Maity S , Mazzolini M , Arcangeletti M , Valbuena A , Fabris P , Lazzarino M , Torre V .

	Figure 1. F–D curves from CNGA1 and CNGA1-N2B-HisTag in the closed state.(a) An example of an F–D curve. (b) Transformation of the F–D curve in a into an (F,Lc) plot; if the F–D curve fits well in a piecewise manner using the WLC model, the resulting (F,Lc) plot is composed of a series of almost vertical segments located at the corresponding value of Lc and the upper value of each of these segments corresponds to the value of the force peak F. (c–e) Three different coding schemes of increasing complexity that are all based on the processing of the (F,Lc) plot in b; coding scheme I (c) considers only the location of the force peak (the corresponding value of Lc) and is a binary coding; coding II (d) considers the location and the amplitude of the force peak and is not a binary coding; coding III (e) fits all sample values of F above 30 pN. Red dots represent the final coding of the F–D curves. (f) 22 F–D curves from Cluster 1-CS of CNGA1 (red) detected using bioinformatics analysis and 45 F–D curves from the construct CNGA1-N2B-HisTag (black) in the closed state. The construct with N2B has an initial flat region of 85 nm, followed by force peaks matching the unfolding events observed with the CNGA1 construct. (g) Superimposition of 157 F–D curves obtained from injected oocytes showing the peak force location of CNGA1. The 23F–D curves (violet) end with a force peak with an Lc of ∼234 nm; 134 F–D curves (red) have an additional force peak with an Lc of ∼276 nm. (h) Superimposition of histograms of normalized counts/bin against Lc from the 157 F–D curves of b (now all in red) and 45 F–D curves from the CNGA1-N2B-HisTag construct (black). (i) Histogram of Lc values of force peak (with Gaussian fit for the different peaks) from the F–D curves in g with five peaks located at 96±3 (mean±s.d., n=117), 116±3 (mean±s.d., n=146), 159±3 (mean±s.d., n=157), 189±5 (mean±s.d., n=157), 234±6 (mean±s.d., n=157) nm and the detachment with the probability (P) of 0.75, 0.93, 0.95, 0.98 and 0.97, respectively.
	Figure 2. Unfolding of CNGA1 and CNGA1-N2B-HisTag in the open state.(a) Example of F–D curves obtained from the unfolding of single CNGA1 (blue and cyan) and CNGA1-N2B-HisTag (green) constructs in the open state (19 curves for cluster 1-OS, 8 curves for cluster 2-OS and 1 curve for CNGA1-N2B-HisTag). The construct with N2B has an initial flat region of 85 nm, followed by peaks that correspond to the unfolding events that were observed in the CNGA1 construct. (b) Superimposition of 132 F–D curves that were obtained from injected oocytes in the open state using the curves of Cluster 1-OS as a template. (c) Superimposition of the histograms of normalized counts/bin against Lc from the F–D curves of b (blue) and 32 F–D curves from the CNGA1-N2B-HisTag (green) constructs in the open state. (d) Histogram of Lc values of force peak (with Gaussian fit for the different peaks) from the F–D curves in d with eight peaks located at 54±3 (mean±s.d., n=41), 84±3 (mean±s.d., n=86), 96±3 (mean±s.d., n=59), 116±3 (mean±s.d., n=114), 144±3 (mean±s.d., n=132), 171±3 (mean±s.d., n=132), 189±5 (mean±s.d., n=132), 234±6 (mean±s.d., n=132) nm and the detachment; the probabilities of the unfolding are 0.31, 0.65, 0.45, 0.86, 0.96, 0.95, 0.98 and 0.86, respectively.
	Figure 3. Unfolding of CNB domain.(a) Schematic representation of the unfolding sequence, with the expected Lc values and the relative a.a. length in parenthesis and single module of I27 (cyan triangle), CNB domain (green rectangle), C-linker (green oval), transmembrane domains (cyan rectangles) and mutation P366C (red dots). (b) (F,Lc) plot from a typical curve of P366C-(I27)2-HisTag. The finger print is composed by two force peaks of ∼200 pN separated by ∼28 nm with the detachment peak, with a value of Lc corresponding to unfolding of the cytoplasmic domain from D690 to P366 and the two I27 modules (that is, (D690−P366)·0.4 nm+2·28 nm))=186 nm. (c) Superimposition of the F–D curves of CNGA1 (red), P366C-HisTag (cyan) and P366C-(I27)2-HisTag (black). The F–D curves from the construct P366C-HisTag are shorter than those from CNGA1 and have a detachment event at 120–140 nm. Force peaks with Lc values at 96 and 116 nm are present in the F–D curves from the CNGA1, P366C-HisTag and P366C-(I27)2-HisTag constructs. (d) Different unfolding pathways in the open and closed states; I and II: superposition of the F–D curves from CNGA1 channels (red) and from the construct P366C-(I27)2-HisTag (black) in the closed state. In I, there are two force peaks with Lc values of 96 and 116 nm, but there is only one force peak in II; III and IV: superposition of the F–D curves from CNGA1 channels (blue) and from the constructs P366C-(I27)2-HisTag (green) in the open state. In III there are three force peaks with Lc values of 54, 84 and 116 nm, and in IV there are three force peaks with Lc values of 84, 96 and 118 nm. The force peak with an Lc of 158 is only present in the construct P366C-(I27)2-HisTag. (e) Superimposition of histograms of normalized counts/bin against Lc from the F–D curves obtained from P366C-(I27)2-HisTag (black) and from CNGA1 channels (red) in the closed state (f) same as in e for P366C-(I27)2-HisTag (green) and from CNGA1 channels (blue) in the open state. The data in e,f are from CNGA1 channels as described in Figs 1 and 2.
	Figure 4. Assignment of secondary structure and different domains of CNGA1 channels to the observed force peaks in the closed and open states.The first and second columns show the key features of the functional domains identified in CNGA1 channels according to the available literature2324252627282930; the third column reports the presumed secondary structure1731; the fourth and fifth columns contain the corresponding polypeptide segments and the associated a.a. number; the remaining columns indicate the Lc, ΔLc and force values of the corresponding force peaks in the closed and open states. The length and the polypeptide segments are obtained from the results of Figs 1 and 2. The number of a.a. for the transmembrane domains has been calculated by considering the membrane thickness (∼5 nm or ∼13 a.a.)344. The initial and final a.a. of each polypeptide segment are only indicative (a) average Lc values of the force peaks with s.d. in nanometres, with the number inside the brackets representing the average position of the force peaks in the a.a. sequence. (b) Average ΔLc in nanometres and the corresponding number of a.a. (in brackets). (c) Average value of the unfolding force with the corresponding s.d. in pN. *The unfolding of the cytoplasmic segments in the C-terminal cannot be exactly determined.
	Figure 5. Conformational changes in the transmembrane domain during the gating of CNGA1 channels.(a) Superposition of two representative F–D curves of CNGA1 channels in the closed (red) and open (blue) states. Continuous black lines obtained from the fitting with WLC model. Numbers indicate the corresponding values of Lc. (b) Schematic representation of hypothesized interactions between the transmembrane helices in the closed and open states. Red dots indicate the approximate location of the force peaks and the numbers of the corresponding a.a. (c) Electrophysiological recordings from mutant channel P293A in the presence of Na+ (black) and ethylammonium (EA+) (green) at ±60 mV. In the black box, the homology model of the S4 and S6 transmembrane domains of the CNGA1 channel based on the molecular structure of the Kv1.2 channel; the conserved P293 is indicated in green. (d) Superposition of two representative F–D curves for the CNGA1 channel (red) and for the P293A (green), both in the closed state. The green arrow, corresponding to the value of Lc=159 nm, indicates the differences between the two constructs. (e) Superimposition of histograms of normalized counts/bin against Lc from the 157 F–D curves of CNGA1 (red, the same as in Fig. 1h) and 21 F–D curves from mutant channel P293A in the closed state (green), both in the closed state. Arrow as in d.
	Figure 6. Variability of the unfolding of the transmembrane segments in the closed state.(a) A representative F–D curve in the closed state (the same as in Fig. 5a). Continuous black lines obtained from the fitting with WLC model. Horizontal lines indicate the corresponding values of ΔLc and ΣΔLc. (b) Histogram of ΔLcclosed(3/4) (bin 2 nm) in the closed state between the force peaks with Lc values around 159 to 189 nm. Colours correspond to different clusters of F–D curves with values of ΔLcclosed(3/4) around 26, 34 and 50 nm. (c) Superposition of representative F–D curves of the three clusters of b and the relative number of stretched a.a. (d–f) Schematic representation of the corresponding unfolding pathways. Small red dots or circles indicate the position of the force peaks, and the number is the corresponding value of Lc. The a.a. that presumably corresponds to the force peak is also indicated. (g) ΣΔLcclosed(TM) histogram (bin 1 nm) in the closed state for the force peaks as in a with values around 113, 123 and 131 nm. (h) Superposition of representative F–D curves of the three clusters of g and the relative number of stretched a.a.
	Figure 7. Variability of the unfolding of the transmembrane segments in the open state.(a) Histogram of ΔLcopen(3/4) (bin 2 nm) in the open state. (b) Superposition of representative F–D curves of the two clusters of a and the related number of stretched a.a. (c,d) Schematic representation of the corresponding unfolding pathways. Small red dots or circles indicate the position of the force peaks and the number is the corresponding value of Lc. The a.a. that presumably corresponds to the force peak is also indicated. (e) ΣΔLcopen(TM) histogram (bin 1 nm) in the open state. (f) Superposition of representative F–D curves of the two clusters of e and the related number of stretched a.a.

Anderson, Phylogeny of ion channels: clues to structure and function. 2001, Pubmed

Anderson, Phylogeny of ion channels: clues to structure and function. 2001, Pubmed
Andreopoulos, Efficient unfolding pattern recognition in single molecule force spectroscopy data. 2011, Pubmed
Anselmi, Origin of functional diversity among tetrameric voltage-gated channels. 2007, Pubmed
Becchetti, Cyclic nucleotide-gated channels. Pore topology studied through the accessibility of reporter cysteines. 1999, Pubmed , Xenbase
Bosshart, Reference-free alignment and sorting of single-molecule force spectroscopy data. 2012, Pubmed
Bosshart, The transmembrane protein KpOmpA anchoring the outer membrane of Klebsiella pneumoniae unfolds and refolds in response to tensile load. 2012, Pubmed
Carrion-Vazquez, Mechanical and chemical unfolding of a single protein: a comparison. 1999, Pubmed
Clayton, Structure of the transmembrane regions of a bacterial cyclic nucleotide-regulated channel. 2008, Pubmed
Contreras, Gating at the selectivity filter in cyclic nucleotide-gated channels. 2008, Pubmed
Craven, CNG and HCN channels: two peas, one pod. 2006, Pubmed
Derebe, Structural studies of ion permeation and Ca2+ blockage of a bacterial channel mimicking the cyclic nucleotide-gated channel pore. 2011, Pubmed
Dietz, Protein structure by mechanical triangulation. 2006, Pubmed
Dietz, Exploring the energy landscape of GFP by single-molecule mechanical experiments. 2004, Pubmed
Engel, Structure and mechanics of membrane proteins. 2008, Pubmed
Flynn, A cysteine scan of the inner vestibule of cyclic nucleotide-gated channels reveals architecture and rearrangement of the pore. 2003, Pubmed
Flynn, Conformational changes in S6 coupled to the opening of cyclic nucleotide-gated channels. 2001, Pubmed , Xenbase
Ge, Locating an extracellular K+-dependent interaction site that modulates betaine-binding of the Na+-coupled betaine symporter BetP. 2011, Pubmed
Gordon, Direct interaction between amino- and carboxyl-terminal domains of cyclic nucleotide-gated channels. 1997, Pubmed
Gorostiza, Molecular handles for the mechanical manipulation of single-membrane proteins in living cells. 2005, Pubmed , Xenbase
Grandbois, How strong is a covalent bond? 1999, Pubmed
Higgins, Molecular architecture of a retinal cGMP-gated channel: the arrangement of the cytoplasmic domains. 2002, Pubmed
Hoffmann, Single molecule force spectroscopy using polyproteins. 2012, Pubmed
Jan, A superfamily of ion channels. 1990, Pubmed
Kaupp, Cyclic nucleotide-gated ion channels. 2002, Pubmed
Kaupp, Primary structure and functional expression from complementary DNA of the rod photoreceptor cyclic GMP-gated channel. 1989, Pubmed , Xenbase
Kawamura, Kinetic, energetic, and mechanical differences between dark-state rhodopsin and opsin. 2013, Pubmed
Kedrov, Deciphering molecular interactions of native membrane proteins by single-molecule force spectroscopy. 2007, Pubmed
Kedrov, Differentiating ligand and inhibitor interactions of a single antiporter. 2006, Pubmed
Kotamarthi, Multiple unfolding pathways of leucine binding protein (LBP) probed by single-molecule force spectroscopy (SMFS). 2013, Pubmed
Kowal, Ligand-induced structural changes in the cyclic nucleotide-modulated potassium channel MloK1. 2014, Pubmed
Kusch, Role of the S4-S5 linker in CNG channel activation. 2010, Pubmed
Li, Reverse engineering of the giant muscle protein titin. 2002, Pubmed
Li, Atomic force microscopy reveals the mechanical design of a modular protein. 2000, Pubmed
Liu, Constraining ligand-binding site stoichiometry suggests that a cyclic nucleotide-gated channel is composed of two functional dimers. 1998, Pubmed
Lolicato, Tetramerization dynamics of C-terminal domain underlies isoform-specific cAMP gating in hyperpolarization-activated cyclic nucleotide-gated channels. 2011, Pubmed , Xenbase
Maity, Corrigendum: Conformational rearrangements in the transmembrane domain of CNGA1 channels revealed by single-molecule force spectroscopy. 2017, Pubmed
Marchesi, Gating of cyclic nucleotide-gated channels is voltage dependent. 2012, Pubmed
Mari, Gating of the MlotiK1 potassium channel involves large rearrangements of the cyclic nucleotide-binding domains. 2011, Pubmed
Matulef, Multimerization of the ligand binding domains of cyclic nucleotide-gated channels. 2002, Pubmed , Xenbase
Mazzolini, Gating in CNGA1 channels. 2010, Pubmed
Mazzolini, The analysis of desensitizing CNGA1 channels reveals molecular interactions essential for normal gating. 2009, Pubmed , Xenbase
Mazzolini, Movement of the C-helix during the gating of cyclic nucleotide-gated channels. 2002, Pubmed , Xenbase
Mazzolini, A comparison of electrophysiological properties of the CNGA1, CNGA1tandem and CNGA1cys-free channels. 2008, Pubmed , Xenbase
Müller, Vertebrate membrane proteins: structure, function, and insights from biophysical approaches. 2008, Pubmed
Murata, Novel biosensor for the rapid measurement of estrogen based on a ligand-receptor interaction. 2001, Pubmed
Nair, Movements of native C505 during channel gating in CNGA1 channels. 2009, Pubmed
Nair, Locking CNGA1 channels in the open and closed state. 2006, Pubmed , Xenbase
Nair, Conformational rearrangements in the S6 domain and C-linker during gating in CNGA1 channels. 2009, Pubmed
Ng, The S4-S5 linker acts as a signal integrator for HERG K+ channel activation and deactivation gating. 2012, Pubmed
Oesterhelt, Unfolding pathways of individual bacteriorhodopsins. 2000, Pubmed
Orsini, Atomic force microscopy characterization of Xenopus laevis oocyte plasma membrane. 2006, Pubmed , Xenbase
Orsini, Intermittent contact mode AFM investigation of native plasma membrane of Xenopus laevis oocyte. 2009, Pubmed , Xenbase
Peng, Kinetic partitioning mechanism governs the folding of the third FnIII domain of tenascin-C: evidence at the single-molecule level. 2011, Pubmed
Peng, Atomic force microscopy reveals parallel mechanical unfolding pathways of T4 lysozyme: evidence for a kinetic partitioning mechanism. 2008, Pubmed
Rief, Reversible unfolding of individual titin immunoglobulin domains by AFM. 1997, Pubmed
Rosenbaum, Dissecting intersubunit contacts in cyclic nucleotide-gated ion channels. 2002, Pubmed , Xenbase
Santacroce, Atomic force microscopy imaging of Xenopus laevis oocyte plasma membrane purified by ultracentrifugation. 2008, Pubmed , Xenbase
Santacroce, Imaging of Xenopus laevis oocyte plasma membrane in physiological-like conditions by atomic force microscopy. 2013, Pubmed , Xenbase
Schünke, Structural insights into conformational changes of a cyclic nucleotide-binding domain in solution from Mesorhizobium loti K1 channel. 2011, Pubmed
Shuart, Molecular mechanism for 3:1 subunit stoichiometry of rod cyclic nucleotide-gated ion channels. 2011, Pubmed
Tanuj Sapra, Detecting molecular interactions that stabilize native bovine rhodopsin. 2006, Pubmed
Tetreault, Characterization of a novel cyclic nucleotide-gated channel from zebrafish brain. 2006, Pubmed , Xenbase
Valbuena, On the remarkable mechanostability of scaffoldins and the mechanical clamp motif. 2009, Pubmed
Verbelen, The NTA-His6 bond is strong enough for AFM single-molecular recognition studies. 2007, Pubmed
Watanabe, Molecular mechanics of cardiac titin's PEVK and N2B spring elements. 2002, Pubmed
Yarov-Yarovoy, Structural basis for gating charge movement in the voltage sensor of a sodium channel. 2012, Pubmed
Yu, Overview of molecular relationships in the voltage-gated ion channel superfamily. 2005, Pubmed
Zagotta, Structural basis for modulation and agonist specificity of HCN pacemaker channels. 2003, Pubmed
Zocher, Cholesterol increases kinetic, energetic, and mechanical stability of the human β2-adrenergic receptor. 2012, Pubmed