BACK TO DIAGRAM CFTR REVIEW PAGE
NUCLEOTIDE BINDING DOMAIN 2 REVIEW
YTEGGNAILENISFSISPGQRVGLLGRTGSGKSTLLSAFLRLLNTEGEIQIDGVSWDSITLQQ
WRKAFGVIPQKVFIFSGTFRKNLDPYEQWSDQEIWKVADEVGLRSVIEQFPGKLDFVLVD
GGCVLSHGHKQLMCLARSVLSKAKILLLDEPSAHLDPVTYQIIRR
(amino acids 1219 thru 1386)
The nucleotide binding domains are the most conserved portions of CFTR, and NBD2 shares similarity with NBD1 in sequence (~30%), however the ultimate function of these two domains in channel gating seems to differ considerably. While both NBD domains have been shown to bind and hydrolyze ATP, NBD1 (along with ICLs 1 and 2) plays more of a role in channel opening, while NBD2 (and ICLs 3 and 4) are involved in channel closing. Usually, mutations in NBD2 serve to prolong the open state of CFTR, for example the substitution mutation K1250A in the Walker A motif. When considering the actions of the NBDs, it should be remembered that cross-talk probably occurs between them, because a mutation in one NBD domain can have large effects in the overall ATPase activity of pure CFTR. Many believe that the CFTR ion channel evolved from a bacterial ABC transporter, and at that time first functioned as a pump in bacteria around 100 million years ago. (Pumps can be distinguished from ion channels in that pumps require the use of ATP as an energy source in order to move molecules against a concentration gradient. But channels on the other hand use already established gradients and move molecules without the need for ATP hydrolysis as an energy source) A question might then be raised: why would CFTR need two domains (i.e. two NBDs) specialized for binding and hydrolyzing ATP if it does not use ATP in the same way a pump would use it? Perhaps the ancesteral version of CFTR managed to somehow relegate its two nucleotide binding domains into a more regulatory role: that of opening and closing the channel. It appears that the NBDs do function today more as "gatekeepers" rather than as transducers of chemical energy into mechanical energy, as they would in a pump. Also, it is possible that the NBDs may be involved in regulating other ion channels, such as ORCC, in the cell as well as regulating CFTR itself. This might explain the presence of such an unusual change that the NBDs have undergone, perhaps even before vertebrates evolved.
For structure information, see NBD1.
THEORY AND PAST EXPERIMENTAL RESULTS:
The two NBD domains of the multi-drug resistance protein MDR are symmetrical in sequence identity, while those of CFTR are not, suggesting different roles for the NBD domains of CFTR. They may operate "in series" during gating.
Single ATP binding occurs at each of the NBD domains following phosphorylation of the R-domain. Hydrolysis of ATP at NBD1 is necessary for channel opening while NBD2 is more involved in channel closing. CFTR can be locked into various kinetic states using non-hydrolyzable analogs of ATP. Most likely, steps corresponding to ATP hydrolysis also correspond to the irreversible steps in gating which keep the process moving in the right direction. The fact that mutations in NBD1 also affect channel closing has been used to infer "cross-talk" between the two NBDs. The solved NBD structure from bacteria is known to form a dimer both in the crystal structure as well as in vivo. This is shown to occur along the edge of Arm I. Purified CFTR binds ATP w/Km of 0.3mM and hydrolyzes it at a rate of ~ 1 molecule ATP/second, which is similar to rate of channel opening and closing. NBDs have Walker A and B motifs. So do other ATPases.
NBD2 appears to have lower ATP hydrolase activity compared to NBD1 (in vitro assays), but nonetheless still believed to hydrolyze ATP in the same manner, while producing a different response in gating. It may shut off the channel when hydrolysis occurs. Mutagenesis and loss of the entire second half of CFTR indicate the second half is more involved with inhibition of the channel.
Peptides synthesized based on the NBD sequences have been shown to interact with the R domain and intracellular loops (yeast 2 hybrid and coimmunoprecipitation studies). They also have ATPase activity.
In both NBDs, GTP has been found to be able to replace ATP to an extent, but not so with non-hydrolyzable analogs. ADP competitively competes with ATP at NBD2.
The fact that CFTR responds to cellular concentrations of ATP indicates that CFTR is responsive to the energy levels of the cell. This would allow the cell to reduce ion transport in times of energy need.
Mutations in this domain as well as NBD1 and the R domain have been shown to alter gating rather than kinetics.
The K1250A mutation in NBD2, besides being able to prolong channel openings, is able to inhibit channel opening, which has previously been believed to be under control of NBD1 only. This implies considerable cross-talk between the two domains is taking place. Other members of the ABC transporter family have NBDs that are believed to dimerize or tetermerize.
ADP competes with ATP and therefore is an inhibitor. But when mutations were made in NBD2, this effect was reduced and channel activity increased. But mutations at NBD1 had no effect, which means ADP competes apparently only at NBD2. This constitutes still more evidence that the two NBDs function differently.
It is possible that closing of the channel is the result of the NBDs releasing the hydrolysis products ADP and/or Pi rather than hydrolysis itself. As yet, there is no way to know whether or not they remain bound to the NBDs after hydrolysis.
Phosphorylation of the R-domain of CFTR also increases the ATPase activity of the NBDs.
Walker A motif is G-X-X-G-X-G-K-T/S and Walker B motif is R/K-X(7-8)-h(4)-D. Synthetic peptides made from the sequence of NBD2 are able to bind ATP in vitro. The structure of the NBD from the E.coli ribose ABC transporter is known and makes it likely that NBD1 is actually 60 residues into R domain as originally defined. This probably means the acid-end of NBD2 will also need to be redefined.
There is evidence for cross-talk between the NBDs. Mutations in the Walker A motif of either is able to decrease ability of Magnesium-ATP to open the channel. Also, some mutations which destabilize the open state have been found in NBD1 while corresponding ones in NBD2 will prolong the open state. Perhaps there is overlap in function as well.
Mutation of the aspartate responsible for Mg++ binding increases channel open time. NBD2 seems to have a clearer role than NBD1. NBD1 with a mutation in Walker A lysine is not able to completely abolish channel activity, as would be expected if it was solely responsible for channel opening.
The tyrosine kinase inhibitor genstein increases activation of CFTR. 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.
Amino acids 1208 thru 1399 have been expressed as a fusion protein with maltose-binding protein and shown to hydrolyze ATP with an estimated turnover rate of 0.01/sec and are more on the order of GTPase activity reported for G-proteins rather than other ATPases. However, the high-resolution crystal structure of the NBD of RbsA indicates that the NBDs of ABC transporters are unique and unlike G-proteins or other ATPases. CFTR's NBD1 was easily threaded onto the structure.
The K1250M mutation occurs in the Walker A motif in some CF mutant alleles. This mutation does not affect channel opening, suggesting NBD1 is more important for this than NBD2. However it slows channel closing. Other mutations in NBD2 which are similar in effect include G1349 and D1370.
The sequence LSGGQ is known as the ABC "signature sequence". It is found between the Walker A and B motifs. It may help couple ATP hydrolysis with transport. The second glycine is the site of a natural mutation in CF. This mutation can occur in either of the NBDs. The crystal structure of RbsA shows LSGGQ is located in an alpha helix which is NOT part of the catalytic site and may be able to interact with the transmembrane helices.
There are less natural mutations in NBD2 than NBD1. This is consistent with the view that the NBD1 domain has a more important role due to its involvement in channel opening, or the fact that processing mutations occur at NBD1.
Nonhydrolyzable analogs of ATP do not open CFTR channel by themselves, but in presence of some ATP they probably bind to NBD2 and inhibit channel closing by inhibiting hydrolysis. NBD2 may normally act as a kind of "timer" switch, telling the channel when to shut off. This has been compared to the known actions of cellular GTPases. These proteins are responsible for controlling the duration of activation of their downstream effectors.
Mutations G1244E, G1349D and S1255P in this domain are processed normally in epithelial cells, and are able to mediate normal chloride transport, however they are associated with CF, specifically pancreatic insufficiency. Other mutant CFTRs like this include G178R, A1067T, and I148T.
The signature sequence LSGGQ in NBD1 is present as LSHGH in NBD2. There are 17 mutations in the LSGGQ sequence and Walker B motif of NBD1 which cause CF, but only 4 in the corresponding region of NBD2. However in the Walker A motif of NBD1, there is only 1 CF-causing mutation, while in NBD2 there are 5.
Recent Evidence: More Pieces to the Puzzle?
Chen et al. in 2001 found by doing CFTR sequence comparisons among species that the NBD2 domain probably extends from 1225 to 1417. And that the acid-terminus of CFTR's NBD2 domain is less functionally restrained throughout the ABC transporter family. They concluded that this "..lends further weight to the unique structure of NBD1". Mol. Biol. Evol. 18(9):1771-1788 2001
Berger, et al (U.Iowa) presented evidence at the Biophysical Society Meeting on 2/2001 that Q1291 may face the gamma-phosphate of ATP. They used crystal structures of the NBDs of HisP and LivF as models to design experiments to test the theory. They examined the chloride currents of CFTR in presence/absence of PPi added along with ATP and found that mutant Q1291A CFTR showed only a ~15% stimulation (compared to 100% in wild-type) in presence of PPi. However, all other aspects of gating were normal. They concluded: "These results suggest that Q1291 may be a phosphate sensor that mediates the stimulatory effects of PPi."
The K1250A mutant CFTR channel, once open, can stay open for
minutes. Zhou, et al. reported in April, 2001 that
"Flickering block of K1250A-CFTR channels was voltage dependent since the
open probability within an opening burst decreased as the membrane was
hyperpolarized." The concluded "...these
results suggest that the resident time of the blocker is prolonged by conditions
(i.e. hyperpolarization or the absence of external permeant anions) that deplete
Cl- in the CFTR pore. Results from macroscopic current noise analysis of both
wild-type CFTR and K1250A-CFTR channels further confirm the voltage dependence
and Cl- sensitivity of the fast flickery block observed with single-channel
analysis. We conclude that the voltage dependence of the flickery block in CFTR
is mainly due to the voltage-dependent occupancy of an anion-binding site in the
channel pore by trans-anions. The blocker acquires a voltage-dependent off rate
through an electrostatic interaction with Cl- in the
pore." J Physiol 2001 Apr 15;532(Pt 2):435-48
The function of NBD2 may be only to prolong the open state, and NBD1 may be able to both open and close the channel. This theory, proposed by Zeltwanger et al., is based on mutagenesis and ATP concentration studies at NBD2. Zeltwanger, et al., J. Gen. Physiol. 113, 541-554 (1999)