Cystic Fibrosis

A number sign (#) is used with this entry because cystic fibrosis is caused by homozygous or compound heterozygous mutation in the cystic fibrosis conductance regulator gene (CFTR; 602421) on chromosome 7q31.

Cystic fibrosis (CF) is classically described as a triad of chronic obstructive pulmonary disease, exocrine pancreatic insufficiency, and elevation of sodium and chloride concentration in sweat. Almost all males with CF are infertile due to congenital bilateral absence of the vas deferens. The disorder is associated with decreased longevity (summary by Cutting, 2002).

For discussion of a phenotype consisting of bronchiectasis with or without elevated sweat chloride caused by mutation in the genes encoding the 3 subunits of the epithelial sodium channel, see BESC1 (211400).

The mildest extreme of CF is represented by patients not diagnosed until middle age (Scully et al., 1977). The phenotypic variability in CF was analyzed by Sing et al. (1982). In an inbred kindred in North Carolina, a mild form of cystic fibrosis was described by Knowles et al. (1989). There was 1 instance of mother-daughter involvement, the mother being related to her husband. One of the presumed homozygotes was a 62-year-old woman. Another was her 52-year-old sister, the mother of the affected proposita. The daughter was an intensive care nurse, the mother of a normal daughter. Manifestations in the family were predominantly pulmonary; pancreatic exocrine insufficiency was not a conspicuous feature, especially in the older patients.

The 2 subgroups defined by the A and C haplotypes of polymorphisms closely linked to the CF locus on chromosome 7, reported by Estivill et al. (1987), have clinical differences in terms of the frequency of meconium ileus, pseudomonas infections, and pancreatic disease (Woo, 1988).

Gasparini et al. (1990) described a RFLP DNA marker closely linked to the CF locus which showed an allelic correlation with severity of the disorder: the genotype 2/2 was associated with severe disease; the genotype 1/2 was overrepresented in patients with very mild clinical manifestations, including pancreatic insufficiency, absence of meconium ileus, and absence of Pseudomonas colonization.

Meconium Ileus

Allan et al. (1981) showed that sibs tend to show recurrence of meconium ileus as a feature of cystic fibrosis. The distal intestinal obstruction syndrome is a 'meconium ileus equivalent' that occurs in adolescents and adults with CF. It is the consequence of the abnormally viscid mucofeculant material in the terminal ileum and right colon, where the fecal stream is normally liquid. Typical features are recurrent episodes of RLQ pain with palpable mass in the right iliac fossa. Symptoms are exacerbated by eating.

Mornet et al. (1988) determined the haplotype associated with cystic fibrosis in 41 families using 4 DNA probes, all of which are tightly linked to the CF gene. In 17 of the families an affected child had meconium ileus, and in the other 24 families there was a child without meconium ileus. A different haplotype was associated with the 2 types of families, suggesting that multiple allelism, i.e., different mutations at the same locus, accounts for CF with or without meconium ileus.

Liver Disease

Gaskin et al. (1988) found that 96% of patients with cystic fibrosis and evidence of liver disease had biliary tract obstruction, usually a stricture of the distal common bile duct. All patients without liver disease had normal intrahepatic and common-duct excretion of tracer.

Bilton et al. (1990) described a case of cystic fibrosis complicated by common bile duct stenosis.

Gabolde et al. (2001) showed that the presence of cirrhosis in patients with cystic fibrosis is significantly associated with either homozygous or compound heterozygous mutations in the MBL2 gene (154545), which encodes mannose-binding lectin (MBL). The authors compared 216 patients homozygous for the delta-F508 mutation (602421.0001) and found that 5.4% of those homozygous or compound heterozygous for wildtype mannose-binding lectin had cirrhosis, while 30.8% of those homozygous or compound heterozygous for mutant alleles had cirrhosis (p = 0.008).

Approximately 3 to 5% of patients with cystic fibrosis develop severe liver disease defined as cirrhosis with portal hypertension. Bartlett et al. (2009) performed a 2-stage case control study enrolling patients with CF and severe liver disease with portal hypertension from 63 CF centers in the United States as well as 32 in Canada and 18 outside of North America. In the first stage, 124 patients with CF and severe liver disease, enrolled between January 1999 and December 2004, and 843 control patients without CF-related liver disease (all assessed at greater than 15 years of age) were studied by genotyping 9 polymorphisms in 5 genes previously studied as modifiers of liver disease in CF. In the second stage, the 2 genes that were positive from the first stage were tested in an additional 136 patients with CF-related liver disease, enrolled between January 2005 and February 2007, and in 1,088 with no CF-related liver disease. The combined analysis of the initial and replication studies by logistic regression showed CF-related liver disease to be associated with the SERPINA1 Z allele (107400.0011) (odds ratio = 5.04; 95% confidence interval, 2.88-8.83; p = 1.5 x 10(-8)). Bartlett et al. (2009) concluded that the SERPINA1 Z allele is a risk factor for liver disease in CF. Patients carrying the Z allele are at greater risk (odds ratio = approximately 5) of developing severe liver disease with portal hypertension.

Pancreatic Insufficiency

Approximately 15% of CF patients do not have pancreatic insufficiency, i.e., are 'pancreatic sufficient.' Kerem et al. (1989) performed linkage disequilibrium and haplotype association studies of patients in 2 clinical subgroups, one pancreatic insufficient (PI) and the other pancreatic sufficient (PS). Significant differences were found in allelic and haplotype distributions in the 2 groups. The data suggested that most of the CF-PI patients were descendants of a single mutational event at the CF locus, whereas the CF-PS patients resulted from multiple, different mutations. Corey et al. (1989) commented on the intrafamilial concordance for pancreatic insufficiency in CF.

Devoto et al. (1989) studied the allele and haplotype frequencies of 5 polymorphic DNA markers near the CF locus in 355 CF patients from Belgium, the German Democratic Republic, Greece, and Italy who were divided into 2 groups according to whether or not they were taking supplementary pancreatic enzymes. The distributions of alleles and haplotypes revealed by 2 of the probes were always different in patients with or without pancreatic insufficiency in all the populations studied. In the case of 1 haplotype that was present in 73% of all the CF chromosomes in their sample, they found homozygosity in only 28% of patients without pancreatic insufficiency as contrasted with 64% who were homozygous and had pancreatic insufficiency. Like other workers, they concluded that this indicated that pancreatic insufficiency and sufficiency are associated with different mutations at the CF locus.

Ferrari et al. (1990) studied the distribution of haplotypes based on 8 polymorphic DNA markers linked to CF in 163 Italian patients and correlated the findings with clinical presentation. Among 19 pancreatic sufficient patients, 6 (31.6%) showed at least 1 copy of a rare phenotype which was present in only 16 of 138 patients (11.6%) with pancreatic insufficiency. In addition, only 5 pancreatic sufficient patients were homozygous for the common 2,1 haplotype as compared with 88 patients (63.8%) with pancreatic insufficiency. Kristidis et al. (1992) likewise found intrafamilial consistency of the pancreatic phenotype, whether pancreatic sufficient or insufficient. Furthermore, the PS phenotype occurred in patients who had 1 or 2 mild CFTR mutations, such as arg117-to-his (602421.0005), arg334-to-trp (602421.0034), arg347-to-pro (602421.0006), ala455-to-glu (602421.0007), and pro574-to-his (602421.0018), whereas the PI phenotype occurred in patients with 2 severe alleles, such as phe508-to-del (602421.0001), ile507-to-del (602421.0002), gln493-to-ter (602421.0003), gly542-to-ter (602421.0009), arg553-to-ter (602421.0014), and trp1282-to-ter (602421.0022).

Borgo et al. (1993) commented on the phenotypic intrafamilial heterogeneity displayed by an Italian family in which 3 sibs, 2 of whom were dizygotic twins, were compound heterozygotes for the delF508 (602421.0001) and the 1717,-1,G-A splicing mutation (602421.0008). While close intrafamilial concordance was found for exocrine pancreatic phenotype, the pulmonary phenotype varied widely. They suggested that interaction of the CFTR protein with tissue-specific proteins or the action of modifier loci (which may be operationally identical possibilities) plays a role in intrafamilial variability.

Barreto et al. (1991) concluded that the father of a girl with severe CF also had CF but was mildly affected. The child was homozygous for the delta-F508 mutation associated with haplotype B; the father was a compound heterozygote for this mutation and a second CF mutation associated with haplotype C. Perhaps it should not be surprising that some patients with cystic fibrosis have no pancreatic lesions (Oppenheimer, 1972).

Sharer et al. (1998) and Cohn et al. (1998) demonstrated that heterozygosity for CFTR mutations can lead to 'idiopathic' chronic pancreatitis, especially when the mutation is associated with the 5T allele of the variable number of thymidines in intron 8 of the CFTR gene.

Pulmonary Disease

Pier et al. (1996) provided an experimental explanation for the susceptibility of CF patients to chronic Pseudomonas aeruginosa lung infections. They found that cultured human airway epithelial cells expressing the delta-F508 allele of the CFTR gene were defective in uptake of P. aeruginosa compared with cells expressing the wildtype allele. P. aeruginosa lipopolysaccharide-core oligosaccharide was identified as the bacterial ligand for epithelial cell ingestion; exogenous oligosaccharide inhibited bacterial ingestion in a neonatal mouse model, resulting in increased amounts of bacteria in the lungs. The authors concluded that CFTR may normally contribute to a host-defense mechanism that is important for clearance of P. aeruginosa from the respiratory tract.

Ernst et al. (1999) identified unique lipopolysaccharide structures synthesized by P. aeruginosa within CF patient airways. P. aeruginosa synthesized lipopolysaccharide with specific lipid A structures, indicating unique recognition of the CF airway environment. CF-specific lipid A forms containing palmitate and aminoarabinose were associated with resistance to cationic antimicrobial peptides and increased inflammatory responses, indicating that they are likely to be involved in airway disease.

Because mannose-binding lectin (MBL), encoded by the MBL2 gene (154545), is a key factor in innate immunity, and lung infections are a leading cause of morbidity and mortality in CF, Garred et al. (1999) investigated whether MBL variant alleles, which are associated with recurrent infections, might be risk factors for CF patients. In 149 CF patients, different MBL genotypes were compared with respect to lung function, microbiology, and survival to end-stage CF (death or lung transplantation). The lung function was significantly reduced in carriers of MBL variant alleles when compared with normal homozygotes. The negative impact of variant alleles on lung function was especially confined to patients with chronic Pseudomonas aeruginosa infection. Burkholderia cepacia infection was significantly more frequent in carriers of variant alleles than in homozygotes. The risk of end-stage CF among carriers of variant alleles increased 3-fold, and the survival time decreased over a 10-year follow-up period. Moreover, by using a modified life table analysis, Garred et al. (1999) estimated that the predicted age of survival was reduced by 8 years in variant allele carriers when compared with normal homozygotes.

Davies et al. (2000) found that MBL binds to Burkholderia cepacia, an important pathogen in patients with CF, and leads to complement activation, but that this was not the case for Pseudomonas aeruginosa, the more common colonizing organism in CF. Davies et al. (2000) suggested that patients with CF and mannose-binding lectin deficiency would be at a particularly high risk of B. cepacia colonization. The lack of binding to P. aeruginosa suggests that the effect of this organism on lung function in patients with MBL-deficient CF reflects a role for MBL, either in intercurrent infections with other organisms, or in the inflammatory process.

In an association study involving 112 patients with cystic fibrosis, Yarden et al. (2004) found that patients with the MBL2 A/O or O/O genotypes were more likely to have a more severe pulmonary phenotype than patients with the A/A genotype (p = 0.002). No association was found between the MBL2 genotype and the age at first infection with P. aeruginosa. Yarden et al. (2004) concluded that it is very likely that MBL2 is a modulating factor in cystic fibrosis.

Tarran et al. (2001) stated that there is controversy over whether abnormalities in the salt concentration or volume of airway surface liquid (ASL) initiate CF airway disease. Using CF mouse nasal epithelia, they showed that an increase in goblet cell number was associated with decreased ASL volume rather than abnormal Cl- concentration. Aerosolization of osmolytes in vivo failed to raise ASL volume. Osmolytes and pharmacologic agents were effective in producing isotonic volume responses in human airway epithelia but were typically short acting and less effective in CF cultures with prolonged volume hyperabsorption and mucus accumulation. These data showed that therapies can be designed to normalize ASL volume without producing deleterious compositional changes in ASL, and that therapeutic efficacy will likely depend on development of long-acting pharmacologic agents and/or an increased efficiency of osmolyte delivery.

In 69 Italian patients with CF due to homozygosity for the delF508 mutation in the CFTR gene (F508del; 602421.0001), De Rose et al. (2005) found that those who also carried the R131 allele of the immunoglobulin Fc-gamma receptor II gene (FCGR2A; see 146790.0001) had a 4-fold increased risk of acquiring chronic Pseudomonas aeruginosa infection (p = 0.042). De Rose et al. (2005) suggested that FCGR2A locus variability contributes to this infection susceptibility in CF patients.

Emond et al. (2012) used exome sequencing and an extreme phenotype study design to discover genetic variants influencing Pseudomonas aeruginosa infection in cystic fibrosis. Forty-three individuals with early age of onset of chronic P. aeruginosa infection (all below the tenth percentile of age at onset), and the 48 oldest individuals who had not reached chronic P. aeruginosa infection (all past the mean age of onset) were sequenced. After Bonferroni adjustment, a single gene, DCTN4, was significantly associated with time to chronic P. aeruginosa infection (naive P = 2.2 x 10(-6); adjusted P = 0.025). Twelve of the 43 individuals in the early extreme sample carried a missense variant in DCTN4, 9 a phe349-to-leu substitution (F349L; rs11954652) and 3 a tyr270-to-cys substitution (Y270C; rs35772018). None of the 48 individuals in the late P. aeruginosa extreme sample had either missense variant. Subsequently, 696 individuals with varied CFTR genotypes were studied. Seventy-eight participants were heterozygous and 9 were homozygous for the F349L (614758.0001) mutation; 15 were heterozygous for the Y270C (614758.0002) mutation; 1 individual was heterozygous for both mutations. The presence of at least 1 DCTN4 missense variant was significantly associated with both early age of first P. aeruginosa-positive culture (p = 0.01, hazard ratio = 1.4) and with early age of onset of chronic P. aeruginosa infection (p = 0.004, hazard ratio = 1.9). The risk was highest in individuals with less selective bias toward a P. aeruginosa-negative history, i.e., children enrolled before 1.5 years of age and 103 enrollees who participated in the study despite a history of P. aeruginosa-positive cultures. No significant interaction was found between CFTR genotypes and DCTN4 mutations, although power to detect such an interaction was low.

Infertility

Oppenheimer et al. (1970) suggested that characteristics of cervical mucus may account for infertility in females with cystic fibrosis. Congenital bilateral absence of the vas deferens (CBAVD; 277180) is a usual cause of male infertility in cystic fibrosis. It also occurs with CFTR mutations in heterozygous state, especially when associated with the polymorphic number of thymidines in intron 8, specifically the 5T allele.

Carcinoma

Siraganian et al. (1987) pointed to adenocarcinoma of the ileum in 3 males with cystic fibrosis. The diagnosis was made between ages 29 and 34 years.

From a pancreatic adenocarcinoma developing in a 26-year-old patient with cystic fibrosis due to the phenylalanine-508 deletion, Schoumacher et al. (1990) established a cell line in which the cells showed morphologic and chemical characteristics typical of pancreatic duct cells and showed physiologic properties of CF cells. Schoumacher et al. (1990) suggested that the cell line, which had been stable through more than 80 passages over a 2-year period, could serve as a continuous cell line for studies of the CF defect. Bradbury et al. (1992) demonstrated that the CFTR protein is involved in cAMP-dependent regulation of endocytosis and exocytosis. In a study of pancreatic cancer cells derived from a CF patient, they found that plasma membrane recycling did not occur until normal CFTR was provided.

Neglia et al. (1995) performed a retrospective cohort study of the occurrence of cancer in 28,511 patients with cystic fibrosis from 1985 through 1992 in the United States and Canada. The number of cases observed was compared with the number expected, calculated from population-based data on the incidence of cancer. They also analyzed proportional incidence ratios to assess the association between specific cancers and cystic fibrosis in Europe. The final results indicated that although the overall risk of cancer in patients with cystic fibrosis is similar to that of the general population, there is an increased risk of digestive tract cancers. They recommended that persistent or unexplained gastrointestinal symptoms in CF patients should be carefully investigated.

Patients with cystic fibrosis have altered levels of plasma fatty acids. Affected tissues from cystic fibrosis knockout mice show elevated levels of arachidonic acid and decreased levels of docosahexaenoic acid. Freedman et al. (2004) performed studies of fatty acids in nasal and rectal biopsy specimens, nasal epithelial scrapings, and plasma from 38 patients with cystic fibrosis, and found alterations in fatty acids similar to those in the knockout mice.

Other Features

Delayed puberty is common among individuals with cystic fibrosis and is usually attributed to chronic disease and/or poor nutrition. However, delayed puberty has been reported as a feature of CF even in the setting of good nutritional and clinical status (Johannesson et al., 1997).

Recessive inheritance of cystic fibrosis was first shown clearly by Lowe et al. (1949). Roberts (1960) collected family data which appeared to him inconsistent with the quarter ratio expected of a recessive trait. Bulmer (1961) pointed out, however, that when proper correction is made for ascertainment bias, the observed proportions may agree with those expected for a recessive trait.

Rather than estimating the frequency of the CF gene from the square root of the incidence figure, Danks et al. (1983) used the frequency of CF in first cousins. The estimate of gene frequency was 0.0281 as contrasted with 0.0198 (based on direct count). Danks et al. (1983) suggested that the disparity between the 2 estimates might be the existence of 2 gene loci, each with a frequency of 0.0140 for the CF gene and a heterozygote frequency of 1 in 36. Thus, in Victoria, Australia, 1 in 18 persons might be heterozygous at one or the other locus. Later, however, the authors published a retraction and concluded that they had no evidence of more than 1 locus.

For risk analysis in cystic fibrosis, Edwards and Miciak (1990) proposed a simple procedure called the 'slash sheet.' They pointed out that the various methods of estimating genetic risk fall into 2 main groups: first, enumerating all possibilities and excluding those inconsistent with the tests, a simple procedure in small families, and second, using conditional arguments. The latter approach uses Bayes theorem. The former approach, Edwards and Miciak (1990) pointed out, follows a procedure advanced in 1654 by Pascal, following correspondence with Fermat, on the problem of the Chevalier de Mere, now known as the 'problem of points.' Two noblemen were gambling, and, while one was winning, the other was called away and the game was abandoned. How should the stakes be divided? Edwards and Miciak (1990) noted that 'genetic risk is merely an unfinished game of chance.'

See Hodge et al. (1999) for a discussion of calculation of CF risk in a fetus with 1 identified mutation in CFTR and echogenic bowel.

Park et al. (1987) concluded that CF is distal to and on the 5-prime side of MET. They determined this by in situ hybridization on metaphase and prometaphase chromosomes of normal lymphocytes as well as lymphoblastoid cells containing a t(5;7)(q35;q22). Normal cells showed clustering of MET grains to 7q31. Furthermore, in the lymphoblastoid cell line, there was significant labeling within the 5q+ chromosome, confirming that MET is located distal to 7q22 with most grains clustered at 7q31. Somatic cell hybrids containing the derivative 7 showed on Southern analysis that the 3-prime portion of the MET gene, but not the 5-prime portion, was located there; thus, MET is at the translocation breakpoint. Studies in another cell line with a 7q32 translocation breakpoint indicated that MET is located at or proximal to 7q32. A break at this site was accompanied by loss of 3 markers within 1 cM of CF, suggesting that if MET is at the breakpoint on 7q31, CF is located distally.

In the course of studying a case of cystic fibrosis, Spence et al. (1988) discovered what appeared to be a case of uniparental disomy: the father did not contribute alleles to the propositus for markers near the CF locus or for centromeric markers on chromosome 7. High-resolution cytogenetic analysis was normal, and the result could not be explained by nonpaternity or a submicroscopic deletion. Uniparental disomy could be explained by various mechanisms such as monosomic conception with subsequent chromosome gain, trisomic conception followed by chromosome loss, postfertilization error, or gamete complementation. Patients with more than one genetic disorder might be suspected of having isodisomy, which should also be suspected in cases of an apparent new mutation leading to a recessive disorder when only 1 parent is heterozygous, and in cases of females affected with X-linked recessive disorders. Engel (1980) appears to have originated the concept of uniparental disomy and resulting isodisomy. Voss et al. (1988, 1989) also demonstrated uniparental disomy for chromosome 7 in a patient with cystic fibrosis.

Mayo et al. (1980) attempted to map the cystic fibrosis gene by study of CF x mouse cell hybrids and examination for production of the cystic fibrosis mucociliary inhibitor. The strongest chance of assignment was for chromosome 4. Scambler et al. (1985) found that the albumin locus labeled by a DNA clone did not segregate with CF or with any of 6 other chromosome 4 markers. They estimated that about half the length of chromosome 4 was accounted for by the markers used. Eiberg et al. (1984) found a hint of linkage to F13B (134580); the maximum lod score was 1.71 at a recombination fraction of 0.05 for males and females combined. Linkage with 56 other genetic markers was negative (Eiberg et al., 1984). Eiberg et al. (1985) showed that cystic fibrosis and paraoxonase (PON; 168820) are linked; the maximum lod score was 3.70 at theta = 0.07 in males and 0.00 in females.

Tsui et al. (1985) found that the CF locus is linked to that of a DNA marker which is also linked to the PON locus, which in turn by independent evidence is linked to CF, thus closing the circle. The DNA marker was provisionally called D0CRI-917. The interval between the marker and PON was about 5 cM and the interval between it and CF about 15 cM. Whether the order is marker--PON--CF or PON--marker--CF was not certain; the former order was favored by 9:5 odds. Knowlton et al. (1985) reported that the anonymous probe D0CRI-917, linked to CF with about 15% recombination, is located on chromosome 7. White et al. (1985) showed very tight linkage to the MET oncogene (164860), which was assigned to the midportion of 7q. Wainwright et al. (1985) reported tight linkage also to the gene for another anonymous DNA probe, pJ3.11, which was assigned to 7cen-7q22. The closely linked probes pJ3.11 and MET are sufficiently informative to permit carrier detection in 80% of families in which there is a living CF child and unaffected sibs (Farrall et al., 1986). Scambler et al. (1985) showed that the COL1A2 gene (120160) is linked to CF (maximum lod for the sexes combined = 3.27 at a male recombination fraction of 0.08 and a female recombination fraction of 0.15.) PON and CF show recombination frequency of about 10%. CF is about 10 cM from both TCRB (see 186930) and COL1A2. TCRB and COL1A2 are not closely linked; thus, CF lies between them in the proximal part of 7q22. Wainwright et al. (1986) presented linkage data for COL1A2 versus CF (lod = 3.58 at theta = 0.10), TCRB versus CF (lod = 2.20 at theta = 0.15) and TCRB versus PON (all lods negative). Based on combined linkage data from 50 informative 2-generation families, Buchwald et al. (1986) concluded that CF is 19 cM from COL1A2, which is located at 7q21.3-q22.1. COL1A2 is closely linked to D7S15 and to PON. The probable order is COL1A2--D7S15--PON--CF. The regional localization of CF is 7q22.3-q23.1. Linkage of cystic fibrosis to various DNA markers and/or classical markers was reported in a series of articles by Beaudet et al. (1986), White et al. (1986), Bowcock et al. (1986), Farrall et al. (1986), Tsui et al. (1986), Spence et al. (1986), and Watkins et al. (1986). In Amish/Mennonite/Hutterite kindreds, Klinger et al. (1986) and Watkins et al. (1986) found close linkage with markers on chromosome 7, consistent with locus homogeneity for the defect causing CF in the populations that had been examined to date.

Estivill et al. (1987) identified a candidate for the cystic fibrosis locus by using a 'rare-cutter cosmid library.' They found a genomic region with the characteristics of an HTF island in high linkage disequilibrium with CF. The fact that the sequence was conserved throughout mammalian evolution strengthens the view that this is the CF gene. HTF islands, standing for HpaII tiny fragments, have a sequence length of between 500 and 1000 bp and often include the first exons as well as upstream sequences 5-prime to coding genes (Bird, 1986; Brown and Bird, 1986). These HTF islands are regions of DNA rich in the nonmethylated dinucleotide CpG and contain clusters of sites for CpG-methylation-sensitive restriction enzymes. (There are about 30,000 HTF islands in the human genome.) Estivill et al. (1987) stated that 94% of the chromosomes are of haplotype B, which is present in only 34% of the chromosomes in the general population. In 127 Italian families, Estivill et al. (1988) studied linkage disequilibrium of markers at the locus containing the CpG-enriched methylation-free island designated D7S23. In a search for deletions by means of field inversion gel electrophoresis (FIGE), Morreau et al. (1988) analyzed DNA from 10 cystic fibrosis patients representing 19 different CF chromosomes. No differences were detected after digestion of the samples with 2 different restriction enzymes and hybridization with 4 different probes. The authors estimated that the percentage of deletions occurring within the CF region is less than 15.2% (95% confidence interval, N = 19). The fact that no patient with a combination of cystic fibrosis and a genetic syndrome due to a second affected locus in close vicinity to the CF locus has been described suggests that deletions are rare. Beaudet et al. (1989) found strong linkage disequilibrium between the CF locus and closely linked markers on chromosome 7. By in situ hybridization Duncan et al. (1988) mapped 2 DNA sequences closely linked to the CF locus to 7q31.3-q32. This is a more distal location than had been inferred from previous data.

Using cystic fibrosis and published CF haplotypes as the test bed, Collins and Morton (1998) illustrated how allelic association can be efficiently combined with linkage evidence to identify a region for positional cloning of a disease gene.

For an extensive discussion of the molecular genetics of cystic fibrosis and a listing of allelic variants of the CFTR gene, see 602421.

Collins (1992) gave an update concerning the molecular biology of CF and the therapeutic implications thereof.

O'Sullivan and Freedman (2009) reviewed the clinical features, pathogenesis, diagnosis, molecular genetics, and current state of gene therapy in CF.

Vitale et al. (1986) found close linkage of the CF gene and the MET locus in 12 unrelated Italian cystic fibrosis families, thus supporting their hypothesis of genetic homogeneity based on the analysis of consanguineous marriages among 624 couples of CF parents. Lander and Botstein (1986) and Romeo et al. (1986) discussed further the consanguinity method for studying heterogeneity in cystic fibrosis. Estivill et al. (1987) used their haplotype data to argue against genetic heterogeneity at the CF locus. They proposed that the great majority of CF mutations found in the population arose from an original mutational event which occurred in the Caucasian population after racial divergence in man.

Nonclassic forms of CF have been associated with mutations that reduce but do not eliminate the function of the CFTR protein. Mekus et al. (1998) described a patient with a nonclassic CF phenotype in whom no CFTR mutations could be found. Groman et al. (2002) assessed whether alteration in CFTR function is responsible for the entire spectrum of nonclassic CF phenotypes. Extensive genetic analysis of the CFTR gene was performed in 74 patients with nonclassic CF. Furthermore, they evaluated 2 families that each included a proband without identified CFTR mutations and a sib with nonclassic CF to determine whether there was linkage to the CFTR locus and to measure the extent of CFTR function in the sweat gland and nasal epithelium. Of the 74 patients studied, Groman et al. (2002) found that 29 had 2 mutations in the CFTR gene (i.e., were either homozygous or compound heterozygous at the CFTR locus), 15 had 1 mutation, and 30 had no mutations. A genotype of 2 mutations was more common among patients who had been referred after screening for a panel of common CF-causing mutations that had identified 1 mutation than among those who had been referred after screening had identified no such mutations. Comparison of clinical features and sweat chloride concentrations revealed no significant differences among patients with 2, 1, or no CFTR mutations. Haplotype analysis in the 2 families in which 2 sibs had nonclassic CF showed no evidence of linkage to CFTR. Although each of the affected sibs had elevated sweat chloride concentrations, measurements of cAMP-mediated ion and fluid transport in the sweat gland and nasal epithelium demonstrated the presence of functional CFTR. Groman et al. (2002) concluded that factors other than mutations in the CFTR gene can produce phenotypes clinically indistinguishable from nonclassic CF caused by CFTR dysfunction.

Because proteinase-antiproteinase imbalances are common in both CF and alpha-1-antitrypsin deficiency (613490), Meyer et al. (2002) investigated the hypothesis that the common AAT deficiency alleles PI Z (107400.0011) and PI S (107400.0013) contribute to pulmonary prognosis in CF. In 269 CF patients from southern Germany, they determined the serum concentrations of AAT (107400) and C-reactive protein (CRP; 123260) by nephelometry and screened for the common AAT deficiency alleles by PCR and restriction enzyme digest. The onset of chronic bacterial colonization by P. aeruginosa was correlated with the AAT phenotypes PI MM, PI MS, and PI MZ. Only 3 of 9 (33%) CF patients diagnosed with either PI MS or PI MZ had developed chronic P. aeruginosa lung infection earlier in their lives; the remaining 6 PI MS or PI MZ patients showed a later onset of chronic P. aeruginosa lung infection. The results suggested that PI MS and PI MZ are not associated with a worse pulmonary prognosis in CF.

Mekus et al. (2003) examined modifying factors in CF by studying 34 highly concordant and highly discordant delF508 homozygous sib pairs selected from a group of 114 pairs for extreme disease phenotypes by nutritional and pulmonary status. They were typed for SNPs and short tandem repeat polymorphisms (STRPs) in a 24-cM CFTR-spanning region. Allele frequencies differed significantly at D7S495, located within a 21-cM distance 3-prime of CFTR, comparing concordant mildly affected, concordant severely affected, and discordant sib pairs. A rare haplotype of 2 SNPs within the leptin gene promoter (LEP; 164160) was found exclusively among the concordant mildly affected pairs. All concordant sib pairs shared the paternal delF508 chromosome between CFTR and D7S495, while the cohort of discordant sib pairs inherited equal proportions of recombined and nonrecombined parental chromosomes. Mekus et al. (2003) concluded that disease manifestation in CF is modulated by loci in the partially imprinted region 3-prime of CFTR that determine stature, food intake, and energy homeostasis, such as the Silver-Russell syndrome (180860) candidate gene region and LEP.

There is great variability of pulmonary phenotype and survival in cystic fibrosis, even among patients who are homozygous for the most prevalent mutation, delF508 (602421.0001). Although environmental influences may modify clinical disease, there is probably additional genetic variation (i.e., modifier genes) that contribute to the expression of the final phenotype. Drumm et al. (2005) studied variants of 10 genes previously reported as modifiers in cystic fibrosis in 2 studies with different patient samples. They first tested 808 patients who were homozygous for the delF508 mutation and were classified as having either severe or mild lung disease. Significant allelic and genotypic associations with phenotype were seen only for TGFB1 (190180), the gene encoding transforming growth factor beta-1, particularly the -509 and codon 10 polymorphisms. The odds ratio was about 2.2 for the highest-risk TGFB1 genotype (codon 10 CC; 190180.0007) in association with the phenotype of severe lung disease. In the replication (second) study, Drumm et al. (2005) tested 498 patients, with various CFTR genotypes and a range of values for forced expiratory volume in 1 second (FEV1), for an association of the TGFB1 codon 10 CC genotype with low FEV1. This replication study confirmed the association of the TGFB1 codon 10 CC genotype with more severe lung disease.

Buranawuti et al. (2007) determined the genotype of 4 variants of 3 putative CF modifier genes (TNF-alpha-238; TNF-alpha-308, 191160.0004; TGF-beta-509; and MBL2 A/O) in 3 groups of CF patients: 101 children under 17 years of age, 115 adults, and 38 nonsurviving adults (21 deceased and 17 lung transplant after 17 years of age). Genotype frequencies among adults and children with CF differed for TNF-alpha-238 (G/G vs G/A, p = 0.022) and MBL2 (A/A vs O/O, p = 0.016), suggesting that MBL2 O/O is associated with reduced survival beyond 17 years of age, whereas TNF-alpha-238 G/A appears to be associated with an increased chance of surviving beyond 17 years of age. When adults with CF were compared to nonsurviving adults with CF, genotype frequencies of both genes differed (TNF-alpha 238 G/G vs G/A, p = 0.0015; MBL2 A/A vs O/O, p = 0.009); the hazard ratio for TNF-alpha-238 G/G versus G/A was 0.25 and for MLB2 O/O versus A/A or A/O was 2.5. Buranawuti et al. (2007) concluded that TNF-alpha-238 G/A and MBL2 O/O genotypes appear to be genetic modifiers of survival in patients with CF.

In a study of 1,019 Canadian pediatric CF patients, Dorfman et al. (2008) found a significant association between earlier age of first P. aeruginosa infection and MBL2 deficiency (onset at 4.4, 7.0, and 8.0 years for low, intermediate, and high MBL2 groups according to MBL2 genotype, respectively; p = 0.0003). This effect was amplified in patients with the high-producing genotypes of TGFB1, including variant C of codon 10. MBL2 deficiency was also associated with more rapid decline of pulmonary function, most significantly in those homozygous for the high-producing TGFB1 genotypes (p = 0.0002). However, although TGFB1 affected the modulation of age of onset by MBL2, there was no significant direct impact of TGFB1 codon 10 genotypes alone. The findings provided evidence for a gene-gene interaction in the pathogenesis of CF lung disease, whereby high TGFB1 production enhances the modulatory effect of MBL2 on the age of first bacterial infection and the rate of decline of pulmonary function.

Using quantitative transmission disequilibrium testing of 472 CF patient/parent trios, Bremer et al. (2008) found significant transmission distortion of 2 TGFB1 SNPs, -509 (rs1800469) and codon 10 (rs1982073), when patients were stratified by CFTR genotype. Although lung function and nutritional status are correlated in CF patients, there was no evidence of association between the TGFB1 SNPs and variation in nutritional status. A 3-SNP haplotype (CTC) composed of the -509 SNP C allele, the codon 10 T allele, and a 3-prime SNP rs8179181 C allele was highly associated with increased lung function in patients grouped by CFTR genotype. Bremer et al. (2008) concluded that TGFB1 is a modifier of CF lung disease, with a beneficial effect of certain variants on the pulmonary phenotype.

To identify genetic modifiers of lung disease severity in cystic fibrosis, Gu et al. (2009) performed a genomewide single-nucleotide polymorphism scan in 1 cohort of cystic fibrosis patients, replicating top candidates in an independent cohort. This approach identified IFRD1 (603502) as a modifier of cystic fibrosis lung disease severity. IFRD1 is a histone deacetylase-dependent transcriptional coregulator expressed during terminal neutrophil differentiation. Neutrophils, but not macrophages, from Ifrd1-null mice showed blunted effector function, associated with decreased NF-kappa-B p65 (RELA; 164014) transactivation. In vivo, IFRD1 deficiency caused delayed bacterial clearance from the airway, but also less inflammation and disease--a phenotype primarily dependent on hematopoietic cell expression, or lack of expression, of IFRD1. In humans, IFRD1 polymorphisms were significantly associated with variation in neutrophil effector function. Gu et al. (2009) concluded that IFRD1 modulates the pathogenesis of cystic fibrosis lung disease through the regulation of neutrophil effector function.

Association with Epithelial Sodium Channel Subunits

Stanke et al. (2006) genotyped 37 delF508 homozygous sib pairs for markers on chromosome 12p13, encompassing the epithelial sodium channel (ENaC) subunit A (SCNN1A; 600228) and TNF-alpha receptor (TNFRSF1A; 191190) genes, and chromosome 16p12, encompassing the SCNN1B (600760) and SCNN1G (600761) genes, as potential CF disease modifiers. Transmission disequilibrium was observed at SCNN1G and association with CF phenotype intrapair discordance was observed at SCNN1B. Family-based and case-control analyses and sequencing uncovered an association of the TNFRSF1A intron 1 haplotype with disease severity. Stanke et al. (2006) suggested that the SCNN1B, SCNN1G, and TNFRSF1A genes may be modulators of CF disease by affecting changes in airway surface liquids and host inflammatory responses.

Fajac et al. (2008) screened the SCNN1B gene in 55 patients with idiopathic bronchiectasis (see 211400) who had 1 or no mutations in the CFTR gene and identified heterozygosity for 3 missense mutations in the SCNN1B gene (see, e.g., 600760.0015) in 5 patients, 3 of whom also carried a heterozygous mutation in CFTR (602421.0001 and 602421.0086). Fajac et al. (2008) concluded that variants in SCNN1B may be deleterious for sodium channel function and lead to bronchiectasis, especially in patients who also carry a mutation in the CFTR gene.

Viel et al. (2008) analyzed the SCNN1B and SCNN1G genes in 56 adult patients with classic CF and discordance between their respiratory phenotype and CFTR genotype, including 38 patients with a severe genotype and an unexpectedly mild lung phenotype, and 18 patients with a mild genotype and severe lung phenotype. Three patients carried at least 1 missense mutation in SCNN1B or SCNN1G, but analysis of sodium channel function by nasal potential difference (PD) measurements did not support that the variants were functional. Viel et al. (2008) concluded that variation in SCNN1B and SCNN1G genes do not modulate disease severity in the majority of CF patients.

Azad et al. (2009) identified several rare SCNN1A polymorphisms with an increased incidence in patients with a cystic fibrosis-like phenotype and 1 or no CFTR mutations versus controls, including several patients with no CFTR mutation who were heterozygous for a hyperactive variant (W493R; 600228.0007). The authors hypothesized that given the CF-carrier (3.3%) and the W493R-carrier (3.1%) frequency in some populations, there ma be a polygenic mechanism of disease involving CFTR and SCNN1A in some patients.

Mutesa et al. (2009) analyzed the CFTR gene in 60 unrelated Rwandan children who had CF-like symptoms and identified heterozygosity for a CFTR mutation in 5 patients (none were homozygous). Sequencing of the ENaC subunits revealed heterozygous mutations in the SCNN1A and SCNN1B genes in 4 patients, respectively, whereas the remaining patient was heterozygous for a mutation in both SCNN1B and SCNN1G. Among the 55 patients who were negative for mutation in CFTR, only polymorphisms were found in the ENaC genes. Mutesa et al. (2009) concluded that some cases of CF-like syndrome in Africa may be associated with mutations in CFTR and ENaC genes.

Frizzell (1987) pointed out that cystic fibrosis is of interest to neuroscientists because it appears to be a disease of ion channels. It is apparently not the conduction properties of ion channels that are affected, but rather their gating by chemical agonists. These conductance pathways appear to be unique to epithelial cells in which salt and water transport rates are governed by cyclic AMP and calcium-dependent regulatory processes.

Decrease in fluid and salt secretion is responsible for the blockage of exocrine outflow from the pancreas and the accumulation of heavy dehydrated mucous in the airways. In sweat glands, salt reabsorption is defective. This is the basis of the folkloric anecdote that the midwife would lick the forehead of the newborn and, if the sweat tasted abnormally salty, predict that the infant was destined to die of pulmonary congestion and its side effects. Quinton (1983) and Knowles et al. (1983) first suggested that the primary defect of cystic fibrosis may be in chloride transport. Widdicombe et al. (1985) demonstrated a cyclic AMP-dependent transepithelial chloride current in normal but not CF epithelia. The pathophysiology of cystic fibrosis, specifically the impermeability of epithelia to chloride ion, was reviewed by Welsh and Fick (1987).

Landry et al. (1989) purified several proteins from kidney and trachea that exhibit chloride channel activity when they are reconstituted into artificial phospholipid bilayer membranes. One or more of these proteins may turn out to be all or part of the secretory chloride channel that is defective in CF. Using antibodies against CFTR peptides, Marino et al. (1991) demonstrated that the CFTR molecule is located in and confined to the apical domain of pancreatic centroacinar and intralobular duct cells. From this they concluded that the proximal duct epithelial cells play a key role in the early events leading to pancreatic insufficiency in CF and that apical chloride transport by these cells is essential for normal pancreatic secretory function. Jetten et al. (1989) created a stable human airway epithelial cell line by retroviral transformation of CF airway epithelium. They found that it maintains the defect in the secretory chloride channel. Rich et al. (1990) expressed the CFTR gene in cultured cystic fibrosis airway epithelial