FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2097823677> ?p ?o ?g. }

• W2097823677 endingPage "362" @default.
• W2097823677 startingPage "353" @default.
• W2097823677 abstract "Article15 January 1998free access Three amino acids in the C-linker are major determinants of gating in cyclic nucleotide-gated channels Xiangang Zong Xiangang Zong Institut für Pharmakologie und Toxikologie, Technische Universität München, Biedersteinerstrasse 29, 80802 München, Germany Search for more papers by this author Helmut Zucker Helmut Zucker Max-Planck-Institut für Psychiatrie, EDV-Gruppe, Am Klopferspitz 18a, 82152 Martinsried, Germany Search for more papers by this author Franz Hofmann Franz Hofmann Institut für Pharmakologie und Toxikologie, Technische Universität München, Biedersteinerstrasse 29, 80802 München, Germany Search for more papers by this author Martin Biel Corresponding Author Martin Biel Institut für Pharmakologie und Toxikologie, Technische Universität München, Biedersteinerstrasse 29, 80802 München, Germany Search for more papers by this author Xiangang Zong Xiangang Zong Institut für Pharmakologie und Toxikologie, Technische Universität München, Biedersteinerstrasse 29, 80802 München, Germany Search for more papers by this author Helmut Zucker Helmut Zucker Max-Planck-Institut für Psychiatrie, EDV-Gruppe, Am Klopferspitz 18a, 82152 Martinsried, Germany Search for more papers by this author Franz Hofmann Franz Hofmann Institut für Pharmakologie und Toxikologie, Technische Universität München, Biedersteinerstrasse 29, 80802 München, Germany Search for more papers by this author Martin Biel Corresponding Author Martin Biel Institut für Pharmakologie und Toxikologie, Technische Universität München, Biedersteinerstrasse 29, 80802 München, Germany Search for more papers by this author Author Information Xiangang Zong1, Helmut Zucker2, Franz Hofmann1 and Martin Biel 1 1Institut für Pharmakologie und Toxikologie, Technische Universität München, Biedersteinerstrasse 29, 80802 München, Germany 2Max-Planck-Institut für Psychiatrie, EDV-Gruppe, Am Klopferspitz 18a, 82152 Martinsried, Germany *Corresponding author. E-mail: [email protected] The EMBO Journal (1998)17:353-362https://doi.org/10.1093/emboj/17.2.353 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The activation of cyclic nucleotide-gated (CNG) channels is a complex process comprising the initial ligand binding and a consecutive allosteric transition from a closed to an open configuration. The cone and olfactory CNG channels differ considerably in cyclic nucleotide affinity and efficacy. In each channel, the cyclic nucleotide-binding site is connected to the last transmembrane segment of the channel by a linker peptide (C-linker) of ∼90 amino acids. Here we report that replacement of three amino acids in the cone C-linker by the corresponding amino acids of the olfactory channel (I439V, D481A and D494S) profoundly enhanced the cAMP efficacy and increased the affinities for cAMP and cGMP. Unlike the wild-type cone channel, the mutated channel exhibited similar single-channel kinetics for both cGMP and cAMP, explaining the increase in cAMP efficacy. We thus conclude that the identified amino acids are major determinants of channel gating. Introduction Cyclic nucleotide-gated (CNG) channels comprise a multi-gene family of Ca2+-permeable non-selective cation channels that are activated directly by the binding of cAMP or cGMP (Zufall et al., 1994; Finn et al., 1996; Zagotta and Siegelbaum, 1996). CNG channels were detected originally in photoreceptor (Fesenko et al., 1985; Yau and Nakatani, 1985) and olfactory neurons (Nakamura and Gold, 1987), where they play key roles in transducing sensory information into changes in membrane potential. CNG channels have also been identified in a variety of non-sensory cells (Distler et al., 1994; Biel et al., 1995), including kidney (Ahmad et al., 1992; Biel et al., 1994; Karlson et al., 1995), testis (Weyand et al., 1994; Biel et al., 1994), heart (Biel et al., 1994; Ruiz et al., 1996) and several parts of the brain (Dryer and Henderson, 1991; Bönigk et al., 1996; Kingston et al., 1996; Bradley et al. 1997; Sautter et al., 1997). However, the physiological role of CNG channels is not known in most of these cells. Native CNG channels are supposed to have a tetrameric structure consisting of α and β subunits with unknown stoichiometry (Liu et al., 1996; Varnum and Zagotta, 1996; for recent reviews, see Biel et al., 1996; Finn et al. 1996). In contrast to the β subunits (Chen et al. 1993; Bradley et al., 1994; Liman and Buck, 1994), the α subunits induce the formation of homomeric CNG channels when heterologously expressed in a variety of expression systems (Kaupp et al., 1989; Dhallan et al., 1990; Bönigk et al., 1993; Biel et al., 1994). The olfactory CNG channel differs from rod and cone photoreceptor channels in its response to cyclic nucleotides (Zagotta and Siegelbaum, 1996): (i) it has a 15- to 100-fold higher apparent affinity for both cGMP and cAMP than the rod and cone channel and (ii) it is fully activated by cGMP and cAMP (i.e. cGMP and cAMP have the same efficacy for this channel) whereas cAMP acts as a partial agonist of photoreceptor channels activating only a fraction of the current induced by cGMP (i.e. cGMP has a higher efficacy than cAMP). CNG channel activation involves a complex cascade of conformational changes leading from closed to open channel configuration (Goulding et al., 1994; Gordon and Zagotta, 1995b). In agreement with a cyclic allosteric model of channel activation, it recently has been demonstrated (Tibbs et al. 1997) that CNG channels transiently open in the absence of any cyclic nucleotide. Cyclic nucleotides strongly enhance channel activation by binding more tightly to the open state than to the closed state, thereby stabilizing the open conformation. Cyclic nucleotide binding takes place within a stretch of 130 amino acids in the carboxy-terminus (Kaupp et al., 1989), which is homologous to the cyclic nucleotide-binding domain (CNBD) of protein kinases and the CAP protein of Escherichia coli (Shabb and Corbin, 1992). A threonine residue (T560) within the putative β7-roll (Altenhofen et al., 1991) and a glutamic acid residue (D604) within the C-α helix (Varnum et al., 1995) of the CNBD have been identified which critically affect the ligand-binding step and the cyclic nucleotide discrimination, respectively. However, since both amino acids are conserved in mammalian olfactory and photoreceptor channels, they cannot explain the differences observed in the response to cyclic nucleotides. Moreover, mutagenesis studies indicate that domains outside the CNBD, including the amino-terminus (Goulding et al., 1994; Gordon and Zagotta, 1995b) and the cytoplasmatic linker connecting the sixth transmembrane segment (Broillet and Firestein, 1996, 1997), also influence CNG channel activation. In the present study, we have investigated chimeric channels constructed between the bovine cone photoreceptor channel (CNG3; Biel et al., 1994) and the rabbit olfactory channel (CNG2; Biel et al., 1993) to narrow down sequences important for the different cAMP efficacies in both types of channels. Unlike previous studies, we used CNG3 instead of the rod channel (CNG1; Kaupp et al., 1989) to construct chimeras since its structural and functional relationship to CNG2 is higher than that of CNG1. We expected that this closer relationship might circumvent intrinsic problems arising from combining domains of more distantly related proteins and thereby would allow single amino acids which are critical for channel activation to be pinpointed. Results Figure 1 shows the response of the CNG3 and CNG2 channel to cGMP and cAMP as measured in excised inside-out patches of transfected HEK293 cells. CNG3 differed in two major aspects from CNG2. (i) It was only partially activated by cAMP. The maximal current (Imax) induced by a saturating cAMP concentration corresponded to ∼20% of the Imax induced by cGMP (Figure 1A and C, Table I). By contrast, CNG2 was fully activated by both cAMP and cGMP (Figure 1B and D, Table I). (ii) CNG 3 revealed a 15- to 20-fold lower apparent affinity than CNG2 for both cyclic nucleotides (Figure 1C and D, Table I). However, the ratio of the apparent affinities for cGMP and cAMP, which is a measure of the ligand selectivity, was comparable for both channels (CNG2: Ka[cAMP]/Ka[cGMP] = 57; CNG3: Ka[cAMP]/Ka[cGMP] = 87; see Table I). To confirm that cAMP acted as a partial agonist in CNG3 and a full agonist in CNG2, we measured the dose–response curves of cGMP alone and in combination with fixed concentrations of cAMP (Figure 1E and F). As expected for a competitive interaction between a partial and a full agonist, cAMP inhibited the current of CNG3 that would have been elicited by subsaturating concentrations of cGMP (Figure 1E). The inhibition depended on the concentration of cAMP. For example, the current induced by 30 μM cGMP, which corresponds to ∼90% of Imax, was reduced by 1 mM cAMP to ∼65% of Imax whereas it was reduced to 50% of Imax by 2 mM cAMP. By contrast, cAMP and cGMP responses were additive for the CNG2 channel, which indicated that both ligands were full agonists of this channel (Figure 1F). Figure 1.Differential activation of CNG2 and CNG3 by cyclic nucleotides. (A and B) Top: transmembrane folding model of the CNG3 (A) and CNG2 channel (B). The transmembrane segments (1–6), the pore region (P) and the cyclic nucleotide-binding domain (CNBD) are indicated. The C-linker connects the S6 segment with the CNBD. Bottom: current traces of CNG3 (A) and CNG2 (B) induced in excised inside-out patches by saturating concentrations of cGMP and cAMP. The membrane potential was +60 mV. (C and D) Dose–response curves of CNG3 (C) and CNG2 (D) for cGMP (○) and cAMP (●). The curves were calculated from the Hill equation with the following parameters: activation of CNG3 by cGMP, ν = 1.65, Ka = 12.5 μM (n = 11), by cAMP, ν = 1.78, Ka = 995 μM (n = 14); activation of CNG2 by cGMP, ν = 1.57, Ka = 0.83 μM (n = 15), by cAMP, ν = 1.56, Ka = 55 μM (n = 17). (E and F) Dose–response curves of CNG3 (E) and CNG2 (F) for cGMP alone (○) and in the presence of 1 or 0.05 mM (●) and 2 or 0.1 mM (▿) of cAMP. Solid lines for current activation in the presence of both ligands were calculated with the Dean equation (see Materials and methods) with the following parameters: for CNG3 at 1 mM cAMP, Ka = 17.9 μM, Kb = 2012 μM, α = 1.02, β = 0.15 (n = 7); at 2 mM cAMP, Ka = 21.3 μM, Kb = 1520 μM, α = 1.03, β = 0.26 (n = 8); for CNG2 at 0.05 mM cAMP, Ka = 0.83 μM, Kb = 64.5 μM, α = 0.99, β = 1.00 (n = 8); at 0.1 mM cAMP, Ka = 0.97 μM, Kb = 49.2 μM, α = 1.00, β = 1.03 (n = 5). The fits for activation by cGMP alone were taken from (C) and (D). Download figure Download PowerPoint Table 1. Ratio of maximal currents measured in the presence of saturating concentrations of cAMP (IcAMP) and cGMP (IcGMP) and apparent ligand affinities (Ka) of wild-type and chimeric CNG channels Channels IcAMP/IcGMP (%) Ka (μM) Ka(cAMP)/Ka(cGMP) cGMP cAMP CNG3 18.9 ± 1.64 (19) 12.6 ± 0.92 (11) 1095 ± 80 (14) 87 CNG2 101 ± 0.76 (11) 0.90 ± 0.21 (15) 50.9 ± 9.0 (17) 57 Ch2 19.9 ± 3.95 (20) 11.4 ± 2.08 (12) 595 ± 35 (12) 52 Ch3 95.6 ± 4.16 (10) 5.80 ± 1.02 (11) 239 ± 31 (6) 41 Ch4 93.3 ± 5.18 (12) 2.65 ± 0.43 (12) 61.8 + 8.6 (11) 23 Ch8 98.1 ± 3.04 (12) 2.53 ± 0.31 (8) 305 ± 53 (7) 120 Ch9 35.1 ± 3.83 (10) 13.2 ± 2.42 (7) 512 ± 31 (8) 39 Ch12 77.6 ± 2.27 (18) 4.64 ± 0.60 (8) 534 ± 64 (7) 115 Ch13 42.2 ± 3.80 (13) 9.86 ± 0.75 (8) nd nd Ch20 37.1 ± 2.66 (16) 10.7 ± 1.35 (6) nd nd Ch21 81.0 ± 2.73 (12) 3.45 ± 0.42 (8) nd nd Ch24 81.1 ± 2.73 (13) 6.17 ± 0.56 (9) nd nd Ch25 70.4 + 4.94 (12) 8.85 ± 0.51 (10) 482 ± 35 (7) 55 Ch26 46.5 ± 3.95 (10) 7.24 ± 0.45 (9) nd nd Ch28 57.3 ± 2.98 (11) 5.14 ± 0.55 (7) nd nd Ch30 80.5 ± 1.96 (13) 4.97 ± 0.38 (10) 625 ± 52 (13) 126 Ch35 60.0 ± 4.83 (8) nd 790 ± 50 (8) nd Ch36 83.6 ± 2.20 (7) nd 435 ± 44 (7) nd Ch37 88.0 ± 2.89 (8) nd 429 ± 55 (13) nd Ch38 91.8 ± 4.59 (4) nd 657 ± 41 (7) nd IcAMP and IcGMP were measured at +60 mV in the presence of 10 mM cAMP and 0.3 mM cGMP, respectively. The Ka values were taken at +60 mV by fits to the Hill equation. Numbers in parentheses indicate the number of patches tested. nd, not determined. Identification of amino acid residues affecting cAMP efficacy To identify channel domains responsible for the different cAMP efficacies of CNG2 and CNG3, we constructed chimeric channels based on the CNG3 transmembrane core and variable parts of CNG2. As a measure of the relative cAMP efficacy (EcAMP), we determined the ratio of the currents obtained at saturating concentrations of cAMP and cGMP (IcAMP/IcGMP). In an initial round of mutations, we replaced large parts of CNG3 with the corresponding sequences of CNG2 (Figure 2; Table I). Replacement of the cytoplasmic amino-terminal segment (Ch2) did not substantially change EcAMP, whereas the replacement of the complete carboxy-terminus, which includes the CNBD and the linker between the S6 segment and the CNBD (C-linker), resulted in a chimera (Ch3) which was fully activated by cAMP. Replacement of both amino- and carboxy-termini did not alter EcAMP with respect to Ch3. The replacement of CNG3 channel domains by corresponding CNG2 domains did not alter the voltage dependence of channel activation. The ratio of currents evoked at −60 mV and +60 mV was ∼0.8, which is identical to the value reported for the wild-type CNG3 channel (Biel et al., 1994). We next subdivided the carboxy-terminus of Ch3 into two parts. Ch8 contains only the C-linker and the first part of the CNBD which is highly conserved between CNG2 and CNG3 (only three exchanges; see Figure 3A). The complementary chimera (Ch9) contains the CNBD and the carboxy-terminal end of CNG2. The currents of both chimeras are shown in Figure 2. Ch8 was activated equally well by cAMP and cGMP (EcAMP = 0.98) whereas Ch9 was only activated slightly better by cAMP than was CNG3. Similarly to Ch3, Ch8 revealed a significant increase in the apparent affinity for both cAMP and cGMP (Table I). By contrast, the exchange of amino- and carboxy-terminal sequences between CNG2 and CNG3 did not alter the ligand selectivity substantially. However, chimeras containing the olfactory amino-terminus (Ch2), the olfactory CNBD (Ch3, Ch9) or both domains of the olfactory channel (Ch4) revealed, similarly to the wild-type CNG2 channel, a somewhat smaller difference in Ka (∼20- to 60-fold) than the wild-type CNG3 channel (90-fold) and Ch8 (120-fold) (see Table I). Taken together, these results suggested that the amino-terminus and the CNBD of CNG2 increased apparent ligand affinities and weakened the ligand discrimination but had only a small influence on EcAMP. On the other hand, the high EcAMP of Ch8 clearly indicated that the major determinant of EcAMP was localized within the C-linker. Figure 2.The C-linker determines the efficacy of cAMP. Left panel: schematic representation of the chimeras 2, 3, 4, 8 and 9. Channel domains originating from CNG3 are illustrated by thin white lines, domains derived from CNG2 are shown by thick black lines. Middle panel: current traces of the chimeras induced by saturating concentrations of cGMP and cAMP at ±60 mV. Right panel: ratio between the current amplitudes elicited by saturating concentrations of cAMP and cGMP. Download figure Download PowerPoint Figure 3.Three amino acids within the C-linker are major determinants of cAMP efficacy. (A) Comparison of the sequence of the C-linker and the first part of the CNBD (A-α helix to β3-roll) of wild-type CNG3 and CNG2 channels and CNG channel chimeras. The numbering refers to the sequence of CNG3. In CNG2 and chimeras Ch8–Ch37, only those amino acids are shown which differ from the sequence of CNG3. Black lines indicate sequence identities with CNG3. (B) Ratio of the current amplitudes induced by saturating concentration of cAMP and cGMP at +60 mV for wild-type and chimeric CNG channels. Download figure Download PowerPoint Figure 3A shows the sequence of the C-linker and the first part of the CNBD of CNG3 compared with the corresponding sequence of CNG2 which is present in Ch8. In total, there are 20 amino acid residues exchanged, 17 residues within the C-linker and three within the CNBD, two of which are conservative exchanges (K530R and R536K). To identify single amino acids affecting EcAMP, we systematically replaced amino acid stretches or single residues of Ch8 with the corresponding sequences of CNG3. The sequences of some of the resulting chimeric channels are shown in Figure 3A. All constructed chimeras were expressed reasonably well. The ratio of IcAMP/IcGMP as a parameter for EcAMP is shown in Figure 3B and Table I. We first reduced the size of the exchanged sequence with respect to Ch8. Ch12 differs from Ch8 in that it has only exchanged the C-linker of CNG3 against that of CNG2 whereas the complete CNBD of CNG3 is present. This reduction slightly decreased EcAMP from 0.98 to ∼0.8, suggesting that the influence of the CNBD is small. By contrast, Ch13, which is distinguished from Ch12 only by a difference in the first four amino acids, revealed a strongly reduced EcAMP (0.42). The difference in EcAMP between Ch12 and Ch13 could be attributed completely to the exchange of I439V. Ch21, which is identical to Ch13 but additionally contains the I439V exchange, revealed the same EcAMP as Ch12 (∼0.8). The three other residues (S428, A438 and Q441) had no influence on channel activation by cAMP. The I439V exchange was necessary but not sufficient to explain the high EcAMP of Ch12, since Ch20, which contains this exchange but lacks the C-terminal half of the C-linker, revealed a strongly reduced EcAMP (0.37) with respect to Ch12. Extensive investigation of a variety of chimeras resulted in the identification of two additional amino acids within the C-linker (D481A and D494S) which also participate in determining EcAMP. Ch30, which differs from the wild-type CNG3 channel only by three exchanges (I439V, D481A and D494S), revealed the same EcAMP (∼0.8) as Ch12. All three exchanges were necessary in combination since the omission of any of the exchanges resulted in a significant reduction in EcAMP (see Ch26, Ch28 and Ch35 in Figure 3 and Table I). We next tried to increase EcAMP further by systematically exchanging the five amino acids in Ch30 which vary between Ch8 (EcAMP = 0.98) and Ch12 (EcAMP = 0.8). However, the remaining difference in EcAMP could not be attributed to a single residue. The biggest effect was seen with the exchange of A520Q (Ch37) in the β1-roll of the CNBD which resulted in an additional 10% increase in EcAMP (EcAMP = 0.88). Functional properties of Ch30 Since the exchange of three amino acids in the CNG3 C-linker giving rise to Ch30 represented the minimal requirement for the induction of high EcAMP, we subsequently analysed this chimera in more detail. EcAMP is defined as the ratio of Imax at saturating concentrations of cAMP and cGMP. Thus, the increase in EcAMP may result from a relative decrease in the maximal cGMP current of Ch30 with respect to that of CNG3. We could exclude this possibility since the maximal cGMP current was not significantly different (P >0.05) between CNG3 (Imax = 749 ± 119 pA, range 60–2250 pA, n = 19) and Ch30 (Imax = 589 ± 143 pA, range 105–1680 pA, n = 13). The increase in EcAMP could also be explained by other mechanisms, including an increase in the cAMP affinity, the alteration of the voltage dependence of channel opening or changes in the channel gating process. Figure 4 shows that changes in ligand affinity and voltage dependence are unlikely to explain the strong enhancement of the cAMP-induced current observed in Ch30 (Figure 4A). Ch30 revealed a slight increase in the apparent affinity for cAMP with respect to CNG3 (Figure 4B, Table I); however, an even bigger increase in affinity was seen for cGMP, resulting in a similar affinity ratio of both channels (Ka[cAMP]/Ka[cGMP] was 87 for CNG3 and 125 for Ch30). As has been shown for CNG2 and CNG3, cGMP and cAMP competed for the same binding site in Ch30 (Figure 4C). The major difference between CNG3 and Ch30 consisted of the increase in EcAMP changing cAMP from a partial to a nearly full agonist. Finally, the shape of the current–voltage relationship curves was not substantially changed between Ch30 (Figure 4D) and CNG3 (Figure 4E), indicating that the three amino acid exchanges present in Ch30 did not affect the voltage dependence of channel activation. Figure 4.Functional properties of the Ch30. (A) Current traces of Ch30 induced by saturating concentrations of cGMP and cAMP at +60 mV. (B)Dose–response curves for cGMP (○) and cAMP (●). The curves were fitted with the Hill equation using the following parameters. cGMP, ν = 2.18, Ka = 4.83 μM, cAMP, ν = 1.56, Ka = 581 μM. The dotted lines represent dose–response curves for cGMP and cAMP of the wild-type CNG3 channel. (C) Dose–response curves of Ch30 for cGMP alone (○) and in the presence of 0.5 mM cAMP (●). The solid lines for current activation in the presence of 0.5 mM cAMP are best fits of the Dean equation using the following parameters: Ka = 2.9 μM, Kb = 782 μM, α = 0.96, β = 0.83. The curve of cGMP alone was taken from (B). For comparison, the dose–response curves of the CNG3 channel for activation by cGMP and by a combination of cGMP + 1 mM cAMP are included as dotted lines. (D and E) Current–voltage relationships of Ch30 (D) and CNG3 (E). Download figure Download PowerPoint Single-channel analysis of Ch30 The above results suggested that the mutations in the C-linker might affect channel gating. We therefore measured the single-channel current of Ch30 and compared it with the current of CNG2 and CNG3 (Figure 5; Table II). The single-channel current of CNG3 induced by cGMP and cAMP revealed prominent differences (Figure 5A). With cGMP, two kinds of opening events were visible, a very brief opening mode with an open time constant of <1 ms and a long opening mode with a time constant of ∼5 ms. By contrast, when the current was activated by cAMP, only brief openings were present. At high concentrations of cAMP, the opening of multiple channels was observed (Figure 5A, sixth trace) whereas long openings did not occur. This finding suggested that the ligand specificity of the opening modes is an intrinsic property of the CNG3 channel and does not depend on the ligand concentration. Figure 5.Single-channel currents of CNG3 (A), CNG2 (B) and Ch30 (C) recorded at +60 mV. Left panels: current traces induced in the same patch by cGMP (cGMP concentrations: A, 1 μM; B, 0.1 μM; C, 0.3 μM) or cAMP (cAMP concentrations: A, 100 μM; B, 10 μM; C, 30 μM). For each channel, the third current trace for cAMP was recorded in the presence of a high concentration of cAMP (cAMP concentrations: A, 300 μM; B, 100 μM; C, 300 μM). The continuous lines indicate the closed state of the respective channels. Right panels: open time distribution of currents activated by cGMP and cAMP for CNG3 (A), CNG2 (B) and Ch30 (C). The cGMP concentrations were the same as used in the left panel. cAMP was used at the low concentration as shown in the left panel. The open times were fitted by two exponentials for all currents, with the exception of the CNG3 current induced by cAMP, which was fitted by one exponential (for values of time constants see Table II). Download figure Download PowerPoint Table 2. Single-channel kinetics of CNG channels Channels/ligand Time constants (ms) τo1 τo2 τc1 τc2 CNG2/cGMP 0.88 ± 0.14 13.7 ± 1.11 0.94 ± 0.28 29.6 ± 6.49 CNG2/cAMP 0.75 ± 0.11 11.9 ± 0.82 0.97 ± 0.39 31.8 ± 7.67 CNG3/cGMP 0.72 ± 0.06 4.55 ± 0.58 1.92 ± 0.70 26.2 ± 10.6 CNG3/cAMP 0.48 ± 0.05 − 2.31 ± 0.45 24.1 ± 7.08 Ch30/cGMP 0.57 ± 0.01 3.34 ± 0.37 2.30 ± 0.56 17.8 ± 1.37 Ch30/cAMP 0.54 ± 0.02 2.71 ± 0.24 1.90 ± 0.81 13.7 ± 1.52 Time constants for open (τo1, τo2) and closed (τc1, τc2) states were fitted from open and closed time distributions as shown in Figure 5. For each channel, three individual patches were measured. The single-channel current of the CNG2 channel was identical with both ligands (Figure 5B). It was characterized by very long openings which were at least three times as long as those observed with CNG3. In addition, brief openings similar to those observed in CNG3 were also present. The amplitude of the CNG2 single-channel current did not depend on the activating ligand (IcAMP = 3.15 ± 0.08 pA; IcGMP = 3.02 ± 0.08 pA at +60 mV; n = 3) but differed slightly from that of CNG3 (IcAMP = 2.46 ± 0.07 pA; n = 3; IcGMP = 2.42 ± 0.06 pA; n = 3), most likely reflecting the presence of different pore regions in both channels. Figure 5C shows single-channel currents recorded from Ch30. The opening mode was very similar for both cyclic nucleotides, and closely resembled that of the cGMP-activated CNG3 channel. It was characterized by a mixture of very brief and longer opening events. Thus, Ch30 resembled CNG2 on the single-channel level in that it did not discriminate between cGMP and cAMP. However, Ch30 differed from CNG2 in that the very long opening events which are typical for CNG2 were absent. In addition, the amplitude of the single-channel current of Ch30 corresponded to that of CNG3 (IcAMP = 2.40 ± 0.06 pA; IcGMP = 2.39 ± 0.05 pA at +60 mV; n = 3), reflecting the presence of the same pore region in both channels. From these results, we concluded that the exchange of three amino acids giving rise to Ch30 increased EcAMP by allowing longer openings in the presence of cAMP. Inhibitory effect of the CNG3 C-linker on the CNG2 current If the C-linker of CNG2 increases EcAMP when introduced into CNG3, the replacement of the C-linker of CNG2 with the corresponding sequence of CNG3 should conversely diminish EcAMP. Figure 6 shows currents activated in the reverse chimera 25 (Ch25) which is based on the CNG2 core but contains the C-linker of CNG3. As expected, Ch25 discriminated between the cyclic nucleotides, being only partially activated by cAMP (Figure 6A). However, the effect of the C-linker was less than in the opposite case, leading to a reduction of EcAMP from 1.0 in CNG2 to 0.7 in Ch25. In parallel to the decrease in EcAMP, Ch25 exhibited a 10-fold decrease in apparent affinities for both cGMP and cAMP, bringing the Ka values of Ch25 and Ch30 for cGMP and cAMP close to each other (Table I). The inhibitory influence of the CNG3 C-linker could also be demonstrated at the single-channel level (Figure 6B). As can be seen clearly from the current traces, the presence of the CNG3 C-linker significantly reduced the open time for cAMP with respect to that observed with cGMP. In particular, the very long openings which are typical for the wild-type CNG2 channel were completely missing in Ch25. Figure 6.Activation of the inverse chimera (Ch 25) by cGMP and cAMP. (A) Left: schematic representation of Ch25. The C-linker and the first part of the CNBD originating from CNG3 are illustrated in white. Right: current traces induced by saturating concentrations of cGMP and cAMP at +60 mV. (B) Left panels: single-channel currents induced in the same patch by 3 μM cGMP and 100 μM cAMP at +60 mV. Right panels: open time distribution of the current induced by 3 μM cGMP and 100 μM cAMP. The solid lines represent the best fit by two exponentials. The time constants were as follows: for cGMP: τo1 = 1.03 ms, τo2 = 14.5 ms; for cAMP: τo1 = 1.03 ms, τo2 = 9.40 ms. Download figure Download PowerPoint Discussion In this study, we provide evidence that the linking region (the C-linker) connecting the last transmembrane domain with the CNBD of CNG channels is an important determinant of channel activation. The replacement of the cone (CNG3) channel C-linker by that of the olfactory (CNG2) channel resulted in a chimera (Ch8) which, like the olfactory channel, was almost equally well activated by cAMP and cGMP, whereas the wild-type cone channel is only poorly activated by cAMP. Conversely, the reverse chimera (Ch25) containing the cone C-linker within an olfactory channel backbone revealed a significantly reduced efficacy of channel activation by cAMP. The increase and decrease in the cAMP efficacy in Ch8 and Ch25, respectively, was paralleled by an increase and decrease in the apparent affinities for both cAMP and cGMP but not by a substantial change in the ligand selectivity, suggesting that the C-linker affects the channel gating process. The exchange of the CNBD and the cytoplasmic amino-terminus slightly increased the apparent cAMP affinity but had no major influence on the cAMP efficacy (EcAMP). This finding contrasts with recent studies on the activation of chimeras constructed between the rod photoreceptor and the olfactory channel, which identified the amino-terminus and the first two transmembrane segments of the channel as being especially influential in affecting channel activation (Goulding et al., 1994; Gordon and Zagotta, 1995b). This discrepancy could arise from the fact that cone and olfactory channels are both structurally and functionally more related to each other than are rod and olfactory channels. Thus, the amino-termini of cone and olfactory channels might be functionally equivalent whereas those of rod and olfactory channels are not. The finding that the C-linker has no major influence on the ligand selectivity is in line with the recent studies showing that an aspartate residue within the C-α helix of the CNBD (Varnum et al., 1995), which is present in both the cone and olfactory channel, is the ma" @default.
• W2097823677 created "2016-06-24" @default.
• W2097823677 creator A5038091806 @default.
• W2097823677 creator A5060768208 @default.
• W2097823677 creator A5089190614 @default.
• W2097823677 creator A5091874127 @default.
• W2097823677 date "1998-01-15" @default.
• W2097823677 modified "2023-10-16" @default.
• W2097823677 title "Three amino acids in the C-linker are major determinants of gating in cyclic nucleotide-gated channels" @default.
• W2097823677 cites W1487066117 @default.
• W2097823677 cites W1658856691 @default.
• W2097823677 cites W1966001834 @default.
• W2097823677 cites W1966631841 @default.
• W2097823677 cites W1970706920 @default.
• W2097823677 cites W1972334416 @default.
• W2097823677 cites W1976320202 @default.
• W2097823677 cites W1991785322 @default.
• W2097823677 cites W1993461423 @default.
• W2097823677 cites W1993910833 @default.
• W2097823677 cites W2001365834 @default.
• W2097823677 cites W2004561786 @default.
• W2097823677 cites W2004670954 @default.
• W2097823677 cites W2013348596 @default.
• W2097823677 cites W2019328466 @default.
• W2097823677 cites W2025364863 @default.
• W2097823677 cites W2033434600 @default.
• W2097823677 cites W2034176125 @default.
• W2097823677 cites W2034899978 @default.
• W2097823677 cites W2043628792 @default.
• W2097823677 cites W2048419229 @default.
• W2097823677 cites W2050203483 @default.
• W2097823677 cites W2056221479 @default.
• W2097823677 cites W2060836378 @default.
• W2097823677 cites W2078895135 @default.
• W2097823677 cites W2079854101 @default.
• W2097823677 cites W2082789276 @default.
• W2097823677 cites W2083689768 @default.
• W2097823677 cites W2087767751 @default.
• W2097823677 cites W2089103245 @default.
• W2097823677 cites W2089253451 @default.
• W2097823677 cites W2092506990 @default.
• W2097823677 cites W2098094063 @default.
• W2097823677 cites W2101873472 @default.
• W2097823677 cites W2106781808 @default.
• W2097823677 cites W2144529674 @default.
• W2097823677 cites W2152887599 @default.
• W2097823677 cites W2167689408 @default.
• W2097823677 doi "https://doi.org/10.1093/emboj/17.2.353" @default.
• W2097823677 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/1170386" @default.
• W2097823677 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/9430627" @default.
• W2097823677 hasPublicationYear "1998" @default.
• W2097823677 type Work @default.
• W2097823677 sameAs 2097823677 @default.
• W2097823677 citedByCount "74" @default.
• W2097823677 countsByYear W20978236772012 @default.
• W2097823677 countsByYear W20978236772013 @default.
• W2097823677 countsByYear W20978236772014 @default.
• W2097823677 countsByYear W20978236772017 @default.
• W2097823677 countsByYear W20978236772020 @default.
• W2097823677 countsByYear W20978236772021 @default.
• W2097823677 crossrefType "journal-article" @default.
• W2097823677 hasAuthorship W2097823677A5038091806 @default.
• W2097823677 hasAuthorship W2097823677A5060768208 @default.
• W2097823677 hasAuthorship W2097823677A5089190614 @default.
• W2097823677 hasAuthorship W2097823677A5091874127 @default.
• W2097823677 hasBestOaLocation W20978236772 @default.
• W2097823677 hasConcept C104317684 @default.
• W2097823677 hasConcept C111919701 @default.
• W2097823677 hasConcept C12554922 @default.
• W2097823677 hasConcept C194544171 @default.
• W2097823677 hasConcept C2778524494 @default.
• W2097823677 hasConcept C2780557392 @default.
• W2097823677 hasConcept C41008148 @default.
• W2097823677 hasConcept C512185932 @default.
• W2097823677 hasConcept C515207424 @default.
• W2097823677 hasConcept C54355233 @default.
• W2097823677 hasConcept C55493867 @default.
• W2097823677 hasConcept C86803240 @default.
• W2097823677 hasConceptScore W2097823677C104317684 @default.
• W2097823677 hasConceptScore W2097823677C111919701 @default.
• W2097823677 hasConceptScore W2097823677C12554922 @default.
• W2097823677 hasConceptScore W2097823677C194544171 @default.
• W2097823677 hasConceptScore W2097823677C2778524494 @default.
• W2097823677 hasConceptScore W2097823677C2780557392 @default.
• W2097823677 hasConceptScore W2097823677C41008148 @default.
• W2097823677 hasConceptScore W2097823677C512185932 @default.
• W2097823677 hasConceptScore W2097823677C515207424 @default.
• W2097823677 hasConceptScore W2097823677C54355233 @default.
• W2097823677 hasConceptScore W2097823677C55493867 @default.
• W2097823677 hasConceptScore W2097823677C86803240 @default.
• W2097823677 hasIssue "2" @default.
• W2097823677 hasLocation W20978236771 @default.
• W2097823677 hasLocation W20978236772 @default.
• W2097823677 hasLocation W20978236773 @default.
• W2097823677 hasLocation W20978236774 @default.
• W2097823677 hasOpenAccess W2097823677 @default.
• W2097823677 hasPrimaryLocation W20978236771 @default.
• W2097823677 hasRelatedWork W1973037262 @default.

• next