Communicated by Claude Férec
Mutation In Brief
Disrupted posttranscriptional regulation of the cystic fibrosis transmembrane conductance regulator (CFTR) by a 5′UTR mutation is associated with a cftr-related disease†
Article first published online: 19 SEP 2011
© 2011 Wiley-Liss, Inc.
Volume 32, Issue 10, pages E2266–E2282, October 2011
How to Cite
Lukowski, S. W., Bombieri, C. and Trezise, A. E. O. (2011), Disrupted posttranscriptional regulation of the cystic fibrosis transmembrane conductance regulator (CFTR) by a 5′UTR mutation is associated with a cftr-related disease. Hum. Mutat., 32: E2266–E2282. doi: 10.1002/humu.21545
- Issue published online: 19 SEP 2011
- Article first published online: 19 SEP 2011
- Accepted manuscript online: 11 AUG 2011 08:50AM EST
- Manuscript Accepted: 12 MAY 2011
- Manuscript Received: 2 NOV 2010
- University of Queensland
- cystic fibrosis;
- 5′ untranslated region;
- upstream open reading frame, CFTR-related disease, disseminated bronchiectasis
Cystic fibrosis (CF) is characterized as a single-gene disorder with a simple, autosomal recessive mode of inheritance. However, translation of cystic fibrosis transmembrane conductance regulator (CFTR) genotype into CF phenotype is influenced by nucleotide sequence variations at multiple genetic loci, and individuals heterozygous for CFTR mutations are predisposed to a range of CFTR-related conditions, such as disseminated bronchiectasis. CF disease severity and CFTR-related conditions are more akin to complex, multifactorial traits, which are increasingly being associated with mutations that perturb gene expression. We have identified a patient with disseminated bronchiectasis, who is heterozygous for a single-nucleotide substitution in the CFTR 5′ untranslated region (UTR) (c.-34C>T). The c.-34C>T mutation creates an upstream AUG codon and upstream open reading frame that overlaps, and is out of frame with, the CFTR protein coding sequence. Using luciferase reporter constructs, we have shown that the c.-34C>T mutation decreases gene expression by 85–99%, by reducing translation efficiency and mRNA stability. This is the first CFTR regulatory mutation shown to act at a posttranscriptional level that reduces the synthesis of normal CFTR (Class V), and reaffirms the importance of regulatory mutations as a genetic basis of multifactorial phenotypes. ©2011 Wiley-Liss, Inc.
Cystic fibrosis (CF) is a multisystem, lethal, autosomal recessive disease that is caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR; MIM no. 602421) gene. A total of 80–90% of CF patients are either homozygous, or compound heterozygous, for the p.Phe508del in-frame deletion mutation [Kerem et al., 1989; Riordan et al., 1989; Rommens et al., 1989]. The CF mutation database (www.genet.sickkids.on.ca) currently lists over 1,400 CF disease-causing mutations, and a further 281 DNA sequence variations or polymorphisms in CFTR. While CF affects the gastrointestinal tract, sweat glands, and reproductive systems, it is the airway system disease of CF patients that is responsible for shortened median survival of 37.4 years (www.cff.org, 2008).
The translation of CFTR genotype to phenotype is not straightforward. Patients homozygous for p.Phe508del exhibit the full spectrum of CF lung disease [Cutting, 2010]. Also, individuals that are heterozygous for a CF disease-causing mutation are predisposed to a range of complex diseases in which multiple genetic and environmental factors play a role. These complex CFTR-related diseases (CFTR-RDs) include: (1) congenital bilateral absence of vas deferens (CBAVD) [Casals et al., 2000; Jarvi et al., 1998], (2) disseminated bronchiectasis (DB) [Bombieri et al., 1998; Casals et al., 2004; Girodon et al., 1997; King et al., 2004], (3) pancreatitis [Chen and Ferec, 2009; Cohn et al., 2005; Noone et al., 2001], and (4) allergic bronchopulmonary aspergillosis (ABPA) [Miller et al., 1996; Tzetis et al., 2001] of unknown etiology.
Understanding the genetic basis of complex diseases, such as the CFTR-RDs, and quantitative traits is a major challenge in current biology [Mackay et al., 2009]. What is becoming apparent is that the spectrum of mutations that underlie complex traits is (1) shifted toward missense mutations that subtly alter protein function [Buratti et al., 2007; Pagani et al., 2003], and (2) predominated by mutations in regulatory sequence elements such as 5′ and 3′ untranslated regions (UTRs), promoters, intronic elements, and noncoding RNAs [Mattick, 2009; Segat et al., 2010]. This work focuses on how the perturbation of CFTR expression may contribute to the development of complex CFTR-RDs.
The molecular regulatory mechanisms controlling the expression of CFTR, and their role in CF and CFTR-RD pathogenesis, are not fully understood. Most research aimed at determining the regulatory mechanisms underlying CFTR expression has focused on CFTR transcriptional control and posttranslational modifications of the CFTR protein. To date, the posttranscriptional regulation of CFTR mRNA stability and translation has received little attention. Through detailed evaluation of the vertebrate CFTR 5′UTR sequences, we have identified several putative regulatory elements known to modulate the mRNA translation efficiency and stability of other genes. These posttranscriptional regulatory elements include (1) upstream open reading frames (uORFs), (2) upstream AUG codons (uAUGs), (3) consensus sites for RNA-binding proteins, and (4) RNA secondary structures [Davies et al., 2004; Trezise, 2006].
uORFs are sequences within mRNA 5′UTRs, defined by an uAUG and an in-frame stop codon, which may either be restricted to the 5′UTR, or overlap the main protein-coding region (pORF) as an out-of-frame peptide sequence. Bioinformatic analysis has found that 45% of human genes contain uORFs or uAUGs that could posttranscriptionally modulate gene expression [Iacono et al., 2005]. Reporter gene constructs have been used to demonstrate that the uORFs present in about 75 different mammalian mRNA transcripts are functional posttranscriptional regulatory elements that, in general, repress translation of the main protein-coding region and may also destabilize the mRNA. Also, mutations that create or destroy uORFs can substantially alter the expression of downstream protein-coding regions. It is somewhat surprising that, of over 111,000 known disease-causing mutations listed in the Human Gene Mutation Database (www.hgmd.cf.ac.uk), to date, there are just six known uORF-creating mutations that have been functionally validated and linked to the development of human diseases [Calvo et al., 2009; Liu et al., 1999].
Human CFTR uses spatially and temporally regulated transcription start site (TSS) choice in cells and tissues associated with CF and CFTR-RDs [Davies et al., 2004; White et al., 1998]. Two CFTR transcription initiation sites predominate in airway and intestinal epithelial cells. The first, CFTR-132, is used in the human fetal lung, and the second, CFTR-69, is used in both adult lung and intestinal epithelia (Fig. 1A and C). Together, these transcripts comprise over 90% of CFTR mRNA expressed in epithelial cells—the sites of the majority of CF disease [White et al., 1998]. TSSs are numbered relative to the A of the CFTR protein AUG (pAUG) nucleotides, which is designated +1. This numbering reflects cDNA sequence and is consistent with current human gene nomenclature [den Dunnen and Antonarakis, 2000]. We identified a single-uORF encoding an 18-amino acid peptide and in-frame stop codon in the CFTR-132 5′UTR. The CFTR-69 5′UTR begins at the termination codon of the CFTR-132 uORF and both transcripts encode a stable RNA stem loop (Fig. 1C). The inclusion of the uORF within the mature CFTR mRNA is dependent upon the selection of the -132 TSS, which is most likely associated with a developmental regulatory switch previously described by White et al. .
We have identified a patient with a CFTR-RD, DB, who is heterozygous for a single-nucleotide sequence substitution (c.-34C>T) in the CFTR 5′UTR. This mutation creates a 36 codon uORF that overlaps, and is out-of-frame with, the CFTR protein AUG codon (pAUG). Using luciferase reporter constructs, we have established that the CFTR 5′UTR c.-34C>T mutation inhibits translation initiation from the CFTR pAUG codon and reduces mRNA stability, reducing overall gene expression by 85–99%. To our knowledge, this is both the first report to experimentally demonstrate a functional effect of a CFTR 5′UTR mutation, and the first CFTR regulatory mutation shown to act at a posttranscriptional level that reduces the synthesis of normal CFTR (Class V).
Materials and Methods
Population Study and Genotyping
Denaturing gradient gel electrophoresis and automated sequencing were used to identify the CFTR 5′UTR c.-34C>T mutation in a single patient with no additional CFTR mutations. Clinically, the patient was diagnosed with DB and analysis of the patients sweat chloride levels yielded negative results. Nucleotide numbering reflects cDNA numbering with +1 corresponding to the A of the ATG translation initiation codon in the reference sequence, according to journal guidelines (www.hgvs.org/mutnomen). The initiation codon is codon 1.
The human CFTR mRNA sequence (NM_000492.3) and the CFTR c.-34C>T variant mRNA sequence were analyzed using the ORF Finder software package available at the NCBI website to confirm the mechanism of the predicted CFTR c.-34C>T ORF (http://www.ncbi.nlm.nih.gov/projects/gorf/). These sequences were also analyzed using the TransFac public database (www.gene-regulation.com) to detect any transcription factor binding sites either destroyed by, or arising from, the c.-34C>T mutation. The region of the mutation in the CFTR-132 and CFTR-69 mRNA sequences was further analyzed using the Geneious software package, version 4.0 (http://www.geneious.com) [Drummond et al., 2008].
Prediction of RNA Secondary Structures
The prediction of RNA secondary structures within the human CFTR 5′UTR mRNA was performed using the mFold program, version 3.2 [Mathews et al., 1999; Zuker, 2003]. The mFold server is located at http://frontend.bioinfo.rpi.edu/applications/mfold/.
Generation of Reporter Constructs and Site-Directed Mutagenesis
Primers used for cloning and site-directed mutagenesis are listed in Table 1. Transcript-specific 5′-tailed primers were used to amplify the CFTR-132 and CFTR-69 5′UTRs, up to and including the second codon of the CFTR protein coding sequence, from HEK293 genomic DNA. Both the forward and reverse primers included 5′-tailed ends that were complementary to the multiple cloning site and the luciferase CDS (luc2) of the pGL4.13 (Promega) luciferase reporter vector, respectively. A series of fusion constructs were created, linking the CFTR 5′UTR and the first two CFTR codons directly to the third codon of the luc2 gene (pGL4.13-CFTR-132 and pGL4.13-CFTR-69). This was done to assess translation initiation in the context of the native CFTR pAUG. These CFTR 5′UTR-luc2 fusion sequences were cloned into the multiple cloning region of pGL4.13 between HindIII and XbaI restriction sites. Control constructs were created by fusing the 5′UTR and first two codons of the human beta-globin (HBB) gene with the third codon of luc2 coding region of the pGL4.13 vector.
|#||Primer sequence (5′-3′)||Length (nt)||ID|
Within this series, a range of site-directed mutagenesis constructs were generated using the Phusion Site-Directed Mutagenesis kit (Finnzymes). Back-to-back primers were designed such that the common forward primer incorporated the single-nucleotide mutation (c.-34C>T) and a 5′UTR-specific (either CFTR-132 or CFTR-69) reverse primer facilitated the linear amplification of the experimental construct. The resulting PCR products were cloned in DH5α Sub-Cloning Efficiency Competent Cells (Invitrogen) and ampicillin-resistant colonies were screened for the correct mutation by forward and reverse sequencing of the plasmid DNA. The negative control was created through conversion of the pGL4.13-HBB 5′UTR start codon to a stop codon (AUG>UAG). By design, the pGL4.13-HBB AUG>UAG control construct does not allow translation initiation from the luc2-AUG and is therefore unable to produce the luciferase protein. To test the ability of the c.-34C>T uAUG to support translation, the uAUG and the second codon of the c.-34C>T uORF was linked directly to the third codon of the luc2 region. Site-directed mutagenesis was used to delete the intervening sequence, including the CFTR-AUG and second codon.
Transient Transfection and Luciferase Reporter Assays
HT29 and HEK293 cells were cultured in RPMI-1640 and Dulbecco's Modified Eagle Medium (Gibco), respectively, and supplemented with 10% newborn calf serum and 0.5% penicillin/streptomycin (Gibco). One day before the transfection, each cell line was plated at ∼0.7 × 105 cells into opaque 96-well plates. Five hours prior to the transfection, the media in both cell lines was replaced with antibiotic-free OptiMEM medium (Gibco). The plasmids (0.2 µg) were co-transfected with 0.2 µg of Renilla luciferase vector (pHRL-TK; Promega) using the Fugene HD transfection reagent (Roche) as per the manufacturers protocol. After 6 hr of incubation at 37°C, the OptiMEM transfection media was replaced with RPMI-1640 and DMEM maintenance media (HT29 and HEK293, respectively). Firefly and Renilla luciferase activities were measured in each cell line 48 hr posttransfection using the Dual-Glo Luciferase Assay (Promega) in a BMG Fluostar fluorescence/luminescence detector. Each transfection was performed with three technical replicates per assay and each assay was repeated at least three times for each 5′UTR-luc2 construct.
Molecular Beacons (MBs)
Luc2-(CGCGA-ACTTCCCATTTGCCACCCGG-TCGCG) and GAPDH-specific (CCGGC-TGGCACCGTCAAGGCTGAGAAC- GCCGG) MBs probes were based upon a 5′ 6-FAM (fluorophore) and 3′ BHQ-1 (Black Hole Quencher) modifications (Sigma). Probes were designed such that the probe:template hybrid annealing temperatures were approximately 5–8°C above the desired quantitative real-time PCR (qPCR) annealing temperature, allowing both to be analyzed simultaneously. MB stem regions were designed to self-anneal with high specificity, while remaining nonspecific against target mRNAs. In addition, the MB stem regions dissociate approximately 5–8°C above the annealing temperature of the qPCR reaction.
Quantitative Real-Time PCR
HT29 cells were grown to subconfluency in 12-well tissue culture plates (BD Falcon) prior to transient transfection of selected constructs, as previously described. Inhibition of transcription was achieved by supplementing transfected cells with Actinomycin D (5 µg/ml; Sigma) 48 hr posttransfection. Total RNA was extracted at 0, 3, 6, and 9 hr using the InnuPrep RNA Mini kit (Analytik Jena). Purity and quantity of extracted RNA was measured using a Nanodrop-1000 spectrophotometer (Thermo Fisher Scientific). Total RNA was subjected to Turbo DNA-free DNaseI treatment (Ambion) for 1 hr at 37°C and EDTA inactivated for 10 min at 75°C. Ten microliter of total RNA was added to reverse transcription reactions (±RT), which were performed with the High-Capacity Reverse Transcription kit (Applied Biosystems), as per the manufacturers instructions. One microliter of cDNA was added to each qPCR reaction in a total volume of 25 µl (2X Brilliant Blue II PCR Mastermix [Stratagene]; 400 nM MB probe; 1 µM primers; dH2O). qPCR reactions were performed on a Mx3000P real-time PCR instrument (Stratagene) and the cycling conditions were optimized for both probe/primer sets (95°C for 10 min, followed by 40 cycles at 95°C for 1 min, 52°C for 30 sec, and 72°C for 1 min). Luc2-specific (F: 5′-ATGTACCGCTTCGAGGAGG-3′; R: 5′-CGATCTCGTGCAAGTTGC-3′) and GAPDH-specific (F: 5′-TGTTGCCATCAATGACCCCTT-3′; R: 5′-CTCCACGACGTA-CTCAGCG-3′) primers were used in conjunction with transcript-specific MBs probes to measure relative gene expression.
Luciferase activity of the CFTR-132 wild-type construct was designated as the wild-type expression level and was arbitrarily set at 100%. Activity readings from all other constructs were expressed as a percentage of the CFTR-132 wild-type expression level. Luciferase (luc2) activity was normalized against transfection efficiency, which was assessed by co-transfection of a Renilla luciferase construct (pHRL-TK; Promega). The differences in relative luciferase activity of the experimental and control constructs were compared using Student's unpaired t-test analysis in GraphPad Prism, version 5.00 for Mac OS X (GraphPad Software, San Diego, CA, www.graphpad.com). Results are expressed as the mean ± SE.
Raw qPCR data were prepared for analysis using the MxPro v4.0 software (Stratagene) and subsequently analyzed in both Excel 2008 (Microsoft) and GraphPad Prism. Reaction efficiencies for luc2 and GAPDH were 91.3% and 91.4%, respectively. Each data set (±ActD) was calibrated against the relevant time zero data point and mRNA loading was normalized to GAPDH levels for each time-point per construct. All qPCR experiments were biologically replicated at least two times (n ≥ 2) and each was technically duplicated. mRNA decay kinetics was ascertained using linear regression and their significance was determined using paired t-tests. Relative steady-state mRNA quantities were compared using paired t-tests following normalization to the CFTR-132 wild-type construct. For all analyses, results were considered statistically significant when p ≤ 0.05, and are expressed as the mean ± SE.
The sequence variation, a C to T substitution at position -34 (c.-34C>T; Fig. 1B and D), was identified within the CFTR 5′UTR of a 15-year-old patient with DB. The function of the 5′UTR c.-34C>T polymorphism was unknown at the time of identification. This mutation was found once out of 50 chromosomes of children with DB and was never observed in 104 chromosomes of chronic obstructive pulmonary disease (COPD) patients, in 46 chromosomes of adult DB patients, in 106 chromosomes of CF patients, and in 600 control chromosomes.
By initial visual inspection of the sequence context of the CFTR c.-34C>T mutation, we recognized that the c.-34C>T mutation creates a novel uAUG (ACG>AUG) and initiates a uORF that both overlaps the normal CFTR pAUG codon, and is out-of-frame with the CFTR protein coding sequence. Bioinformatic analysis, using publicly available ORF Finder software (www.ncbi.nlm.nih.gov/projects/gorf/), of the c.-34C>T CFTR mRNA sequence confirmed these initial findings. The c.-34C>T uORF extends through the first exon of CFTR and terminates at an in-frame stop codon within exon 2, potentially preventing translation initiation at the CFTR pAUG. Ribosomes initiating translation at the c.-34C>T uORF will translate an out-of-frame, 35 amino acid peptide (MALAGTPAPERPCRGRLWKRPALSPNFFS-AGPDQF*; Fig. 1D) and then either detach from the mRNA at the uORFs translation termination codon or continue scanning. Multiple, short, out-of-frame ORFs, located in exons 3–6, would be encountered before any scanning ribosome reached the next available in-frame AUG codon in exon 6. We predicted that a small percentage of ribosomes will fail to recognize the c.-34C>T uAUG codon and may initiate translation at the CFTR pAUG to produce a low level of functional cftr. The amount of functional CFTR chloride channel produced would be critically dependent upon the proportion of scanning ribosomes that initiate translation at the c.-34C>T uAUG. Based on our model, the higher the proportion of ribosomes initiating translation at the c.-34C>T uAUG, the lower the amount of functional CFTR chloride channel that will be produced.
We tested this model by assessing the ability of the wild-type and c.-34C>T mutant CFTR 5′UTRs to support translation initiation at the CFTR pAUG. Using pGL4.13 luciferase reporter constructs, we created a series of CFTR 5′UTR variants, which included the first two codons of the CFTR transcript, linked to the third codon of the firefly luciferase 2 reporter gene (luc2; Fig. 2). Maintaining the endogenous CFTR pAUG Kozak context in these constructs will most faithfully reflect the behavior of ribosomes on the native CFTR mRNA. The luciferase activity of these constructs was measured in the HT29 (human intestinal epithelial origin) and HEK293 (human embryonic kidney origin) cell lines. These cell lines were selected based on their CFTR expression profiles, CFTR TSS use and widespread use in other studies of CFTR expression and function. Human intestinal epithelium and HT29 cells express relatively high levels of endogenous CFTR mRNA (compared to other sites of in vivo CFTR expression), while CFTR expression in HEK293 is negligible [Trezise, 2006]. CFTR is known to use both tissue specific and developmentally regulated TSS selection [Davies et al., 2004; White et al., 1998]. The endogenous CFTR TSSs have been mapped in HT29 cells. The major CFTR TSS in HT29 cells exhibits microheterogeneity [Yoshimura et al., 1991] and co-locates with the -69 CFTR TSS used in vivo in human adult lung and small intestine [White et al., 1998]. HEK293 cells have been widely used as a control cell line in studies of CFTR expression and function since the original report of no detectable CFTR mRNA in HEK293 cells [Chou et al., 1991]. No additional tissue samples were available from the patient in which the c.-34C>T mutation was identified, precluding any in vivo analysis.
CFTR-69 5′UTR Directs Less Efficient Translation Initiation at the CFTR pAUG than the CFTR-132 5′UTR
Using luc2 reporter assays, we measured the relative capacities of the wild-type CFTR-132 and CFTR-69 transcript 5′UTRs to support translation from the native CFTR pAUG. Compared to the wild-type CFTR-132 5′UTR (arbitrarily designated 100% luciferase activity), the CFTR-69 5′UTR reduced translation initiation at the CFTR pAUG by 54.1% (p = 0.00750) and 62.3% (p < 0.0001) in HT29 and HEK293 cells, respectively, (Fig. 3A). To determine whether the observed reduction in luc2 reporter activity was due to a translational mechanism, as predicted, or instead due to altered mRNA stability, qPCR analysis was performed on HT29 cells transfected with either wild-type CFTR-132-luc2 or CFTR-69-luc2 constructs. While the average steady-state levels of CFTR-69-luc2 mRNA were lower (36%) than the average steady state levels of the CFTR-132-luc2 (Fig. 3B), this difference was not statistically significant (p = 0.347). In addition, the mRNA decay kinetics of the two wild-type CFTR 5′UTR-luc2 transcripts were similar (p = 0.0618; Fig. 3C). These results suggest that reduced luc2 reporter activity from the CFTR-69-luc2 transcripts is primarily due to decreased translation initiation from the CFTR-69 pAUG.
The CFTR 5′UTR c.-34C>T Substitution is a Functional Mutation that Inhibits Translation Initiation at the CFTR pAUG and Normal Protein Synthesis
When we compared the c.-34C>T-CFTR-132-luc2 mutant construct against the wild-type CFTR-132-luc2 construct, we observed large reductions in luc2 activity of 84.4% (p < 0.0001) and 79.1% (p < 0.0001) in HT29 and HEK293 cell lines, respectively, (Fig. 4A). qPCR analysis compared both steady-state levels and mRNA half-lives of wild-type CFTR-132-luc2 and c.-34C>T-CFTR-132-luc2 transcripts. Compared with the wild-type CFTR-132 5′UTR, introduction of the c.-34C>T mutation had no significant effect on steady-state mRNA abundance (Fig. 4B) or the rate of mRNA decay (Fig. 4C).
Luciferase activity from the c.-34C>T-CFTR-69-luc2 construct, compared to the wild-type CFTR-69-luc2 construct, was essentially eliminated, being reduced by 98.5% (p < 0.0001) and 98% (p < 0.0001) in HT29 and HEK293 cell lines, respectively, (Fig. 5A). We investigated the impact of the c.-34C>T mutation, in the context of the CFTR-69 5′UTR, on the steady-state levels and the stabilities of the CFTR-69-luc2 and c.-34C>T-CFTR-69-luc2 transcripts. Introduction of the c.-34C>T mutation into the CFTR-69 5′UTR resulted in a 65% decrease in mRNA abundance (p = 0.0346; Fig. 5B), and a 14.3% reduction in mRNA half-life (p = 0.0373; Fig. 5C).
These results show that the c.-34C>T mutation has differential effects on mRNA translation efficiency, and mRNA stability and abundance depending on the CFTR 5′UTR context. In the context of the CFTR-132 5′UTR, the c.-34C>T mutation reduces overall luc2 activity by 80–85%, primarily by reducing the efficiency of translation initiation at the CFTR pAUG codon, without significantly altering either transcript abundance or stability. While, in the context of the CFTR-69 5′UTR, the c.-34C>T mutation essentially eliminates luc2 activity by reducing transcript abundance, stability, and translation initiation at the CFTR pAUG codon.
Efficient Translation Initiation at the Upstream AUG Created by the c.-34C>T Mutation
Our model postulates that the uAUG codon created by the c.-34C>T mutation recruits scanning ribosomes and efficiently initiates translation at this novel uAUG, thus substantially reducing translation initiation at the CFTR pAUG codon. To further test this aspect of our hypothesis, we created a construct based upon the CFTR-69 backbone (c.-34C>T-luc2), in which the second codon of the c.-34C>T uORF was fused in-frame to the third codon of the luc2 reporter gene in the pGL4.13 vector. Compared to translation initiation at the CFTR pAUG in the wild-type CFTR-69 construct, the c.-34C>T uAUG codon was equally efficiently recognized by scanning ribosomes and supported 100% (p = 0.987), and 132% (p = 0.370) translation initiation in HT29 and HEK293 cell lines, respectively, (Fig. 5A).
The difference between the two wild-type CFTR-132 and CFTR-69 5′UTRs in supporting translation from the CFTR pAUG is likely due to the differential presence of multiple posttranscriptional regulatory elements. Elements such as RNA secondary structure, RNA-binding protein sites, regulatory RNA binding sites, uORFs and uAUGs, that are present within the UTRs of many mRNAs, have been shown to be important posttranscriptional regulators of mRNA stability and translation. Such elements have been shown to modulate translational efficiency by recruiting trans-acting factors [Rouault et al., 1988], altering mRNA secondary structures and interactions [Mehta et al., 2006], and regulating ribosome occupancy on translated uORFs [Arava et al., 2005; McCarthy, 1998]. The observed reduction in translation initiation at the CFTR pAUG codon in the CFTR-69-luc2 mRNA is consistent with previous observations by ourselves [White et al., 1998] and others [Sehgal et al., 1996a, 1996b] of decreased CFTR-69 mRNA expression and CFTR chloride channel function in adult lung in vivo, compared to levels of CFTR-132 mRNA present in fetal lung. In this work, we have shown that mRNA transcripts containing the wild-type CFTR-69 5′UTR are less stable, and their protein coding sequence is less efficiently translated, than transcripts containing the CFTR-132 5′UTR, resulting in overall reduced gene expression.
We identified a single-nucleotide polymorphism in the CFTR 5′UTR (CFTR c.-34C>T) in a 15-year-old patient with DB produced by chronic airway infection and inflammation. We found that the CFTR c.-34C>T mutation results in a knockdown/null CFTR allele and not a polymorphic wild-type CFTR allele. Introduction of the c.-34C>T mutation produces an inhibitory 5′UTR that severely reduces translation initiation at the CFTR pAUG and, in the CFTR-69 context, exposes downstream sequences to enhanced mRNA decay mechanisms. We predicted that the mechanism of action of the c.-34C>T mutation is via the creation of a novel uAUG codon that is efficiently recognized by scanning ribosomes which then initiate translation of a uORF that both overlaps, and is out-of-frame with, the CFTR pAUG codon. Efficient translation initiation at the c.-34C>T-created uAUG then effectively blocks translation initiation at the CFTR pAUG codon, resulting in a mutant CFTR allele that eliminates 85–99% of protein expression. This report provides experimental evidence that supports this hypothesized mechanism of action of the c.-34C>T mutation.
We have demonstrated that the c.-34C>T-created uAUG codon is efficiently recognized by scanning ribosomes, which then initiate translation of this novel uORF. Fusion of the second codon of the c.-34C>T-created uORF to the third codon of the luc2 reporter, maintaining the c.-34C>T uAUG Kozak context, produced robust luciferase activity. The c.-34C>T uAUG codon initiated translation at least as efficiently as the CFTR pAUG codon (Fig. 5A: CFTR(-69)c.-34C>T:luc2 compared with CFTR(-69)WT). We confirmed that the observed luc2 activity genuinely resulted from translation initiation at the c.-34C>T uAUG by two strategies. A negative control construct, in which the reporter gene pAUG codon was mutated to a UAG stop codon, produced no luciferase activity and demonstrated the necessity of the full-length luc2 coding region to produce functional luciferase protein. Also, the uAUG codon produced by the c.-34C>T mutation is the only AUG codon present in the CFTR(-69)c.-34C>T:luc2 construct that could result in the translation of the full-length, functional luciferase protein. The robust luc2 reporter activity produced by the CFTR(-69)c.-34C>T:luc2 fusion construct must therefore be due to efficient translation at the c.-34C>T-created uAUG codon.
An alternative interpretation of this data is that the c.-34C>T mutation creates or destroys a transcription factor binding site, and is the underlying cause of altered reporter activity. We believe that this possibility is unlikely for a number reasons. Firstly, using the TransFac public database (www.gene-regulation.com) of transcription factor binding site consensus sequences, we interrogated the wild-type and mutant CFTR 5′UTR sequences for the presence or absence of any unique transcription factor binding sites arising from the c.-34C>T mutation. No differences in predicted transcription factor binding sites were found between the wild-type and c.-34C>T mutant CFTR 5′UTR sequences. Secondly, only the CFTR-69 5′UTR reduced mRNA stability and steady state mRNA levels in the presence of the c.-34C>T mutation, despite the c.-34C>T mutation being encoded within both the CFTR-132 and CFTR-69 5′UTRs. Lastly, the quantitative reductions in luciferase reporter activity (85% by the c.-34C>T mutation in CFTR-132 5′UTR context and 99% in CFTR-69 5′UTR context) were substantially greater than any changes in steady state mRNA levels, which either were not significantly different (CFTR-132 5′UTR) or reduced to a lesser extent and with accompanying reductions in mRNA stability (CFTR-69 5′UTR). For these reasons, and because the uAUG codon is efficiently recognized by scanning ribosomes, we believe the most likely conclusion is that the CFTR c.-34C>T mutation is acting at a posttranscriptional level.
This work demonstrates that the presence of a single-nucleotide substitution (c.-34C>T), of previously unknown function, in the CFTR 5′UTR is sufficient to initiate a perturbed posttranscriptional regulatory mechanism that significantly inhibits translation of the downstream protein coding region and decreases the stability of the mRNA. As the CFTR c.-34C>T mutation introduces an efficiently recognized uAUG codon into the CFTR 5′UTR, it is perhaps surprising that any translation initiation occurs at the CFTR pAUG codon. We have shown that translation at the CFTR pAUG codon occurs at 15% of wild-type levels in the context of the CFTR-132 5′UTR. A number of possible mechanisms could result in the failure of scanning small rRNA subunits to recognize and initiate translation at the c.-34C>T uAUG in the CFTR-132 5′UTR and continue scanning, termed “leaky” scanning [Wang and Rothnagel, 2004]. Such small rRNA subunits may continue scanning and then recognize the CFTR pAUG and initiate translation of the protein coding sequence. The particular nucleotide sequence surrounding each AUG codon, referred to as the Kozak context of the AUG codon [Kozak, 1986a], influences the proportion of scanning small rRNA subunits that recognize the AUG codon and initiate translation. Recent analysis of genomic data from 47 eukaryotic species has identified strong biases in favor of nucleotides -3A/G, -2A/C, and +5C (where the A of the AUG codon is designated +1) surrounding efficiently recognized AUG codons [Nakagawa et al., 2008]. Several functional studies have shown that the -3A/G nucleotide has the major influence on enhancing translation initiation [Kochetov, 2005; Kozak, 1986b; Lukaszewicz et al., 2000]. Figure 6 shows the nucleotide sequence contexts of the c.-34C>T-created uAUG codon and the CFTR pAUG codon in comparison to the “preferred” sequence context surrounding AUG codons of highly expressed eukaryotic genes [Nakagawa et al., 2008]. Both the c.-34C>T uAUG and the CFTR pAUG match the preferred nucleotide context in two of the three critical nucleotide positions. This is consistent with our functional data showing that both AUG codons support translation initiation with equal efficiency (Fig. 5A). The “2 out of 3” preferred sequence context of the c.-34C>T uAUG codon is also consistent with some “leaky” scanning, leading to a low level of translation initiation from the CFTR pAUG. Our demonstration that the c.-34C>T uAUG substantially reduces expression of the downstream protein coding sequence provides further support for the recent findings of Calvo et al. , that uORF translation can substantially reduce protein expression and may have broader implications as a model for human disease pathogenesis.
We also found that the c.-34C>T mutation had a different effect on transcript stability in the context of the CFTR-69 5′UTR (14% decrease in stability) compared to the CFTR-132 5′UTR (no change in stability). We speculate that the different effects of the c.-34C>T mutation in the two CFTR 5′UTR contexts may be due to interaction with known translation regulatory elements, such as the endogenous uORF at −120 bp, that are only present in the CFTR-132 5′UTR. There may be other, unidentified regulatory elements that are differentially present in the CFTR-132 and CFTR-69 5′UTRs and which may contribute to the differential effects of the c.-34C>T mutation. Further experiments will be required to uncover the mechanism of the differential effects of the c.-34C>T mutation in the two CFTR 5′UTR contexts.
The evidence provided by this study indicates that the CFTR c.-34C>T mutation meets the classification criteria of a Class V CFTR mutation. Class V CFTR mutations result in normal protein production, albeit at reduced levels and typically include promoter mutations that inhibit transcription, mutations that disrupt splicing motifs and amino acid substitutions that result in inefficient protein maturation [Nissim-Rafinia et al., 2006]. We have demonstrated that the CFTR c.-34C>T mutation results in inefficient protein translation and reduced mRNA stability via the action of a novel uAUG codon.
The CFTR c.-34C>T mutation was identified in a patient diagnosed with DB, a CFTR-RD. There was no evidence of mutation in the patient's second CFTR allele. Single-allele CFTR mutations have been identified at a higher frequency in patients with CFTR-RDs, such as CBAVD, DB, pancreatitis, and ABPA, than in the general population. Of these, heterozygosity for the CFTR M470V allele has been implicated in the development of asthma, DB, and COPD of unknown etiology [Casals et al., 2004; Tzetis et al., 2001]. Also, heterozygosity for other CFTR mutations, such as ΔF508, R117H, and R75Q, has been linked to the development of CBAVD and bronchiectasis [Wilschanski et al., 2006; Ziedalski et al., 2006]. We have shown that the c.-34C>T mutation disrupts normal posttranscriptional regulation of mRNA stability and translation, resulting in a dramatic loss of gene expression. The c.-34C>T mutation was identified in a heterozygous patient in the absence of any other CFTR mutations, suggesting that this CFTR regulatory mutation may contribute to the patient's CFTR-RD.
Clearly, there is not a direct correlation between single-mutant CFTR allele carrier status and CFTR-RD, as many carriers of CF disease-causing mutations are asymptomatic. There is increasing evidence for a multifactorial mode of inheritance of CFTR-RD, with environmental factors, CFTR mutation carrier status and additional mutations in other genes all thought to contribute to overall disease phenotype. One such CF-disease modifier gene mutation was recently identified in the endothelin receptor A gene (EDNRA). A 3′UTR sequence variation, that increased pulmonary EDNRA expression and airway inflammation, was significantly associated with decreased pulmonary function in ΔF508 homozygous CF patients [Darrah et al., 2009]. Sequence variations in other genes with roles in inflammation/immune function, such as TGFB1 [Drumm et al., 2005], IL8 [Hillian et al., 2008], and IFRD1 [Gu et al., 2009], have also been identified as CF modifier genes. CF modifier genes are also likely to play a role in the development of CFTR-RDs in the presence of single-allele CFTR mutations, such as the c.-34C>T mutation described here. In addition, Nissim-Rafinia et al.  suggested that CFTR polymorphisms or sequence variations in patients with CFTR-RDs, but without additional CFTR mutations, should be redefined as “mutations resulting in CFTR-RDs.”
We have identified a novel C to T mutation in the CFTR 5′ UTR in a patient with DB and shown that this mutation eliminates 85–99% of expression via perturbed posttranscriptional gene regulation. The c.-34C>T mutation in the CFTR 5′UTR creates an AUG codon upstream and out-of-frame with the CFTR protein AUG. We have demonstrated that this uAUG codon is efficiently recognized by scanning ribosomes, effectively eliminating translation initiation from the CFTR protein AUG.
This is the first report to establish a functional link between a 5′UTR mutation in CFTR and dramatically altered posttranscriptional regulation, which is associated with a CFTR-RD. Our data exemplifies the importance and broad potential of uAUGs and uORFs in the development of human disease. Combined with environmental and/or genetic factors, the c.-34C>T 5′UTR mutation may promote CF or CFTR-related lung disease and should be considered in CFTR mutation detection, particularly in idiopathic cases. Patients with CFTR-RDs, such as DB and COPD, would benefit from additional mutation screening to include the CFTR 5′UTR, to ascertain the extent to which CFTR regulatory mutations contribute to these common, multifactorial human diseases.
The authors would like to thank Dr. Carlo Castellani (Cystic Fibrosis Center, Verona Hospital, Verona, Italy) for the recruitment of DB patients in the initial population study. This research has been supported by the University of Queensland. SWL was the recipient of a University of Queensland Research Scholarship.
- 2005. Dissecting eukaryotic translation and its control by ribosome density mapping. Nucleic Acids Res 33:2421–2432. , , , .
- 1998. Complete mutational screening of the CFTR gene in 120 patients with pulmonary disease. Hum Genet 103:718–722. , , , , , , .
- 2007. Aberrant 5′ splice sites in human disease genes: mutation pattern, nucleotide structure and comparison of computational tools that predict their utilization. Nucleic Acids Res 35:4250–4263. , , , , , , .
- 2009. Upstream open reading frames cause widespread reduction of protein expression and are polymorphic among humans. Proc Natl Acad Sci USA 106:7507–7512. , , .
- 2000. Heterogeneity for mutations in the CFTR gene and clinical correlations in patients with congenital absence of the vas deferens. Hum Reprod 15:1476–1483. , , , , , , , , , and others.
- 2004. Bronchiectasis in adult patients: an expression of heterozygosity for CFTR gene mutations? Clin Genet 65:490–495. , , , , , , , , , .
- 2009. Chronic pancreatitis: genetics and pathogenesis. Annu Rev Genomics Hum Genet 10:63–87. , .
- 1991 Characterization of the promoter region of the cystic fibrosis transmembrane conductance regulator gene. J Biol Chem 266:24471–24476. , , .
- 2005. Increased risk of idiopathic chronic pancreatitis in cystic fibrosis carriers. Hum Mut 26:303–307. , , , , , , , , .
- 2010. Modifier genes in Mendelian disorders: the example of cystic fibrosis. Ann N Y Acad Sci 1214:57–69. .
- 2009. EDNRA variants associate with smooth muscle mRNA levels, cell proliferation rates, and cystic fibrosis pulmonary disease severity. Physiol Genomics 41:71–77. , , , , , , , , , and others.
- 2004. Cardiac expression of the cystic fibrosis transmembrane conductance regulator involves novel exon 1 usage to produce a unique amino-terminal protein. J Biol Chem 279:15877–15887. , , , .
- 2000. Mutation nomenclature extensions and suggestions to describe complex mutations: a discussion. Hum Mut 15:7–12. , .
- 2005. Genetic modifiers of lung disease in cystic fibrosis. N Engl J Med 353:1443–1453. , , , , , , , , , and others.
- 2008. Geneious v4.0. , , , , , , , , .
- 1997. CFTR gene mutations in adults with disseminated bronchiectasis. Eur J Hum Genet 5:149–155. , , , , , , , , , and others.
- 2009. Identification of IFRD1 as a modifier gene for cystic fibrosis lung disease. Nature 458:1039–1042. , , , , , , , , , and others.
- Group CFGMS. 2008. Modulation of cystic fibrosis lung disease by variants in interleukin-8. Genes Immun 9:501–508. , , , , , , ,
- 2005. uAUG and uORFs in human and rodent 5′ untranslated mRNAs. Gene 349:97–105. , , .
- 1998. Heterogeneity of reproductive tract abnormalities in men with absence of the vas deferens: role of cystic fibrosis transmembrane conductance regulator gene mutations. Fertil Steril 70:724–728. , , , , , , , , , and others.
- 1989. Identification of the cystic fibrosis gene: genetic analysis. Science 245:1073–1080. , , , , , , , .
- 2004. Role of CFTR mutations in adult bronchiectasis. Thorax 59:357–358. , , , , , .
- 2005. AUG codons at the beginning of protein coding sequences are frequent in eukaryotic mRNAs with a suboptimal start codon context. Bioinformatics 21:837–840. .
- 1986a. Influences of mRNA secondary structure on initiation by eukaryotic ribosomes. Proc Natl Acad Sci USA 83:2850–2854. .
- 1986b. Point mutations define a sequence flanking the AUG initiator codon that modulates translation by eukaryotic ribosomes. Cell 44:283–292.
- 1999. Mutation of the CDKN2A 5′ UTR creates an aberrant initiation codon and predisposes to melanoma. Nat Genet 21:128–132. , , , , , , .
- 2000. In vivo evaluation of the context sequence of the translation initiation codon in plants. Plant Sci 154:89–98. , , , , .
- 2009. The genetics of quantitative traits: challenges and prospects. Nat Rev Genet 10:565–577. , , .
- 1999. Expanded sequence dependence of thermodynamic parameters improves prediction of RNA secondary structure. J Mol Biol 288:911–940. , , , .
- 2009. The genetic signatures of noncoding RNAs. PLoS Genet 5:e1000459.
- 1998. Posttranscriptional control of gene expression in yeast. Microbiol Mol Biol Rev 62:1492–1553.
- 2006. Derepression of the Her-2 uORF is mediated by a novel post-transcriptional control mechanism in cancer cells. Genes Dev 20:939–953. , , .
- 1996. Cystic fibrosis transmembrane conductance regulator (CFTR) gene mutations in allergic bronchopulmonary aspergillosis. Am J Hum Genet 59:45–51. , , , , , , , .
- 2008. Diversity of preferred nucleotide sequences around the translation initiation codon in eukaryote genomes. Nucleic Acids Res 36:861–871. , , , , .
- 2006. The CFTR gene: structure, mutations and specific therapeutic approaches. In: Bush A, Alton EWFW, Davies JC, Griesenbach U, Jaffe A, editors. Cystic fibrosis in the 21st century. Cape Town: Karger. , , .
- 2001. Cystic fibrosis gene mutations and pancreatitis risk: relation to epithelial ion transport and trypsin inhibitor gene mutations. Gastroenterology 121:1310–1319. , , , , , .
- 2003. New type of disease causing mutations: the example of the composite exonic regulatory elements of splicing in CFTR exon 12. Hum Mol Genet 12:1111–1120. , , , , , , , .
- 1989. Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA. Science 245:1066–1073. , , , , , , , , , and others.
- 1989. Identification of the cystic fibrosis gene: chromosome walking and jumping. Science 245:1059–1065. , , , , , , , , , and others.
- 1988. Binding of a cytosolic protein to the iron-responsive element of human ferritin messenger RNA. Science 241:1207–1210. , , , , .
- 2010. Analysis of DEFB1 regulatory SNPs in cystic fibrosis patients from North-Eastern Italy. Int J Immunogenet. , , , , , , .
- 1996a. Isolation of differentially expressed cDNAs during ferret tracheal development: application of differential display PCR. Exp Lung Res 22:419–434. , , , .
- 1996b. Developmental expression patterns of CFTR in ferret tracheal surface airway and submucosal gland epithelia. Am J Respir Cell Mol Biol 15:122–131. , , .
- 2006. Exquisite and multilevel regulation of CFTR expression. In: Bush A, Alton EWFW, Davies JC, Griesenbach U, and Jaffe A, editors. Cystic fibrosis in the 21st century. Cape Town: Karger. .
- 2001. CFTR gene mutations–including three novel nucleotide substitutions—and haplotype background in patients with asthma, disseminated bronchiectasis and chronic obstructive pulmonary disease. Hum Genet 108:216–221. , , , , , , , .
- 2004. 5′-untranslated regions with multiple upstream AUG codons can support low-level translation via leaky scanning and reinitiation. Nucleic Acids Res 32:1382–1391. , .
- 1998. Tissue-specific in vivo transcription start sites of the human and murine cystic fibrosis genes. Hum Mol Genet 7:363–369. , , .
- 2006. Mutations in the cystic fibrosis transmembrane regulator gene and in vivo transepithelial potentials. Am J Respir Crit Care Med 174:787–794. , , , , , , , , , .
- 1991. The cystic fibrosis gene has a “housekeeping”-type promoter and is expressed at low levels in cells of epithelial origin. J Biol Chem 266:9140–9144. , , , , , , .
- 2006. Prospective analysis of cystic fibrosis transmembrane regulator mutations in adults with bronchiectasis or pulmonary nontuberculous mycobacterial infection. Chest 130:995–1002. , , , , .
- 2003. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res 31:3406–3415.