BACK TO DIAGRAM CFTR REVIEW PAGE
R DOMAIN REVIEW
FESCVCKLMANKTRILVTSKMEHLKKADKILILHEGSSYFYGTFSELQNLQPDFSSKLMG
CDSFDQFSAERRNSILTETLHRFSLEGDAPVSWTETKKQSFKQTGEFGEKRKNSILNPINSI
RKFSIVQKTPLQMNGIEEDSDEPLERRLSLVPDSEQGEAILPRISVISTGPTLQARRRQSVLN
LMTHSVNQGQNIHRKTTASTRKVSLAPQANLTELDIYSRRLSQETGLEISEEINEEDLKECF
FDDMESIPAVTTWNTYLRYITVHKSL
(amino acids 587 thru 859 includes the R-domain somewhere, probably from amino acids 647 to 835)
It is the presence of the R-domain which sets CFTR apart from other ABC transporters. This addition to the CFTR structure has made understanding the function of the R-domain somewhat difficult and ambiguous. In fact, it now appears that the R-domain has more than one function. In addition to it's inhibitory role in channel gating (when unphosphorylated) the R-domain seems to enable the NBDs to bind and hydrolyze ATP more efficiently (when the R-domain is phosphorylated). It seems that the early ancestor of CFTR found a way of adding a new layer of complexity by incorporating the R-domain into its structure. Surprisingly, the only areas of the R-domain that are highly conserved across species are the PKA phosphorylation consensus sequences. It is probably not by chance that the R-domain has become situated in between the two NBDs in the primary structure of CFTR because this unusual domain seems to be capable of interacting with both NBD1 and NBD2, possibly at the same time. This R-domain review is divided up into 4 sections: Theory and Past Experimental Results, Summary of Mutagenesis Experiments, More R-Domain Facts, and Recent Evidence: More Pieces to the Puzzle?
THEORY AND PAST EXPERIMENTAL RESULTS:
The exact boundary between the R-domain and NBD1 still hasn't been mapped precisely. The carboxy terminal boundary of NBD1 could be from residue 589 to 670. The amino boundry probably doesn't extend past amino acid 708 because deletion mutants further in the N-direction do not produce a channel that is processed correctly and doesn't open in the presence of ATP. Conductive properties are also altered.
It is currently believed that both allosteric as well as electrostatic mechanisms are used by the R-domain in channel gating. The NBDs and R domains coprecipitate when expressed in the baculovirus protein expression system. The R-domain is phosphorylated by more than one type of kinase, including PKA and PKC. (1991) There are 2 PKC sites in the R-domain and sites probably exist for cGMP-dependent protein kinase, and Ca++/calmodulin-dependent kinase. 11 serines are phosphorylated in vivo by PKA (10 on R domain). This implies that a "graded response" occurs during gating. Apparently only 6 sites are heavily used and any 3 are expendable at one time. In 1991, 2 sites for phosphorylation were found to be inhibitory (S768 is one). Stimulation with PKC alone does not cause the same level of activation caused by PKA. CFTR is also somehow activated by actin polymerization, which may be able to take the place of PKA-stimulated activation of CFTR (i.e. the R-domain).
Phosphodiesterase enzymes play an essential and as yet uncharacterized role in deactivating the channel due to their ability to remove phosphates from the R-domain.
There are 5 serines (660, 700, 737, 795, 813) phosphorylated by PKA in cells that express CFTR normally. But there are a total of 10 phosphorylated in vitro by PKA (most of which are probably phosphorylated in vivo). In vitro, serines 660, 700, 712, 737, 753, 768, 795, and 813 are phosphorylated by PKA. Also, serine 422 in NBD1-R-domain peptide and serine 670 in an R-domain peptide have been shown to be phosphorylated, but not in vivo. Access of PKA to phosphorylation sites is influenced by the site itself (dibasic, monobasic, etc) and probably the structure it assumes in CFTR (secondary, tertiary or even quartenary structure). The structure of the R-domain has been shown to change (CD spect.) upon phosphorylation at certain sites. It's possible therefore that phosphorylation at one site can influence PKA's accessibility at other sites. Multiple bands on SDS-PAGE is also used. In vitro, it has been found that PKA phosphorylates at only 5 consensus sites in purified R-domain. PKC was able to phosphorylate at 2 places, S686 and S700. It's possible PKC potentiates PKA phosphorylation based on studies where PKC activators were added to cells before PKA activation. An increase in overall phosphorylation was seen as opposed to PKA alone.
Dulhanty and Riordan have proposed a model of the R-domain which has two distinct domains, residues 587 to 672 (called R1) and 679 to 798 (R2). R1 is the most conserved sub-domain between species while R2 is not. This is true even when considering the many consensus sequences for phosphorylation in R2. Interestingly, CF mutations in R1 cause misfolding of CFTR. These researchers propose that only the R1 sub-domain has a well-defined structure. The R domain may be considered as consisting of 2 distinct halves: a c-terminal and an n-terminal one. Sequence comparison of 10 species showed areas most conserved. 14 viral polymerase proteins are said to show limited homology to fragments of the R domain.
The R-domain also has sites for cGMP TypeII protein kinase phosphorylation. PKC phosphorylation may be needed prior to PKA to activate channel in vivo. Dephosphorylation is via the phosphatases PP2C and PP2A. PP2C has been cross-linked and coimmunoprecipitated with CFTR. Adding phosphorylated R-domains to mutant CFTR lacking R-domains, but w/ATP present increases channel open probability. The conclusion most often drawn from this is that the R-domain causes the NBDs to pick up ATP easier. Once the channel is open, it is no longer dependent on PKA activity to stay open (assuming no phosphatase activity is present). However, if ATP is removed, the channels close.
A mobility shift of recombinant R-domain upon phosphorylation has been attributed to a change in conformation. This predicted conformational change has also been detected by CD (circular dichroism). CD has been used to show a decrease in alpha-helical content of R domain upon phosphorylation by PKA.
A Possible Conclusion: When the R-domain in wild-type CFTR is unphosphorylated, it somehow is able to inhibit the NBD1, perhaps directly by keeping it from either binding or hydrolyzing ATP. It may also "get in the way" of the NBD1 as the NBD tries to open the channel pore like normal. When the R-domain is phosphorylated, it may then somehow undergo a conformational change which allows it to interact with the same NBD1, but at a different site, and stimulate it to perform its usual gate opening responsibilities (i.e. pick up ATP and hydrolyze it, and communicate this to the "gate" to cause channel opening). It may also help the NBD to interact with the pore region more efficiently.
The simplest view is that the R domain functions as a channel inhibitor until phosporylated by PKA, whereupon it changes it's conformation to open the pore to chloride. The next step involves the NBDs. Binding and cleavage(?) of ATP must occur at the NBD1 for the channel to open all the way to chloride. The channel remains open until the same thing occurs at NBD2. When this occurs, the R domain becomes accessible to phosphatases and becomes inactivated by dephosphorylaton. Phosphorylation of the R-domain could have two effects: 1) release of the steric inhibition and 2) stimulation of the NBDs to pick up ATP (and hydrolyze it?).
There is currently no single theory about how the R domain functions which is generally accepted. Some models show the R domain interacting directly with the two NBDs, kind of like a bridge which joins them. It implies that phosphorylation of the R domain directly influences NBD binding of ATP. One possibility is that the R domain is analogous to a receptor (or exchange factor) for G-proteins, and it increases the rate NTP exchange takes place at the NBDs.
Residues 760-783 include a highly conserved portion of the R-domain shown to be important for constitutive activity of the channel. This motif consists of amino acids R765-R766-Q767-S768-V769-L760. Three other conserved motifs are K696-R697-K698, R709-K710, and R764-R765-R766. These positively charged residues may interact with negatively charged residues within the N-terminal tail. In a paper by Chen et al, they state that "...these motifs are conserved even within the most evolutionarily divergent R domain sequences." Mol. Biol. Evol. 18(9):1771-1788. 2001
SUMMARY Of EXPERIMENTS: Deletion of the R gave only 1/3 normal conductance of Cl- in absence of cAMP but better conductance when it was present. This was the first suggestion that NBD2 and the R-domain interact. Adding R-domains to mutant CFTR w/o R domains has no effect unless phosphorylated. Changing serines to alanines one by one in R domain changes sensitivity to activating conditions in a manner "highly dependent on site". It is possible to replace the serines with negatively charged amino acids like aspartate and the channel will open as if phosphorylated.
Lack of a change in CFTR gating does not necessarily mean mutagenesis has no effect. It's always possible changes are too small to detect with patch clamping. A key question to keep in mind when considering mutagenesis information to understand R-domain interactions is the following: is the interaction between the R-domain and the NBDs a result of a change in the conformation of the R-domain, an increase in the negative charge due to phosphorylation of it, or both?
The data which supports the notion that the R-domain acts normally to inhibit CFTR while at rest include: its partial deletion relieves the inhibitory effect; so too does changing the serines which are phorphorylated by PKA to aspartic acids (i.e. increasing the negative charge on the domain). Also, adding to R-deletion mutants the soluble unphosphorylated R-domains causes inhibition again. The data which indicates that the R-domain can also act as a stimulator include: the activity of CFTR when it's phosphorylated is higher than of mutants which do not need phosphorylation, as well as the addition of soluble phosphorylated R-domains increase the rate of activation of CFTR-deltaR and S660A.
CFTR with only a portion of its R-domain (amino acids 708 thru 835 deleted) functions as a normal channel except that it does not require PKA to be activated. But the open probability is only 1/3 normal CFTR. The amino acids which appear expendable are also not conserved in other ABC proteins. Extensive deletion of R domain results in non-functional channels. Loss of the part of the R-domain also has the effect of suppressing a mutation at NBD2.
When the R-domain peptide consisting of amino acids 590 thru 858 was added to wild-type CFTR, it required phosphorylation to activate it. When it was unphosphorylated, it inhibited it. When purified unphosphorylated R-domains alone were added to CFTR lacking R-domains, it had no effect on the channel's activity. But when phosphorylated, it increased the open probability. This contradicts other studies.
When all 9 PKA consensus sites were mutated, there was still substantial activation. Most of this residual activity has been explained by a 10th site located outside the R-domain and just N-terminal of NBD1 at serine 422. The serines at positions 660, 700, 737, 795, and 813 seem to be the most important for activation of CFTR by PKA. A key question is why does phosphorylation at serines 737 and 7680 appear to be inhibitory? And why does phosphorylation by PKC (at serines 684 and 790) seem to potentiate activation by PKA? When all 10 possible serine and threonine phosphorylation sites in the R-domain were mutated to alanines, there was still residual activity of CFTR when activated by PKA. It was shown that a serine at position 753 was mainly responsible. When mutated to an alanine, the activity decreased significantly. The affinity of the NBDs for ATP is decreased when the R-domain is not phosphorylated.
Mutating 8 serines to aspartic acids activated the channel, proving negative charge is important. However, the resulting channel activity was still decreased compared to wild-type hinting that additional effects are provided by phosphorylation, such as a conformational change.
When the Atlantic salmon's two CFTR genes are compared with each other on the amino acid level, the r-domains show the greatest sequence divergence of the entire protein (~85% similarity).
When the neutral adduct N-ethylmaleimide was covalently attached to C832, it stimulated chloride channel activity as well as increasing the ability of ATP to stimulate CFTR. Since this residue is not phosphorylated, it suggests that perhaps a change in conformation of the R domain is necessary for activation. Studies using CD spectroscopy also suggest a conformational change takes place upon phosphorylation.
The R-domain has a relatively high percentage of charged amino acid residues (~28%). Of the 24 basic residues, 16 of them are at PKA phosphorylation sites. The acidic residues tend to be clustered between residues 817-838. Increasing R-domain phosphorylation increases the open probability of the channel. When the R-domain is expressed as a soluble peptide (708-831), it has a helical content of about 5%, with the remainder being random coil. Changes in pH, salt, or phosphorylation status does not seem to affect this.
There may be some interactions among the various serines when phosphorylated. Evidence for this is the following: both serine-660 or serine-813 need to be present for wild-type channel activity. Also, serine-737 is unable to stimulate or inhibit channel activity unless other serines are present. It is possible that sequential phosphorylation and/or concomitant phosphorylation are needed for full activity.
There are 8 prolines in the human R-domain, 3 of which are highly conserved. When 3 (P740, 750, 759) are mutated simultaneously to alanines, there was a significant increase in activity compared to wild type. However, when the cis-trans peptidyl-prolyl isomerase cyclophilin A was added, only wild type but not the mutated channel changed its activity (because the mutant was already active).
The R-domain does not simply block the channel, as was at first thought in the early 1990s, since deletion mutants are inactive until ATP is added.
It still isn't known whether ATP binding in and of itself is able to cause conformational changes, or whether hydrolysis is also necessary. If so, then these steps would be reversible.
Consensus sequences for PKA are as follows: RR/KXS/T > RXXS/T = RXS/T CFTR has 10 sites for phosphorylation by PKA which have 2 basic positively charged residues in the consensus sequence. Only 8 appear to be actually phosphorylated by PKA, however. 8 of them are phosphorylated on serines (S660, S686, S700, S712, S737, S768, S795, and S813) while two are at threonines (T788 in the R domain) and one in front of NBD1 (S422). There are 2 which are not phosphorylated by PKA: T788 and S422.
The amount of time CFTR spends in the open state is increased when phosphorylated at all sites. An increased concentration of ATP is needed to activate CFTR when it has lost it's PKA sites. Sensitivity of CFTR to ATP is influenced by phosphorylation state of R-domain. Somehow the R-domain interacts either directly or indirectly with the NBDs. Not all R-domain partial deletions are functional. But most are active (10-20% of wild type phosphorylated by PKA) when only ATP is present. Because addition of just the unphosphorylated R-domain to these mutants does not block PKA-independent activity, the R-domain may not simply be blocking the pore. Most researchers now believe that there are 2 activities associated with the R-domain: 1) to inhibit activity of the channel. This inhibition is relieved by PKA activity. 2) to stimulate the channel activity when phosphorylated.
Electrophysiological evidence suggests there are at least 3 functionally distinct phosphorylated states which may be due to different actions of specific protein phosphatases on phosphorylated CFTR.
One view is that R-domain function is not dependent on which but rather the total number of serines phosphorylated. This implies the interaction is primarily electrostatic rather than the result of a conformational change. A problem with this model is that not all serines seem to be equivalent. Also, several phosphoforms can be distinguished using channel function as a criteria. It may be that the R-domain is predominantly random-coil in structure and does not adopt a well-ordered structure until it makes contact with the rest of CFTR.
Serines 737 and 768 are inhibitory when phosphorylated and the ion channel is not fully phosphorylated at other serines. When amino acids 760-835 were removed, constitutive activity was the result (i.e. they must be inhibitory, normally). More specifically, residues 760-783 seem to be the most inhibitory. When peptides of this amino acid stretch were added (or moved to the c-terminus), however, there was still constitutive activity. This implies that these inhibitory residues need to be present in their normal place in the protein (i.e. between the two NBDs).
PKC with or without calcium in vitro phosphorylates the R-domain at serines 686 (a PKC-specific site which is also phosphorylated in intact cells treated with phorbol ester by PKA) and 790. Serines 660 and 700 are normally phosphorylated by PKA, but can be by PKC. There was no change in secondary structure (via CD) when PKC was used. PKC applied directly to CFTR in excised patches activates it, but often only weakly. PKC also activates CFTR as efficiently as PKA does when phorbol ester is added to intact cells (but not in cardiac myocytes or pancreatic duct cells). It consistently potentiates CFTR to PKA in both excised and cell intact studies. PKC does this even in mutants lacking PKC sites known from in vitro studies. It may phosphorylate at serine 790 which then allows PKA to phosphroylate at sites it normally has trouble with. It is possible PKC is activating CFTR to PKA indirectly, such as phosphorylation of cytoskeltal elements.
Only 4 of the 11 isoforms of PKC known are activated by calcium. CFTR can be activated in vivo by either type.
cGMP-dependent protein kinases have also been found to activate CFTR. The do so at sites which appear to overlap PKA sites. They phosphorylate at over 5 sites. The two isoforms studied were PKGI and PKGII. PKGII is myristoylated and localizes to the cell membrane. This may turn out to be the mechanism by which pathogenic strains of E. coli activate CFTR in the intestinal epithelium. It's been shown that the enterotoxin secreted by the bacteria stimulates guanylyl cyclase to produce cGMP and causes diarrhea. Note: cholera toxin activates CFTR via cAMP pathway. PKGII knockout mice appear to be resistant to the E. coli above.
High levels of cGMP could conceivably promiscuously "cross-activate" CFTR by activating PKA.
CFTR which has been purified can be activated by CaM (calmodulin kinase I). It has not been tried in vivo yet.
The tyrosine kinase inhibitor genstein increases activation of CFTR by prolonging the open state. Recently, phosphorylation has been detected on tyrosine of CFTR (anti-phosphotyrosine antibody), probably by the tyrosine kinase p60(c-src). Src can activate CFTR when PKA is not present in excised patches, which seems to contradict the finding that genstein works by inhibiting this kinase. It is now believed that genstein upregulates CFTR channel activity in another way, perhaps by binding directly to it and stabilizing the open state at NBD2. But when genistein is added in high concentrations (over 50 uM) it has an inhibiting effect. It therefore may bind at a lower affinity site on CFTR within the pore. A possible explanation as to why genistein increases the phosphorylation of CFTR possibly by stabilizing the open configuration at NBD2, which may make CFTR less attractive of a substrate for phosphatases.
Dephorphroylated R-domain probably acts more to prevent ATP hydrolysis rather than ATP binding. This is implied because wild-type CFTR with a dephorphorylated R-domain is unable to open, even in presence of ATP and because mutants lacking most of R-domain will open when ATP is present. The R-domain does NOT form a plug because CFTR w/o it will stay closed until ATP is added. Even wild-type CFTR with highly phosphorylated R-domains remain closed w/o ATP. Direct measurements now show PKA activated CFTR increases ATP hydrolysis. CFTR lacking R-domain have slightly increased activity when phosphorylated but not dephosphorylated R-domains are added. Therefore, the R-domain may have 2 functions: 1) inhibition by dephorphroylated R-domains, and 2) increase of ATP binding at NBDs when R-domain is phorphroylated. It should be remembered that removal of most of the R-domain creates a structure which may be constrained such that open probability of the channel decreases. Some of the best evidence that channel gating is tied to ATP hydrolysis and not just ATP binding comes from the fact that Mg++ must always be present, even in extremely high concentrations of ATP. An important question remains: why are specific regions within the R-domain required for inhibition, and yet these same regions are not enough in and of themselves to cause inhibition? This is shown by the fact that these residues are not inhibitory when transferred to other parts of the protein.
RD1 is the N-terminal third of the R domain and is highly conserved, while RD2 is the large central region of the R domain has less rigid structural requirements. But the fact that two of the four main phosphorylation sites are excluded from RD2 implies that RD2 is not the only part that is "functional". Amino acids C647 to D836 stand up to comparison tests of CFTR sequence conservation studies between 15 different species as being most important in R-domain function. The fact that portions of the R-domain are able to stimulate channel activity suggests that the R-domain does not normally have a well-defined tertiary structure. And the lack of amino acid sequence homology among species also suggests lack of a well-defined structure. For example, there is a lack of conserved hydrophobic amino acids, which are often needed for forming highly ordered domains (for example, the NBDs). An important clue to R-domain structure could be that the positions of the PKA sites are conserved across species, and yet the structure itself seems to be predominantly random-coil in nature.
While CFTR is the only member of the ABC Transporter family to have an R-domain, the P-gp protein has a region in between NBD1 and its second membrane spanning domain, which includes two PKA consensus sequences (motifs). It was found to be necessary for ATP hydrolysis and drug transport. An unrelated sequence was used instead, and this restored activity of the pump. When the normal linker residues were put into CFTR where amino acids 780-830 were, it restored channel activity to wild-type levels.
Recent Evidence: More Pieces of the Puzzle?
A short peptide segment with a net negative charge of -9 (amino acids 817-838 called "NEG2") and predicted helical tendency within the R domain has been shown to play a critical role in CFTR chloride channel function. Xie et al. from Case Western Reserve University in Cleveland, Ohio in April, 2002 reported that when they deleted NEG2 region from CFTR, it completely eliminated the PKA dependence of channel activity, and when they added exogenous NEG2 peptide, it interacted with CFTR to in their words "...exert both stimulatory and inhibitory effects on the channel function." When they scrambled the NEG2 sequence to remove helical tendencies it inhibited channel function, but did not stimulate it. They got the same result when they inserted a proline to disrupt helical tendency. However, when six of the negatively charged carboxylic acid residues were replaced with neutral charges, reducing the net negative and increasing helical propensity, they found "...the peptide stimulated CFTR channel function, but did not inhibit." They went on to speculate that "...the NEG2 region interacts with other cytosolic domains of CFTR to control opening and closing transitions of the chloride channel." J Biol Chem 2002 Apr 11
In May, 2001 Wei et al, reported their investigation into the structural basis of CFTR inhibition of CaCC (a calcium-activated chloride channel) in bovine pulmonary artery endothelium (which does not normally express CFTR). It should be noted that inhibition of CaCC does not require CFTR activation, but is potentiated when CFTR is activated. They transiently infected these endothelial cells with various CFTR gene constructs and then used patch clamp to assess the functional interaction between the two chloride channels, CFTR and CaCC. They found that CaCC was stimulated by the addition of ATP. They state "...the inhibitory effect of CFTR was conserved when the PDZ binding motif was deleted. In contrast, both the CFTR activity-independent and -dependent inhibition of CaCC were abolished when the C-terminal part of the regulatory (R)-domain of CFTR was deleted (CFTR-delta R780-830). The activity-dependent inhibition of CaCC, but not the activity-independent inhibition, could be rescued by introducing the multiple drug resistance (MDR)-1 mini-linker in place of the deletion (CFTR-delta R-linker). It is concluded that the C-terminal part of the R-domain is an important determinant for CFTR-CaCC interaction." Pflugers Arch 2001 May;442(2):280-5
In May, 2000 Ostedgaard expressed soluble R-domain (708-831) and found phosphorylated version is functional but predominantly random coil (used CD and limited proteolysis). The CD spectra of phosphorylated and nonphosphorylated domains were both similarly random in structure, implying that phosphorylation doesn't change the structure greatly. They state that: "the random nature may explain the seemingly complex way in which phosphorylation regulates CFTR activity."
Not all sites phosphorylated by kinases have consensus sequences. This could be explained by the possibility that these "cryptic" consensus sequences must be read from carboxy to amino terminal to see it. Both may have similar shapes.
Baldursson in November of 2000 (of Howard Hughes Medical Institute) mutated four of the following serines known to be phosphorylated in vivo in an attempt to define the functional role of regulatory domain phosphoserines : Ser(660), Ser(737), Ser(795), and Ser(813) (S-Quad-A), and got a decreased cAMP-stimulated current. They suggest that these four serines account for most of the phosphorylation-dependent response. Especially serines Ser(660) or Ser(813) when mutated by themselves significantly decreased current and therefore are key in phosphorylation-dependent stimulation. It was also found that neither Ser(660) nor Ser(813) alone increased current back to normal; in fact, both residues were needed for this. Changing Ser(737) to alanine increased current above wild-type levels, suggesting that phosphorylation of Ser(737) may inhibit current in wild-type CFTR.
In October, 2000 Baldursson, Ostedgaard, Rokhlina, Cotten, and Welsh (of Howard Hughs Medical Institute) found that "several portions of the R domain conferred phosphorylation-stimulated activity" and that "this was true whether the R domain sequences were present in their normal location or were translocated to the C-terminus." They also found that some parts of the R domain could be deleted without inducing constitutive activity. But when they deleted residues 760-783, the channels opened without needing to be phosphorylated. When they translocated the R domain to the C-terminus, it did not prevent constitutive activity. They therefore suggest that "different parts of the phosphorylated R domain can stimulate activity, and that their location within the protein is not critical." "In contrast, prevention of constitutive activity required a short specific sequence that could not be moved to the C-terminus. These results are consistent with a recent model of an R domain composed primarily of random coil in which more than one phosphorylation site is capable of stimulating channel activity, and net activity reflects interactions between multiple sites in the R domain and the rest of the channel." J Biol Chem 2000 Oct 18
Csanady, Chan, Seto-Young, Kopsco, Nairn, and Gadsby (of Rockefeller University) in September, 2000 expresses "split" CFTR channels in oocytes. A single cut (between residues 633 and 634) was made just before the R domain, as well as split channels with a single cut (between residues 835 and 837) just after the R domain, and from which the entire R domain (residues 634-836) between those two cut sites was omitted was made. They state that "the channels cut before the R domain had characteristics almost identical to those of WT channels, except for less than twofold shorter open burst durations in the presence of PKA. Channels cut just after the R domain were characterized by a low level of activity even without phosphorylation, strong stimulation by PKA, enhanced apparent affinity for ATP as assayed by open probability, and a somewhat destabilized binding site for the locking action of the nonhydrolyzable ATP analog AMPPNP. Split channels with no R domain (from coexpression of CFTR segments 1-633 and 837-1480) were highly active without phosphorylation, but otherwise displayed the characteristics of channels cut after the R domain, including higher apparent ATP affinity, and less tight binding of AMPPNP at the locking site, than for WT. Intriguingly, severed channels with no R domain were still noticeably stimulated by PKA, implying that activation of WT CFTR by PKA likely also includes some component unrelated to the R domain. As the maximal opening rates were the same for WT channels and split channels with no R domain, it seems that the phosphorylated R domain does not stimulate opening of CFTR channels; rather, the dephosphorylated R domain inhibits them." J Gen Physiol 2000 Sep;116(3):477-500
In August, 2000 King and Sorscher (UAB) coexpressed the R-domain with Delta1-836 CFTR (a carboxyl hemi-CFTR beginning immediately after the R-domain) in order to ascertain whether binding of the R-domain occurs with other parts of the channel. Using coimmunoprecipitation methods, they found that the R-domain binds to the Delta1-836 construct. They write: "Our results indicate that the R-domain binds CFTR residues after amino acid 836 and that this binding facilitates phosphorylation and CFTR activation. We have also characterized a subdomain within CFTR (residues 723-837) that is necessary for PKA-dependent constitutive activation." "...these experiments demonstrate that constitutive CFTR activity can be accomplished by at least two mechanisms: (1) direct modulation of the R-domain to abrogate PKA regulation and (2) modifications that increase R-domain susceptibility to steady-state phosphorylation through PKA." Biochemistry 2000 Aug 15;39(32):9868-75
At the Biophysical Society Meeting on 2/2001, Mutsuhiro et al. presented evidence that among the R-domain mutants they engineered into CFTR, only partial R-domain mutants del(760-783) were not constitutively active without PKA stimulation, "suggesting an inhibitory role to these residues." They also translocated amino acids 760-783 to the C-terminus of CFTR and found that it was unable to inhibit channel activity, suggesting that these amino acids normally interact with other parts of the R-domain. It was also noted that upon translocation of these amino acids to the C-terminus of a CFTR deletion mutant lacking amino acids 708-835 (at the R-domain) that PKA was better at stimulating current than the other deletion mutants. They concluded from this that 760-783 amino acids "may play both stimulatory and inhibitory roles. [Our] results support the conclusion that multiple phosphoserines in an unstructured R domain regulate channel function."
The mutations E193K, D648V, H949Y, and R1070Q are all CF-causing mutations that are associated with pancreatic sufficiency. They have no effect on chloride transport but reduce bicarbonate transport by 50-60 %. Nature 3/1/01 pgs 94-96
In 1999, Naren, et al. provided strong evidence that the R-domain and N-terminus of CFTR interact. Science 286, 544-48
One of the paradigms used to describe how the R-domain might affect CFTR channel opening and closing has been based upon channel gating of the potassium channel Shaker. It has been believed that they open and close via the "ball and chain" model, whereby an intracellular ball-like domain of the channel is able to move up to the opening of the pore and close it by plugging it. New evidence based upon site-directed mutagenesis and crystallography by MacKinnon et al. has revealed that the N-terminal tail portion of the "ball" domain alone is able to snake its way deeper into the pore and block it from potassium travel. They show that electrostatic and hydrophobic events drive the closing. The amino terminus of the tail domain is positively charged at normal pH inside the cell, and the central cavity inside the pore is organized to favor cation binding. The rest of the tail peptide is hydrophobic and binds to the rest of the interior of the cavity, which is also hydrophobic. Because the R-domain of CFTR does not have a free N-terminal peptide tail as the potassium channel does, this may mean a rethinking of how the R-domain of CFTR, when unphosphorylated, is able to block chloride ion flow. Nature 411, 657(2001).