Lipovsek M et al. (2014), Tracking the molecular evolution of calcium per...

XB-ART-51048

Mol Biol Evol 2014 Dec 01;3112:3250-65. doi: 10.1093/molbev/msu258.

Show Gene links Show Anatomy links

Tracking the molecular evolution of calcium permeability in a nicotinic acetylcholine receptor.

Lipovsek M , Fierro A , Pérez EG , Boffi JC , Millar NS , Fuchs PA , Katz E , Elgoyhen AB .

???displayArticle.abstract???
Nicotinic acetylcholine receptors are a family of ligand-gated nonselective cationic channels that participate in fundamental physiological processes at both the central and the peripheral nervous system. The extent of calcium entry through ligand-gated ion channels defines their distinct functions. The α9α10 nicotinic cholinergic receptor, expressed in cochlear hair cells, is a peculiar member of the family as it shows differences in the extent of calcium permeability across species. In particular, mammalian α9α10 receptors are among the ligand-gated ion channels which exhibit the highest calcium selectivity. This acquired differential property provides the unique opportunity of studying how protein function was shaped along evolutionary history, by tracking its evolutionary record and experimentally defining the amino acid changes involved. We have applied a molecular evolution approach of ancestral sequence reconstruction, together with molecular dynamics simulations and an evolutionary-based mutagenesis strategy, in order to trace the molecular events that yielded a high calcium permeable nicotinic α9α10 mammalian receptor. Only three specific amino acid substitutions in the α9 subunit were directly involved. These are located at the extracellular vestibule and at the exit of the channel pore and not at the transmembrane region 2 of the protein as previously thought. Moreover, we show that these three critical substitutions only increase calcium permeability in the context of the mammalian but not the avian receptor, stressing the relevance of overall protein structure on defining functional properties. These results highlight the importance of tracking evolutionarily acquired changes in protein sequence underlying fundamental functional properties of ligand-gated ion channels.

???displayArticle.pubmedLink??? 25193338
???displayArticle.pmcLink??? PMC4245820
???displayArticle.link??? Mol Biol Evol
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
Genes referenced: acta4 adm chrna4 gnl3

???attribute.lit??? ???displayArticles.show???

	Fig. 1. Chick α9 containing receptors do not activate the oocyte IClCa. Representative traces of responses evoked by 100 μM ACh in oocytes expressing either heteromeric (A), homomeric (B) or hybrid (C), nAChRs, before (left—black trace) and after (right—gray trace) a 3-h incubation with the fast calcium chelator BAPTA-AM (Vhold = −70 mV; [Ca2+]extracellular = 1.8 mM). Traces are representative of n = 4–12 per group.
	Fig. 2. Homomeric and hybrid receptors have different relative calcium permeabilities. (A) Representative I–V curves obtained by application of voltage ramps (−120 to +50 mV, 2 s) at the plateau of the response to 100 μM ACh in oocytes superfused with NMG+-based solutions containing different Ca2+ concentrations (0.5 mM, gray and 5 mM, black) for oocytes expressing either rat α9 (upper left), chicken α9 (upper middle), or chicken α10 (upper right) nAChRs. Insets: Magnification near the Erev. (B) Plot of Erev values as a function of extracellular Ca2+ concentration for rat α9, chicken α9, and chicken α10 homomeric nAChRs. Erev values for rat α9α10 (red) and chicken α9α10 (blue) are shown for comparison. Values are mean ± SEM of 4–12 experiments per group. Solid lines are fit to the GHK equation (see Materials and Methods). (C) Idem as in (A) for oocytes expressing rat α9/chicken α10 hybrid nAChRs. (D) Idem as in (B) for the rat α9/chicken α10 hybrid nAChR.
	Fig. 3. The α9 nAChR subunit defines the extent of calcium permeability of α9α10 receptors. (A) Percentage response to 100 μM ACh after 3-h incubation with the fast calcium chelator BAPTA-AM for heteromeric, homomeric, and hybrid nAChRs. The percentage of the initial response remaining after BAPTA incubation was determined for each oocyte individually and then averaged for each receptor. Values are mean ± SEM. (B) Relative calcium permeability (pCa/pNa) as determined from fitting Erev data, as a function of extracellular calcium concentration, to the GHK equation extended to include divalent cations. Values are mean ± SEM. Note that whenever a rat α9 subunit is present calcium permeability is high and whenever a chicken α9 subunit is present the calcium permeability is low.
	Fig. 4. Ancestral sequence reconstruction allocates nonsynonymous substitutions to the mammalian lineage. (A) Amino acid residues present at the 42 clade-specific sites for the rat and chicken α9 extant sequences and eutheria, diapsida, and amniote predicted ancestral α9 sequences. Red, mammalian residue; blue, sauropsid residue; green, tetrapod residue. Note that the majority of the residues present in the predicted ancestral amniote sequence corresponds to sauropsid residues. (B) Structure of the Torpedo californica nAChR (2bg9; Unwin 2005) showing, in blue, the homologous location of the sites that present clade-specific nonsynonymous substitutions and are positioned along the ion conduction pathway. Clade-specific sites K433, L435, K436, T441, N442, S443, and S446 of the MA α-helix are also highlighted in blue. Two subunits were removed for a better view of the channel. The pore-lining transmembrane domain 2 α-helices are highlighted in pink. (C) Schematic phylogeny depicting the evolutionary history of the six sites analyzed by site-directed mutagenesis: 110 and 127 at the extracellular vestibule, and 24′, 11′, 7′, and −4′ at, or flanking the TM2 domain (rat α9 and Miller 1989 numbering). For each site, the character state (amino acid) and the posterior probability of the marginal reconstruction are shown. Red, mammalian residue; blue, sauropsid residue; white, residue present only in monotremes.
	Fig. 5. Mutations in the extracellular vestibule and the exit of the channel pore alter calcium permeability of the rat α9α10 receptor. (A) Representative traces of responses evoked by 100 μM ACh in oocytes expressing rat α9α10 single mutant receptors: rat α9D110Nα10 (left panel), rat α9S127Fα10 (middle panel), and rat α9 A-4′Dα10 (right panel), before (left—black trace) and after (right—gray trace) a 3-h incubation with the fast calcium chelator BAPTA-AM (Vhold = −70 mV; [Ca2+]extracellular = 1.8 mM). Traces are representative of n = 4–6 per group. (B) Plot of Erev values as a function of extracellular Ca2+ concentration for rat α9D110Nα10 (left panel), rat α9S127Fα10 (middle panel), and rat α9 A-4′Dα10 (right panel) single mutant receptors. Erev values for rat α9α10 (red) and chicken α9α10 (blue) are shown for comparison. Values are mean ± SEM of 5–11 experiments per group. Solid lines are fit to the GHK equation (see Materials and Methods).
	Fig. 6. MD simulation shows calcium interaction with residues in the extracellular vestibule. (A) Force associated with hydrated calcium passage through the channel vestibule, as a function of simulation time. Red arrows indicate force peaks that denote interactions with channel residues at different times along the trajectory. (B–D) Structure of the homology model of the rat α9α10 receptor showing the passage of a calcium ion (purple) through the channel vestibule at different simulation times. Residues D110 and S127, analyzed by site-directed mutagenesis, are highlighted in light blue. Other residues interacting with the calcium ion are highlighted in yellow (α9 subunit) and green (α10 subunit). Dark gray, α9 subunits; light gray, α10 subunit. Two α10 subunits and water molecules were removed for a better view of the vestibule. A close-up of the interactions is shown in the middle panels. Dotted blue lines indicate interactions of the calcium ion with channel residues; the corresponding interaction distances are shown alongside in Å. (B) 0.1 ns: calcium interacts with residues D110 and E129 of the α9 subunit and with residues F38 and Y41 of one of the removed α10 subunits. (C) 0.7 ns: calcium interacts with D124 and D125 of the α9 subunit. (D) 1.1 ns: calcium interacts with residues E294 of the α9 subunit and E293 of the α10 subunit, located in the respective TM2–TM3 loops, and with residues N74 of the α9 subunit and D70 and N73 of the α10 subunit, located in the respective β1–β2 loops.
	Fig. 7. The rat α9N110D/S127F/A-4′Dα10 triple mutant shows avian-like low calcium permeability. (A) Representative traces of responses evoked by 100 μM ACh in oocytes expressing the rat α9N110D/S127F/A-4′Dα10 triple mutant receptor, before (left—black trace) and after (right—gray trace) a 3-h incubation with the fast calcium chelator BAPTA-AM (Vhold = −70 mV; [Ca2+]extracellular = 1.8 mM). Traces are representative of n = 6. (B) Plot of Erev values as a function of extracellular Ca2+ concentration for the rat α9N110D/S127F/A-4′Dα10 triple mutant receptor. Erev values for rat α9α10 (red) and chicken α9α10 (blue) are shown for comparison. Values are mean ± SEM of 7–11 experiments per group. Solid lines are fit to the GHK equation (see Materials and Methods).
	Fig. 8. Mammalian-specific mutations do not increase the calcium permeability of the chicken α9α10 receptor. Plot of Erev values as a function of extracellular Ca2+ concentration for chicken α9N110Dα10 and chicken α9F12SFα10 extracellular vestibule mutants (A) and chick α9D-4′Aα10 TM1–TM2 loop mutant and chicken α9D110N/F127S/D-4′Aα10 triple mutant receptor (B). Erev values for rat α9α10 (red) and chicken α9α10 (blue) are shown for comparison. Values are mean ± SEM of 5–11 experiments per group. Solid lines are fit to the GHK equation (see Materials and Methods).
	Fig. 9. Homology modelling shows that point amino acid substitutions in the extracellular domain have long range effects on the electrostatic potential. (A) Homology model of the rat α9α10 wild-type receptor showing, in a gray scale, the electrostatic potential at a longitudinal plane midsagital to the vestibule. Location of the plane is shown on an upper view of the receptor (left panel). Only the subunits at the plane of the section are shown. The dotted black line denotes the location of the central Z axis. The dotted light blue lines denote the location of the three sites of calcium interaction corresponding to 0.1 -, 0.7 - and 1.1-ns simulation times, as depicted in figure 6. Position of the two amino acid substitutions studied is denoted by purple asterisks. (B) Electrostatic potential determined at the central Z axis for rat and chicken α9α10 wild-type and mutant receptors, spanning the entire length of the vestibule. Purple asterisks denote the location of substituted residues 110 and 127.

References [+] :

Baker, Electrostatics of nanosystems: application to microtubules and the ribosome. 2001, Pubmed