2026-09-16RELRCSBINITIAL_RELEASEREPOSITORYINITIAL_RELEASEREPOSITORYRCSBRCSB2025-09-032026-09-162026-09-162026-09-16Cystic Fibrosis FoundationHUNT13XX0, HUNT18G0, HUNT20G0, 004400G222-Hunt, 007997G224-Hunt, FRANK16XX0, FRANK18G0United StatesV-shaped (channel-formed), ATP-bound, VX809-bound conformation of wild-type human CFTR (composite map from PHENIX based on consensus and local refinement maps from cryoSPARC)Hunt JFPaige ASBaranwal JCohen BMGoldberg PMWang CLoughlin BJKappes JCYang ZJiang FGovaerts COvertus MRich Zcystic fibrosis, CFTR, nanobody, protein folding, MEMBRANE PROTEINHunt JFPaige ASBaranwal JCohen BMGoldberg PMWang CLoughlin BJKappes JCYang ZJiang FGovaerts COvertus MRich ZNanobody-Driven Stabilization Synergistically Rescues F508del-CFTR
and Reveals an Alternative Active State of the ChannelTo Be Published0353EMD-72496consensus EM volumeConsensusEMD-72495focused EM volumeCORE1EMD-72493focused EM volumeCORE2EMD-72356focused EM volumeAHD1EMD-72492focused EM volumeAHD2EMD-72494focused EM volumeWalkerB1EMD-72491focused EM volumeWalkerB2EMD-77407focused EM volumeNBD1EMD-77409focused EM volumeNBD29y4tFULLOVERLAPEMDBEMD-72496consensus EM volumeConsensusEMDBEMD-72495focused EM volumeCORE1EMDBEMD-72493focused EM volumeCORE2EMDBEMD-72356focused EM volumeAHD1EMDBEMD-72492focused EM volumeAHD2EMDBEMD-72494focused EM volumeWalkerB1EMDBEMD-72491focused EM volumeWalkerB2EMDBEMD-72497associated EM volumeV-shaped (channel-formed), ATP-bound, VX809-bound conformation of wild-type human CFTR (composite map from PHENIX based on consensus and local refinement maps from cryoSPARC)EMDBEMD-77407focused EM volumeNBD1EMDBEMD-77409focused EM volumeNBD2Wild-type human CFTR solubilized in digitonin and cholesterol-hemisuccinate in the presence of 2 mM MgATPWild-type human CFTR solubilized in digitonin and cholesterol-hemisuccinate in the presence of 2 mM MgATP010.1725Wild type human Cystic Fibrosis Transmembrane Conductance Regulator (hCFTR)11Homo sapiensCystic fibrosis transmembrane conductance regulatorHomo sapiens0.17273498399999998The sequence assignment for residues 637-640, which was guided by an AlphaFold3 model, is tentative. The lipid model is based on inspection of density features exceeding the size of a water molecule at a 6.0 sigma contour level, which were modeled using the map at approximately a 4.5 sigma contour level. All those chemical and chemical fragment assignments are hypotheses guided by inspection of the density and the chemical environment, and none of them are supported by definitive evidence. The residue numbers of the modeled lipids are coded to indicate the confidence level of the chemical assignment and the extent of conservation between models for different hCFTR conformations. Residue numbers in the 4000, 5000, and 6000 series correspond to higher, lower, and lowest levels of confidence in the chemical assignments. Residue numbers in each of those series ending 1-299 designate sites at which binding of an equivalent lipid species is conserved in the model for at least one additional hCFTR conformation, while those ending 301-599 designate sites at which a different lipid species is bound in at least one additional hCFTR conformation. Residue numbers ending 601-899 designate sites not occupied by a lipid species in any other hCFTR conformations structurally characterized to date. Digitonin molecule L5001 is bound in a site at the C-terminus of NBD1 previously demonstrated to interact with hydrophobic drugs in a solution NMR study of the isolated domain (Hudson et al., 2017, Mol. Pharmacol. 92:124). This tentative assignment is supported by the observation of convincing density for digitonin at the same site in NBD-associated hCFTR structures, although the density in this structure, which exceeds the size of a water molecule at 6.5 sigma contour level, only approximately matches digitonin. All density features exceeding the size of a water molecule at a 6.0 sigma contour level have been modeled except for thirteen features proximal to the Alpha Helical Subdomains in NBD1 and NBD2 and the C-terminus of NBD1; these features exceed the size of a water molecule at contour levels from 6.0-8.5 sigma. The features proximal to the C-terminus of NBD1 likely represents residual density from partially ordered conformations of residues 641-668 at the N-terminus of the R Region. This inference is based on the observation that filtering the map to low resolution shows continuous density at this site that matches the shape of an alpha-helical hairpin formed by this protein segment in AlphaFold3 models. The features proximal to the Alpha Helical Subdomains in NBD1 and NBD2 seem likely to represent a combination of residual density from other disordered segments of hCFTR and noise from the local refinement algorithm used to generate the final composite map used for modeling.1Cricetulus griseusLEVOMQRSPLEKASVVSKLFFSWTRPILRKGYRQRLELSDIYQIPSVDSADNLSEKLEREWDRELASKKNPKLINALRRCFFWR
FMFYGIFLYLGEVTKAVQPLLLGRIIASYDPDNKEERSIAIYLGIGLCLLFIVRTLLLHPAIFGLHHIGMQMRIAMFSLI
YKKTLKLSSRVLDKISIGQLVSLLSNNLNKFDEGLALAHFVWIAPLQVALLMGLIWELLQASAFCGLGFLIVLALFQAGL
GRMMMKYRDQRAGKISERLVITSEMIENIQSVKAYCWEEAMEKMIENLRQTELKLTRKAAYVRYFNSSAFFFSGFFVVFL
SVLPYALIKGIILRKIFTTISFCIVLRMAVTRQFPWAVQTWYDSLGAINKIQDFLQKQEYKTLEYNLTTTEVVMENVTAF
WEEGFGELFEKAKQNNNNRKTSNGDDSLFFSNFSLLGTPVLKDINFKIERGQLLAVAGSTGAGKTSLLMMIMGELEPSEG
KIKHSGRISFCSQFSWIMPGTIKENIIFGVSYDEYRYRSVIKACQLEEDISKFAEKDNIVLGEGGITLSGGQRARISLAR
AVYKDADLYLLDSPFGYLDVLTEKEIFESCVCKLMANKTRILVTSKMEHLKKADKILILHEGSSYFYGTFSELQNLQPDF
SSKLMGCDSFDQFSAERRNSILTETLHRFSLEGDAPVSWTETKKQSFKQTGEFGEKRKNSILNPINSIRKFSIVQKTPLQ
MNGIEEDSDEPLERRLSLVPDSEQGEAILPRISVISTGPTLQARRRQSVLNLMTHSVNQGQNIHRKTTASTRKVSLAPQA
NLTELDIYSRRLSQETGLEISEEINEEDLKECFFDDMESIPAVTTWNTYLRYITVHKSLIFVLIWCLVIFLAEVAASLVV
LWLLGNTPLQDKGNSTHSRNNSYAVIITSTSSYYVFYIYVGVADTLLAMGFFRGLPLVHTLITVSKILHHKMLHSVLQAP
MSTLNTLKAGGILNRFSKDIAILDDLLPLTIFDFIQLLLIVIGAIAVVAVLQPYIFVATVPVIVAFIMLRAYFLQTSQQL
KQLESEGRSPIFTHLVTSLKGLWTLRAFGRQPYFETLFHKALNLHTANWFLYLSTLRWFQMRIEMIFVIFFIAVTFISIL
TTGEGEGRVGIILTLAMNIMSTLQWAVNSSIDVDSLMRSVSRVFKFIDMPTEGKPTKSTKPYKNGQLSKVMIIENSHVKK
DDIWPSGGQMTVKDLTAKYTEGGNAILENISFSISPGQRVGLLGRTGSGKSTLLSAFLRLLNTEGEIQIDGVSWDSITLQ
QWRKAFGVIPQKVFIFSGTFRKNLDPYEQWSDQEIWKVADEVGLRSVIEQFPGKLDFVLVDGGCVLSHGHKQLMCLARSV
LSKAKILLLDEPSAHLDPVTYQIIRRTLKQAFADCTVILCEHRIEAMLECQQFLVIEENKVRQYDSIQKLLNERSLFRQA
ISPSDRVKLFPHRNSSKCKSKPQIAALKEETEEEVQDTRLLEENLYFQGGGGSGGSWSHPQFEKAAAGGGSGGGSWSHPQ
FEKP135695.6.1.6DigitoninHomo sapiens0.0012293119999999998The lipid model is based on inspection of density features in the transmembrane region exceeding the size of a water molecule at a 5.5 sigma contour level, which were modeled using the map at a 4.0 sigma contour level. The chemical and chemical fragment assignments at these sites are hypotheses guided by inspection of the density and the chemical environment. These assignments are not supported by any other experimental evidence, and they should all be considered placeholders for significant density features. Modeled lipids and lipid fragments were given residue numbers encoding to the confidence level of the chemical assignment (5000 series for high, 6000 series for medium, and 7000 series for low). A putative digitonin molecule modeled as residue 8001 is located in an extramembranous cavity on the surface of NBD1 that was previously shown by NMR to interact with hydrophobic drugs (https://pubmed.ncbi.nlm.nih.gov/28546419/). This prominent density feature was modeled because a similar feature is generally observed at the same site in maps for other conformations of human CFTR that have well-defined density for NBD1. Five smaller density features proximal to the surfaces of the NBDs that exceeded the 5.5 sigma threshold were not modeled because similar features are not observed in other conformations, and they likely represent noise in the local refinements of the subdomains of the NBDs that were used to generate the composite map used for modeling. A density feature proximal to the protein surface modeled as peptide chain P seems likely to represent a segment of the disordered glycosylated extracellular surface loop spanning residues 887-910.2AJPCHOLESTEROL0.00038665410CLRO-[(R)-{[(2R)-2,3-bis(octadecanoyloxy)propyl]oxy}(hydroxy)phosphoryl]-L-serine0.0007920752P5SPHOSPHATIDYLETHANOLAMINE0.00073403899999999992PTYOLEIC ACID0.0002824611OLAPENTADECANE0.000212414999999999981MYSPALMITIC ACID0.000256423999999999941PLMDODECANE0.0001703352D121-PALMITOYL-2-LINOLEOYL-SN-GLYCERO-3-PHOSPHOCHOLINE0.000758061CPLN-OCTANE0.0001142291OCTHEXANE8.6175e-051HEXADENOSINE-5'-TRIPHOSPHATE0.0005071812ATPMAGNESIUM ION2.4305e-052MGLumacaftor0.000452406999999999941VX8singleParticleparticle1.57.5200.0NaClSodium Chloride3.0MgCl2Magnesium Chloride50.0C4H11NO3Tris buffer0.06C56H92O29Digitonin2.0C10H16N5O13P3ATPQuantifoil R0.6/1GOLD300GOLDHOLEY50.0PLASMA CLEANINGOTHERThe grid was treated in a Solarus Plasma Cleaner 950 (Gatan Inc., USA) for 25 sec with O2/H2 flow-rates of 27.5/6.4 sccm and 15 W cleaning power.ETHANE100277FEI VITROBOT MARK IVTFS KRIOSFLOOD BEAMBRIGHT FIELDFIELD EMISSION GUN300100.02.70.22.9105000.0FEI TITAN KRIOS AUTOGRID HOLDERNITROGEN85.090.0GIF Quantum ER20GATAN K3 BIOQUANTUM (6k x 4k)576040921111102.558.0Movies comprised 40 frames collected in 2.5 seconds.12493670Particle count after low-resolution template picking, which was followed by light 2D cleaning. That process, which included all 2D classes roughly resembling views of CFTR, yielded 702,919 particles that were input into iterative heterorefinement against a set of protein classes and 6 decoy volumes, resulting in 252,534 particles (42%) being assigned to the final set of protein classes.cryoSPARC4.3.1Patch CTF correctionPHASE FLIPPING AND AMPLITUDE CORRECTION1BACK PROJECTION3.14FSC 0.143 CUT-OFFcryoSPARC4.7.081674MAXIMUM LIKELIHOODcryoSPARC4.3.1MAXIMUM LIKELIHOODcryoSPARC4.3.1463346.0cryoSPARC4.7.0Real space refinement in PHENIX v 2.1_6048emd_72497_half_map_2.map.gz1IMAGE STORED AS FLOATING POINT NUMBER (4 BYTES)
320
320320
0
00320320320265.6265.6265.690.090.090.0XYZ-9.45148400000000121.710995-0.034388730.63318420.830000040.830000040.83000004AUTHORemd_72497_half_map_1.map.gz1IMAGE STORED AS FLOATING POINT NUMBER (4 BYTES)