Probing the Structural Mechanism of Partial Agonism in Glycine

Subscriber access provided by University of Newcastle, Australia

Article

Probing the structural mechanism of partial agonism in glycine receptors using the fluorescent artificial amino acid, ANAP Ming S. Soh, Argel Estrada-Mondragon, Nela Durisic, Angelo Keramidas, and Joseph W Lynch ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.6b00926 • Publication Date (Web): 25 Jan 2017 Downloaded from http://pubs.acs.org on January 26, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Chemical Biology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Biology

TOC graphic 107x128mm (300 x 300 DPI)

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 20

Probing the structural mechanism of partial agonism in glycine receptors using the fluorescent artificial amino acid, ANAP Ming S. Soh1, Argel Estrada-Mondragon1, Nela Durisic1, Angelo Keramidas1 & Joseph W. Lynch1,2* 1

Queensland Brain Institute, The University of Queensland, Brisbane, Queensland 4072, Australia.

2

School of Biomedical Sciences, The University of Queensland, Brisbane, Queensland 4072,

Australia. *Corresponding author: Email: [email protected] Phone: +617 3346 6375

Abstract.

The efficacy of an agonist at a pentameric ligand-gated ion channel is determined by the rate at which it induces a conformational change from the resting closed state to a pre-open (‘flip’) state. If the ability of an agonist to promote this isomerization is sufficiently low, then it becomes a partial agonist. As partial agonists at pentameric ligand-gated ion channels show considerable promise as therapeutics, understanding the structural basis of the resting-flip state isomerization may provide insight into therapeutic design. Accordingly, we sought to identify structural correlates of the resting-flip conformational change in the glycine receptor chloride channel. We used nonsense suppression to introduce the small, fluorescent amino acid, 3-(6-acetylnaphthalen-2-ylamino)-2aminopropanoic acid (ANAP), into specific sites in the extracellular and transmembrane domains. Then, under voltage-clamp conditions in Xenopus oocytes, we simultaneously quantified current and fluorescence responses induced by structurally-similar agonists with high, medium and low efficacies (glycine, β-alanine and taurine, respectively). Analyzing results from nine ANAPincorporated sites, we show that glycine receptor activation by agonists with graded efficacies manifests structurally as correspondingly graded movements of the β1−β2 loop, the β8−β9 loop and the Cys-loop from the extracellular domain and the TM2−TM3 linker in the transmembrane domain. We infer that the resting-flip transition involves an efficacy-dependent molecular reorganization at the extracellular-transmembrane domain interface that primes receptors for efficacious opening.


1

Page 3 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Glycine receptor (GlyR) chloride channels mediate inhibitory neurotransmission in the spinal cord and brainstem. They are members of the pentameric ligand-gated ion channel (pLGIC) family that also includes the nicotinic acetylcholine receptor (nAChR), GABA type-A receptor (GABAAR) and 5-hydroxytryptamine type-3 receptor (5-HT3R). Each pLGIC subunit consists of an extracellular domain, four α-helical transmembrane domains (TM1-TM4), and a large intracellular loop connecting TM3 and TM4. The agonist binding pocket is located at the extracellular subunit interface and is lined by six domains: loops A-C from the (+) side and loops D-F from the (-) side of the interface1. The recently published molecular structures of the α1 and α3 GlyRs in the closed (strychnine-bound) and open (glycine-bound) states2,3,4 now allow us to study GlyR structure and activation mechanisms in unprecedented detail. Resolving pLGIC activation mechanisms is important for understanding how they function in health and disease, and for identifying structural changes that might be exploited during drug development. The initial step in the activation process involves the agonist binding in its pocket5,6. This induces loop C to clamp around the agonist7,8 and induce a global conformational change in the extracellular domain that culminates in the loop connecting the first and second β-strands (i.e., the β1−β2 loop) and the conserved Cys-loop undergoing a differential movement2. These loops protrude towards the transmembrane domain where they intercalate with the TM2-TM3 linker and the pre-TM1 domain. Agonist-binding thereby induces both the TM2-TM3 linker and the pre-TM1 to move radially outwards, allowing TM2 to relax away from the pore to create a patent ion permeation pathway8,9. pLGIC activation thus progresses as a ‘conformational wave’ from the agonist-binding site and to the transmembrane domain5,6. Agonist efficacy (E) is a measure of the ease with which a bound agonist induces channel activation10. Assuming activation proceeds in a single step from the ligand-bound resting closed state, then E = (opening rate constant)/(closing rate constant). This model also predicts that the maximum fraction of receptors in the open state at a high agonist concentration is given by E/(1+E) and the EC50 of the agonist is inversely proportional to (1+E). Thus, if the E values of two agonists are high (>10), then they are both full agonists although the agonist with the lower E exhibits a higher EC50. As E reduces to values β-alanine > taurine23, whereas (in accordance with basic allosteric theory10) their EC50 values follow the inverse sequence: taurine > β-alanine > glycine24-26. The agonists are similar in structure and share a common binding orientation within the α1 GlyR binding site27,28. They also evoke identical single-channel conductances18,23 and induce similar open state configurations of the TM2-TetM3 linker29, suggesting a common open state structure. We interpret our data on the premise that functional differences such as open probabilities and equilibrium constants that govern state transitions18,27 must have physical correlates. We will interpret differences in maximum fluorescence changes (∆Fmax) elicited by different agonists that induce comparable maximum current (∆Imax) values as evidence of these structural correlates. To ACS Paragon Plus Environment

4


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 20

facilitate comparison between agonists, the ∆Imax and ∆Fmax values elicited by β-alanine and taurine are normalized to those elicited by glycine via the relations, RI = (∆Imax(β-ala or tau))/(∆Imax(gly)) and RF = (∆Fmax(β-ala or tau))/(∆Fmax(gly)), respectively, where R refers to ratio and subscripts I and F refer to current and fluorescence, respectively. These values are summarized in Table S4 for all mutants.

Concentration-response experiments confirm the rank order of agonist efficacy in GlyR mutants. We characterized the concentration-response relationships for glycine, β-alanine and taurine at the wild-type (WT) and mutant GlyRs. Sample voltage-clamp recordings and mean concentration-response curves for the α1WT GlyR (Fig. S3A,B, Table S5) confirm that their EC50 values follow the sequence, taurine > β-alanine > glycine. Averaged results for α1L22’ANAP, α1M140ANAP and α1P185ANAP GlyRs (Fig. S3C-E, Table S5) confirm that they exhibit the same rank order of EC50 values. Nevertheless, some mutations (Q-26’ANAP, R19’ANAP and L22’ANAP) also converted β-alanine and taurine into partial agonists, with RI values for β-alanine being greater than taurine (Table S4). This indicates that the mutations do not change the rank order of agonist efficacies, but rather suggests that they impair a common gating transition that reduces the efficacies of all agonists in parallel. Agonist-dependent conformational changes in the β1−β2 loop. When the sulfhydryl-tagged rhodamine derivative, 2-((5(6)-tetramethylrhodamine)carboxylamino)ethyl methanethiosulfonate (MTS-TAMRA), was attached to A52C in the α1 GlyR β1−β2 loop, it reported ∆Fmax values that correlated with agonist efficacy26. We therefore hypothesized that an agonist-specific conformational change occurred near the β1−β2 loop. This loop, which contributes to the interface between the extracellular and transmembrane domains, lies in a crevice surrounded by the conserved Cys-loop, the TM2-TM3 linker and the β8−β9 loop from the adjacent subunit (Fig. 2AC). Fig. 2C shows a comparison of strychnine- and glycine-bound α1 GlyR structures2, allowing us to resolve the nature of glycine-mediated conformational changes in this region. In the closed state, the β1−β2 loop residue, Met56, is sandwiched between Met140 in the Cys-loop and Pro185 in the β8−β9 loop. In the glycine-bound state, the Cys-loop and Pro185 in the β8−β9 loop increase their separation by almost 3 Å and the respective distances between the α-carbons of Met56, Pro185 and Met140 all increase. Simultaneously, the β1−β2 loop undergoes a downward movement towards the TM2-TM3 linker, whereby Thr54 in the β1−β2 loop strengthens its contact with Pro23’ in the TM2-TM3 linker which is thought to help slide the TM2-TM3 away from the central pore axis to open the channel2. To determine whether the magnitudes of these movements are agonist-specific, we measured ∆F responses at receptors that incorporated ANAP at Ala52, Thr54, Met56, Met140 and Pro185. We ACS Paragon Plus Environment

5

Page 7 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


also measured ∆F responses at receptors that incorporated ANAP at Asn203 loop C of the glycinebinding site, and at three sites (Gln-26’, Arg19’ and Leu22’) (Fig. 2A,B) in the transmembrane domain. Our first experiment was to quantitate the ∆I and ∆F concentration-response relationships for glycine, β-alanine and taurine at the α1A52ANAP GlyR. Examples of ∆I and ∆F responses recorded simultaneously together with the averaged concentration-relationships for each agonist are displayed in Fig. 3A-C with mean EC50 and nH values summarized in Table S6. The ∆I and ∆F concentration-response relationships were overlapping for all three agonists. The agonist ∆I EC50 values of α1A52ANAP GlyR were all significantly increased relative to unmutated α1 GlyR values (Tables S4, S5). We then quantified the ∆Imax and ∆Fmax values at the α1T54ANAP and α1M56ANAP GlyRs induced by saturating concentrations of glycine (10 mM), β-alanine (10 mM) and taurine (30 mM). The α1T54ANAP GlyR responded like the α1A52ANAP GlyR with the three agonists evoking similar ∆Imax values while the ∆Fmax values were directly correlated with agonist efficacy (Fig. 3E, Table S4). Notably, the ∆Fmax values at α1M56ANAP GlyRs exhibited the reverse agonist-dependence. That is, taurine produced the largest ∆Fmax, followed by β-alanine and finally glycine (Fig. 3F, Table S4). Collectively, these results support the hypothesis that the magnitude of a conformational change in the vicinity of the β1−β2 loop is correlated with agonist efficacy. This correlation can be positive or negative, depending on the location of the probe.

Agonist-dependent conformational changes in the glycine-binding site. Based on structural analysis of Torpedo nAChRs, Unwin originally proposed that agonist binding induces loop C to clasp around the bound agonist7. This mechanism has recently been confirmed in the α1 GlyR2. It has been proposed that this movement corresponds to an agonist-induced pre-open state in the nAChR30. To determine whether loop C movements are agonist-dependent, we incorporated ANAP at Asn203 at the tip of loop C (Fig. 2A). Examples of ∆I and ∆F concentration-response relationships from individual oocytes expressing α1N203ANAP GlyRs together with the averaged concentration-response curves for each agonist are displayed in Fig. 4A-C with mean EC50 and nH values summarized in Table S6. As was the case for the α1A52ANAP receptors, the ∆I and ∆F concentration-response relationships were overlapping for all three agonists. The agonist sensitivities of α1N203ANAP GlyR were modestly reduced relative to unmutated α1 GlyR values (Tables S4,S5), suggesting minimal interference with agonist binding mechanisms. We observed a downward trend in both ∆Imax and ∆Fmax values from glycine > β-alanine > taurine (Fig. 4D, Table


6


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 20

S4), but as these changes were relatively minor (only the taurine induced ∆Fmax value reached significance) we infer that structural changes as detected by ∆Fmax at loop C are not agonist specific.

Agonist-dependent conformational changes in the Cys-loop and the β8−β9 loop. As outlined above, Met140 (in the Cys-loop) and Pro185 (in the β8−β9 loop) both lie close to Met56 in the β1−β2 loop (Fig. 2C). Given that M56ANAP reports a ∆Fmax that correlates inversely with agonist efficacy (Fig. 3F), we quantified the ∆Imax and ∆Fmax values at both the α1M140ANAP and α1P185ANAP GlyRs induced by saturating concentrations of glycine (10 mM), β-alanine (10 mM) and taurine (30 mM). At the α1M140ANAP GlyR, the ∆Imax values decreased modestly, with those elicited by glycine being the greatest, followed by β-alanine and taurine, whereas the ∆Fmax values exhibited a strong agonist-specific reduction in the same rank order as that for current (Fig. 5A). This result is similar to that observed for the α1A52ANAP GlyR. The α1P185ANAP GlyR responded like the α1M56ANAP GlyR whereby taurine produced the largest ∆Fmax, followed by β-alanine and finally glycine, whereas the respective ∆Imax values were similar (Fig. 5B, Table S4). Thus, both labeled sites reported agonistspecific ∆Fmax changes.

Agonist-dependent conformational changes in the transmembrane domain. It was shown that a disulfide bond between cysteines introduced at Leu22’ and Thr54 induced α1 GlyRs to form a preactivated ‘locally closed’ conformation characterized by spontaneous channel activations and an equalization of glycine, β-alanine and taurine induced current amplitude31. In support of this, the distance between the α-carbons of Thr54 and Leu22’ decreases from 11.6 to 5.5 Å as the channel transitions from the strychnine- to the glycine-bound conformations2. Thus, it seems plausible that agonist efficacy could be encoded via a differential movement of the β1−β2 loop towards Leu22’. We therefore investigated the effects of saturating glycine, β-alanine and taurine on ∆Imax and ∆Fmax responses in the α1L22’ANAP GlyR. The mean ∆Imax and ∆Fmax declined dramatically with in the sequence, glycine > β-alanine > taurine, implying that maximum open probability was proportional to agonist efficacy (Fig. 5C, Table S4). However, normalizing ∆Fmax responses to the respective

∆Imax values to control for the variation in open probability revealed no correlation between agonist efficacy and ∆Fmax at the Leu22’ labeled site. Thus, results from this mutant are not easily interpreted. We have recently demonstrated that Gln-26’ at the top of TM1 and Arg19’ at the top of TM2 interact energetically with each other32,33. The distance between these residues changes during channel opening2, and the resulting change in their interaction energy is thought to make a major contribution to the net energy required to power activation33. Given that Gln-26’ and Arg19’ both ACS Paragon Plus Environment

7

Page 9 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


sense a key final gating conformational change, we introduced ANAP individually at each site to determine whether this conformational change was correlated with agonist efficacy. The effects of saturating agonist concentrations at the α1Q−26’ANAP and α1Q19’ANAP GlyRs are summarized in Fig. 5D,E and Table S4. In both cases, ∆Imax values declined with reducing agonist efficacy. However,

∆Fmax values at the α1Q−26’ANAP GlyR were agonist-independent, whereas they declined with agonist efficacy in α1Q19’ANAP GlyRs. We infer that the TM2-TM3 linker, as demonstrated by the data from α1L22’ANAP and α1R19’ANAP GlyRs encodes the agonist efficacy via differential structural changes. In contrast, the segment containing the Gln-26’ residue is likely downstream of the resting-flip transition.

Discussion and conclusions. We infer that ANAP is incorporated efficiently into functional GlyRs in a site-specific manner. The sites that showed ∆Fmax changes that were proportional to agonist efficacy are all located at the extracellular-transmembrane interface. The observation that the changes in ∆Fmax follow a pattern that corresponds to the equilibrium constants of the resting-open (flip) transition18 argues strongly that the ‘flip’ transition is encoded by differential movements of these segments. Given that pLGIC activation proceeds from the binding pocket to the transmembrane pore5 as a wave of structural rearrangements, we infer that any conformational changes experienced by the segments that contain Asn203 and Gln-26’ flank the movements associated with the resting-flip transition. Indeed, the unchanging ∆Fmax values for all three agonists argues strongly that the segment harboring the Gln-26’ position (extracellular end of TM1) is downstream of the resting-flip transition where the structure of the permeation pathway is similar for all agonists. When attached to Ala52, Thr54 or Met140, ANAP reported ∆Fmax values that were directly correlated with agonist efficacy. In contrast, when attached to Met56 or Pro185, ANAP reported

∆Fmax values that correlated inversely with agonist efficacy. As with a positive correlation, a negative correlation shows that the conformational change is agonist-dependent. Thus, results from these five sites prompt us to conclude that glycine, β-alanine and taurine induced conformational changes that correlated with their agonist efficacies in the vicinity of the β1−β2 loop. Channel activation is driven by energy released by the agonist binding interaction. A full agonist imparts more energy for this process than does a partial agonist34. However, it is not clear for any pLGIC member whether (or how) this energy difference is encoded into distinct conformational changes in the binding site. Agonist-induced closure of loop C is one of the largest conformational changes that occurs in this region2,7 and there is structural evidence from acetylcholine binding proteins for a correlation between agonist efficacy and the degree of loop C


8


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

closure

Page 10 of 20

35

. However, thermodynamic analysis of nAChRs incorporating mutations to loop C

residues found no evidence for loop C closure being relevant to activation36,37. Consistent with this later result, we found no convincing evidence for conformational changes in the α1N203ANAP GlyR that might correlate with agonist efficacy differences. Our data allow some inferences to be drawn about the structural basis of the efficacy-dependent conformational change in the β1−β2 loop. Structural analysis reveals that Met56 in the β1−β2 loop and Pro185 in the β8−β9 loop of the neighbouring subunit are closely apposed in the strychninebound state but move apart in the glycine-bound state (Fig. 2C). Our finding that ANAP attached at either Met56 or Pro185 yielded the same inverse relationship of ∆Imax and ∆Fmax is satisfying as it suggests that ANAP attached at either site reports the same conformational change and implies that the β1−β2 and β8−β9 loops are coupled during activation. A very similar conclusion can be drawn about A52 in the β1−β2 loop and Met140 in the Cys-loop: their sidechains face towards each other, their α-carbons move apart in the open state and ANAP attached at either site reported a ∆Fmax that was directly proportional to agonist efficacy. It is notable that the ‘flip’ transition is concerted. It describes a global conformational switch in the receptor16,18. This would require strong inter-subunit interactions. As the β1−β2 and β8−β9 loops interact across the subunit interface, these movements may be responsible for propagating agonist-mediated conformational changes between subunits, as previously suggested38. As noted above, Thr54 in the β1−β2 loop faces directly towards the TM2-TM3 linker. Indeed, the α-carbon separation between Thr54 and Leu22’ decreases from 11.6 to 5.5 Å as the channel opens2. The importance of this in the GlyR activation process has been revealed by experiments that crosslinked Thr54C and Leu22’C and found that the resulting receptor exhibited similar current amplitudes for glycine, β-alanine and taurine31. This provides additional structural basis of the flip transition and suggests that receptor activation by different agonists may be encoded by a differential movement of the β1−β2 loop towards the TM2-TM3 linker. A recent study proposed that an open state electrostatic interaction between the β1−β2 loop and the pre-TM1 domain stabilizes a GLIC receptor state with high agonist affinity39. These residues correspond to Glu53 and Arg218, respectively, in the α1 GlyR. As the point of closest contact between Arg218 and Glu53 decreases from 6.2 to 5.3 Å as the GlyR opens2, such an interaction seems plausible in the GlyR. Unfortunately, as mutations to Glu217 and Arg218 are poorly tolerated40,41, we were unable to test this interaction directly. However, in a previous conventional VCF study we demonstrated that that the fluorescently tagged Q219C residue elicited identical

∆Imax and ∆Fmax responses to saturating concentrations of glycine, β-alanine and taurine26, arguing against an agonist efficacy-dependent conformational change in this region. In the present study we ACS Paragon Plus Environment

9

Page 11 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


investigated Q-26’ANAP, located eight residues below Arg218 at the top of TM1, and observed no agonist efficacy-dependent ∆Fmax. Thus, the available evidence does not favour an agonistdependent movement towards pre-TM1 as a structural basis of agonist. In conclusion, we propose that a resting-flip state conformation change with a magnitude proportional to the agonist affinity increase from the resting to flip states occurs in the microenvironment of the β1−β2 loop. Our evidence suggests the closed-flip transition involves a lateral conformational change that alters the relative orientations of the β1−β2 loop, the β8−β9 loop and the conserved Cys-loop. We found no evidence that it involves a differential movement of the β1−β2 loop towards the pre-TM1 domain. We also found no evidence for a correlation between agonist efficacy and the degree of loop C closure. As the β1−β2 and β8−β9 loops interact across the subunit interface, these agonist-mediated conformational changes may contribute to the intersubunit cooperativity that is a hallmark of pLGIC activation42,43. Our results are consistent with a model whereby the agonist efficacy-dependent reorganization of the extracellular domain side of the interface promotes the GlyR into a conformation from which channel gating occurs with equivalent high efficacies for the three glycinergic agonists.

Methods Reagents and solutions. The pANAP plasmid was obtained from Addgene (plasmid #48696). ANAP was a kind gift of Dr Stephan Pless (University of Copenhagen). All other reagents were obtained from Sigma. Glycine, β-alanine and taurine were stored as 100 mM stocks in water and dissolved into the recording solution on the day of the experiment. The calcium-free OR-2 solution contained 82.5 mM NaCl, 2 mM KCl, 1 mM MgCl2.6H2O and 5 mM HEPES (pH 7.4). The ND96 storage solution contained 96 mM NaCl, 2 mM KCl, 1 mM MgCl2.6H2O, 1.8 mM CaCl2, 5 mM HEPES, 50 µg/ml gentamicin, 2.5 mM sodium pyruvate and 0.5 mM theophylline (pH 7.4). The ND96 recording solution was similar to this except it did not contain pyruvate, theophylline or gentamicin.

Molecular biology. Human GlyR α1 cDNA was subcloned into the pGEMHE vector. Plasmid DNAs were purified using NucleoBond® Xtra Midi (Macherey-Nagel) or DNA-spinTM Plasmid DNA Extraction Kit (iNtRON). All of the amber stop-codon (TAG) mutants used in this study were produced by site-directed mutagenesis reactions (QuikChange, Stratagene) using the human α1 GlyR sequence as the template. The mRNAs were then produced using the mMESSAGE


10


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 20

mMACHINE Kit (Ambion Inc.). The concentrations of resultant mRNAs were measured using NanoDrop 1000 Spectrophotometer (Thermo Scientific).

Oocyte preparation and receptor expression. All animal handling procedures were approved by the

University

of

Queensland

Animal

Ethics

Committee

(approval

number:

QBI/059/13/ARC/NHMRC). Adult female Xenopus laevis frogs (Xenopus Express) were anaesthetized (1.3 mg/ml MS-222) and oocytes were surgically removed via a small incision to the abdomen. Oocytes were then defolliculated with 1.5 mg/ml collagenase (Sigma) for 2 h. Free oocytes were rinsed with calcium-free OR-2 solution and mature stage V or VI oocytes were isolated and injected first with 200 ng µl-1 of the pANAP plasmid using a Nanoliter 2000 microinjector (WPI Inc.). Injected oocytes were then incubated in the ND96 storage solution at 16 °C. On the following day, 2 mM ANAP and ~17 ng of GlyR mRNA were injected separately into the pANAP-expressed oocytes. The injected volumes ranged from 20–40 nl. Oocytes were used in experiments 2–3 d later.

Voltage-clamp fluorometry. We employed an inverted microscope (Ti-S, Nikon Instruments) equipped with a DAPI filter set (49000, Chroma Technology) and a CFI Fluor 40X water immersion objective (N.A. 0.80, MRF07420, Nikon Instruments). A Lambda LS 175 W xenon arc lamp served as a light source and was coupled to the microscope via a liquid light guide (Sutter Instruments). Fluorescence was detected using a H7360-03 photomultiplier (Hamamatsu Photonics) coupled to a PMT400R photomultiplier subsystem (Ionoptix). Oocytes were placed securely in the bath and were continually perfused with ND96 solution. Agonists were dissolved in the same solution and were applied to the oocytes via a gravity-fed perfusion system. Currents were recorded via two-electrode voltage clamp using 0.2-2 MΩ resistance glass pipettes filled with 3 M KCl. Oocytes were voltage-clamped at −40 mV and currents were recorded using a Gene Clamp 500B amplifier (Molecular Devices). ∆I and ∆F signals were digitized at 2 kHz via a Digidata 1322A interface and pCLAMP 9.2 software (Molecular Devices).

Confocal microscopy and imaging analysis. Fluorescence imaging was carried out on a LSM 510 META (Zeiss) inverted laser scanning microscope in lambda scan mode and equipped with a water immersion objective lens (LD C-Apochromat 40x/1.1 W Corr, Zeiss). The blue diode laser (405 nm) was used as the excitation source and the emission intensity was recorded within a pre-set wavelength range (425–571 nm). Image capture and analysis were performed using ZEN 2009 (Zeiss) and ImageJ software.


11

Page 13 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Data analysis. For analysis and display, ∆F signals were digitally filtered at 1–2 Hz with an eightpole Bessel filter. Concentration-responses were fitted using the Hill equation (Prism 6, GraphPad). Unless stated otherwise, data were compared using Student’s t-test and considered significant when p

Probing the Structural Mechanism of Partial Agonism in Glycine

Recommend Documents