Alignment of transmembrane regions in the cystic fibrosis transmembrane conductance regulator chloride channel pore

Different transmembrane (TM) α helices are known to line the pore of the cystic fibrosis TM conductance regulator (CFTR) Cl(-) channel. However, the relative alignment of these TMs in the three-dimensional structure of the pore is not known. We have used patch-clamp recording to investigate the accessibility of cytoplasmically applied cysteine-reactive reagents to cysteines introduced along the length of the pore-lining first TM (TM1) of a cysteine-less variant of CFTR. We find that methanethiosulfonate (MTS) reagents irreversibly modify cysteines substituted for TM1 residues K95, Q98, P99, and L102 when applied to the cytoplasmic side of open channels. Residues closer to the intracellular end of TM1 (Y84-T94) were not apparently modified by MTS reagents, suggesting that this part of TM1 does not line the pore. None of the internal MTS reagent-reactive cysteines was modified by extracellular [2-(trimethylammonium)ethyl] MTS. Only K95C, closest to the putative intracellular end of TM1, was apparently modified by intracellular [2-sulfonatoethyl] MTS before channel activation. Comparison of these results with recent work on CFTR-TM6 suggests a relative alignment of these two important TMs along the axis of the pore. This alignment was tested experimentally by formation of disulfide bridges between pairs of cysteines introduced into these two TMs. Currents carried by the double mutants K95C/I344C and Q98C/I344C, but not by the corresponding single-site mutants, were inhibited by the oxidizing agent copper(II)-o-phenanthroline. This inhibition was irreversible on washing but could be reversed by the reducing agent dithiothreitol, suggesting disulfide bond formation between the introduced cysteine side chains. These results allow us to develop a model of the relative positions, functional contributions, and alignment of two important TMs lining the CFTR pore. Such functional information is necessary to understand and interpret the three-dimensional structure of the pore.     

Inhibition of cystic fibrosis transmembrane conductance regulator chloride channel currents by arachidonic acid

Chloride permeation through the cystic fibrosis transmembrane conductance regulator (CFTR) Cl- channel is inhibited by a number of different classes of organic anions which are able to enter and block the channel pore from its cytoplasmic end. Here I show, using patch clamp recording from CFTR-transfected baby hamster kidney cell lines, that the cis-unsaturated fatty acid arachidonic acid also inhibits CFTR Cl- currents when applied to the cytoplasmic face of excised membrane patches. This inhibition was of a relatively high affinity compared with other known CFTR inhibitors, with an apparent Kd of 6.5 +/- 0.9 microM. However, in contrast with known CFTR pore blockers, inhibition by arachidonic acid was only very weakly voltage dependent, and was insensitive to the extracellular Cl- concentration. Arachidonic acid-mediated inhibition of CFTR Cl- currents was not abrogated by inhibitors of lipoxygenases, cyclooxygenases or cytochrome P450, suggesting that arachidonic acid itself, rather than some metabolite, directly affects CFTR. Similar inhibition of CFTR Cl- currents was seen with other fatty acids, with the rank order of potency linoleic > or = arachidonic > or = oleic > elaidic > or = palmitic > or = myristic. These results identify fatty acids as novel high affinity modulators of the CFTR Cl- channel.     

State-dependent access of anions to the cystic fibrosis transmembrane conductance regulator chloride channel pore

The cystic fibrosis transmembrane conductance regulator (CFTR) Cl(-) channel is gated by intracellular factors; however, conformational changes in the channel pore associated with channel activation have not been identified. We have used patch clamp recording to investigate the state-dependent accessibility of substituted cysteine residues in the CFTR channel pore to a range of cysteine-reactive reagents applied to the extracellular side of the membrane. Using functional modification of the channel current-voltage relationship as a marker of modification, we find that several positively charged reagents are able to penetrate deeply into the pore from the outside irrespective of whether or not the channels have been activated. In contrast, access of three anionic cysteine-reactive reagents, the methanesulfonate sodium (2-sulfonatoethyl)methanesulfonate, the organic mercurial p-chloromercuriphenylsulfonic acid, and the permeant anion Au(CN)(2)(-), to several different sites in the pore is strictly limited prior to channel activation. This suggests that in nonactivated channels some ion selectivity mechanism exists to exclude anions yet permit cations into the channel pore from the extracellular solution. We suggest that activation of CFTR channels involves a conformational change in the pore that removes a strong selectivity against anion entry from the extracellular solution. We propose further that this conformational change occurs in advance of channel opening, suggesting that multiple distinct closed pore conformations exist.     

Thiocyanate as a probe of the cystic fibrosis transmembrane conductance regulator chloride channel pore

Immediately following exposure to thiocyanate (SCN-)-containing solutions, the cystic fibrosis conductance regulator Cl- channel exhibits high unitary SCN conductance and anomalous mole fraction behaviour, suggesting the presence of multiple anion binding sites within the channel pore. However, under steady-state conditions SCN-conductance is very low. Here I show, using patch clamp recording from CFTR-transfected mammalian cell lines, that under steady-state conditions neither SCN- conductance nor SCN- permeability show anomalous mole fraction behaviour. Instead, SCN conductance, permeability, and block of Cl- permeation can all be reproduced by a rate theory model that assumes only a single intrapore anion binding site. These results suggest that under steady-state conditions the interaction between SCN- and the CFTR channel pore can be understood by a simple model whereby SCN- ions enter the pore more easily than Cl-, and bind within the pore more tightly than Cl-. The implications of these findings for investigating and understanding the mechanism of anion permeation are discussed.     

Molecular determinants of anion selectivity in the cystic fibrosis transmembrane conductance regulator chloride channel pore

Ionic selectivity in many cation channels is achieved over a short region of the pore known as the selectivity filter, the molecular determinants of which have been identified in Ca(2+), Na(+), and K(+) channels. However, a filter controlling selectivity among different anions has not previously been identified in any Cl(-) channel. In fact, because Cl(-) channels are only weakly selective among small anions, and because their selectivity has proved so resistant to site-directed mutagenesis, the very existence of a discrete anion selectivity filter has been called into question. Here we show that mutation of a putative pore-lining phenylalanine residue, F337, in the sixth membrane-spanning region of the cystic fibrosis transmembrane conductance regulator (CFTR) Cl(-) channel, dramatically alters the relative permeabilities of different anions in the channel. Specifically, mutations that reduce the size of the amino acid side chain present at this position virtually abolish the relationship between anion permeability and hydration energy, a relationship that characterizes the anion selectivity not only of wild-type CFTR, but of most classes of Cl(-) channels. These results suggest that the pore of CFTR may indeed contain a specialized region, analogous to the selectivity filter of cation channels, at which discrimination between different permeant anions takes place. Because F337 is adjacent to another amino acid residue, T338, which also affects anion selectivity in CFTR, we suggest that selectivity is predominantly determined over a physically discrete region of the pore located near these important residues.     

Slow conversions among subconductance states of cystic fibrosis transmembrane conductance regulator chloride channel.

The cystic fibrosis transmembrane conductance regulator (CFTR) chloride channel exhibits multiple subconductance states. To study the regulation of conductance states of the CFTR channel, we expressed the wild-type CFTR protein in HEK 293 cells, and isolated microsomal membrane vesicles for reconstitution studies in lipid bilayer membranes. A single CFTR channel had a dominant conductance of 7.8 pS (H), plus two sub-open states with conductances of approximately 6 pS (M) and 2.7 pS (L) in 200 mM KCl with 1 mM MgCl2 (intracellular) and 50 mM KCl with no MgCl2 (extracellular), with pH maintained at 7.4 by 10 mM HEPES-Tris on both sides of the channel. In 200 mM KCl, both H and L states could be measured in stable single-channel recordings, whereas M could not. Spontaneous transitions between H and L were slow; it took 4.5 min for L-->H, and 3.2 min for H-->L. These slow conversions among subconductance states of the CFTR channel were affected by extracellular Mg; in the presence of millimolar Mg, the channel remained stable in the H state. Similar phenomena were also observed with endogenous CFTR channels in T84 cells. In high-salt conditions (1.5 M KCl), all three conductance states of the expressed CFTR channel, 12.1 pS, 8.2 pS, and 3.6 pS, became stable and seemed to gate independently from each other. The existence of multiple stable conductance states associated with the CFTR channel suggests two possibilities: either a single CFTR molecule can exist in multiple configurations with different conductance values, or the CFTR channel may contain multimers of the 170-kDa CFTR protein, and different conductance states are due to different aggregation states of the CFTR protein.     

Dual activity of aminoarylthiazoles on the trafficking and gating defects of the cystic fibrosis transmembrane conductance regulator chloride channel caused by cystic fibrosis mutations

A large fraction of mutations causing cystic fibrosis impair the function of the cystic fibrosis transmembrane conductance regulator (CFTR) chloride channel by causing reduced channel activity (gating defect) and/or impaired exit from the endoplasmic reticulum (trafficking defect). Such defects need to be treated with separate pharmacological compounds termed potentiators and correctors, respectively. Here, we report the characterization of aminoarylthiazoles (AATs) as compounds having dual activity. Cells expressing mutant CFTR were studied with functional assays (fluorescence-based halide transport and short circuit current measurements) to assess the effect of acute and chronic treatment with compounds. We found that AATs are effective on F508del, the most frequent cystic fibrosis mutation, which is associated with both a gating and a trafficking defect. AATs are also effective on mutations like G1349D and G551D, which cause only a gating defect. Evaluation of a panel of AAT analogs identified EN277I as the most effective compound. Incubation of cells expressing mutant CFTR with EN277I caused a strong stimulation of channel activity as demonstrated by single channel recordings. Compounds with dual activity such as AATs may be useful for the development of effective drugs for the treatment of cystic fibrosis.     

Cystic fibrosis with three mutations in the cystic fibrosis transmembrane conductance regulator gene

Summary Three mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene were discovered in a pancreas-insufficient patient with cystic fibrosis (CF) who displayed an uncommon combination of almost normal chloride concentration in sweat tests and typical symptoms of gastrointestinal and pulmonary disease. The R553Q mutation was found on the maternal F508-CFTR gene. Codon 553 is located within a consensus motif of the ATP-binding cassette transport proteins at a less conserved position. Other members of this protein superfamily contain a glutamine instead of arginine at the homologous position, suggesting a modulating rather than disease-causing role of the R553Q mutation in CFTR. The amplification refractory mutation system did not detect the R553Q mutation in a further 65 normal, 113 F508, and 91 non-F508 CF chromosomes. The index case carried the R553X nonsense mutation on the paternal chromosome. The R553X mutation was present on a further 9 out of 86 German nonF508 CF chromosomes linked with the XV2c-KM19Mp6d9-J44-GATT haplotypes 2-2-2-1-1 and 1-1-2-1-2. The location of R553X on separate haplotypes including both alleles of the intragenic GATT repeat suggests an ancient and/or multiple origins of the R553X mutations. The association of the genotype of the CFTR mutation and the clinical phenotype was assessed for the patients carrying the related genotypes F508/F508 (n = 80), F508/R553X (n = 9) and F508-R553Q/R553X (n = 1). In compound heterozygotes, the median chloride concentration in pilocarpine iontophoresis sweat tests was significantly lower than in the F508 homozygotes (P < 0.01). The patient groups were significantly different with respect to the distributions of the centiles for height (P < 0.001) and weight (P < 0.01) as the most sensitive predictors of the course and prognosis in CF. Growth retardation was more pronounced in the compound heterozygotes.     

Voltage-dependent gating of the cystic fibrosis transmembrane conductance regulator Cl- channel

When excised inside-out membrane patches are bathed in symmetrical Cl--rich solutions, the current-voltage (I-V) relationship of macroscopic cystic fibrosis transmembrane conductance regulator (CFTR) Cl- currents inwardly rectifies at large positive voltages. To investigate the mechanism of inward rectification, we studied CFTR Cl- channels in excised inside-out membrane patches from cells expressing wild-type human and murine CFTR using voltage-ramp and -step protocols. Using a voltage-ramp protocol, the magnitude of human CFTR Cl- current at +100 mV was 74 +/- 2% (n = 10) of that at -100 mV. This rectification of macroscopic CFTR Cl- current was reproduced in full by ensemble currents generated by averaging single-channel currents elicited by an identical voltage-ramp protocol. However, using a voltage-step protocol the single-channel current amplitude (i) of human CFTR at +100 mV was 88 +/- 2% (n = 10) of that at -100 mV. Based on these data, we hypothesized that voltage might alter the gating behavior of human CFTR. Using linear three-state kinetic schemes, we demonstrated that voltage has marked effects on channel gating. Membrane depolarization decreased both the duration of bursts and the interburst interval, but increased the duration of gaps within bursts. However, because the voltage dependencies of the different rate constants were in opposite directions, voltage was without large effect on the open probability (Po) of human CFTR. In contrast, the Po of murine CFTR was decreased markedly at positive voltages, suggesting that the rectification of murine CFTR is stronger than that of human CFTR. We conclude that inward rectification of CFTR is caused by a reduction in i and changes in gating kinetics. We suggest that inward rectification is an intrinsic property of the CFTR Cl- channel and not the result of pore block.     

Curcumin stimulates cystic fibrosis transmembrane conductance regulator Cl- channel activity

Compounds that enhance either the function or biosynthetic processing of the cystic fibrosis transmembrane conductance regulator (CFTR) Cl(-) channel may be of value in developing new treatments for cystic fibrosis (CF). Previous studies suggested that the herbal extract curcumin might affect the processing of a common CF mutant, CFTR-DeltaF508. Here, we tested the hypothesis that curcumin influences channel function. Curcumin increased CFTR channel activity in excised, inside-out membrane patches by reducing channel closed time and prolonging the time channels remained open. Stimulation was dose-dependent, reversible, and greater than that observed with genistein, another compound that stimulates CFTR. Curcumin-dependent stimulation required phosphorylated channels and the presence of ATP. We found that curcumin increased the activity of both wild-type and DeltaF508 channels. Adding curcumin also increased Cl(-) transport in differentiated non-CF airway epithelia but not in CF epithelia. These results suggest that curcumin may directly stimulate CFTR Cl(-) channels.     

Rectification of cystic fibrosis transmembrane conductance regulator chloride channel mediated by extracellular divalent cations.

We report here distinct rectification of the cystic fibrosis transmembrane conductance regulator (CFTR) chloride channel reconstituted in lipid bilayer membranes. Under the symmetrical ionic condition of 200 mM KCl (with 1 mM MgCl2 in cis intracellular and 0 MgCl2 in trans extracellular solutions, pH in both solutions buffered at 7.4 with 10 mM HEPES), the inward currents (intracellular-->extracellular chloride movement) through a single CFTR channel were approximately 20% larger than the outward currents. This inward rectification of the CFTR channel was mediated by extracellular divalent cations, as the linear current-voltage relationship of the channel could be restored through the addition of millimolar concentrations of MgCl2 or CaCl2 to the trans solution. The dose responses for [Mg]zero and [Ca]zero had half-dissociation constants of 152 +/- 72 microM and 172 +/- 40 microM, respectively. Changing the pH buffer from HEPES to N-tris-(hydroxymethyl)methyl-2-aminoethanesulfonic acid did not alter rectification of the CFTR channel. The nonlinear conductance property of the CFTR channel seemed to be due to negative surface charges on the CFTR protein, because in pure neutral phospholipid bilayers, clear rectification of the channel was also observed when the extracellular solution did not contain divalent cations. The CFTR protein contains clusters of negatively charged amino acids on several extracellular loops joining the transmembrane segments, which could constitute the putative binding sites for Ca and Mg.     

Effect of ATP-sensitive K+ channel regulators on cystic fibrosis transmembrane conductance regulator chloride currents.

The cystic fibrosis transmembrane conductance regulator (CFTR) is a Cl- channel that is regulated by cAMP-dependent phosphorylation and by intracellular ATP. Intracellular ATP also regulates a class of K+ channels that have a distinct pharmacology: they are inhibited by sulfonylureas and activated by a novel class of drugs called K+ channel openers. In search of modulators of CFTR Cl- channels, we examined the effect of sulfonylureas and K+ channel openers on CFTR Cl- currents in cells expressing recombinant CFTR. The sulfonylureas, tolbutamide and glibenclamide, inhibited whole-cell CFTR Cl- currents at half-maximal concentrations of approximately 150 and 20 microM, respectively. Inhibition by both agents showed little voltage dependence and developed slowly; > 90% inhibition occurred 3 min after adding 1 mM tolbutamide or 100 microM glibenclamide. The effect of tolbutamide was reversible, while that of glibenclamide was not. In contrast to their activating effect on K+ channels, the K+ channel openers, diazoxide, BRL 38227, and minoxidil sulfate inhibited CFTR Cl- currents. Half-maximal inhibition was observed at approximately 250 microM diazoxide, 50 microM BRL 38227, and 40 microM minoxidil sulfate. The rank order of potency for inhibition of CFTR Cl- currents was: glibenclamide < BRL 38227 approximately equal to minoxidil sulfate > tolbutamide > diazoxide. Site-directed mutations of CFTR in the first membrane-spanning domain and second nucleotide-binding domain did not affect glibenclamide inhibition of CFTR Cl- currents. However, when part of the R domain was deleted, glibenclamide inhibition showed significant voltage dependence. These agents, especially glibenclamide, which was the most potent, may be of value in identifying CFTR Cl- channels. They or related analogues might also prove to be of value in treating diseases such as diarrhea, which may involve increased activity of the CFTR Cl- channel.     

Novel regulation of cystic fibrosis transmembrane conductance regulator (CFTR) channel gating by external chloride

The cystic fibrosis transmembrane conductance regulator (CFTR) is vital for Cl(-) and HCO(3)(-) transport in many epithelia. As the HCO(3)(-) concentration in epithelial secretions varies and can reach as high as 140 mm, the lumen-facing domains of CFTR are exposed to large reciprocal variations in Cl(-) and HCO(3)(-) levels. We have investigated whether changes in the extracellular anionic environment affects the activity of CFTR using the patch clamp technique. In fast whole cell current recordings, the replacement of 100 mm external Cl(-) ((Cl(o)(-))) with HCO(3)(-), Br(-), NO(3)(-), or aspartate(-) inhibited inward CFTR current (Cl(-) efflux) by approximately 50% in a reversible manner. Lowering Cl(o)(-) alone by iso-osmotic replacement with mannitol also reduced Cl(-) efflux to a similar extent. The maximal inhibition of CFTR current was approximately 70%. Raising cytosolic calcium shifted the Cl(-) dose-inhibition curve to the left but did not alter the maximal current inhibition observed. In contrast, a reduction in the internal [Cl(-)] neither inhibited CFTR nor altered the block caused by reduced Cl(o)(-). Single channel recordings from outside-out patches showed that lowering Cl(o)(-) markedly reduced channel open probability with little effect on unitary conductance. Together, these results indicate that alterations in Cl(o)(-) alone and not the Cl(-)/HCO(3)(-) ratio regulate the gating of CFTR. Physiologically, our data have implications for current models of epithelial HCO(3)(-) secretion and for the control of pH at epithelial cell surfaces.     

Asymmetric structure of the cystic fibrosis transmembrane conductance regulator chloride channel pore suggested by mutagenesis of the twelfth transmembrane region

The cystic fibrosis transmembrane conductance regulator (CFTR) Cl(-) channel contains 12 membrane-spanning regions which are presumed to form the transmembrane pore. Although a number of findings have suggested that the sixth transmembrane region plays a key role in forming the pore and determining its functional properties, the role of other transmembrane regions is currently not well established. Here we assess the functional importance of the twelfth transmembrane region, which occupies a homologous position in the carboxy terminal half of the CFTR molecule to that of the sixth transmembrane region in the amino terminal half. Five residues in potentially important regions of the twelfth transmembrane region were mutated individually to alanines, and the function of the mutant channels was examined using patch clamp recording following expression in mammalian cell lines. Three of the five mutations significantly weakened block of unitary Cl(-) currents by SCN(-), implying a partial disruption of anion binding within the pore. Two of these mutations also caused a large reduction in the steady-state channel mean open probability, suggesting a role for the twelfth transmembrane region in channel gating. However, in direct contrast to analogous mutations in the sixth transmembrane region, all mutants studied here had negligible effects on the anion selectivity and unitary Cl(-) conductance of the channel. The relatively minor effects of these five mutations on channel permeation properties suggests that, despite their symmetrical positions within the CFTR protein, the sixth and twelfth transmembrane regions make highly asymmetric contributions to the functional properties of the pore.     

Glycosylation and the cystic fibrosis transmembrane conductance regulator

The cystic fibrosis transmembrane conductance regulator (CFTR) has been known for the past 11 years to be a membrane glycoprotein with chloride channel activity. Only recently has the glycosylation of CFTR been examined in detail, by O'Riordan et al in Glycobiology. Using cells that overexpress wild-type (wt)CFTR, the presence of polylactosamine was noted on the fully glycosylated form of CFTR. In the present commentary the results of that work are discussed in relation to the glycosylation phenotype of cystic fibrosis (CF), and the cellular localization and processing of ΔF508 CFTR. The significance of the glycosylation will be known when endogenous CFTR from primary human tissue is examined.     

Arg352 is a major determinant of charge selectivity in the cystic fibrosis transmembrane conductance regulator chloride channel

The cystic fibrosis transmembrane conductance regulator forms an anion-selective channel. We previously showed that charge selectivity, the ability to discriminate between anions and cations, occurs near the cytoplasmic end of the channel. The molecular determinants of charge selectivity, however, are unknown. We investigated the role of Arg352, a residue flanking the predicted cytoplasmic end of the M6 segment, in the mechanism of charge selectivity. We determined the Cl- to Na+ permeability ratio (PCl/PNa) from the reversal potential measured in a 10-fold NaCl gradient. For the wild type, PCl/PNa was 36 (range of 28-51). For the R352H mutant, PCl/PNa was dependent on cytoplasmic pH. At pH 5.4, the PCl/PNa was 33 (range of 27-41), similar to that of the wild type, but at pH 7.2, where the histidine should be largely uncharged, PCl/PNa was 3 (range of 2.9-3.1). For the R352C and R352Q mutants, PCl/PNa was 7 (range of 6-8) and 4 (range of 3.5-4.4), respectively. Furthermore, Na+ which does not carry a significant fraction of the current through the wild type is measurably conducted through R352Q. Thus, the charge of the side chain at position 352 is a strong determinant of charge selectivity. In the wild type, the positive charge on Arg352 contributes to an electrostatic potential in the channel that forms a barrier to cation permeation. Mutation of Arg352 did not alter the halide selectivity sequence. Selectivity among halides must involve other residues.     

Direct comparison of the functional roles played by different transmembrane regions in the cystic fibrosis transmembrane conductance regulator chloride channel pore

The cystic fibrosis transmembrane conductance regulator (CFTR) Cl(-) channel contains 12 transmembrane (TM) regions that are presumed to form the channel pore. However, little is known about the relative functional contribution of different TM regions to the pore. We have used patch clamp recording to investigate the functional consequences of point mutations throughout the six transmembrane regions in the N-terminal part of the CFTR protein (TM1-TM6). A range of specific functional assays compared the single channel conductance, anion binding, and anion selectivity properties of different channel variants. Overall, our results suggest that TM1 and -6 play dominant roles in forming the channel pore and determining its functional properties, with TM5 perhaps playing a lesser role. In contrast, TM2, -3, and -4 appear to play only minor supporting roles. These results define transmembrane regions 1 and 6 as major contributors to the CFTR channel pore and have strong implications for emerging structural models of CFTR and related ATP-binding cassette proteins.     

Domain-domain associations in cystic fibrosis transmembrane conductance regulator

Cystic fibrosis is caused by mutations inthe cystic fibrosis transmembrane conductance regulator (CFTR) gene.CFTR is a chloride channel whose activity requires protein kinaseA-dependent phosphorylation of an intracellular regulatory domain(R-domain) and ATP hydrolysis at the nucleotide-binding domains (NBDs).To identify potential sites of domain-domain interaction within CFTR,we expressed, purified, and refolded histidine (His)- andglutathione-S-transferase (GST)-tagged cytoplasmic domainsof CFTR. ATP-binding to his-NBD1 and his-NBD2 was demonstrated bymeasuring tryptophan fluorescence quenching. Trypticdigestion of in vitro phosphorylated his-NBD1-R and in situphosphorylated CFTR generated the same phosphopeptides. An interactionbetween NBD1-R and NBD2 was assayed by tryptophan fluorescencequenching. Binding among all pairwise combinations of R-domain, NBD1,and NBD2 was demonstrated with an overlay assay. To identifyspecific sites of interaction between domains of CFTR, an overlay assaywas used to probe an overlapping peptide library spanning allintracellular regions of CFTR with his-NBD1, his-NBD2, andGST-R-domain. By mapping peptides from NBD1 and NBD2 that bound toother intracellular domains onto crystal structures for HisP, MalK, andRad50, probable sites of interaction between NBD1 and NBD2 wereidentified. Our data support a model where NBDs form dimers with theATP-binding sites at the domain-domain interface.