Molecular and Functional Basis of Cystic Fibrosis in Indian Patients: Genetic, Diagnostic and Therapeutic Implications

Introduction

Cystic fibrosis (CF MIM#219700) is the most common lethal autosomal recessive mono-genic disease in Caucasian population with an average prevalence of one in 2500 live birth with carrier frequency 1 in 25 individuals (1). Notwithstanding, CF is still thought to be very rare in Indian population (2). However, published reports indicate that CF is probably far more common in people of Indian origin than previously thought but is underdiagnosed or missed in majority of cases. Dr. Bhakoo reported first case of CF in 1968 from PGIMER, Chandigarh (3). Later, we estimated the relative frequencies of various genetic disorders, which were found, with 7.56% frequency (4). The mutations in the CF transmembrane conductance regulator gene (CFTR or ABC C7; MIM # 602421), which are responsible for both classic and non-classic presentation of the disease (CAVD, MIM # 22180) (5, 6). The CFTR gene was identified and cloned about two decades ago (7, 8). The CFTR gene is located on the long arm of chromosome 7 (region q31 - q32) encompassing 250 kb and comprising 27 exons (7-9). It encodes a trans-membrane protein of 1480- amino acids that function as a CAMP- regulated chloride channel in exocrine epithelia (10). Hydropathy plot analysis disclosed that CFTR protein composed of two motifs, each containing a membrane spaning domain (MSD) that is composed of six tansmembrane helices and nucleotide binding domain (NBD) that contains sequence predicted to interact with ATP (II). MSD-NBD motifs are linked by a unique domain termed as regulatory domain (R) that contains multiple phosphorylation sites and many charged amino acids. The carbonyl terminal, consisting of threonine, arginine, and leucine (TRL) which is anchored through a PDZ type – binding interaction with the cytoskeleton (11, 12) (Fig. 1).

Close

Fig. 1:

Hypothesized structure of CFTR showing proposed structure of CFTR protein.

Export to PPT

More than 2000 sequence variants have been identified in the CFTR gene (https://www.genet.sickkids.on.cafchrome.html) and many of them have been implicated in a variety of CFTR related pathologic conditions such as respiratory distress, pancreatic insufficiency, meconium ileus and congenital absence of the vas deferens (CAVD) (13). Delta 508F is the most common CFTR mutation worldwide, which is present upto 92% of patients with CF. However, this frequency varies among countries and ethnic groups (https://www.cftrscience.com). All these mutations are classified into six classes based on the mechanism of the disruption of CFTR (14) and the absence of the CFTR protein at the apical plasma membrane which include defective protein synthesis, impaired protein maturation leading to protein degradation, defective regulation of CFTR channel activity, altered ionic selectivity and conductance, lowered CFTR mRNA amount and decreased protein stability.

CAVD is characterized as an isolated urogenital form of CF which is associated with normal spermatogenesis, but absence of sperm in the ejaculate due to lack of vas deferens (15). More than 95% of males with CF are infertile due to obstructive azoospermia caused by absent, atrophic or fibrotic Wolffian duct structure (16). The p. Phe 508 del, p. Arg 117H is and T5 allele were identified as most common CFTR mutations in caucassians associated with CAVD phenotype (16, 17). Despite advances in understanding the pathophysiology of CF, there are still many inexplicable differences in its clinical association with CAVD. The role of intensifying the severity of classical CF is well established (17, 18). Transforming growth factor (TGF β), codon 10 polymorphism (rs180470) and codon 25 polymorphism (rs1800471) as well as endothelial receptor type- A (EDNRA) gene polymorphism have been previously established in enhancing the severity of classical CF (17, 19). Molecularly, established mutations including preferentially novel and rare mutations is urgently needed for functional characterization at cellular level, so that, therapeutic molecules could be developed to target underlying molecular defect. Additionally, the cellular and functional data on theses mutations can improve CF genetic counseling. Recent advances of targeted molecular therapies and high throughput screening have resulted in multiple drug therapies that target many important mutations in the CFTR protein (20).

In this manuscript, we provide the work done on CF at Postgraduate Institute of Medical Education and Research, Chandigarh with special reference to its diagnostic and molecular characterization of spectrum of mutations from both classical and non-classical forms of CF. Haplotype association have been used to trace the origin and age of different CF mutations worldwide (21). There is no information available in this regard from Indian subcontinent. Therefore, haplotype study was also carried out using associated intragenic and extragenic marker haplotypes. The study was also performed to investigate whether genetic modifiers of CF lung disease also predispose to CAVD in association with CFTR mutations. We also provide the latest results and current progress of CFTR modulators for the treatment of cystic fibrosis, focusing on potentiators of CFTR channel gating and p.Phe508del CFTR mutation. Special emphasis is placed on the molecular basis of understanding these new therapies.

Molecular Diagnosis of Cystic Fibrosis and its Underlying Pathogenesis

The diagnosis of CF has vast implications for patients and their families. The broad spectrum of clinical disease and reports of over 2,000 different mutations have made the CF diagnosis difficult (22). The guidelines for CF diagnosis were established by the Cystic Fibrosis Foundation for diagnosis of both infants with positive newborn screening (NBS) findings and older patients presenting with an indistinct clinical picture. CF Foundation proposed the following diagnostic algorithm state: the diagnosis of CF should be based on the presence of one or more characteristic clinical features viz. respiratory, gastrointestinal or gastrourinary symptom, a history of CF in a sibling or a positive NBS test, plus laboratory evidence of an abnormality in the CFTR gene or protein (23, 24). Either biological evidence of channel dysfunction such as an abnormal sweat chloride test or nasal potential difference and identification of CF disease causing mutation on each allele of the CFTR gene are acceptable evidence of a CFTR abnormality. Newborn screening depends on initial analysis of fetal blood for high values of immunoreactive trypsinogen (IRT) followed by genetic testing or repeat (25).

The sweat chloride test remains the initial test of choice and gold standard for CF diagnosis despite its limitations. Sweat chloride test was developed by Gibson-Cooke in 1959 (25) . The test is performed via pilocarpine inotophoresis, which is used to stimulate sweat gland secretion. The sweat is collected and analyzed for chloride concentration. The underlying mechanism of elevated levels of sweat sodium was demonstrated by Roeve et al (26) (Fig. 2).

Close

Fig. 2:

Mechanism understanding elevated sodium and chloride levels in the sweat of CF patients. Left panel shows cartoon of a sweat gland with sweat duct projecting out of the sweat gland through dermis into the epidermis. Middle panel shows role of CFTR and ENaC in maintaining normal salt concentration in sweat. Last panel shows dysfunctional CFTR leading to elevated salt concentration in sweat.

Export to PPT

Sweat chloride concentration above 60 mmol per liter (>70 mmol/L in adolescent and adults) is suggestive of cystic fibrosis. In normal individuals and carriers of the CF gene, the mean sweat chloride concentrations are 30 mmol/L.

Cut-off value for infants is 30 mmol/L (27) . A positive sweat test is diagnostic as long as it is performed on adequate amount of sweat (>100 mg) in CF. A second test for CFTR function such as nasal potential difference measurement or analysis of rectal mucosal biopsy is recommended if sweat test is equivocal (28) .

DNA analysis in establishing CF diagnosis is most useful for those individuals with sweat chloride values in the intermediate range. Two or more disease causing CFTR mutations should be located on different alleles as CF is an autosomal recessive disease. CF Foundation recommends testing for the 23 CF mutation panel developed by the American College of Medical Genetics (ACMG). These mutations have been demonstrated to cause sufficient loss of CFTR function to confer CF disease and are therefore noted as conclusive genetic evidence for diagnosis of CF (23). Indirect non-invasive parameters available for assessing the pancreatic function test are serum immunoreactive trypsin (IRT), fecal chymotrypsin, stool immunoreactive human lipase and stool fat excretion as a gold standard (29) .

Spectrum of Mutations in Classical Cystic Fibrosis

A prospective cohort study was designed to establish a spectrum of mutations in CFTR gene. In this study, 1005 suspected CF subtypes were subjected for sweat chloride measurement. Notably, 45 subjects were diagnosed as CF on the basis of classical clinical phenotype and elevated sweat chloride. On the basis of clinical symptoms and autopsy reports that confirmed the diagnosis of CF in three patients with normal sweat chloride were also included. Besides, 2 infants with raised IRT were included as CF subjects.

Demographics and family history of CF subjects revealed that average age at diagnosis in CF subjects was 62.8 months compared to 9.3 months at age of presentation. Male to female ratio was 35.15. Notwithstanding, females had a higher median age at diagnosis (33 months) as compared to males (18 months). History of previous sib death due to respiratory or gastrointestinal problems was observed in 9 families. Consanguinity was noticed in three and history of similar gastrointestinal or respiratory complaints was recorded in another 4 families. In 1 case, the patient's mother had a history of chronic pancreatitis and diabetes (30).

Clinical Findings

The usual presentations include failure to thrive (94%), malabsoption (82%), chronic cough (90%), recurrent or persistent pneumonia (79%) and meconium ileus (10%). Mean Schwachman-Kidczyski score was 57.34±12.07. The median age of CF patients with P.aeruginosa colonization was 39 months compared to 8 months with S.aureus. Pancreatic insufficiency was present in 37 patients (77%). Mean IRT levels in CF patients were 176.43±141.23 ug/L. Determination of fecal fat revealed steatorrhoea (7.36±4.56 g/24h) in 8 patients out of 15 tested (53.33%). Notably, all patients with liver disease (n = 6) and meconium ileus (n = 5) had pancreatic insufficiency.

Sweat Chloride

Median sweat chloride value in CF subjects was 86.25 mEq/L. All individual patient values were mean of two/three repeat sweat chloride estimation on different days. Amount of sweat chloride collected was more than 100 mg on each occasion (198±30 mg). In 2 infants, adequate sweat could not be collected. Fig. 3 shows individual sweat chloride values of CF (n = 48) and non-CF (n = 956) subjects. Median sweat chloride value of 86.25 mEq/L in CF was significantly elevated as compared to 25.5 meq/L in non-CF individuals. In India, there are only few centers where sweat testing is performed.

Close

Fig. 3:

Comparison of sweat chloride values between CF and non-CF cases.

Export to PPT

Identification and Molecular Characterization of Mutations in the CFTR Gene from Classical CF

The spectrum of mutations in CFTR gene from classical CF at PGI is listed in Table 1. In this study, 100 unrelated CF chromosomes were first analyzed for most common delta F508 mutation, which was found with 27% of CF chromosomes R117H, R553X, N1303K and G551D were identified by ARMS on a total of five chromosomes.

By restriction digestion, 3849+10 KbC- T was found in 3 CF subjects with normal sweat chloride. Using SSCP and subsequently DNA sequencing, mutations were molecularly characterized in 31 of the remaining 65 CF chromosomes which include nine missense mutations, viz. L69H, S158N, Q493L, Y517C, V520F, I530L, S549N, E1329Q and Y1381H, one insertion mutation (1792 insA), three splice site mutations (876-6del4, 1525-1G-A, 3120+1G-A), two deletion mutation (1161 delC, 3986 delC), and one nonsense mutation (L218X). S549N and 1525-1G-A were the second most common mutations observed in our population (5% each) Among them, 8 novel mutations, viz. L69H, Q493L, S158N, I530L, E1329Q, 1792 InsA and 3986-3987 delC were identified in one patient each. Notably, none of novel mutation was found in 50 control subjects.

It is noteworthy here that output productions score for 3 novel mutation (S158N, I530L and E1329Q) was found 0.05, representing deleterious effects using molecular modeling and Bioinformatics CMMB. The most common CF mutation worldwide is the deletion of a phenylalanine residue at position 508. A review of all genotyped South Asian patients showed that delta 508 is identified in 19% to 44% CF alleles (31-33), which is lower than the reported frequency of 66% worldwide CF population. In our study, delta F 508 represents only 27% of the analyzed and 41% of the identified CF allele (30). The striking characteristic of the mutation spectrum is absence of some mutations common in the Mediterranean and European population (34). S549N and 1525-1G-A were the second most common mutations, followed by 3849+10 kbC- T. This corroborates the study of Shastri et al (35) which has recommended testing Indian CF patient for deltaF508,1161 delC, 3849+10 kbC- T and S549N. Absolute linkage between delta F508 and KM.19-GATT -TUB9-M470V-T854T haplotype (2-2-1-1-1) predicts a relatively recent appearance of delta F508 in Indian CF patients (28). Low frequency of delta F508 mutation and detection of eight novel and thirteen rare mutation imitate a heterogeneous spectrum of mutation in Indian CF patients. Notwithstanding, CF mutation frequencies and haplotypes vary widely among different populations and haplotype heterogeneity has been found to be greater in population with lower delta F508 frequencies (36). We documented NBD1 and NBD2 as the hotspots identified in CFTR protein in CF patients at our institute (Table 1).

Identification and Molecular Characterization of Mutations in the CFTR Gene from Non -Classical CF (CAVD)

Association of Cystic Fibrosis Genetic Modifiers with Congenital Bilateral Absence of the Vas Deferens

To investigate whether genetic modifiers of CF lung disease also predispose to CAVD in association with CFTR mutations, we tested the hypothesis that polymorphisms of TGF-β1 and EDNRA polymorphism might play a role in penetrance of CAVD. These polymorphisms have been reported to modify CF lung disease, and might also contribute as genetic modifiers of CAVD. Sixty CAVD subjects and fifty controls were investigated for candidate polymorphism TGF-β1 or EDNRA associated with a more serve lung phenotype among CF disease. Importantly, CC genotype (SNP r5335) of EDNRA and TT genotype (SNP r51982073) TGF- β1 were associated with CAVD to occur more frequently among CF individuals (37). TGF- β1 is the best described modifier of CF associated with pulmonary phenotype. In reference to our study, the human vas deferens, epididymis and seminal vesicles develop from the Wolffian duct. Notably, TGF- β1 and related signaling pathways play an important role in normal Wolffian duct development and differentiation (44, 45). On the other hand, other modifier of CF lung disease (EDNRA) is also associated with CAVD.

EDNRA has been implicated in normal formation of mammalian nervous system, the anorectum, and craniofacial structures such as mandible (46-48). EDNRA plays an important role during development of the vas deferens which could be due to loss of the vas in the setting of CFTR insufficiency. This is the first study in reference to the Indian population where we identify possible involvement of these pathways in an atypical CF- related condition, namely CAVD. Recently, it has been documented TGF- βi receptor and P 38 MAPK signaling reduce CFTR activity and probably enhance the effect associated with heterogeneous CFTR mutations and may result in male infertility (49).

Functional Characterization and Pharmacological Correction of Novel and Rare Mutations

The striking characteristic of the identified mutations at our center were the presence of severe rare mutations but the absence of the most common mutations identified in the Mediterranean and European population, which clearly represented the heterogeneous spectrum of CFTR mutations at our Institute. These rare mutations are called orphan mutation because of their very low incidence. Moreover, rare familiar mutations cannot provide sufficient information on the phenotype due to these mutations, viz. L69H, F87I, S118P, G126S, H139Q, F157C, F494L, E543A, S549N, Y852F and D1270E from both classical CF patients and CAVD patients (40). It is noteworthy here that the cellular and function data on these mutations can improve CF genetic counseling. For functional characterization and therapeutic implications for these CF rare mutations, baby hamster kidney cells are an adherent cell line used for this study (50). Functional findings revealed that L69H was found as a novel class II CF mutation. The trafficking to the plasma membrane of L69H- CFTR is abnormal as corroborated by western blot analysis. Further confocal microscopic imaging showed the abundance of L69H mutated CFTR protein in the endoplasmic reticulum and absence on the plasma membrane. In view of above finding, the processing pathway of this variant is similar to that of F508 del-CFTR mutation which is the most common mutation worldwide. Misfolded F508 del- CFTR is retained by the ER and undergoes ubiquitination, thereby accelerating its proteome degradation and reducing F508 del-CFTR trafficking in the plasma membrane (51,52). The model of 3D structure of CFTR which is predicted using the Sav1866 experimental 3D structure. L69H is studied in a short cytosolic a- helix in the N-terminus of CFTR preceding the first transmembrane helix TMI of MSD that is called "elbow helix", which is conserved feature among the ABC exporter family. Possibly, L69H mutation may perturb the network formed hydrophobic side chains of the cytosolic extension of MSD1TMI helices. MSDI-NBDI linker (1368) might thus play an important role for MSD1 folding and destabilization of this domain at the membrane. Another important S549N located in the LSGGQ signature motif of NBD1 and may alter the hydrolysis of ATP to regulate channel activity (53) as a result S549N classified as class III CF mutation (54). Normal maturation of S459 N-CFTR protein was observed in our study as reported by others (54). Further, the study was conducted to evaluate the effect of some potentiator viz Miglustat, Isolab and VX-809 to increase the activity of defective CFTR protein due to presence of these mutations. Ivacaftor (VX-770) is an investigational, orally bioavailable agent which was shown to augment the chloride transport activity of G551D-CFTR protein in vitro (54). In our study, the pharmacological corrector VX809 was found to activate L69H in BHK-21 cells in similar to that of delta F508 (50). The activation of L69H and delta F508 were both significantly corrected in terms of maturation and translocation of CFTR protein to cell membrane when cells were treated with VX809 potentiator. Whereas, S549N mutation classify as type III mutation was also corrected as its translocation to plasma membrane (50). It is noteworthy here that VX809 molecule is FDA approved, so that this potentiator can be used as a therapeutic molecule to rescue from type II and III mutations associated with Indian CF patients.

Current Progress in Treatment of Basic Defects in CF

CFTR modulator therapies have been directed towards specific disease causing mutations and the molecules pathways that underlie these causes. CFTR mutations grouped into conservative classes have led to the development of specific approaches towards treating the molecular effect in CF (Table 3) (55-58). Non-sense mutations in CFTR have been shown to be rescued with application of a compound ataluren, derived from aminoglycoside antibiotic that can induce read- through of premature termination codons. This investigational drug is in phase III clinical trial (55). Two new molecules have been developed for correction, VX809 and potentiator, VX770, of the delta F508 mutations. Some of class III mutants respond to VX770 by increasing the chlorides transport. CFTR potentiator Ivacaflor (Kalydeco, VX-770, vertex pharmaceuticals, Boston, MA, USA) FDA and European regulatory authorities approved drug to treat CF patient with a class III G551D mutation (56-58). Recent advances of targeted molecular therapies and high throughput screening have resulted in multiple drug therapies that target many important mutations in the CFTR protein (59).

In summary, these findings represent an important milestone in the development of treatments designed to improve CFTR protein function as a means of addressing the underlying cause of cystic fibrosis.

Future Directions

In Indian scenario, the diagnosis of CF is important since sweat chloride measurement using pilocarpine iontophoresis is limited to only few centers despite a large population in this country. Sweat chloride measurement technique facility is must at every district level to pick up the CF population. The most reliable is the sweat induction by pilocarpine iontophoresis, followed by sweat chloride collection on a gauze or filter paper. Sweat chloride concentration in the CF patient is in intermediate range. In these cases, gene sequencing is essential to provide positive CF diagnosis as well as, mutations specific therapies which have become available recently (59, 60).

Other mechanisms of corrector therapy are being developed. Recently, the direct and indirect modulation of the nitric oxide (NO) pathway has been investigated as a possible corrector mechanism in delta F508-CFTR (58, 59). If future drug combinations are sufficiently robust to correct CFTR function in individuals with only one copy Phe508 del, this modulator will ideally provide clinical benefit to over 90% of all patients with CF. Thus, other approaches such as drugs that read through premature stop mutation and gene replacement or editing must continue (61).