|View Full Text||PubReader|
|Abstract||Print this Article|
|PMC||Export to Citation|
|Article as PDF||Open Access|
Exp Neurobiol 2021; 30(5): 319-328
Published online October 31, 2021
© The Korean Society for Brain and Neural Sciences
Minwoo Wendy Jang1,2, Tai Young Kim2, Kushal Sharma3, Jea Kwon1,2, Eunyoung Yi3 and C. Justin Lee1,2*
1KU-KIST Graduate School of Converging Science and Technology, Korea University, Seoul 02841,
2Center for Cognition and Sociality, Institute for Basic Science (IBS), Daejeon 34141,
3College of Pharmacy and Natural Medicine Research Institute, Mokpo National University, Mokpo 58554, Korea
Correspondence to: *To whom correspondence should be addressed.
TEL: 82-42-878-9150, FAX: 82-42-878-9151
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
The TMEM43 has been studied in human diseases such as arrhythmogenic right ventricular cardiomyopathy type 5 (ARVC5) and auditory neuropathy spectrum disorder (ANSD). In the heart, the p.(Ser358Leu) mutation has been shown to alter intercalated disc protein function and disturb beating rhythms. In the cochlea, the p.(Arg372Ter) mutation has been shown to disrupt connexin-linked function in glia-like supporting cells (GLSs), which maintain inner ear homeostasis for hearing. The TMEM43-p.(Arg372Ter) mutant knock-in mice displayed a significantly reduced passive conductance current in the cochlear GLSs, raising a possibility that TMEM43 is essential for mediating the passive conductance current in GLSs. In the brain, the two-pore-domain potassium (K2P) channels are generally known as the “leak channels” to mediate background conductance current, raising another possibility that K2P channels might contribute to the passive conductance current in GLSs. However, the possible association between TMEM43 and K2P channels has not been investigated yet. In this study, we examined whether TMEM43 physically interacts with one of the K2P channels in the cochlea, KCNK3 (TASK-1). Utilizing co-immunoprecipitation (IP) assay and Duolink proximity ligation assay (PLA), we revealed that TMEM43 and TASK-1 proteins could directly interact. Genetic modifications further delineated that the intracellular loop domain of TMEM43 is responsible for TASK-1 binding. In the end, gene-silencing of
Keywords: TMEM43, KCNK3, Cochlea, Protein interaction domains and motifs, Ion transport
TMEM43 was initially identified
The passive conductance channels, which give rise to a low membrane resistance and linear I-V relationship, are known as the unique property that defines mature astrocytes in the brain [9-11]. Our group has formally revealed the molecular identity of the astrocytic passive conductance channels as a dimer of TWIK-1 and TREK-1 subunits of the two-pore potassium (K2P) channels . In the brain, passive conductance channels are critically involved in K+ ion homeostasis  and volume regulation [14, 15]. In the cochlea, the endocochlear potential of +80 mV is generated by maintaining a high K+ concentration in the endolymph, and this potential is critical for the activation of hair cells . Therefore, various channel proteins in GLSs should be thoroughly examined to understand their role in establishing and maintaining K+ ion homeostasis and their associated role in hearing and speech discrimination. Furthermore, the TMEM43-mediated large passive conductance current in GLSs was reminiscent of the typical passive conductance current observed in hippocampal astrocytes . Accordingly, we hypothesized that there is a possibility of TMEM43 protein interacting with one of the K2P channels in the cochlea.
There are 15 known members of the K2P family, each containing two-pore (P) loops and four transmembrane (TM) domains . They have a unique extracellular cap structure formed by two TM1-P1 linkers, containing a cysteine for dimerization [18, 19]. Interestingly, this cysteine residue is not present in the TASK-1/3/5 subfamily , suggesting a disulfide-bridge-free binding in the TASK channels. In addition, the K2P channels are regulated by diverse stimuli such as oxygen tension, pH, lipids, mechanical stretch, neurotransmitters, and G protein-coupled receptors . Among the various 15 K2P channels, we searched for candidate K2P subunits that would most likely interact with TMEM43. In the end, we narrowed it down to KCNK3 (TASK-1), which exhibits the most similar properties as the TMEM43-involved currents, sensitive to external acidosis and displaying linear rectification [2, 21]. Although TMEM43 is expected to play critical roles in mammals, studies on this protein are sparse. In this study, we provide a piece of additional information regarding TMEM43 protein that it interacts with one of the K2P channels, KCNK3 (TASK-1), in the cochlea.
All experimental procedures were conducted according to protocols approved by the directives of the Animal Care and Use Committee of the Institutional Animal Care and Use Committee of IBS (Daejeon, Republic of Korea). C57BL/6 mouse pups of Postnatal day 5~8 (p5~p8) were purchased from Raon Bio (Youngin, Republic of Korea) or DBL (Eumseong, Republic of Korea) and were sacrificed on the delivered day.
C57BL/6 pups of embryonic day 15 were used. Embryos were dissociated from amniotic sac, and fetuses were put into a sterile petri dish with HBSS. They were decapitated, and skull and meninges were removed on ice. Only cortex was collected in 50 ml tube with HBSS and washed 3 times with 20 ml of HBSS to clear the debris. The cortex was incubated in 5 ml HBSS containing 0.1% trypsin for 5 min in a 37℃ water bath (shaking every 1 min). Afterward, 5 ml FBS (1X volume) was added to the tube, and the cortex was dissociated by pipetting. The cells were collected using a cell strainer (40 μm) and centrifuge (800 g, 3 min). The supernatant was removed, and the cell pellet was resuspended in Neurobasal A medium (10888022, Gibco). 1×104 cells were plated on 0.1 mg/ml Poly-L-Ornithine hydrobromide (PLO) (P3655, Sigma-Aldrich)-coated 24 well plates and incubated in the incubator (37℃, 5% CO2). The neuron culture was maintained by replacing the half of medium with fresh medium: Neurobasal A medium+2% B27 (17504-044, Gibco)+1% Glutamax (35050-061, Thermofisher Scientific)+0.5% P/S (250 ul/50 ml) (SV30010, Hyclone) every 2~3 days.
TMEM43 (Myc-DDK-tagged)-Human transmembrane protein 43 (TMEM43) (NM_024334.2) was purchased from OriGene (RC200998), KCNK3 (Myc-DDK-tagged) (TASK-1) (NM_002246.3) was purchased from OriGene (RC215155), GJB2 (NM_004004) Human Tagged ORF Clone was purchased from OriGene (RC202092), and Cx30-msfGFP was purchased from Addgene (69019). The coding sequences of these genes were cloned into CMV-MCS-IRES2-EGFP or CMV-MCS-IRES2-dsRed plasmid vector using BglII/XmaI sites. TMEM43 truncation mutations were made using the EZchange site-directed mutagenesis kit (EZ004S, Enzynomics).
Human embryonic kidney (HEK) 293T cells were purchased from ATCC (CRL-3216). The cell line has been tested for mycoplasma contamination. HEK293T cell was cultured in DMEM (10-013, Corning) supplemented with 10% heat-inactivated fetal bovine serum (10082-147, Gibco) and 10,000 units/ml penicillin-streptomycin (15140-122, Gibco) at 37℃ in a humidified atmosphere of 95% air and 5% CO2. According to the manufacturer’s protocol, the transfection of expression vectors was performed with Effectene Transfection Reagent (Effectene, 301425, Qiagen). One day before performing the experiments, HEK293T cells were transfected with each DNA 1 μg per 35 mm dish.
DNA transfected HEK293T cells, or cochlear tissues of an average of 20 mice (p5~p8) per sample were extracted and homogenized with PierceTM IP Lysis Buffer (87787, Thermo Fisher Scientific) and HaltTM Protease and Phosphatase.
Inhibitor Cocktail (100X) (78446, Thermo Fisher Scientific). Equal amounts of precleared cell lysates were incubated with mouse anti-TASK-1 (KCNK3) monoclonal (1 μg, ab186352, Abcam) antibody or control IgG (1 μg) overnight. Protein A/G-Agarose beads (20422, Thermo Fisher Scientific) were added to the mixtures and further incubated with rotation for 1 h at 4℃, followed by a wash with lysis buffer. Bound proteins were eluted from the beads with SDS-PAGE sample buffer, and western blotting was performed with rabbit anti-FLAG (1:500, 2368s, Cell Signaling) or rabbit anti-TMEM43 (1:100, NBP1-84132, Novus).
Duolink® In Situ Red Starter Kit Mouse/Rabbit (DUO92101, Sigma-Aldrich) and far-red detection kit (DUO92013, Sigma-Aldrich) were used. On the experimental day, cochlea tissue (p5~p8) was obtained, fixed with 4% paraformaldehyde, and washed with 0.3% Triton-X100 containing PBS. After blocking, the sample was incubated with rabbit anti-TMEM43 polyclonal (1:100, NBP1-84132, Novus) and mouse anti-TASK-1 (KCNK3) monoclonal (1:100, ab186352, Abcam) at 4℃, overnight. The next day, the sample was incubated with anti-rabbit MINUS and anti-mouse PLUS probe, ligase, and polymerase sequentially. DNA strands participate in rolling circle DNA synthesis only when two probes are in close proximity (<40 nm). Fluorescent-labeled complementary oligonucleotide probes were observed under Zeiss confocal microscopy.
Neonatal mouse (C57/BL6, P5) cochlear turns were isolated in ice-cold sterile Hanks’ Balanced Salt Solution (HBSS). Cochlear segments were attached on Cell-Tak (354240, Corning) coated coverslips and incubated overnight in DMEM/F12 medium containing 1% fetal bovine serum, 5 µg/ml ampicillin, B27 and N2 (37°C, 5% CO2 humidified incubator). Upon confirming stable attachment, the tissues were treated with either control or shRNA carrying lentivirus diluted in culture medium (1:1,000) for 48 h. The medium was then replaced with a fresh one. After an additional 24 h of incubation, the tissues were subjected to further examination.
The virus infection on cultured cochlea tissue was confirmed via the presence of mCherry signal. For immunostaining, cochlea tissue was fixed in ice-cold 4% paraformaldehyde, for 10 min and incubated in blocking/permeabilizing buffer (PBS with 5% goat serum and 0.25% Triton-X100). Then, the preparations were incubated overnight at 4°C with rabbit anti-TMEM43 polyclonal (1:100, NBP1-84132, Novus) and mouse anti-TASK-1 (KCNK3) monoclonal (1:100, ab186352, Abcam) diluted in the blocking/permeabilizing buffer. After 3 washes, the cochlear turns were reacted with donkey anti-rabbit Alexa Fluor 594 (1:500, 711-585-152, Jackson ImmunoResearch) donkey anti-mouse Alexa Fluor 488 (1:500, 715-545-150, Jackson ImmunoResearch) in blocking/permeabilizing buffer for 1 h at room temperature. Samples were then rinsed once with blocking/permeabilizing buffer and twice with PBS. Using Fluorsave reagent (345789, Calbiochem), the tissues were mounted on glass slides and covered with a coverslip. Specific immunolabeling was examined under Zeiss confocal microscope. No immunoreactivity was found when the primary antibodies were omitted.
Cultured cochlea tissue infected with lentivirus was used. Recordings were done in HEPES buffer containing (mM): 144 NaCl, 5.8 KCl, 1.3 CaCl2, 2 MgCl2, 10 HEPES, 0.7 NaH2PO4, and 5.6 D-glucose (pH 7.4 was adjusted with NaOH). Glia-like supporting cells located below the inner hair cell layer were whole-cell patch clamped, and current traces were elicited by 1-sec ramps descending from +100 mV to -100 mV with -60 mV holding potential. Recording electrodes (7~11 MΩ) supplemented with (mM): 126 K-Gluconate, 5 HEPES, 0.5 MgCl2, and 10 BAPTA (pH adjusted to 7.3 with KOH) were advanced through tissue under positive pressure. Slice chamber was mounted on the stage of an upright Hamamatsu digital camera viewed with a 60X water immersion objective with infrared differential interference contrast optics using Imaging Workbench Software. Electrical signals were digitized and sampled with Digidata 1320A and Multiclamp 700B amplifier (Molecular Devices) using pCLAMP 10.2 software. Data were sampled at 10 kHz and filtered at 2 kHz. Glass pipettes were pulled from a micropipette puller (P-97, Sutter Instrument), and all experiments were conducted at room temperature 20~22℃.
Off-line analysis was carried out using Clampfit version 10.4.1.10 and GraphPad Prism version 7.02 software. When comparing between two samples, the significance of data was assessed by Student’s two-tailed unpaired t-test. One-way ANOVA with Kruskal-Wallis test with Dunn's posthoc test was used to compare 3 samples, as the data did not pass the normality test. Significance levels were given as: N.S. p>0.05, *p<0.05, **p<0.01, ***p<0.001 and #p<0.0001.
In order to examine whether TMEM43 interacts with TASK-1, human
To further examine the direct interaction of TMEM43 and TASK-1 in the
We next investigated the interacting domain of TMEM43 that is required for interaction with TASK-1. We predicted protein structure of TMEM43 with AlphaFold, the state-of-the-art deep neural network-based method with atomic level accuracy (Fig. 2A) . The bioinformatic analyses predicted TMEM43 protein with 4 TM domains and a large intracellular loop (Loop1) containing an intramembrane (IM) domain [2, 3, 25]. Based on the predictions, we firstly made a deletion mutation of Loop1, which lies between TM1 and TM2 region, and linked the two TM with three GGS linkers (Fig. 2B). The co-IP assay revealed that the interaction between TMEM43-ΔLoop1 and TASK-1 was disrupted (Fig. 2C). We next targeted the IM domain (Fig. 2B) and found that the interaction between TMEM43-ΔIM with TASK-1 was intact (Fig. 2D). These findings indicate that TMEM43 intracellular loop (Loop1) is necessary for TASK-1 interaction. The exact binding domain at TMEM43 Loop1, other than IM domain, should be further examined.
We subsequently examined TMEM43 interaction with TASK-1 in the cochlea tissue. The cochlea tissue was obtained from the inner ear of C57BL/6 pups (Fig. 3A). Consistent with the heterologous expression system results, immunoprecipitates from cochlear homogenates with anti-TASK-1 antibody showed a positive TMEM43 signal (Fig. 3B). Interestingly, the co-IP band size for TMEM43 was at around 60 kDa in the cochlea tissue (Fig. 3B), which was different from 47 kDa in the heterologous system (Fig. 1B). This discrepancy of size gap may arise from the absence of many proteins required for proper protein processing (such as chaperone and trafficking proteins) in cell lines but not in native environments. Furthermore, it has been demonstrated that co-translational and post-translational modification, in particular glycosylation, can also affect many proteins in membrane trafficking and functional gating properties [26, 27]. In fact, we have observed two TMEM43 bands, at size 43 kDa and 60 kDa from the cochlea lysates. However, after co-IP with anti-TASK-1 antibody, the major TMEM43 band appeared at 60 kDa but not at 43 kDa. This result is different from the result of the previous study , which showed TMEM43 band size at 43 kDa after co-IP with anti-Cx30 antibody. Therefore, these results together suggest that Cx30 likely interacts with the native TMEM43 (43 kDa) while TASK-1 preferentially associates with the post-translationally modified form of TMEM43 (60 kDa). Identification of these possible chaperones and membrane trafficking proteins or post-translational modifications of TMEM43 in GLSs awaits future investigations.
We next performed immunohistochemistry with the two antibodies to analyze TMEM43 and TASK-1 protein expression in the intact cochlea tissue. The confocal micrographs of the organs of Corti demonstrated a strong co-localization of TMEM43 and TASK-1 immunoreactivities in the cochlear GLSs (Fig. 3C). Furthermore, the Duolink PLA was done in the cochlea tissue using anti-TMEM43 and anti-TASK-1 antibodies. Similar to the
TMEM43 has been reported to mediate the passive conductance current in the cochlear GLSs . Accordingly, we tested if TASK-1 is also involved in the cochlear passive conductance current. To begin with, we developed a mouse
In conclusion, our findings provide molecular and physiological lines of evidence manifesting that TMEM43 protein interacts with a K2P channel, TASK-1. The congruent results from Co-IP and Duolink PLA in both
This work was supported by the Institute for Basic Science (IBS), Center for Cognition and Sociality (IBS-R001-D2) to C.J.L. The graphical abstract was created with BioRender.com.