The cystic fibrosis transmembrane conductance regulator (CFTR) is a 1480 amino acid membrane bound glycoprotein with a molecular mass of 170,000. It is a member of the ATP binding cassette (ABC)superfamily of proteins which includes several clinically important proteins such as P-glycoprotein (P-gp), multidrug resistance associated protein and the TAP transporters.
The protein is comprised of two, six span membrane bound regions each connected to a nuclear binding factor which binds ATP. Between these two units is an R-domain which is comprised of many charged amino acids. The R-domain is a unique feature of CFTR within the ABC superfamily.
Nuclear Binding Domains (NBD)
Many of the mutations identified in CF occur in the first nucleotide binding domain (NBD1), while very few occur in NBD2. This is a common feature of the ABC superfamily and indicates a separate role for the two binding domains. The most common mutation in CF deltaF508 occurs in NBD1. This results in a 3 bp deletion and the loss of a phenylalanine residue. The deletion causes a protein trafficking defect. If this defect is overcome then the protein can form a functional channel. This can be brought about by overexpression of CFTR or by culturing cells at > 30 oC. The NBDs contain a number of highly conserved motifs predicted to bind and hydrolyse ATP. Site directed mutagenesis at these motifs have indicated that ATP binds to both NBDs to control the gating of the channel.
The R domain of CFTR is encoded by exon 13 and it spans the region between NBF1 and the second transmembrane region. It contains several potential sites for phosphorylation by cAMP dependent PKA or PKC. The activity of CFTR as an ion channel depends upon phosphorylation of the R domain and binding of ATP to the nuclear binding domains. The N terminal portion of the R domain (RD1) is highly conserved between species but there is a lower degree of conservation between the rest of the domain (RD2).
A mere 4 % of the CFTR protein is found in the extracellular loops (see the gene sequence and structure section). The loops are designated according to the membrane spanning regions they connect, M1-M2, M3-M4, M5-M6, M7-M8, M9-M10 and M11-M12 (always odd to even). With the exception of the M1-M2 and the M7-M8 these extracellular domains are very short. The M7-M8 loop contains two N-linked glycosylation sites.
19 % of the CFTR protein make up the twelve transmembrane domains (M1 - M12). These domains have been shown to be comprised of typical a-helical secondary structure. Many of the residues within these regions form the channel lining residues and have a major role in the regulation of pore function. Six positively charged residues within the transmembrane domains [K95 (M1), R134 (M2), R334 (M6), K335 (M6), R347 (M6) and R1030 (M10] that are well conserved across species. Two of these are associated with mutations causing CF, R334Q/W and R347C/H/L/P.
The Intracellular Domains
Sequences within the intracellular loops (ICL1 - 4) have been shown to be important for the processing of CFTR and correct delivery to the cell membrane. Site directed mutagenesis studies in ICL2 and ICL3 have indicated that these sections may be close to the intracellular opening of the CFTR pore. Changes in these regions have been shown to alter the conductance state of the channel.