Spontaneous Thermal Motion of the GABAA Receptor M2 Channel-lining Segments

The γ-aminobutyric acid type A (GABAA) receptor channel opening involves translational and rotational motions of the five channel-lining, M2 transmembrane segments. The M2 segment's extracellular half is loosely packed and undergoes significant thermal motion. To characterize the extent of the M2 segment's motion, we used disulfide trapping experiments between pairs of engineered cysteines. In α1β1γ2S receptors the single γ subunit is flanked by an α and β subunit. The γ2M2-14′ position is located in the α-γ subunit interface. γ213′ faces the channel lumen. We expressed either the γ214′ or the γ213′ cysteine substitution mutants with α1 cysteine substitution mutants between 12′ and 16′ and wild-type β1. Disulfide bonds formed spontaneously between γ214′C and both α115′C and α116′C and also between γ213′C and α113′C. Oxidation by copper phenanthroline induced disulfide bond formation between γ214′C and α113′C. Disulfide bond formation rates with γ214′C were similar in the presence and absence of GABA, although the rate with α113′C was slower than with the other two positions. In a homology model based on the acetylcholine receptor structure, αM2 would need to rotate in opposite directions by ∼80° to bring α113′ and α115′ into close proximity with γ214′. Alternatively, translational motion of αM2 would reduce the extent of rotational motion necessary to bring these two α subunit residues into close proximity with the γ214′ position. These experiments demonstrate that in the closed state the M2 segments undergo continuous spontaneous motion in the region near the extracellular end of the channel gate. Opening the gate may involve similar but concerted motions of the M2 segments. The γ-aminobutyric acid type A (GABAA) receptor channel opening involves translational and rotational motions of the five channel-lining, M2 transmembrane segments. The M2 segment's extracellular half is loosely packed and undergoes significant thermal motion. To characterize the extent of the M2 segment's motion, we used disulfide trapping experiments between pairs of engineered cysteines. In α1β1γ2S receptors the single γ subunit is flanked by an α and β subunit. The γ2M2-14′ position is located in the α-γ subunit interface. γ213′ faces the channel lumen. We expressed either the γ214′ or the γ213′ cysteine substitution mutants with α1 cysteine substitution mutants between 12′ and 16′ and wild-type β1. Disulfide bonds formed spontaneously between γ214′C and both α115′C and α116′C and also between γ213′C and α113′C. Oxidation by copper phenanthroline induced disulfide bond formation between γ214′C and α113′C. Disulfide bond formation rates with γ214′C were similar in the presence and absence of GABA, although the rate with α113′C was slower than with the other two positions. In a homology model based on the acetylcholine receptor structure, αM2 would need to rotate in opposite directions by ∼80° to bring α113′ and α115′ into close proximity with γ214′. Alternatively, translational motion of αM2 would reduce the extent of rotational motion necessary to bring these two α subunit residues into close proximity with the γ214′ position. These experiments demonstrate that in the closed state the M2 segments undergo continuous spontaneous motion in the region near the extracellular end of the channel gate. Opening the gate may involve similar but concerted motions of the M2 segments. Fast inhibitory neurotransmission in the central nervous system is largely mediated by the GABAA 3The abbreviations used are: GABAAγ-aminobutyric acid type AAChacetylcholineCFFRCa2+-free frog Ringers solutionCu:phencopper phenanthrolineDTTdithiothreitol. and glycine receptors (1Rabow L.E. Russek S.J. Farb D.H. Synapse. 1995; 21: 189-274Crossref PubMed Scopus (458) Google Scholar). The receptors, members of the Cys loop gene superfamily of neurotransmitter-gated ion channels that includes nicotinic acetylcholine (ACh) and serotonin type 3 (5-HT3) receptors (2Karlin A. Nat. Rev. Neurosci. 2002; 3: 102-114Crossref PubMed Scopus (781) Google Scholar, 3Reeves D.C. Lummis S.C. Mol. Membr. Biol. 2002; 19: 11-26Crossref PubMed Scopus (170) Google Scholar, 4Lester H.A. Dibas M.I. Dahan D.S. Leite J.F. Dougherty D.A. Trends Neurosci. 2004; 27: 329-336Abstract Full Text Full Text PDF PubMed Scopus (361) Google Scholar, 5Olsen R.W. Chang C.S. Li G. Hanchar H.J. Wallner M. Biochem. Pharmacol. 2004; 68: 1675-1684Crossref PubMed Scopus (59) Google Scholar), are formed by the assembly of five homologous subunits around the central channel axis. Each subunit has an ∼200 amino acid extracellular N-terminal domain that forms the agonist binding sites and a similarly sized C-terminal domain with four α-helical, transmembrane segments (M1, M2, M3, and M4). For GABAA receptors formed by expression of α, β, and γ subunits, the most common subunit stoichiometry is two α subunits, two β subunits, and one γ subunit (6Chang Y. Wang R. Barot S. Weiss D.S. J. Neurosci. 1996; 16: 5415-5424Crossref PubMed Google Scholar, 7Tretter V. Ehya N. Fuchs K. Sieghart W. J. Neurosci. 1997; 17: 2728-2737Crossref PubMed Google Scholar, 8Baumann S.W. Baur R. Sigel E. J. Biol. Chem. 2001; 276: 36275-36280Abstract Full Text Full Text PDF PubMed Scopus (188) Google Scholar). Viewed from above, the subunits are arranged counterclockwise in the order βαγβα (9Brejc K. van Dijk W.J. Klaassen R.V. Schuurmans M. van Der Oost J. Smit A.B. Sixma T.K. Nature. 2001; 411: 269-276Crossref PubMed Scopus (1579) Google Scholar). γ-aminobutyric acid type A acetylcholine Ca2+-free frog Ringers solution copper phenanthroline dithiothreitol. The GABAA receptor closed state structure is probably similar to the Torpedo ACh receptor structure that has been solved to 4-Å resolution (10Xu M. Akabas M.H. J. Gen. Physiol. 1996; 107: 195-205Crossref PubMed Scopus (179) Google Scholar, 11Teissere J.A. Czajkowski C. J. Neurosci. 2001; 21: 4977-4986Crossref PubMed Google Scholar, 12Cromer B.A. Morton C.J. Parker M.W. Trends Biochem. Sci. 2002; 27: 280-287Abstract Full Text Full Text PDF PubMed Scopus (173) Google Scholar, 13Miyazawa A. Fujiyoshi Y. Unwin N. Nature. 2003; 423: 949-955Crossref PubMed Scopus (1081) Google Scholar, 14Unwin N. J. Mol. Biol. 2005; 346: 967-989Crossref PubMed Scopus (1418) Google Scholar). The ion channel is largely lined by the M2 transmembrane segments that form an inner ring of five α-helices. The inner ring of helices is surrounded by an outer ring of helices formed by the M1, M3, and M4 segments that separate the M2 segments from the lipid bilayer. In the 4-Å resolution ACh receptor structure, the extracellular halves of the M2 segments appear loosely packed, and the narrow region of the channel, inferred to be the channel gate, is between the 9′ and 14′ 4Positions in the M2 segment are identified by a system of index numbers to facilitate comparison between Cys loop receptors (37Miller C. Neuron. 1989; 2: 1195-1205Abstract Full Text PDF PubMed Scopus (150) Google Scholar). In this system the absolutely conserved basic residue at the N-terminal end of M2 is numbered 0′. Residues toward the C-terminus are numbered 1′, 2′, 3′, etc. Residues toward the N-terminus are numbered -1′, -2′, etc. For the 0′, residues are GABAA receptors α1R255, β1R250, and γ2R265. levels (14Unwin N. J. Mol. Biol. 2005; 346: 967-989Crossref PubMed Scopus (1418) Google Scholar). This provides a static picture of the closed state. Information on the dynamic protein motion in the membrane-spanning domain has been obtained using the substituted cysteine accessibility method (15Akabas M.H. Stauffer D.A. Xu M. Karlin A. Science. 1992; 258: 307-310Crossref PubMed Scopus (595) Google Scholar, 16Xu M. Akabas M.H. J. Biol. Chem. 1993; 268: 21505-21508Abstract Full Text PDF PubMed Google Scholar), disulfide trapping (17Horenstein J. Wagner D.A. Czajkowski C. Akabas M.H. Nat. Neurosci. 2001; 4: 477-485Crossref PubMed Scopus (136) Google Scholar), and fluorescence (18Dahan D.S. Dibas M.I. Petersson E.J. Auyeung V.C. Chanda B. Bezanilla F. Dougherty D.A. Lester H.A. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 10195-10200Crossref PubMed Scopus (77) Google Scholar). Substituted cysteine accessibility method studies of the β1 subunit M2 segment in the presence of GABA showed a high degree of accessibility of the residues above 11′, suggesting loose packing and/or high mobility (19Goren E.N. Reeves D.C. Akabas M.H. J. Biol. Chem. 2004; 279: 11198-11205Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar). In contrast, cysteine substituted at the cytoplasmic end of the channel between 2′ and 6′ had the limited accessibility consistent with that region of the channel being tightly packed with low mobility (10Xu M. Akabas M.H. J. Gen. Physiol. 1996; 107: 195-205Crossref PubMed Scopus (179) Google Scholar, 19Goren E.N. Reeves D.C. Akabas M.H. J. Biol. Chem. 2004; 279: 11198-11205Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar). The cytoplasmic end of the channel contains the size and charge selectivity filters (10Xu M. Akabas M.H. J. Gen. Physiol. 1996; 107: 195-205Crossref PubMed Scopus (179) Google Scholar, 4Lester H.A. Dibas M.I. Dahan D.S. Leite J.F. Dougherty D.A. Trends Neurosci. 2004; 27: 329-336Abstract Full Text Full Text PDF PubMed Scopus (361) Google Scholar). Disulfide trapping experiments between engineered Cys residues substituted for aligned M2 residues at the that the M2 segments undergo translational motion the channel in the absence and the presence of GABA J. Akabas M.H. J. Biol. Chem. 2005; Full Text Full Text PDF PubMed Scopus Google Scholar). In receptors disulfide bonds formed between Cys residues substituted for the 9′ and residues (17Horenstein J. Wagner D.A. Czajkowski C. Akabas M.H. Nat. Neurosci. 2001; 4: 477-485Crossref PubMed Scopus (136) Google Scholar). In contrast, at the 6′ we disulfide bond formation in the presence of GABA, but in between Cys residues in The 6′ disulfide bond formation the spontaneous (17Horenstein J. Wagner D.A. Czajkowski C. Akabas M.H. Nat. Neurosci. 2001; 4: 477-485Crossref PubMed Scopus (136) Google Scholar). We that channel involve a of the M2 segments. on both the and 4-Å resolution of the ACh Unwin (14Unwin N. J. Mol. Biol. 2005; 346: 967-989Crossref PubMed Scopus (1418) Google Scholar, N. Nature. 1995; PubMed Scopus Google also that the M2 segments may rotate channel opening but to a In channels has been that the although other been A. Bezanilla F. Nature. PubMed Scopus Google Scholar, Y. A. J. V. M. R. Nature. 2003; 423: PubMed Scopus Google Scholar). we used disulfide trapping experiments to the spontaneous thermal motion of the M2 segments for of rotational motion. The of a of Cys residues to form a disulfide bond on the presence of an and on the The on the between the in the and the in the region of the Cys We used copper phenanthroline to an the formation of and from K. PubMed Scopus Google Scholar). disulfide bond formation that to in were to disulfide bond This is from disulfide bond formation in the by disulfide bond formation with formation is a of the to form disulfide For a disulfide bond to the Cys α to of one J. Mol. Biol. 1992; PubMed Scopus Google Scholar). is to that formation of a disulfide bond that the of the two α is that in the of thermal motion the to this Disulfide trapping has been used to protein mobility and proximity between residues in both and J. Mol. Biol. 1992; PubMed Scopus Google Scholar, A. W. Sci. 3: PubMed Scopus Google Scholar, M. 1995; PubMed Scopus Google Scholar, J. Proc. Natl. Acad. Sci. U. S. A. 1996; PubMed Scopus Google Scholar, Karlin A. J.A. Proc. Natl. Acad. Sci. U. S. A. 2001; PubMed Scopus Google Scholar). For these experiments we of the that in receptors is a single subunit interface. on substituted cysteine accessibility method experiments in the α subunit we that the 14′ residue in the α-γ subunit (10Xu M. Akabas M.H. J. Gen. Physiol. 1996; 107: 195-205Crossref PubMed Scopus (179) Google Scholar). We the of a Cys substituted for γ214′ to form disulfide bonds with Cys substituted for to residues in the absence of We that disulfide bonds form between γ214′ and sites on the αM2 would that αM2 was to the γ214′ position. demonstrate that formed disulfide bonds with Cys residues that are by on the of the α We also the of γ213′C to form disulfide bonds with to α113′C. The γ213′C formed a disulfide bond with α113′C. and cysteine substitution mutants were using (19Goren E.N. Reeves D.C. Akabas M.H. J. Biol. Chem. 2004; 279: 11198-11205Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar). was in using the high was in and at were from were by in type for were in and to with and in with and were with of a of subunit and in for at solution of GABA in was and at of and were in frog solution and and and for than one at A solution of was in and were in to a of and expressed were at in a at with solution at a of The was to the a 3 with 3 had a of were and using a a and were by of GABA by at of to from were to be the between GABA was Disulfide had was for 3 GABA was to the extent of by the of spontaneously formed disulfide The were with for and two or GABA were both reduce a disulfide bond and with high To that the of to the of was to disulfide bond formation and to by one or both of the engineered Cys we the of a of with high but reduce a disulfide to the of a of was by a of GABA to demonstrate that was to the of The extent of to the of spontaneously formed disulfide bonds was by the is the of and is the to The of was in a similar that is the GABA GABA was used for experiments The and for are the of GABA was by a of and a of The of the was than the GABA with or GABA was for 3 a with a GABA was and the was The of with GABA the GABA a The of the between 3 and on the extent of by the the GABA the were to demonstrate that the had to The of was to the induced by GABA the of The were a of to the and with the is the at is the is the and is the at of the were in using a of by the comparison are the of the of the α1 subunit mutants and with wild-type β1 and The of were than of the other mutants but were for the of these of for 3 had on the of of the single α1 Cys mutants We that of the α1 Cys mutants formed spontaneous disulfide bonds between the two Cys residues a receptor or between the engineered Cys residues and the Cys either of the two Cys mutants used in this γ213′C or with type α1 and β1 subunits also and were by a of and Oxidation by a of to receptors with one of the Cys substitution mutants had on the We that of the α1 or Cys mutants formed disulfide bonds with either the engineered or the Cys Disulfide between of of the α1 Cys mutants with γ214′C and type β1 and For two of the Cys and a of the The in was by a of the A of for 3 the to a similar to the the of We that for these two mutants a disulfide bond had formed spontaneously to the of the The disulfide bond be by and by with The to the with that the was to disulfide bond formation and of the of the engineered Cys to order A disulfide bond also formed spontaneously between γ213′C and α113′C the mutants were expressed in of a significant in This was by in a similar to that above To the that the by the to binding by the engineered we for This had on the Disulfide the and that spontaneously form disulfide bonds with we disulfide bond formation be induced in the by A of to a significant of the This was by a of but was by a of We inferred that a disulfide bond be formed between the and A similar of to either or had on GABA This that disulfide bonds be induced in these for Disulfide the for disulfide bond formation between γ214′C and the α1 Cys and with formed disulfide rate were the of the is but the for the of to disulfide bond formation was The were by the of of on the in The were by the GABA and a of and with a single The for disulfide bond formation in the absence of GABA between γ214′C and both α115′C and α116′C were The for disulfide bond formation between γ214′C and α113′C was than that for the other two pairs This slower is consistent with the of spontaneous disulfide bond formation between this of Cys that the is between this of Cys The of disulfide bond formation were also in the presence of The with the γ214′C were by the presence of for disulfide bond formation between cysteine of GABA of is from mutants and between in the presence and absence of GABA for from the of is from mutants and between in the presence and absence of GABA for from the in a The for disulfide bond formation between the aligned γ213′C and was slower than that for the disulfide bonds formed with For this of residues the in the presence of GABA We disulfide bond formation in the presence of GABA is in the or the To that the were limited by solution we the of the in the we the solution from to a high solution and The in to the from to high solution was with a The for of the The for of the The of the solution on a than the of the disulfide bond solution rates the of the disulfide bond formation GABAA receptor channel involves in the M2 segments in order to the channel gate (10Xu M. Akabas M.H. J. Gen. Physiol. 1996; 107: 195-205Crossref PubMed Scopus (179) Google Scholar, 4Lester H.A. Dibas M.I. Dahan D.S. Leite J.F. Dougherty D.A. Trends Neurosci. 2004; 27: 329-336Abstract Full Text Full Text PDF PubMed Scopus (361) Google Scholar, 14Unwin N. J. Mol. Biol. 2005; 346: 967-989Crossref PubMed Scopus (1418) Google Scholar). The 4-Å resolution structure of the homologous ACh receptor provides a static of the closed state structure but into the mobility of the In the region of the the M2 segments undergo significant spontaneous thermal motion, but the extent of this motion in the channel and the of the motion, or or is (17Horenstein J. Wagner D.A. Czajkowski C. Akabas M.H. Nat. Neurosci. 2001; 4: 477-485Crossref PubMed Scopus (136) Google Scholar, J. Akabas M.H. J. Biol. Chem. 2005; Full Text Full Text PDF PubMed Scopus Google Scholar). In the we the extent of spontaneous thermal motion between engineered Cys residues on the M2 segments of subunits at the extracellular end of the channel gate (14Unwin N. J. Mol. Biol. 2005; 346: 967-989Crossref PubMed Scopus (1418) Google using disulfide These experiments showed that disulfide bonds formed spontaneously between a Cys at γ214′ and a Cys at either the α115′ or the position. The disulfide bond formation rates were similar for these two Cys This that the between the γ214′C and the α115′ and were In a by the presence of copper a disulfide bond was also formed between γ214′C and α113′C. The disulfide bond formation rate was slower between these two residues than between the other two Cys suggesting that the between γ214′ and the α113′ position is than that with the other two αM2 positions. In to the disulfide bonds formed by the disulfide bond formation rate for the bond formed between and was in the presence of GABA with that in the absence of In the presence of GABA the channels undergo between the and We in state disulfide bond formation is We that the between aligned at the is in the presence of The extent of the motion of one M2 segment to the M2 segment be from the of the aligned in the 4-Å resolution structure of the homologous ACh receptor (14Unwin N. J. Mol. Biol. 2005; 346: 967-989Crossref PubMed Scopus (1418) Google Scholar). In a disulfide the the is and that the α is R. N. B. C. 1989; 3: PubMed Scopus Google Scholar, J. Mol. Biol. 1992; PubMed Scopus Google Scholar). In the ACh receptor structure the α between the aligned with and and is and from the protein The between γ214′ and α113′ is The are on the used and by to 3 the of disulfide bond suggesting that motions are by the protein This was in the receptor J. Mol. Biol. 1992; PubMed Scopus Google Scholar). the of an α the and 16′ are by an of the and are by an of A rotational motion of the αM2 segment these residues into close proximity with This would the αM2 segment to rotate on from position in the 4-Å structure ∼80° in one for and in the opposite for from the extracellular the M2 the that would bring α115′ into close proximity with γ214′ is than the that would bring α113′ into close with the M2 segments significant rotational motion, we showed that a disulfide bond form between Cys residues substituted at the 6′ position in subunits (17Horenstein J. Wagner D.A. Czajkowski C. Akabas M.H. Nat. Neurosci. 2001; 4: 477-485Crossref PubMed Scopus (136) Google Scholar). the 6′ the disulfide bond the binding subunit. In order for aligned on M2 segments to into close the M2 segments rotate either in one the or in toward are for in the receptor J. Mol. Biol. 1992; PubMed Scopus Google Scholar), channels A. Bezanilla F. Nature. PubMed Scopus Google Scholar), channels Nature. 2001; PubMed Scopus Google Scholar), W. R. J. Mol. Biol. PubMed Scopus Google Scholar), and the Y. A. E. Y. 2004; PubMed Scopus Google Scholar). we are that the to bring the and than is to for motion a Alternatively, a of translational and rotational motion of the M2 segment would reduce the of rotational motion necessary to bring the γ214′ and and into close proximity A motion of the M2 segment from the central channel toward the outer ring of helices a around an other than the axis. a motion has been for ACh receptor A. Fujiyoshi Y. Unwin N. Nature. 2003; 423: 949-955Crossref PubMed Scopus (1081) Google Scholar, 14Unwin N. J. Mol. Biol. 2005; 346: 967-989Crossref PubMed Scopus (1418) Google Scholar). we and that the cytoplasmic end of the M2 in the of the 2′ is tightly packed and undergoes (19Goren E.N. Reeves D.C. Akabas M.H. J. Biol. Chem. 2004; 279: 11198-11205Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar, S. C. J. Biol. Chem. 2004; 279: Full Text Full Text PDF PubMed Scopus Google Scholar). The extracellular end of the M2 segments may channel from the channel the cytoplasmic end a This would the narrow region of the channel between the 9′ and 14′ A. Fujiyoshi Y. Unwin N. Nature. 2003; 423: 949-955Crossref PubMed Scopus (1081) Google Scholar, 14Unwin N. J. Mol. Biol. 2005; 346: 967-989Crossref PubMed Scopus (1418) Google Scholar). of the disulfide bonds between γ214′ and and formed in the absence of in the closed state of the in the state the M2 segments at the of the end of the channel gate are significant thermal motion. in the closed state the M2 segments are spontaneously a motion between closed and state similar to the motion that channel but in a in the absence of an agonist is at that of the M2 segments would be in the state position spontaneous opening is a M2 segments are but into the state in motion of M2 segments in the state of the The motion in the M2 segments in the closed state may be for the of the The channel S. A. Proc. Natl. Acad. Sci. U. S. A. 2005; PubMed Scopus Google and with an between the closed and state with the binding of two GABA the M2 segments were tightly packed with the to the channel be In of other has been that the of a region is to the extent of protein between that region and the of the protein B. Proc. Natl. Acad. Sci. U. S. A. 2002; PubMed Scopus Google Scholar). the that is necessary for of GABAA receptor channels may that in the closed state the extent of protein between the M2 segments be the dynamic motion that we in the and (19Goren E.N. Reeves D.C. Akabas M.H. J. Biol. Chem. 2004; 279: 11198-11205Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar, J. Akabas M.H. J. Biol. Chem. 2005; Full Text Full Text PDF PubMed Scopus Google Scholar). is that the disulfide bond formation rates between γ214′ and and were similar in both the absence and the presence of In the presence of GABA most of the channels are in the state. the of the disulfide bond formation rates may that the dynamic motion of the M2 segments is similar in the closed and The of in the state may be to the In contrast, the disulfide bond between and formed in the presence of the the disulfide bond formation rate was also in the presence of GABA (17Horenstein J. Wagner D.A. Czajkowski C. Akabas M.H. Nat. Neurosci. 2001; 4: 477-485Crossref PubMed Scopus (136) Google Scholar). is in the region of the channel the rates for residues be similar in both the absence and the presence of GABA, between other pairs of residues the rates be in the presence of This that although the extent of thermal motion of the M2 segments in the closed and may be the channel structure of these two may be the γ213′C to α113′C disulfide we that is between the be between the α subunit and the γ subunit these are residues (10Xu M. Akabas M.H. J. Gen. Physiol. 1996; 107: 195-205Crossref PubMed Scopus (179) Google Scholar, 19Goren E.N. Reeves D.C. Akabas M.H. J. Biol. Chem. 2004; 279: 11198-11205Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar). In we in the presence of GABA, this disulfide is in the or the between γ214′C and is the of residues that formed disulfide bonds with γ214′C and we of disulfide bond formation between these engineered cysteine be with Disulfide trapping pairs of that the in the engineered to with and to disulfide bond formation to are that a from with or in the to rates of disulfide bond In these and experiments that the M2 segments between the and levels undergo continuous spontaneous motion in the closed state. The structure is This motion to involve both rotational and translational These may be to the that the subunits undergo channel opening that may involve the of the M2 segments from the channel axis. The extent of thermal motion is consistent with the loose packing of the extracellular halves of the M2 segments in the structure of the homologous ACh receptor B. Proc. Natl. Acad. Sci. U. S. A. 2002; PubMed Scopus Google Scholar). studies to the in opening the channel gate. We for and and for and also for on this