 2, 73(7):1251-4.

ARTICLES

Multi-ion pore behaviour in the CFTR chloride channel.

Nature, 1993 Nov 4, 366(6450):79-82.

Anderson MP; Berger HA; Rich DP; Gregory RJ; Smith AE; Welsh MJ.

Nucleoside triphosphates are required to open the CFTR chloride channel.

Cell, 1991 Nov 15, 67(4):775-84.

Hwang TC; Nagel G; Nairn AC; Gadsby DC.

Regulation of the gating of cystic fibrosis transmembrane conductance regulator C1 channels by phosphorylation and ATP hydrolysis.

Proceedings of the National Academy of Sciences of the United States of

America, 1994 May 24, 91(11):4698-702.

Baukrowitz T; Hwang TC; Nairn AC; Gadsby DC.

Coupling of CFTR Cl- channel gating to an ATP hydrolysis cycle.

Neuron, 1994 Mar, 12(3):473-82.

Veeze HJ; Halley DJ; Bijman J; de Jongste JC; de Jonge HR; Sinaasappel M.

Determinants of mild clinical symptoms in cystic fibrosis patients.

Residual chloride secretion measured in rectal biopsies in relation to the genotype.

Journal of Clinical Investigation, 1994 Feb, 93(2):461-6.

Engelhardt JF; Yankaskas JR; Ernst SA; Yang Y; Marino CR; Boucher RC; Cohn

JA; Wilson JM.

Submucosal glands are the predominant site of CFTR expression in the human bronchus.

Nature Genetics, 1992 Nov, 2(3):240-8.

Clarke LL; Grubb BR; Yankaskas JR; Cotton CU; McKenzie A; Boucher RC.

Relationship of a non-cystic fibrosis transmembrane conductance regulator-mediated chloride conductance to organ-level disease in Cftr(-/-) mice.

Proceedings of the National Academy of Sciences of the United States of

America, 1994 Jan 18, 91(2):479-83.

Delaney SJ; Rich DP; Thomson SA; Hargrave MR; Lovelock PK; Welsh MJ; Wainwright BJ.

Cystic fibrosis transmembrane conductance regulator splice variants are not conserved and fail to produce chloride channels.

Nature Genetics, 1993 Aug, 4(4):426-31.

Sheppard DN; Rich DP; Ostedgaard LS; Gregory RJ; Smith AE; Welsh MJ.

Mutations in CFTR associated with mild-disease-form Cl- channels with altered pore properties.

Nature, 1993 Mar 11, 362(6416):160-4.

Haws C; Finkbeiner WE; Widdicombe JH; Wine JJ.

CFTR in Calu-3 human airway cells: channel properties and role in cAMP-activated Cl- conductance.

American Journal of Physiology, 1994 May, 266(5 Pt 1):L502-12.

Shen BQ; Finkbeiner WE; Wine JJ; Mrsny RJ; Widdicombe JH.

Calu-3: a human airway epithelial cell line that shows cAMP-dependent Cl- secretion.

American Journal of Physiology, 1994 May, 266(5 Pt 1):L493-501.

Morris AP; Frizzell RA.

Vesicle targeting and ion secretion in epithelial cells: implications for cystic fibrosis.

Annual Review of Physiology, 1994, 56:371-97.

Return to Genotype & Phenotype.

Return to HomePage.