The α subunits of CNG channels of retinal photoreceptors (rod) and olfactory neurons (olf) are proteins that consist of a cytoplasmic NH2 terminus, a transmembrane core region (including the segments S1–S6), and a cytoplasmic COOH terminus. The COOH terminus contains a cyclic nucleotide monophosphate binding domain NBD) that is linked by the C-linker (CL) to the core region. The binding of cyclic nucleotides to the NBD promotes channel opening by an allosteric mechanism. We examined why the sensitivity to cGMP is 22 times higher in olf than in rod by constructing chimeric channels and determining the [cGMP] causing half maximum channel activity (EC50). The characteristic difference in the EC50 value between rod and olf was introduced by the NH2 terminus and the core-CL region, whereas the NBD showed a paradoxical effect. The difference of the free energy difference Δ(ΔG) was determined for each of these three regions with all possible combinations of the other two regions. For rod regions with respect to corresponding olf regions, the open channel conformation was destabilized by the NH2 terminus (Δ(ΔG) = −1.0 to −2.0 RT) and the core-CL region (Δ(ΔG) = −2.0 to −2.9 RT), whereas it was stabilized by the NBD (Δ(ΔG) = 0.3 to 1.1 RT). The NH2 terminus deletion mutants of rod and olf differed by Δ(ΔG) of only 0.9 RT, whereas the wild-type channels differed by the much larger value of 3.1 RT. The results show that in rod and olf, the NH2 terminus, the core-CL region, and the NBD differ by characteristic Δ(ΔG) values that do not depend on the specific composition of the other two regions and that the NH2 terminus generates the main portion of Δ(ΔG) between the wild-type channels.

Modulation of the activity of CNG channels generates the light-induced electrical response of rod photoreceptors and the odor-induced electrical response in olfactory cells. Native rod CNG channels are heteromultimers formed from α and β subunits (Chen et al. 1993, Chen et al. 1994; Körschen et al. 1995). Expression of the α subunit only is sufficient to produce functional homomultimeric channels (Varnum and Zagotta 1996), whereas the β subunit does not form functional channels on its own but imparts properties onto the heteromultimer that are characteristic of native channels (Chen et al. 1993, Chen et al. 1994; Bucossi et al. 1997; Shapiro and Zagotta 1998; Bönigk et al. 1999).

Opening of CNG channels is strongly promoted by the binding of cyclic nucleotides to a binding domain that is ∼120 residues long (Kaupp et al. 1989) and located near the COOH terminus of the channel protein (for review see Finn et al. 1996). This sequence is homologous to cyclic nucleotide monophosphate binding domains of other proteins (Shabb and Corbin 1992). In the past years, it has been shown by several investigators that a sequence of ∼90 residues linking the cyclic nucleotide monophosphate binding domain to the S6 transmembrane segment (C-linker) plays a key role for the different sensitivity to cyclic nucleotides among the CNG channels (Gordon and Zagotta 1995a,Gordon and Zagotta 1995b; Broillet and Firestein 1996; Gordon et al. 1997; Zong et al. 1998; Paoletti et al. 1999). Using chimeric constructs between the bovine rod and the rat olfactory channel, Gordon and Zagotta 1995b observed that the great difference in the [cGMP] causing half maximum channel activity (EC50) is determined by further regions: the NH2 terminus, the S5 segment, and a large part of the core running from S2 to S6, including S5. Also, in rod, specific interactions between the NH2 terminus and the C-linker were reported to be relevant for opening (Gordon et al. 1997), including the formation of a disulfide bond between C35 in the NH2 terminus and C481 in the C-linker.

Despite growing insight into the role of the different channel regions, our present knowledge about the molecular events underlying the gating is only poor. Herein, we screen CNG channels for regions determining the EC50 by constructing chimeric channels between the α subunits of bovine retinal photoreceptors (rod) and olfactory neurons (olf). We show that multiple regions along the sequence between NH2 terminus and C-linker determine the characteristic difference in the EC50, whereas the cyclic nucleotide monophosphate binding domain has a surprising paradoxical effect. Systematic swap of the NH2 terminus, the core region including the C-linker, and the cyclic nucleotide binding domain shows that the difference of the free energy difference Δ(ΔG) between corresponding rod and olf regions is characteristic, independent of the composition of the channel background.

Composition of the Chimeras

The α subunits of bovine rod and olfactory channels (accession No. 51604 and 55010, respectively) and most of the chimeras between these channels were provided by U.B. Kaupp (Institut für Biologische Informationsverarbeitung, Forschungszentrum Jülich, Germany). Fig. 1 summarizes the name, cartoon, and splice sites of all chimeras used. Chimera r1o3r was constructed by a recombinant PCR approach. For this purpose, plasmids encoding r1o and rod were used as templates for the primary amplification of the desired DNA fragments. PCR products were excised from an agarose gel and coupled in frame in a second PCR reaction. The resulting product was subcloned into the SalI/AsuII sites of the plasmid encoding chimera o5r. For the construction of r5oLr, a BstXI site was introduced at the olf/rod splice site of a chimera oLr (the amino acids M1-I367 of olf and V391-D690 of rod; not shown in Fig. 1), which was finally used to combine respective fragments of a chimera r5o (amino acids M1-I295 of rod and F273-P663 of olf; not shown in Fig. 1) and the modified sequence of oLr. The successful exchange of nucleotide sequences was verified by DNA sequencing.

The name of the chimeras was specified in the following way: “o” indicates an olf sequence and “r” indicates a rod sequence. The transmembrane helices, the C-linker, and the NBD are indicated by the abbreviations “1…6,” “L,” and “B,” respectively. Reading from the left roughly indicates the composition of the chimeras in the direction from the N- to the COOH terminus. One of the abbreviations “1…6,” “L,” or “B” before an “r” or “o” indicates that the splice site is located before the respective rod or olf region. For example, “o1rBo” means that the chimera contains an NH2 terminus from olf, the transmembrane segments S1–S6 plus the C-linker from rod, and a nucleotide binding domain from olf.

Oocyte Preparation

Oocytes of Xenopus laevis were prepared as described previously (Benndorf et al. 1999). In brief, ovarian lobes were obtained under anesthesia (0.3% 3-aminobenzoic acid ethyl ester) and transferred to a Petri dish containing the following Barth medium (in mM): 84 NaCl, 1 KCl, 2.4 NaHCO3, 0.82 MgSO4, 0.33 Ca(NO3)2, 0.41 CaCl2, and 7.5 Tris, pH 7.4 (HCl). Oocytes in stages V and VI were prepared by incubation for 20–30 min in a Ca2+-free Barth medium containing either 1 or 2 mg/ml collagenase. Within 2–7 h after isolation and defolliculation, cRNA specific for the respective channel was injected into the oocytes through glass micropipettes. Oocytes were further incubated at 18°C for 2–7 d until experimental use. Before patching, the vitelline membrane of the oocytes was removed after exposing the cells to the following hypertonic “skinning” solution (in mM): 200 aspartate, 20 KCl, 1 MgCl2, 5 EGTA, and 10 HEPES, pH 7.4 (KOH).

Recording Technique

The oocytes were transferred to the experimental chamber that was mounted on the stage of an inverted microscope. The patch pipettes were pulled from borosilicate glass tubing. The glass tubing had an outer diameter of 2.0 mm and an inner diameter of 1.0 mm. The pipette resistance after fire polishing was 1–3 MΩ. The bath solution contained the following (in mM): 140 KCl, 10 NaCl, 1 EGTA, and 5 HEPES, pH 7.4 (KOH). The pipette solution contained the following (in mM): 145 NaCl, 5 KCl, 1 EGTA, and 5 HEPES, pH 7.4 (NaOH). The currents were recorded in inside-out patches with a conventional patch-clamp technique (Hamill et al. 1981). The CNG channels were activated by replacing the bath solution with a respective solution containing cGMP. Each excised patch was first exposed to a saturating [cGMP] to determine the maximum current.

Recording was performed with an amplifier (model Axopatch 200A; Axon Instruments). The currents were filtered at a cut-off frequency of 10 kHz. The holding voltage was generally 0 mV. The membrane voltage was first stepped to −100 mV and then to +100 mV. The pulse duration and the repetition rate of the pulses were chosen such that, at the end of the pulses, the current amplitude was constant. All measurements were performed at room temperature (22–24°C).

Data Acquisition and Analysis

Recording and analysis of the data was performed on a Pentium PC with the ISO2 software (MFK Niedernhausen). All currents were corrected for the capacitive and the very small leak components by subtracting respective currents in the absence of cGMP in the bath solution. The currents considered herein are averages of 5–30 consecutive recordings. The sampling rate was 2–10 kHz. Dose–response relationships were determined from the steady-state current at +100 mV by normalizing the current I at the actual [cGMP] with respect to the current Imax at saturating [cGMP] and fitting the data points with a Hill equation of the form

\begin{equation*}{{\mathrm{I}}}/{{\mathrm{I}}_{{\mathrm{max}}}}={ \left \left[{\mathrm{cGMP}}\right] \right ^{{\mathrm{H}}}}/{ \left \left( \left \left[{\mathrm{cGMP}}\right] \right ^{{\mathrm{H}}}+{\mathrm{EC}}_{50}^{{\mathrm{H}}}\right) \right {\mathrm{,}}}\end{equation*}
1

yielding values for EC50 and H, the [cGMP] of half maximum activity of the channels and the Hill coefficient, respectively. The curves were fitted to the data with appropriate nonlinear approximation algorithms. The Hill coefficients varied between 2.0 and 3.3. In the present report, they were not further considered because we did not detect a systematic dependency.

Statistics

Statistical data are given as mean ± SEM. SEM of Δ(ΔG) (see below) was calculated according to the error propagation law. t test with a significance level P < 0.05 was used to detect significant differences between data.

The Different EC50 Value in rod and olf Is Determined by Multiple Regions between NH2 Terminus and C-linker, Whereas the Cyclic Nucleotide Binding Domain Has a Paradoxical Effect

Fig. 2 shows dose–response relationships and a plot of the resulting EC50 values obtained from chimeras containing rod regions in an olf background. The chimeras r3o, o3r5o, and o5r were designed to probe the influence of the regions S1–S2 (plus the respective NH2 terminus; NT), S3–S4, and S5-P-S6 (plus the respective C-linker [CL], cyclic nucleotide monophosphate [cNMP] binding site, and the adjacent COOH-terminal sequence). Herein the combination of the cNMP-binding site and the respective COOH-terminal sequence is lumped to the cNMP-binding domain (NBD). Transplanting the rod NT-S1-S2 region (chimera r3o) to olf also transfers a large portion of the higher EC50 of rod. This result agrees with findings for a similar chimera between catfish olf and bovine rod (Tibbs et al. 1997). A smaller portion of the higher EC50 was also transferred with the S5-P-S6-CL-NBD region (chimera o5r). The S3–S4 region of rod did not transfer the higher EC50 value of rod to olf (chimera o3r5o). Transfer of the rod NH2 terminus alone (chimera r1o) caused only a 4-fold increase of the EC50 compared with a 14-fold increase in chimera r3o. This result shows that both the NH2 terminus and the S1–S2 region contribute to the different EC50 in rod and olf. We further attempted to differentiate the influence of the C-linker and the NBD. Transfer of the C-linker alone (chimera oLrBo) produced a similar shift of the EC50 as observed in chimera o5r. Therefore, the C-linker is a third region that determines the EC50 values. Quite surprisingly, transfer of the rod NBD (chimera oBr) produced a lower EC50 value than observed in olf.

In Fig. 3, the first six chimeras illustrate effects of olf regions in a rod background. The first four of these chimeras probe the contributions of the NH2 terminus (chimera o1r), S3–S4 region (chimera r3o5r), S5-P-S6 region (chimera r5oLr), and CL-NBD region (chimera rLo). The S1–S2 region is not included because the respective chimera r1o3r did not produce functional channels. Along with the olf NH2 terminus (chimera o1r), a considerable portion of the lower EC50 is transferred from olf to rod. Smaller, but significant effects were generated by the transfer of the CL-NBD region (chimera rLo) and the S5-P-S6 region (chimera r5oLr). As in the reverse chimera, the transfer of the S3–S4 region was without effect (chimera r3o5r). We also tested for the individual effects of the C-linker (chimera rLoBr) and the NBD (chimera rBo). The transfer of the C-linker (chimera rLoBr) lowered the EC50 value much more than the transfer of the larger CL-NBD region (chimera rLo). With respect to the NBD (chimera rBo), we again observed a paradoxical finding: the chimera rBo had a larger EC50 value than rod.

It has been suggested that the NH2- and COOH-terminal regions in both rod and olf undergo an intramolecular interaction that modulates the ease by which channels are opened by cyclic nucleotides (Gordon et al. 1997; Varnum and Zagotta 1997). To study the functional consequences of these interactions, we also constructed double chimeras that contained both the rod NH2 terminus and NBD in an olf background (o1rBo) and, vice versa, the olf NH2 terminus and NBD in a rod background (r1oBr). Both double chimeras produced EC50 values intermediate between the EC50 values from rod and olf (Fig. 3).

NH2 Terminus, NBD, and Core-CL Region Produce Characteristic Shifts of the EC50 Value Independent of the Channel Background

Fig. 4 illustrates schemes of channels composed either of the rod (black boxes) or olf (white boxes) NH2 terminus, core region (S1–S6) plus the C-linker (core-CL), and NBD. The eight chimeras and wild-type channels are ordered from top to bottom by increasing EC50 values (Fig. 1 and Fig. 2). To quantify the energetic effects of corresponding rod and olf regions, we calculated differences in the free energy difference (Δ(ΔG)) between the channels/chimeras according to :

\begin{equation*}{\mathrm{{\Delta}}} \left \left({\mathrm{{\Delta}G}}\right) \right =-{\mathrm{RT\;ln}} \left \left({{\mathrm{EC}}_{{\mathrm{50,1}}}}/{{\mathrm{EC}}_{50{\mathrm{,2}}}}\right) \right {\mathrm{,}}\end{equation*}
2

where R is the molar gas constant and T is the temperature in K. EC50,1 is the EC50 value of the channel/chimera with the rod region, and EC50,2 is the EC50 value of the channel/chimera with the olf region. Hence, a negative Δ(ΔG) value means that a rod region destabilizes the open channel conformation with respect to the corresponding olf region, whereas a positive Δ(ΔG) value means that a rod region stabilizes the open channel conformation with respect to the corresponding olf region. The brackets to the right of the channel schemes and the corresponding bar graphs at the bottom of Fig. 4 indicate the Δ(ΔG) values for the twelve possible comparisons.

The three regions tested produced consistent effects on Δ(ΔG) with all possible combinations of the other two regions. The rod NH2 terminus produced a Δ(ΔG) of −1.0 to −2.0 RT with respect to the olf NH2 terminus and the rod core-CL region produced a Δ(ΔG) of −2.0 to −2.9 RT with respect to the olf core-CL region. These effects correspond to the higher EC50 in rod than in olf. In contrast, the rod NBD caused a Δ(ΔG) of 0.3 to 1.1 RT with respect to the olf NBD. This effect corresponds to the paradoxical higher EC50 for channels with an olf NBD compared with otherwise equal channels with a rod NBD. The latter result is also substantiated by the 2.5 times larger EC50 in chimera rLo compared with chimera rLoBr (Fig. 3). These results show that each of the three regions in rod and olf causes a characteristic Δ(ΔG) value that is largely independent of the combination of the other two channel regions. Therefore, it is suggested that the molecular interactions of these three channel regions are similar in rod and olf and largely independent of the specific composition of the channel background.

Properties of the rod and olf NH2 Terminus Deletion Mutants

CNG channels also open with deleted NH2 terminus (Gordon and Zagotta 1995b; Brown et al. 1998). In rod, deletion of the NH2 terminus caused a small but significant reduction of the EC50 value (30.5 ± 1.1 μM for −1r compared with 45.0 ± 1.6 μM for rod; Fig. 3). In olf, deletion of the NH2 terminus (−1o) considerably increased the EC50 value (13.0 ± 0.7 μM for −1o compared with 2.0 ± 0.1 μM for olf; Fig. 3). Using , Δ(ΔG) for −1r (EC50,1) with respect to −1o (EC50,2) results in only −0.9 RT, which is much less negative than −3.1 RT for rod with respect to olf (Fig. 5). Thus, the main portion of Δ(ΔG) between the wild-type channels is introduced by the NH2 termini.

Fig. 5 also illustrates the comparison of the Δ(ΔG) values of channels/chimeras containing an NH2 terminus with respect to the corresponding deletion mutant containing no NH2 terminus. The main results are as follows. First, both the rod and the olf NH2 terminus increase Δ(ΔG) with respect to −1o. In contrast, with respect to −1r, only the olf NH2 terminus increases Δ(ΔG), whereas the rod NH2 terminus decreases Δ(ΔG). This means, that the rod NH2 terminus stabilizes the open state only in the olf background, whereas in its natural rod background it slightly destabilizes the open state. Second, beside this opposite effect of the rod NH2 terminus in the rod and olf background, Δ(ΔG) between rod and o1r (1.4 RT) is indistinguishable from that between r1o and olf (1.3 RT). This result implies that the free energy for channel opening provided by the NH2 terminus simply adds to the free energy of the channel background.

Effects of the NH2 Terminus and NBD on the Sensitivity of rod and olf to cGMP

Our data are in line with previous results showing that the NH2 terminus is one of the determinants of the characteristic EC50 of CNG channels (Goulding et al. 1994; Gordon and Zagotta 1995b). In rod, Gordon et al. 1997 reported that the formation of a disulfide bond between the NH2 terminus (C35) and the C-linker (C481) stabilizes the open state by a decrease of both the EC50 and the slope of the dose–response relationship. In our experiments, the olf NH2 terminus decreased the EC50 (stabilized the open state) compared with the rod NH2 terminus with all possible combinations of the core-CL region and the NBD (Fig. 4). Based on the fact that the olf NH2 terminus does not contain the respective cysteine, it is concluded that the olf NH2 terminus stabilizes the open state by a mechanism that does not involve the formation of a disulfide bond.

We also observed a stabilization of the open state by the olf NH2 terminus with respect to the NH2 terminus deletion mutants of both rod and olf: In olf, the Δ(ΔG) value with respect to −1o was 1.9 RT, and in the chimera o1r the Δ(ΔG) value with respect to −1r was 1.0 RT (Fig. 5). In contrast, the effect of the rod NH2 terminus was opposite with respect to the two NH2 terminus deletion mutants: in rod the Δ(ΔG) value with respect to −1r was -0.4 RT, whereas in the chimera r1o the Δ(ΔG) value with respect to −1o was 0.6 (Fig. 5). Despite this qualitatively different influence on the Δ(ΔG) value by the rod NH2 terminus, Δ(ΔG) between the rod and olf NH2 terminus was independent of having either a rod or an olf background, suggesting that the interactions of the NH2 terminus with the rod or olf background are similar.

The finding that in rod channels deletion of the NH2 terminus caused a small but significant reduction of the EC50 (30.5 ± 1.1 μM compared with 45.0 ± 1.6 μM) does not agree with the result of Brown et al. 1998 who did not observe a significant change in the respective EC50 value. One explanation for this discrepancy might be that the deletion mutant of Brown et al. 1998 contained eight amino acids of the olfactory channel just before the putative first transmembrane helix that our deletion mutant –1r did not contain.

The paradoxical effect of the NBD on the EC50 (and thus on Δ(ΔG)) is new and surprising. Although the absolute values of Δ(ΔG) between the rod and olf NBD are typically smaller (0.3–1.1 RT; Fig. 4) than those between rod and olf NH2 terminus (1.1–2.0 RT; Fig. 4), the effects of the NBD are consistent with all combinations of the NH2 terminus and the core-CL region. This consistency supports the idea that also this region interacts with well preserved parts of the protein.

Regions within the Core-CL Region Determining the Sensitivity of rod and olf to cGMP

Within the core-CL region, we observed that besides the S1–S2 region and the C-linker, also the S5-P-S6 region determines the larger EC50 value in rod compared with olf. This conclusion has been derived from two results: (1) the S5-P-S6 region of olf in rod (chimera r5oLr) caused a small but significant reduction of the EC50 value compared with rod (Fig. 3); and (2) taking into account the fact that the NBD of olf paradoxically increases the EC50 value, the similarity of the EC50 values of chimera o5r (S5-NBD region of rod in an olf background) and chimera oLrBo (C-linker of rod in an olf background) may be explained only by a decreasing effect of the S5-P-S6 region on the EC50, balancing the effect of the olf NBD.

The S1–S2 region has been reported first by Goulding et al. 1994 to contribute to the EC50. These investigators studied a chimera in which the S1–S2 region (including the S2–S3 linker) of the catfish olf was inserted in a bovine rod background. In our experiments, transfer of the olf S1–S2 region (without the S2–S3 linker) into a rod background did not produce functional channels. This negative result suggests either that the S1–S2 region requires the right S2–S3 linker for normal channel function or that the bovine olf S1–S2 region does not work properly in a bovine rod background, whereas the catfish olf S1–S2 region does. Nevertheless, our results also confirm a relevant role of the S1–S2 region because we observed a 14-fold increase of the EC50 value when inserting the rod NT-S1-S2 region into an olf background (chimera r3o) and only a 4-fold increase of the EC50 when inserting the rod NH2 terminus alone (chimera r1o).

The C-linker has been shown repeatedly and with different approaches to contribute to the characteristic EC50 values of CNG channels. These approaches include studies on chimeric channels between catfish olf and bovine rod (Goulding et al. 1994), between Caenorhabditis elegans tax-4, catfish olf and bovine rod (Paoletti et al. 1999), and between rat olf and bovine rod (Gordon and Zagotta 1995b). Other approaches include point mutations within the C-linker (Zong et al. 1998), modification of sulfhydryl groups within the C-linker (Brown et al. 1998), and potentiation of rod by the binding of transition metal divalent ions to H420 (Gordon and Zagotta 1995b). Our data obtained in chimeric channels of bovine rod and bovine olf further confirm a relevant role of the C-linker for the EC50. The only sequence within the large core-CL region not contributing to the determination of the EC50 is obviously the S3–S4 region because no effect was found in both respective chimeras (o3r5o, Fig. 2; r3o5r, Fig. 3).

Implications for the Interactions of the Channel Regions in the Gating Process

The experimental results summarized in Fig. 4 can be explained by assuming that the open probability of rod and olf CNG channels is governed by the balance of the action of three relatively independent regions: the NH2 terminus, the NBD, and the core-CL region. Corresponding regions in rod and olf show characteristically different actions. These actions may be summarized as follows. First, the NBD exerts the leading effect on channel opening. This effect is strongly promoted by increasing the [cGMP]. The NBD of rod has a slightly larger opening effect than the NBD of olf at all [cGMP]. Second, the core-CL region has a much weaker and cGMP-independent effect on the channel opening compared with the NBD. The core-CL region of olf favors opening more than the core-CL region of rod. Finally, the NH2 termini have differential effects on channel opening: the olf NH2 terminus promotes channel opening with both olf and rod core-CL region. In contrast, the rod NH2 terminus promotes opening only with the olf core-CL region, whereas it promotes closure with the rod core-CL region. All effects of the NH2 terminus are weak compared with the leading effect of the NBD.

Now consider the two extreme cases: at zero cGMP, the open probability is extremely low. Typical reported values are 2.25 × 10−3 for catfish olf and 1.25 × 10−4 for RO133, a rod chimera with the catfish olf pore (Tibbs et al. 1997). The larger spontaneous activity in olf than in rod could be explained by the larger opening effects of the NH2 terminus and the core-CL region. At saturating [cGMP], the NBD generates maximum open probability with all compositions of the NH2 terminus and core-CL region. The observation that olf attains the maximum open probability at much lower [cGMP] than rod could be explained by the larger opening effects of the NH2 terminus and the core-CL region.

The assumption of an independent action of the three regions can also explain the effects of the NH2 termini in complete channels/chimeras with respect to the corresponding NH2 terminus deletion mutants. Consider first the NH2 terminus–deleted rod (−1r): the EC50 would be lower than in rod because of the lacking closing force of the rod NH2 terminus, but it would be higher than in chimera o1r (olf NH2 terminus in a rod background) because of the lacking opening force of the olf NH2 terminus. In the NH2 terminus–deleted olf (−1o), the EC50 would be much higher than in olf because of the lacking opening force of the olf NH2 terminus, and it would be only slightly higher than in chimera r1o (rod NH2 terminus in an olf background) because of the lacking smaller opening force of the rod NH2 terminus (Fig. 5). An independent action of the regions also explains why the Δ(ΔG) between rod and o1r is similar to that between r1o and olf.

We are indebted to U.B. Kaupp for providing the clones of the channels and most of the chimeras and to K. Schoknecht, S. Bernhardt, A. Kolchmeier, and B. Tietsch for excellent technical assistance.

This work was supported by the grants B301-96078 of the state of Thüringen and Be1250/14-1 of the Deutsche Forschungsgemeinschaft to K. Benndorf.

Benndorf
K.
,
Koopmann
R.
,
Eismann
E.
,
Kaupp
U.B.
Gating by cyclic GMP and voltage in the α subunit of the cyclic GMP-gated channel from rod photoreceptors
J. Gen. Physiol.
114
1999
477
489
[PubMed]
Bönigk
W.
,
Bradley
J.
,
Muller
F.
,
Sesti
F.
,
Boekhoff
I.
,
Ronnett
G.V.
,
Kaupp
U.B.
,
Frings
S.
The native rat olfactory cyclic nucleotide-gated channel is composed of three distinct subunits
J. Neurosci.
19
1999
5332
5347
[PubMed]
Broillet
M.C.
,
Firestein
S.
Direct activation of the olfactory cyclic nucleotide-gated channel through modification of sulfhydryl groups by NO compounds
Neuron.
16
1996
377
385
[PubMed]
Brown
R.L.
,
Snow
S.D.
,
Haley
T.L.
Movement of gating machinery during the activation of rod cyclic nucleotide-gated channels
Biophys. J.
75
1998
825
833
[PubMed]
Bucossi
G.
,
Nizzari
M.
,
Torre
V.
Single channel properties of ionic channels gated by cyclic nucleotides
Biophys. J.
72
1997
1165
1181
[PubMed]
Chen
T.-Y.
,
Peng
Y.-W.
,
Dhallan
R.S.
,
Ahamed
B.
,
Reed
R.R.
,
Yau
K.-W.
A new subunit of the cyclic nucleotide-gated cation channel in retinal rods
Nature.
362
1993
764
767
[PubMed]
Chen
T.-Y.
,
Illing
M.
,
Molday
L.L.
,
Hsu
Y.-T.
,
Yau
K.-W.
,
Molday
R.S.
Subunit 2 (or β) of retinal rod cGMP-gated cation channel is a component of the 240-kDa channel-associated protein and mediates Ca2+-calmodulin modulation
Proc. Natl. Acad. Sci. USA.
91
1994
11757
11761
[PubMed]
Finn
J.T.
,
Grunwald
M.E.
,
Yau
K.-W.
Cyclic nucleotide-gated ion channelsan extended family with diverse functions
Annu. Rev. Physiol.
58
1996
395
426
[PubMed]
Gordon
S.E.
,
Zagotta
W.N.
A histidine residue associated with the gate of the cyclic nucleotide-activated channels in rod photoreceptors
Neuron.
14
1995
177
183
a
[PubMed]
Gordon
S.E.
,
Zagotta
W.N.
Localization of regions affecting an allosteric transition in cyclic nucleotide-activated channels
Neuron.
14
1995
857
864
b
[PubMed]
Gordon
S.E.
,
Varnum
M.D.
,
Zagotta
W.N.
Direct interaction between amino- and carboxyl-terminal domains of cyclic nucleotide-gated channels
Neuron.
19
1997
431
441
[PubMed]
Goulding
E.H.
,
Tibbs
G.R.
,
Siegelbaum
S.A.
Molecular mechanism of cyclic-nucleotide-gated channel activation
Nature.
372
1994
369
374
[PubMed]
Hamill
O.P.
,
Marty
A.
,
Neher
E.
,
Sakmann
B.
,
Sigworth
F.J.
Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches
Pflügers Arch.
391
1981
85
100
Kaupp
U.B.
,
Niidome
T.
,
Tanabe
T.
,
Terada
S.
,
Bönigk
W.
,
Stühmer
W.
,
Cook
N.J.
,
Kangawa
K.
,
Matsuo
H.
,
Hirose
T.
Primary structure and functional expression from complementary DNA of the rod photoreceptor cyclic GMP-gated channel
Nature.
342
1989
762
766
[PubMed]
Körschen
H.G.
,
Illing
M.
,
Seifert
R.
,
Sesti
F.
,
Williams
A.
,
Gotzes
S.
,
Colville
C.
,
Müller
F.
,
Dosé
A.
,
Godde
M.
A 240 kDa protein represents the complete β subunit of the cyclic nucleotide-gated channel from rod photoreceptor
Neuron.
15
1995
627
636
[PubMed]
Paoletti
P.
,
Young
E.C.
,
Siegelbaum
S.
C-linker of cyclic nucleotide-gated channels controls coupling of ligand binding to channel gating
J. Gen. Physiol.
113
1999
17
33
[PubMed]
Shabb
J.B.
,
Corbin
J.D.
Cyclic nucleotide-binding domains in proteins having diverse functions
J. Biol. Chem.
267
1992
5723
5726
[PubMed]
Shapiro
M.S.
,
Zagotta
W.N.
Stoichiometry and arrangement of heteromeric olfactory cyclic nucleotide-gated ion channels
Proc. Natl. Acad. Sci. USA.
95
1998
14546
14551
[PubMed]
Tibbs
G.R.
,
Goulding
E.H.
,
Siegelbaum
S.A.
Allosteric activation and tuning of ligand efficacy in cyclic-nucleotide-gated channels
Nature.
386
1997
612
615
[PubMed]
Varnum
M.D.
,
Zagotta
W.N.
Subunit interactions in the activation of cyclic nucleotide-gated ion channels
Biophys. J.
70
1996
2667
2679
[PubMed]
Varnum
M.D.
,
Zagotta
W.N.
Interdomain interactions underlying activation of cyclic nucleotide-gated channels
Science.
278
1997
110
113
[PubMed]
Zong
X.
,
Zucker
H.
,
Hofmann
F.
,
Biel
M.
Three amino acids in the C-linker are major determinants of gating in cyclic nucleotide-gated channels
EMBO J.
17
1998
353
362
[PubMed]

Abbreviations used in this paper: CL, C-linker; cNMP, cyclic nucleotide monophosphate; EC50, [cGMP] causing half maximum channel activity; NBD, cNMP-binding domain; NT, NH2 terminus; olf, olfactory channel; rod, rod channel.