Kawai T et al. (2021), Heterologous functional expression of ascidian ...

XB-ART-58112

J Biol Chem 2021 Jan 01;296:100783. doi: 10.1016/j.jbc.2021.100783.

Show Gene links Show Anatomy links

Heterologous functional expression of ascidian Nav1 channels and close relationship with the evolutionary ancestor of vertebrate Nav channels.

Kawai T , Hashimoto M , Eguchi N , Nishino JM , Jinno Y , Mori-Kreiner R , Aspåker M , Chiba D , Ohtsuka Y , Kawanabe A , Nishino AS , Okamura Y .

???displayArticle.abstract???
Voltage-gated sodium channels (Nav1s) are responsible for the initiation and propagation of action potentials in neurons, muscle, and endocrine cells. Many clinically used drugs such as local anesthetics and antiarrhythmics inhibit Nav1s, and a variety of inherited human disorders are caused by mutations in Nav1 genes. Nav1s consist of the main α subunit and several auxiliary β subunits. Detailed information on the structure-function relationships of Nav1 subunits has been obtained through heterologous expression experiments and analyses of protein structures. The basic properties of Nav1s, including their gating and ion permeation, were classically described in the squid giant axon and other invertebrates. However, heterologous functional expression of Nav1s from marine invertebrates has been unsuccessful. Ascidians belong to the Urochordata, a sister group of vertebrates, and the larval central nervous system of ascidians shows a similar plan to that of vertebrates. Here, we report the biophysical properties of ascidian Ciona Nav1 (CiNav1a) heterologously expressed in Xenopus oocytes. CiNav1a exhibited tetrodotoxin-insensitive sodium currents with rapid gating kinetics of activation and inactivation. Furthermore, consistent with the fact that the Ciona genome lacks orthologous genes to vertebrate β subunits, the human β1 subunit did not influence the gating properties when coexpressed with CiNav1a. Interestingly, CiNav1a contains an ankyrin-binding motif in the II-III linker, which can be targeted to the axon initial segment of mammalian cortical neurons. Our findings provide a platform to gain insight into the evolutionary and biophysical properties of Nav1s, which are important for the development of targeted therapeutics.

???displayArticle.pubmedLink??? 34000300
???displayArticle.pmcLink??? PMC8192821
???displayArticle.link??? J Biol Chem

Species referenced: Xenopus laevis
Genes referenced: ank1 ids map2 nav1 nav2 sh2b2
GO keywords: action potential [+]

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

	Figure 1Molecular phylogenetic characteristics of ascidian Nav1 homologs. A, A molecular phylogenetic tree of Nav1 α subunits. Sequences with homology were collected from public databases to prepare a gap-free alignment of 1148 amino acids. The “maximum-likelihood” tree with the highest log likelihood is depicted. The results of 100 replicates of bootstrap analysis were also depicted, only when the values were larger than 70. NCBI accession numbers of the sequences and genus names from which the sequences were derived are shown. Nav2 channel sequences were used as outgroup. Ciona has three Nav1-like sequences that are closely related to vertebrate Nav1s. 1 and 2 indicate the sequences that have been identified as Nav1-like genemodels in Oikopleura dioica (appendicularian tunicate) genome database (https://www.aniseed.cnrs.fr/). Genemodel IDs of 1 and 2 are OD_K25COV10_DN16743_c0_g1_i1 and OD_K25COV10_DN18236_c0_g1_i7, respectively. Ciona Nav1a is highlighted with arrowhead. B, Consensus amino acid sequences in critical regions, including pore turrets in S5-S6 loops from domain I-IV that determine ion selectivity (black arrowheads) and another associated lining of acidic amino acids (white arrowheads), ankyrin binding motif region in the II-III linker, inactivation latch in the III-IV linker. The amino acids at the pore turrets are highlighted with color [Asp (D), magenta; Glu (E), red; Lys (K), blue; Ala (A), orange; Thr (T), light blue; Met (M), yellow. The core triplet of the inactivation latch (I-F-M) are marked by purple. Amino acids identical to the mammalian Nav1.1 ankyrin-binding motif are indicated by green letters. Cyan residues indicate the amino acids identical to those around the core triplet in the mammalian Nav1.1.
	Figure 2TTX-insensitive sodium current from CiNav1a expressed in Xenopus oocyte. A, Raw traces and I-V curve of CiNav1a by TEVC recording. Only cRNA encoding CiNav1a was microinjected into Xenopus oocyte. Holding potential was -100 mV and depolarizing pulse in 50 ms was stepped to various levels (-90 to 40 mV). The current was isolated by leak subtraction using the P/4 protocol. B, Representative TEVC recordings showing TTX-insensitivity of the current of CiNav1a. CiNav1a is insensitive even to 10 μM TTX, while 1μM TTX is enough to abolish rNav1.4 current. The pulse protocol is the same to A. C, I-V curves of CiNav1a and rNav1.4 current under no TTX, 1 μM or 10 μM TTX) (N=3 for each experiment). p<0.0001 for rNav1.4. p=0.1729 for CiNav1a as analyzed by two-way repeated measures ANOVA. All graphs present mean ± SD.
	Figure 2TTX-insensitive sodium current from CiNav1a expressed in Xenopus oocyte. A, Raw traces and I-V curve of CiNav1a by TEVC recording. Only cRNA encoding CiNav1a was microinjected into Xenopus oocyte. Holding potential was -100 mV and depolarizing pulse in 50 ms was stepped to various levels (-90 to 40 mV). The current was isolated by leak subtraction using the P/4 protocol. B, Representative TEVC recordings showing TTX-insensitivity of the current of CiNav1a. CiNav1a is insensitive even to 10 μM TTX, while 1μM TTX is enough to abolish rNav1.4 current. The pulse protocol is the same to A. C, I-V curves of CiNav1a and rNav1.4 current under no TTX, 1 μM or 10 μM TTX) (N=3 for each experiment). p<0.0001 for rNav1.4. p=0.1729 for CiNav1a as analyzed by two-way repeated measures ANOVA. All graphs present mean ± SD.
	Figure 3Effect of human β1 subunit on activation properties of CiNav1a. A, Representative traces of CiNav1a and rNav1.4 alone, or with human β1 recorded by cut open oocyte recording. Holding potential was -100 mV. Depolarizing step was elicited for 50 ms by 5 mV increment. Traces with 10 mV increment ranging from -60 mV to 40 mV are shown. The current was isolated by leak subtraction using the P/4 protocol. Noise was removed off-line by Gaussian digital low-pass filter at cut-off frequency at 3 kHz. B, I-V curves of CiNav1a and rNav1.4 current with or without hβ1. N=7, 6, 7 and 5 for rNav1.4 alone, rNav1.4 with hβ1, CiNav1a alone and CiNav1a with hβ1, respectively. p=0.7116 for rNav1.4. p=0.7213 for CiNav1a as analyzed by two-way repeated measures ANOVA. C, Activation kinetics (time to peak) of CiNav1a and rNav1.4 plotted against membrane potential. All graphs present mean ± SD. N=7, 6, 7 and 5 for rNav1.4 alone, rNav1.4 with hβ1, CiNav1a alone and CiNav1a with hβ1, respectively. p=0.0010 for rNav1.4. p=0.2817 for CiNav1a as analyzed by two-way repeated measures ANOVA.
	Figure 4Effect of hβ1 on inactivation properties of CiNav1a. A, Effect of hβ1 on steady-state inactivation curve of CiNav1a as examined by cut-open oocyte recordings. rNav1.4 was also analyzed. Pulse protocol was as follows; Interval; 200 ms, prepulse duration; 100 ms, test pulse duration; 20 ms, test pulse voltage; 0 mV, holding potential; -100 mV. N= 5, 7, 7 and 4 for CiNav1a alone, CiNav1a+ hβ1, rNav1.4 alone, rNav1.4+ hβ1, respectively. p=0.3827 for rNav1.4. p=0.0953 for CiNav1a as analyzed by two-way repeated measures ANOVA. B, Inactivation kinetics (time constant of decay phase in fitting by single exponentials) of CiNav1a and rNav1.4 currents plotted against membrane potential as examined by cut-open oocyte recordings. Holding potential was -100 mV and depolarizing pulse in 50 ms was stepped by 5 mV increment with 200 ms interval. The current was isolated by leak subtraction using the P/4 protocol. N=7, 6, 7, 5 for rNav1.4 alone, rNav1.4 with hβ1, CiNav1a alone and CiNav1a with hβ1, respectively. Inactivation kinetics was significantly accelerated by coexpression with hβ1 in rNav1.4 (p<0.0001), whereas there was no significant difference in time constant (p=0.1699) in CiNav1a as analyzed by two-way repeated measures ANOVA. C, Representative traces of CiNav1a and hNav1.5 with or without hβ1 showing the kinetics of recovery from inactivation as examined by TEVC. Holding potential and interval potential was -100 mV. Preconditioning pulse was 0 mV for 100 ms for both types of channels. Test pulse was 0 mV for 50 ms for hNav1.5 and 100 ms for CiNav1a. D, Recovery from inactivation with or without hβ1. Pulse protocol is shown below. hNav1.5 current recovered from inactivation in two phases and fit by two exponential components, whereas CiNav1a current recovered in single phase and fit by single exponential component. Note that recovery from inactivation is accelerated by hβ1 in hNav1.5 (p<0.0001, N=5, 7 for hNav1.5 alone and with hβ1, respectively), whereas there was no significant difference (p=0.1729) in CiNav1a (N=10, 9 for CiNav1a alone and with hβ1, respectively) as analyzed by two-way repeated measures ANOVA. Also note that some of error bars are too small to be discerned in the graphs.
	Figure 5Comparison of effect of TipE between on CiNav1a and fly Para Nav channel studied by TEVC . A, TipE drastically increases the Para current (p=0.0096). Left, representative current traces of Para with or without TipE. Right, I-V curve of Para current with or without TipE. N=7 for each. Holding potential was -100 mV and depolarizing step was applied for 50 ms by 10 mV increment ranging from -90 mV to 40 mV. B, TipE does not modulate CiNav1a current (p=0.3104). Left, representative current traces of CiNav1a with or without TipE. Right, I-V curve of CiNav1a current with or without TipE. N=12 and 13 for CiNav1a and CiNav1a with TipE, respectively. The pulse protocol is the same as in A. Raw traces elicited during steps from -90 mV to 60 mV are superimposed. In A and C, the current was isolated by leak subtraction using the P/4 protocol. The surge to the inward direction was due to leak subtraction. C, Time to peak of CiNav1a current with or without TipE (p=0.3990). D, Decay time constant of CiNav1a current with or without TipE. Values in C and D are derived from single exponential fitting of the data set shown in A and B. p= 0.6038. Two-way repeated measures ANOVA was performed. All graphs present mean ± SD
	Figure 6Neuronal expression of CiNav1a in tailbud embryos and larvae of Ciona. CiNav1a gene expression was detected by whole-mount in situ hybridization in Ciona tailbud embryos and larvae. These panels presented here were the images of samples from three independent trials. Brown-colored stains indicating the expression signals of CiNav1a are detected in the central and peripheral nervous systems. A, Initial tailbud stage (left-side view). B, Early tailbud stage (left-side view). C and D, Mid tailbud stage (left-side view and dorsal view, respectively). E, Late tailbud stage (left-side view). F, Hatched larva (left-side view). All the specimens of embryos and larvae shown in the panels are oriented with anterior to the left. Black arrowheads indicate expression signal in presumptive neurons in the central nervous system of Ciona larvae. Purple and cyan arrowheads indicate epidermal sensory neurons and bipolar tail neurons, respectively, that constitute the peripheral nervous system of larvae. It is of note that all the stained neurons are not necessarily identified and labelled in the panels. Scale bars depict 50 μm. Dashed lines in C and E indicate the borders of combined two photos with different focal planes. Asterisks in F mark dusty materials unintentionally attached to the specimen.
	Figure 7Functional reconstitution of action potentials (APs) in Xenopus oocyte by coexpression of CiNav1a and CiKv1b. Under the configuration of “loose” clamping at -70 mV (see Methods), oocytes expressing CiNav1a and CiKv1b (without any auxiliary subunits) were stimulated with depolarizing step pulses (for 1 ms, 10 mV increment) (top trace). Bottom traces depict representative responses on membrane potential. APs were evoked by the stimulation; when larger pulse was applied, the latency of AP became shorter. Orange-, purple-, and red-colored traces indicate responses to 90, 100, and 110 mV depolarization pulse (for 1 ms), respectively. Similar action potentials were recorded from 38 cells.
	Figure 8Ability of targeting of the II-III linker of CiNav1a to the AIS of rat cortical neurons. A, top, Rat cortical neurons expressing II-III linker of rNav1.2 fused with YFP was stained with the antibody to MAP2, a marker of dendrites and cell bodies. rNav1.2’s II-III linker was localized in the proximity of MAP2 signal. Scale bar = 10 μm. Bottom, Rat cortical neurons expressing II-III linker of rNav1.2 fused with YFP was stained with the antibody to AnkG, a marker of AIS. rNav1.2’s II-III linker was clearly co-localized with AnkG signal. Scale bar = 10 μm. B, top, Rat cortical neurons expressing II-III linker of CiNav1a fused with YFP was stained with the antibody to MAP2, a marker of dendrites and cell bodies. CiNav1a was localized in the proximity of MAP2 signal. Scale bar = 10 μm. Bottom, Rat cortical neurons expressing II-III linker of CiNav1a fused with YFP was stained with the antibody to AnkG, a marker of AIS. CiNav1a was clearly co-localized with AnkG signal. Scale bar = 10 μm. Each data is a representative image in a single trial experiment.

References [+] :

Abitua, The pre-vertebrate origins of neurogenic placodes. 2015, Pubmed