Structural rearrangements underlying K+-channel activation gating

Science

Volumn 285, Issue 5424, 1999, Pages 73-78

  Perozo, Eduardo ^a   Cortes, D Marien ^a   Cuello, Luis G ^a  

^a UNIVERSITY OF VIRGINIA HEALTH SYSTEM   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

POTASSIUM CHANNEL;

ARTICLE; CHANNEL GATING; CHROMATOGRAPHY; MEMBRANE CHANNEL; PH; PRIORITY JOURNAL; SITE DIRECTED MUTAGENESIS; STREPTOMYCES LIVIDANS; STRUCTURE ACTIVITY RELATION;

BACTERIAL PROTEINS; BINDING SITES; CIRCULAR DICHROISM; CYSTEINE; ELECTRON SPIN RESONANCE SPECTROSCOPY; HYDROGEN-ION CONCENTRATION; ION CHANNEL GATING; MODELS, MOLECULAR; POTASSIUM; POTASSIUM CHANNELS; PROTEIN CONFORMATION; PROTEIN STRUCTURE, SECONDARY; RUBIDIUM; SEQUENCE DELETION; STREPTOMYCES;

BACTERIA (MICROORGANISMS); STREPTOMYCES; STREPTOMYCES LIVIDANS;

EID: 0033516494     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.285.5424.73     Document Type: Article

References (49)

1. B. Hille, Ion Channels of Excitable Membranes (Sinaucf, Sunderland, MA, ed 2, 1992); D. J. Aidley and P. R. Stanfield, Ion Channels. Molecules in Action (Cambridge Univ. Press, Cambridge, ed. 1, 1996).
2. B. Hille, Ion Channels of Excitable Membranes (Sinaucf, Sunderland, MA, ed 2, 1992); D. J. Aidley and P. R. Stanfield, Ion Channels. Molecules in Action (Cambridge Univ. Press, Cambridge, ed. 1, 1996).
3. H. Schrempf et al., EMBO J. 14, 5170 (1995).
4. D. A. Doyle et al., Science 280, 69 (1998).
5. L. G. Cuello, J. G. Romero, D. M. Cortes, E. Perozo, Biochemistry 37, 3229 (1998).
6. Although not done at low pH, channel opening appears to depend also on the surrounding lipids [L Heginbotham et al., J. Gen. Physiol. 111, 741 (1998)].
7. C. M. Armstrong, ibid. 54, 553 (1969); ibid. 58, 413 (1971).
8. C. M. Armstrong, ibid. 54, 553 (1969); ibid. 58, 413 (1971).
9. M. Holmgren, P. L. Smith, G. Yellen, ibid. 109, 527 (1997); Y. Liu, M. Holmgren, M. E. Jurman, G. Yellen, Neuron 19, 175 (1997).
10. M. Holmgren, P. L. Smith, G. Yellen, ibid. 109, 527 (1997); Y. Liu, M. Holmgren, M. E. Jurman, G. Yellen, Neuron 19, 175 (1997).
11. Since its development by Hubbell and collaborators [C. Altenbach et al., Science 248, 1088 (1990)],
12. site-directed spin labeling and EPR spectroscopy has emerged as a very powerful approach for studying the structural dynamics of proteins, particularly membrane proteins. It is exquisitely sensitive to changes in mobility, solvent accessibility, and inter-spin distances; see [W. L Hubbell et al., Curr. Opin. Struct. Biol. 8, 649 (1998)].
13. E. Perozo, L. Cuello, D. Cortes, Nature Struct. Biol. 5, 459 (1998).
14. + uptake.
15. D. M. Cortes, L. G. Cuello, E. Perozo, paper presented at the 43rd Annual Meeting of the Biophysical Society, Baltimore, MD, 13 February 1999.
16. G. Chang, R. Spencer, A. Lee, M. Barclay, D. Rees, Science 282, 2220 (1998).
17. L. G. Cuello, C. P. Ptak, E. Perozo, paper presented at the 43rd Annual Meeting of the Biophysical Society, Baltimore, MD, 14 February 1999.
18. P. N. Unwin and P. D. Ennis, Nature 307, 609 (1984); N. Unwin, ibid. 373, 37 (1995).
19. P. N. Unwin and P. D. Ennis, Nature 307, 609 (1984); N. Unwin, ibid. 373, 37 (1995).
20. D. L. Farrens et al., Science 274, 768 (1996).
21. S. K. Aggarwal and R. MacKinnon, Neuron 16, 1169 (1996).
22. CD spectra were obtained between 190 and 250 nm (1-nm interval) in a Jasco J720 spectropolarimeter with a 0.1-mm path length. Samples were in 10% phosphate-buffered saline with 1 mM DDM. We used an average of 10 to 15 individual spectra for each sample. We quantified secondary structures by the self-consistent method [N. Sreerama and R. W. Woody, Anal. Biochem. 209, 32 (1993)], using the program Dicroprot [G. Deleage and C. Geourjon, Comp. Appl. Biosci. 9, 197 (1993)].
23. CD spectra were obtained between 190 and 250 nm (1-nm interval) in a Jasco J720 spectropolarimeter with a 0.1-mm path length. Samples were in 10% phosphate-buffered saline with 1 mM DDM. We used an average of 10 to 15 individual spectra for each sample. We quantified secondary structures by the self-consistent method [N. Sreerama and R. W. Woody, Anal. Biochem. 209, 32 (1993)], using the program Dicroprot [G. Deleage and C. Geourjon, Comp. Appl. Biosci. 9, 197 (1993)].
24. D. M. Cortes and E. Perozo, Biochemistry 36, 10343 (1997).
25. S. A. Tatulian, D. M. Cortes, E. Perozo, FEBS Lett. 423, 205 (1998); J. le Coutre et al., Proc. Natl. Acad. Sci. U.S.A. 95, 6114 (1998).
26. S. A. Tatulian, D. M. Cortes, E. Perozo, FEBS Lett. 423, 205 (1998); J. le Coutre et al., Proc. Natl. Acad. Sci. U.S.A. 95, 6114 (1998).
27. Each mutant was reconstituted into asolectin liposomes at a 500:1 lipid-to-protein ratio (mol/mol). We changed the pH by resuspending each sample in a 10-fold volume excess of buffer (50 mM citrate phosphate buffer), followed by centrifugation at 10,000g in an Airfuge. We obtained X-band continuous wave (CW) EPR spectra by using a loop-gap resonator with 2-mW incident power, 100-kHz modulation frequency, and 1-G modulation amplitude.
28. These spectra contain information about the motional freedom of the spin label and how it is affected by local steric restrictions. In the absence of a formal line shape analysis for each spectrum, the parameter ΔHo, representing the width of the central resonance line, has been successfully used as an empiric measure of probe mobility [H. S. Mchaourab et al., Biochemisty 35, 7692 (1996)]. An increase in ΔHo (a positive ΔΔHo) signals a decrease of motional freedom, perhaps from newly formed tertiary or quaternary contacts. Similarly, a decrease in ΔHo (negative ΔΔHo) indicates an increase in the probe's freedom of movement. For a sample containing multiple spin labels, the spectral line shape can also be affected by the extent of trough-space spin-spin dipolar coupling. Such interactions produce spectral broadening in a distance-dependent manner [G. R. Eaton and S. S. Eaton, in Spin Labeling. Theory and Applications, L. J. Berliner and J. Reuben, Eds. (Plenum, New York, 1989)]. At room temperature, this occurs proportionally for distances as large as 15 to 16 Å and as small as 7 to 8 Å, at which point direct, collisional spin exchange will likely occur. In systems of two spins, these interactions have been used with impressive accuracy to directly estimate interspin distances [M. D. Rabenstein and Y. K. Shin, Proc. Nat. Acad. Sci. U.S.A. 92, 8239 (1995); E. J. Hustedt et al., Biophys. J. 72, 1861 (1997); H. S. Mchaourab et al., Biochemistry 36, 307 (1997); H. J. Steinhoff et al., Biophys. J. 73, 3287 (1997)], but in a system with fourfold symmetry like KcsA dipolar interactions cannot be reliably translated into actual physical distances. Instead, spectral broadening can be used to estimate changes in overall probe proximity in the form of the Ω parameter (9). This is operationally defined as Ω = Ā*/Ā°, where Ā* is the amplitude of the central resonance line (M = 0), normalized to the total number of spins in the sample, of a fully labeled mutant (containing two or more labels), and Ā° is the amplitude of the central resonance line, also normalized to the total number of spins in the sample, of an underlabeled mutant (containing only one label per tetramer). In this report, we have considered a relalive Ω parameter by comparing fully labeled channels in two conformational states, closed and open. Note that under these conditions, the magnitude of Ω can be affected by large changes in probe mobility.
29. These spectra contain information about the motional freedom of the spin label and how it is affected by local steric restrictions. In the absence of a formal line shape analysis for each spectrum, the parameter ΔHo, representing the width of the central resonance line, has been successfully used as an empiric measure of probe mobility [H. S. Mchaourab et al., Biochemisty 35, 7692 (1996)]. An increase in ΔHo (a positive ΔΔHo) signals a decrease of motional freedom, perhaps from newly formed tertiary or quaternary contacts. Similarly, a decrease in ΔHo (negative ΔΔHo) indicates an increase in the probe's freedom of movement. For a sample containing multiple spin labels, the spectral line shape can also be affected by the extent of trough-space spin-spin dipolar coupling. Such interactions produce spectral broadening in a distance-dependent manner [G. R. Eaton and S. S. Eaton, in Spin Labeling. Theory and Applications, L. J. Berliner and J. Reuben, Eds. (Plenum, New York, 1989)]. At room temperature, this occurs proportionally for distances as large as 15 to 16 Å and as small as 7 to 8 Å, at which point direct, collisional spin exchange will likely occur. In systems of two spins, these interactions have been used with impressive accuracy to directly estimate interspin distances [M. D. Rabenstein and Y. K. Shin, Proc. Nat. Acad. Sci. U.S.A. 92, 8239 (1995); E. J. Hustedt et al., Biophys. J. 72, 1861 (1997); H. S. Mchaourab et al., Biochemistry 36, 307 (1997); H. J. Steinhoff et al., Biophys. J. 73, 3287 (1997)], but in a system with fourfold symmetry like KcsA dipolar interactions cannot be reliably translated into actual physical distances. Instead, spectral broadening can be used to estimate changes in overall probe proximity in the form of the Ω parameter (9). This is operationally defined as Ω = Ā*/Ā°, where Ā* is the amplitude of the central resonance line (M = 0), normalized to the total number of spins in the sample, of a fully labeled mutant (containing two or more labels), and Ā° is the amplitude of the central resonance line, also normalized to the total number of spins in the sample, of an underlabeled mutant (containing only one label per tetramer). In this report, we have considered a relalive Ω parameter by comparing fully labeled channels in two conformational states, closed and open. Note that under these conditions, the magnitude of Ω can be affected by large changes in probe mobility.
30. These spectra contain information about the motional freedom of the spin label and how it is affected by local steric restrictions. In the absence of a formal line shape analysis for each spectrum, the parameter ΔHo, representing the width of the central resonance line, has been successfully used as an empiric measure of probe mobility [H. S. Mchaourab et al., Biochemisty 35, 7692 (1996)]. An increase in ΔHo (a positive ΔΔHo) signals a decrease of motional freedom, perhaps from newly formed tertiary or quaternary contacts. Similarly, a decrease in ΔHo (negative ΔΔHo) indicates an increase in the probe's freedom of movement. For a sample containing multiple spin labels, the spectral line shape can also be affected by the extent of trough-space spin-spin dipolar coupling. Such interactions produce spectral broadening in a distance-dependent manner [G. R. Eaton and S. S. Eaton, in Spin Labeling. Theory and Applications, L. J. Berliner and J. Reuben, Eds. (Plenum, New York, 1989)]. At room temperature, this occurs proportionally for distances as large as 15 to 16 Å and as small as 7 to 8 Å, at which point direct, collisional spin exchange will likely occur. In systems of two spins, these interactions have been used with impressive accuracy to directly estimate interspin distances [M. D. Rabenstein and Y. K. Shin, Proc. Nat. Acad. Sci. U.S.A. 92, 8239 (1995); E. J. Hustedt et al., Biophys. J. 72, 1861 (1997); H. S. Mchaourab et al., Biochemistry 36, 307 (1997); H. J. Steinhoff et al., Biophys. J. 73, 3287 (1997)], but in a system with fourfold symmetry like KcsA dipolar interactions cannot be reliably translated into actual physical distances. Instead, spectral broadening can be used to estimate changes in overall probe proximity in the form of the Ω parameter (9). This is operationally defined as Ω = Ā*/Ā°, where Ā* is the amplitude of the central resonance line (M = 0), normalized to the total number of spins in the sample, of a fully labeled mutant (containing two or more labels), and Ā° is the amplitude of the central resonance line, also normalized to the total number of spins in the sample, of an underlabeled mutant (containing only one label per tetramer). In this report, we have considered a relalive Ω parameter by comparing fully labeled channels in two conformational states, closed and open. Note that under these conditions, the magnitude of Ω can be affected by large changes in probe mobility.
31. These spectra contain information about the motional freedom of the spin label and how it is affected by local steric restrictions. In the absence of a formal line shape analysis for each spectrum, the parameter ΔHo, representing the width of the central resonance line, has been successfully used as an empiric measure of probe mobility [H. S. Mchaourab et al., Biochemisty 35, 7692 (1996)]. An increase in ΔHo (a positive ΔΔHo) signals a decrease of motional freedom, perhaps from newly formed tertiary or quaternary contacts. Similarly, a decrease in ΔHo (negative ΔΔHo) indicates an increase in the probe's freedom of movement. For a sample containing multiple spin labels, the spectral line shape can also be affected by the extent of trough-space spin-spin dipolar coupling. Such interactions produce spectral broadening in a distance-dependent manner [G. R. Eaton and S. S. Eaton, in Spin Labeling. Theory and Applications, L. J. Berliner and J. Reuben, Eds. (Plenum, New York, 1989)]. At room temperature, this occurs proportionally for distances as large as 15 to 16 Å and as small as 7 to 8 Å, at which point direct, collisional spin exchange will likely occur. In systems of two spins, these interactions have been used with impressive accuracy to directly estimate interspin distances [M. D. Rabenstein and Y. K. Shin, Proc. Nat. Acad. Sci. U.S.A. 92, 8239 (1995); E. J. Hustedt et al., Biophys. J. 72, 1861 (1997); H. S. Mchaourab et al., Biochemistry 36, 307 (1997); H. J. Steinhoff et al., Biophys. J. 73, 3287 (1997)], but in a system with fourfold symmetry like KcsA dipolar interactions cannot be reliably translated into actual physical distances. Instead, spectral broadening can be used to estimate changes in overall probe proximity in the form of the Ω parameter (9). This is operationally defined as Ω = Ā*/Ā°, where Ā* is the amplitude of the central resonance line (M = 0), normalized to the total number of spins in the sample, of a fully labeled mutant (containing two or more labels), and Ā° is the amplitude of the central resonance line, also normalized to the total number of spins in the sample, of an underlabeled mutant (containing only one label per tetramer). In this report, we have considered a relalive Ω parameter by comparing fully labeled channels in two conformational states, closed and open. Note that under these conditions, the magnitude of Ω can be affected by large changes in probe mobility.
32. These spectra contain information about the motional freedom of the spin label and how it is affected by local steric restrictions. In the absence of a formal line shape analysis for each spectrum, the parameter ΔHo, representing the width of the central resonance line, has been successfully used as an empiric measure of probe mobility [H. S. Mchaourab et al., Biochemisty 35, 7692 (1996)]. An increase in ΔHo (a positive ΔΔHo) signals a decrease of motional freedom, perhaps from newly formed tertiary or quaternary contacts. Similarly, a decrease in ΔHo (negative ΔΔHo) indicates an increase in the probe's freedom of movement. For a sample containing multiple spin labels, the spectral line shape can also be affected by the extent of trough-space spin-spin dipolar coupling. Such interactions produce spectral broadening in a distance-dependent manner [G. R. Eaton and S. S. Eaton, in Spin Labeling. Theory and Applications, L. J. Berliner and J. Reuben, Eds. (Plenum, New York, 1989)]. At room temperature, this occurs proportionally for distances as large as 15 to 16 Å and as small as 7 to 8 Å, at which point direct, collisional spin exchange will likely occur. In systems of two spins, these interactions have been used with impressive accuracy to directly estimate interspin distances [M. D. Rabenstein and Y. K. Shin, Proc. Nat. Acad. Sci. U.S.A. 92, 8239 (1995); E. J. Hustedt et al., Biophys. J. 72, 1861 (1997); H. S. Mchaourab et al., Biochemistry 36, 307 (1997); H. J. Steinhoff et al., Biophys. J. 73, 3287 (1997)], but in a system with fourfold symmetry like KcsA dipolar interactions cannot be reliably translated into actual physical distances. Instead, spectral broadening can be used to estimate changes in overall probe proximity in the form of the Ω parameter (9). This is operationally defined as Ω = Ā*/Ā°, where Ā* is the amplitude of the central resonance line (M = 0), normalized to the total number of spins in the sample, of a fully labeled mutant (containing two or more labels), and Ā° is the amplitude of the central resonance line, also normalized to the total number of spins in the sample, of an underlabeled mutant (containing only one label per tetramer). In this report, we have considered a relalive Ω parameter by comparing fully labeled channels in two conformational states, closed and open. Note that under these conditions, the magnitude of Ω can be affected by large changes in probe mobility.
33. These spectra contain information about the motional freedom of the spin label and how it is affected by local steric restrictions. In the absence of a formal line shape analysis for each spectrum, the parameter ΔHo, representing the width of the central resonance line, has been successfully used as an empiric measure of probe mobility [H. S. Mchaourab et al., Biochemisty 35, 7692 (1996)]. An increase in ΔHo (a positive ΔΔHo) signals a decrease of motional freedom, perhaps from newly formed tertiary or quaternary contacts. Similarly, a decrease in ΔHo (negative ΔΔHo) indicates an increase in the probe's freedom of movement. For a sample containing multiple spin labels, the spectral line shape can also be affected by the extent of trough-space spin-spin dipolar coupling. Such interactions produce spectral broadening in a distance-dependent manner [G. R. Eaton and S. S. Eaton, in Spin Labeling. Theory and Applications, L. J. Berliner and J. Reuben, Eds. (Plenum, New York, 1989)]. At room temperature, this occurs proportionally for distances as large as 15 to 16 Å and as small as 7 to 8 Å, at which point direct, collisional spin exchange will likely occur. In systems of two spins, these interactions have been used with impressive accuracy to directly estimate interspin distances [M. D. Rabenstein and Y. K. Shin, Proc. Nat. Acad. Sci. U.S.A. 92, 8239 (1995); E. J. Hustedt et al., Biophys. J. 72, 1861 (1997); H. S. Mchaourab et al., Biochemistry 36, 307 (1997); H. J. Steinhoff et al., Biophys. J. 73, 3287 (1997)], but in a system with fourfold symmetry like KcsA dipolar interactions cannot be reliably translated into actual physical distances. Instead, spectral broadening can be used to estimate changes in overall probe proximity in the form of the Ω parameter (9). This is operationally defined as Ω = Ā*/Ā°, where Ā* is the amplitude of the central resonance line (M = 0), normalized to the total number of spins in the sample, of a fully labeled mutant (containing two or more labels), and Ā° is the amplitude of the central resonance line, also normalized to the total number of spins in the sample, of an underlabeled mutant (containing only one label per tetramer). In this report, we have considered a relalive Ω parameter by comparing fully labeled channels in two conformational states, closed and open. Note that under these conditions, the magnitude of Ω can be affected by large changes in probe mobility.
34. We obtained power spectra by applying the discrete Fourier transform evaluated for the value ω = ω̂that maximizes P(ω) [J. L. Cornette et al., J. Mol. Biol. 195, 659 (1987); D. Donnelly, J. P. Overington, T. L. Blundell, Prot. Eng. 7, 645 (1994); see (9)].
35. We obtained power spectra by applying the discrete Fourier transform evaluated for the value ω = ω̂that maximizes P(ω) [J. L. Cornette et al., J. Mol. Biol. 195, 659 (1987); D. Donnelly, J. P. Overington, T. L. Blundell, Prot. Eng. 7, 645 (1994); see (9)].
36. In a rigid body movement of an a helix , the overall α periodicity of a given structural property (ΔHo) is normally preserved. Rotations generate a mobility profile that is shifted along the residue axis, whereas tilts and dissociations primarily affect the side of the helix involved in tertiary or quaternary contacts. In either case, the power spectra will show a prominent peak near 100°. If, on the other hand, there are changes in secondary structure, the relative area of the peak near 100° will decrease according to the magnitude of the loss in helical periodicity.
37. This is the resultant vector from the sum of all ΔHo values in polar coordinates calculated from the equation (Equation presented) [D. Eisenberg et al., Proc. Natl. Acad. Sci. U.S.A. 81, 140 (1984)]. The angle θ was obtained as the value of the resultant M(ω) evaluated at ω = 100°, taking an arbitrary residue as a reference point (θ = 0). Δθ was calculated as the angular difference between θ at neutral pH and θ at acidic pH.
38. Abbreviations for amino acid residues are as follows: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; and Y, Tyr.
39. This is because the module of the individual vectors (either ΔHo or Ω) is affected by nonstructurat factors: Ω is nonlinearty dependent with distance and is affected in an unknown way by the presence of the four interacting spin labels; ΔHo values are closely related to the local steric environment at each individual position, which obviously changes upon channel opening. Therefore, the value of the Δθ angle should be taken as a qualitative estimate of the magnitude of the rotation.
40. 1 represents the channel conformation with all four subunits in the resting state and each kinetic step is associated with the change of a given subunit to the active state, leading to the fully open state O. This scheme was used recently to describe the gating kinetics of glutamate receptor channels [C. Rosenmund et al., Science 280, 1596 (1998)].
41. 2 transition represents the measured conformational change that leads to a final, concerted transition to the open state. [F. Bezanilla et al., Biophys. J. 66, 1011 (1994); W. N. Zagotta et al., J. Gen. Physiol. 103, 321 (1994)].
42. 2 transition represents the measured conformational change that leads to a final, concerted transition to the open state. [F. Bezanilla et al., Biophys. J. 66, 1011 (1994); W. N. Zagotta et al., J. Gen. Physiol. 103, 321 (1994)].
43. C. S. Anderson, R. MacKinnon, C. Smith, C. Miller, J. Gen. Physiol. 91, 317 (1988).
44. F. J. Sigworth, Q. Rev. Biophys. 27, 1 (1994).
45. G. Yellen, ibid., in press.
46. M. L. Chapman et al., Biophys. J. 72, 708 (1997); J. Zheng and F. J. Sigworth, J. Gen. Physiol. 110, 101 (1997).
47. M. L. Chapman et al., Biophys. J. 72, 708 (1997); J. Zheng and F. J. Sigworth, J. Gen. Physiol. 110, 101 (1997).
48. A. Nicholls, K. A. Sharp, B. Honig, Proteins 11, 281 (1991).
49. Support by the National Institutes of Health (grants GM54690 and GM57846) is gratefully acknowledged. We are indebted to R. Nakamoto, M. Wiener, and H. Mchaourab and to the members of the Perozo Laboratory (Y.-S. Liu, C. Ptak, and A. Fay) for insightful discussions and to R. Nakamoto, G. Szabo, H. Mchaourab, and D. Cafiso for comments on the manuscript. R. MacKinnon kindly provided the C-terminal His-tagged KcsA construct and C. Ptak participated in some of the radiotracer flux experiments. E.P. is particularly grateful to S. Mochel for support and encouragement. E.P. is a McKnight Scholar.