Biophysical characterization of anion channels in mitochondrion-endoplasmic-reticulum contact sites

Abstract

The mitochondrion-endoplasmic reticulum (ER) contact sites (MERCs, also known as mitochondrial-associated membranes [MAMs]) are specialized regions of the ER that are in close proximity to the mitochondrion. These organelle structures play essential roles in a variety of processes, such as calcium signaling, lipid metabolism, renin-angiotensin-aldosterone system control, the unfolded protein response, and autophagy. MERCs are known to actively participate in ion transport between the ER and mitochondria. Although active calcium channels in MERCs have been detected, limited studies have been carried out to identify or characterize functional anion channels. Here, we tested whether functional anion channels are present in MERCs. We isolated MERCs from mouse organs (heart and brain) and reconstituted them in planar bilayers. The single-channel properties were recorded in the presence of various anion channel blockers or antagonists (IAA-94, DIDS, A9C, and NPPB). We corroborated the presence of anion channels targeted by these drugs using immunoblotting and immunocytochemistry. Biochemical analysis and immunocytochemistry corroborate that CLIC4, CLIC3, and VDACs are present in MERCs. Our results indicate that anion channels are active in MERCs, which could play a pertinent role in intracellular organelle communication.

Significance

MERCs are in focus as they are perceived as Ca²⁺ signaling hotspots for the rapid transport of ions between the ER and mitochondria. The highly dynamic contact site between the two organelles is anchored by mitofusin 2, and this association changes on demand for Ca²⁺. Apart from Ca²⁺ other ion channels in MERCs have not yet been characterized. For the first time, we have shown that Cl⁻ channels are active in MERCs. Cl⁻ ions are implicated in volume regulation, ionic homeostasis, and the fine-tuning of pH. The presence of Cl⁻ currents in MERCs reiterates the physiological significance of Cl⁻ channels in organellar and cellular physiology.

Introduction

Cellular organelles perform essential functions in the cell. Although they were perceived to operate independently, recent evidence suggests that organelles are physically and functionally coupled with other organelles to regulate organelle function and cellular homeostasis (1,2,3,4,5,6,7,8). In these contact sites, some proteins directly tether organelles, and some actively participate in interorganelle communication and function (9). Among these organelle contact sites, connections between mitochondria and the endoplasmic reticulum (ER) were visualized in the 1950s (10), and later widely accepted for interorganelle communications (11). So far, mitochondria are in close contact with lysosomes, lipid droplets, nucleus, peroxisomes, extracellular vesicles, Golgi, plasma membrane, endosomes, and melanosomes (12,13). These contact sites have been implicated in the interorganelle exchange of membranes, lipids, metabolites, and several ions, such as Ca²⁺ (9,14,15,16).

The mitochondrion-ER contact sites (MERCs, also known as mitochondrial-associated membranes [MAMs] ) (17,18,19) are well characterized for Ca²⁺ transport, which in turn impacts mitochondrial bioenergetics, metabolism, and dynamics, as well as on cell death and autophagy (1,14,15,16). While Ca²⁺ is important for intracellular signaling, anions such as Cl⁻ play a vital role in maintaining organelle morphology and ionic homeostasis (20). During Ca²⁺-mediated Ca²⁺ release, Cl⁻ efflux, as well as K⁺ influx, is vital for neutralizing membrane charge and balancing the osmolarity of the ER (20). In addition to volume regulation and membrane potential, Cl⁻ is also implicated in cell excitability and pH regulation of organelles (2,21,22). Several ion channels and transporters are characterized in MERCs (4,23,24,25,26). The majority of these channels and transporters are implicated in the transport of cations such as Ca⁺² (11), K⁺ (23,25), and Fe⁺ (27), and anions such as Cl⁻ (4,24). Among anion channels, voltage-dependent anion channel (VDAC) has been placed in the outer mitochondrial membranes and MERCs (4,24). We have recently discovered a prominent chloride intracellular channel 4 (CLIC4) in the MERCs (26). CLIC4, also known as p64H1 and placed in the ER (28), regulates mitochondrial function and protects the heart from ischemia-reperfusion injury (26). Since contact sites are highly specific and small membrane structures, a lack of technical resources makes it challenging to functionally characterize ion channels and transporters in MERCs.

In this study, we characterized the functional presence of anion channels in MERCs in the heart. Using a combination of electrophysiology and pharmacology approaches, we identified prominent anion currents sensitive to 2-[[(2R)-6,7-dichloro-2-cyclopentyl-2,3-dihydro-2-methyl-1-oxo-1H-inden-5-yl]oxy]-acetic acid (IAA-94), 5-nitro-2-[(3-phenylpropyl)amino]-benzoic acid (NPPB), anthracene-9-carboxylic acid (A9C), and 4,4′-diisothiocyanato-2,2′-stilbenedisulfonic acid disodium salt (DIDS) (29,30,31,32,33). We further probed MERCs with western blots and immunocytochemical methods to identify ion channels in these contact sites. The quantification of anion channels from MERCs will have ramifications for the interpretation of ion fluxes and provide the foundation for future studies on anion channels.

Materials and methods

Animals

All animal studies followed the Guide for the Care and Use of Laboratory Animals (NIH Publication, 8th Edition, 2011) and were approved by the Institutional Animal Care and Use Committee of The Ohio State University. Wild-type 3-month-old CD-1 mice were utilized for MERC isolation.

Isolation and preparation of MERCs

MERCs were prepared as published earlier (23,26,34). Briefly, mouse hearts and brains were isolated, minced, and homogenized in buffer IB-H1 (225 mM mannitol, 75 mM sucrose, 0.5% [w/v] bovine serum albumin [BSA], 0.5 mM EGTA, and 30 mM Tris-HCl [pH 7.4]) using a Potter-Elvehjem homogenizer with seven strokes (whole-organ lysate). The homogenate was centrifuged at 740 × g for 5 min at 4°C. The supernatant was collected and centrifuged again at 740 × g for 5 min at 4°C. The supernatant was then centrifuged at 9000 × g for 10 min at 4°C. The supernatant was removed (cytosol), and the pellet was gently resuspended in ice-cold buffer IB-H2 (225 mM mannitol, 75 mM sucrose, 0.5% [w/v] BSA, and 30 mM Tris-HCl [pH 7.4]) and centrifuged at 10,000 × g for 10 min at 4°C. The supernatant was discarded, and the pellet was resuspended in buffer IB-H3 (225 mM mannitol, 75 mM sucrose, and 30 mM Tris-HCl [pH 7.4]) and centrifuged again at 10,000 × g for 10 min at 4°C. The pellet containing crude mitochondria was resuspended in 55 μL of mitochondria resuspending buffer (MRB) (250 mM mannitol, 0.5 mM EGTA, and 5 mM HEPES [pH 7.4]). The crude mitochondrial pellet was overlaid on 3 mL of 30% (v/v) Percoll prepared in Percoll medium (225 mM mannitol, 1 mM EGTA, and 25 mM HEPES [pH 7.4]) and centrifuged at 95,000 × g for 30 min at 4°C. This step separates the MERCs (upper layer) and mitochondria (lower layer). The lower layer, containing mitochondria, was collected and resuspended in MRB buffer, and centrifuged at 6300 × g for 10 min at 4°C. The pellet obtained after centrifugation was resuspended in radioimmunoprecipitation assay (RIPA) buffer (50 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA-Na₂, 1 mM EGTA-Na₄, 1% [v/v] NP-40, 0.5% [w/v] Na-deoxycholate, 0.1% [w/v] SDS, 1 mM NaF, 1 mM phenylmethylsulfonyl fluoride, and 1 mM Na₃VO₄ [pH 7.4] containing protease inhibitors [1 tablet/50 mL, Roche] and phosphatase inhibitor [1 tablet/10 mL, Roche]), and was then frozen in liquid nitrogen. The upper band containing MERCs was diluted in MRB buffer and further centrifuged at 6300 × g for 10 min at 4°C. The supernatant was collected and centrifuged at 100,000 × g for 1.5 h at 4°C. The MERC pellet was collected, resuspended in RIPA buffer, and frozen in liquid nitrogen. All the cellular fractions, including whole-organ lysate, cytosol, crude mitochondria, purified mitochondrial, and MERC fractions, were subjected to three repeated freeze-thaw cycles and processed for western blot analysis.

Electrophysiology

Ion channels in cardiac MERCs were recorded by planar bilayer approaches, as described earlier (Fig. 1 A) (35,36,37,38,39). The Warner bilayer setup with digitizer 1550 or the Nanion bilayer system (Orbit mini) was utilized for recording single-channel currents. A planar lipid bilayer was formed by 1-palmitoyl-2-oleoyl phosphatidylethanolamine, 1-hexadecanoyl-2-(9Z-octadecenoyl)-sn-glycero-3-phospho-L-serine (sodium salt), and cholesterol in the molar ratio of 4:1:1 (35,36,37), respectively. All the recordings were performed under 150 mM n-methyl-D-glucamine chloride (NMDG-Cl) (symmetrical) solutions. MERC preparations (50 ng/μL) were added to the cis side of the setup or in the recording solution in the Nanion system. Once channel activity was observed, we recorded single-channel activities at various holding potentials ranging from −100 to 100 mV (steps of 20 mV). Data were acquired with Elements Data Reader and/or pClamp 12.5, and analyzed using Clampfit 10.6. Channel blockers or antagonists were added to the cis side of the recording system. Data were plotted using SigmaPlot 12.5.

Figure 1.

Schematic of the procedure for the preparation and incorporation of MERCs into planar bilayers. (A) Mouse heart was harvested, and MERCs were prepared. The MERCs were sonicated and stored at −80°C until use. Planar bilayers were made up of 1-palmitoyl-2-oleoyl phosphatidylethanolamine (POPE), 1-hexadecanoyl-2-(9Z-octadecenoyl)-sn-glycero-3-phospho-L-serine (sodium salt) (POPS), and cholesterol in the molar ratio of 4:1:1. Fifty nanograms of MERCs membrane preparations was added to the planar bilayer (≥100 pF). (B) Single-channel activity was recorded in 150 mM NMDG-Cl (symmetrical) and 1 mM DTT. Several channel currents were observed and marked as 1, 2, 3, and 4.

Immunoblotting

All the cellular fractions from the heart and brain; whole organ lysate, cytosol, crude mitochondria (CM), purified mitochondria (PM), and MERCs were resuspended in RIPA lysis buffer. All the fractions were then snap-frozen in liquid nitrogen and, once thawed, mixed on a rotatory shaker at 4°C for 1 h. After centrifugation at 12,000 × g for 20 min at 4°C, the supernatant was collected in a fresh microcentrifuge tube and 50 μg loaded on 4–20% (w/v) sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis. The gel was transferred to a supported nitrocellulose membrane (Bio-Rad, CA; 1620097) and validated with Ponceau S staining. Furthermore, the membrane was blocked in Intercept Tris-buffered saline (TBS) blocking buffer (LI-COR Biosciences, NE; 927-60001) for 1 h rocking at room temperature and washed three times (5 min each) with TBS. After that, the membrane was probed with various primary antibodies at 4°C overnight on an orbital shaker. The primary antibodies used for probing MERCS were: anti-CLIC4 (Santa Cruz Biotechnology, TX; sc-135739, 0.4 μg/mL), anti-mitofusin-2 (Abcam, United Kingdom; ab205236, 1.4 μg/mL), anti-VDAC (Cell Signaling Technology, 4661S, 1:1000), anti-bestrophin-1 (Alomone Labs, Israel; ABC-001, 1.6 μg/mL), anti-GRP-78 (Abcam, ab21685, 2.0 μg/mL), anti-CLIC1 (Santa Cruz Biotechnology, TX; sc.271051, 0.4 μg/mL), anti-vinculin (Cell Signaling Technology, MA; 13,901, 1:1000), anti-CFTR (Thermo-Fisher Scientific, MA; PA5-121193, 2.2 μg/mL), anti-CLIC3 (Proteintech, IL; 15971-1-AP, 0.46 μg/mL), anti-TMEM16A (Alomone Labs, Israel; ACL-011, 2.0 μg/mL), and anti-ATP5A (Abcam, United Kingdom; ab14748, 2.0 μg/mL). After an overnight incubation, three washes were given with TBS for 10 min each, membranes were incubated with secondary antibodies (goat anti-mouse IgG polyclonal antibody [LI-COR Biosciences, NE; 925-68070 or 925-32210] and/or goat anti-rabbit IgG polyclonal antibody [LI-COR Biosciences, NE; 925-68071 or 925-32211]) for 1 h at room temperature and washed again three times with TTBS (20 mM TBS containing 0.05% Tween-20) for 10 min. The intensity signal on blots was imaged using the Azure Biosystems c600. All the blots were stained with Ponceau S to verify the loading of proteins and the normalization of protein signals (Fig. S3 B).

Isolation of adult cardiomyocytes and cardiac fibroblasts

Cardiomyocytes were isolated from 3-month-old wild-type CD-1 male mice using a direct enzymatic digestion protocol, as previously described (40). Briefly, mice were anesthetized in a 4% (v/v) isoflurane chamber, and a thoracotomy was performed to expose the heart. The descending aorta was excised, and 7 mL of EDTA buffer (130 mM NaCl, 5 mM KCl, 0.5 mM NaH₂PO₄, 10 mM HEPES, 10 mM glucose, 10 mM butanedione monoxime [BDM], 10 mM taurine, and 5 mM EDTA [pH 7.8]) was injected into the right ventricle. The ascending aorta was then clamped, and the heart was transferred to a petri dish containing EDTA buffer.

For adult cardiomyocyte (ACM) isolation, the left ventricle was perfused with 10 mL of EDTA buffer, followed by 3 mL of perfusion buffer (130 mM NaCl, 5 mM KCl, 0.5 mM NaH₂PO₄, 10 mM HEPES, 10 mM glucose, 10 mM BDM, 10 mM taurine, 1 mM MgCl₂ [pH 7.8]). Subsequently, enzymatic digestion was initiated by perfusing 10–20 mL of collagenase enzyme solution containing collagenase type II (0.5 mg/mL), collagenase type IV (0.5 mg/mL), and protease XIV (0.05 mg/mL) prepared in perfusion buffer. Following digestion, the heart chambers were separated, and ventricular tissue was minced into ∼1 mm fragments. Cells were dissociated by gentle trituration, and enzymatic digestion was terminated by adding 5 mL of stop buffer (perfusion buffer supplemented with 5% [v/v] sterile fetal bovine serum). The ACMs were washed three times with perfusion buffer and allowed to sediment via gravity.

Cardiac fibroblasts were isolated from ventricular tissue after enzymatic digestion. All procedures were performed under a laminar flow hood to maintain sterility. Briefly, atria were excised, and ventricles were minced into small fragments in an enzyme buffer (containing 130 mM NaCl, 5 mM KCl, 0.5 mM NaH₂PO₄, 10 mM HEPES, 10 mM glucose, 10 mM BDM, 10 mM taurine, and 1 mM MgCl₂ [pH 7.8]) supplemented with Liberase TH (44 μg/mL). The minced ventricular tissue was subjected to enzymatic digestion in a T-25 culture flask at 37°C for approximately 60 min, with gentle agitation every 20 min. After digestion, the cell suspension was triturated using a 1-mL wide-bore pipette tip to facilitate single-cell dissociation and subsequently centrifuged at 500 × g for 5 min. The resulting cell pellet was resuspended in fibroblast culture medium composed of DMEM/F12 supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin, and 1 μM ascorbic acid. Cells were then seeded into a T-25 flask and incubated for approximately 3 h to allow the selective attachment of fibroblasts. Non-adherent cells were removed by replacing the medium with fresh fibroblast culture medium.

Immunocytochemistry

Isolated ACMs and cardiac fibroblasts were loaded with MitoTracker Red (200 nM) and incubated in 37°C humidifying chamber with 5% (v/v) CO₂ for 10 min. The cells were fixed with 4% (w/v) PFA and permeabilized with 0.5% (v/v) Triton-X-100. ACMs and cardiac fibroblasts were incubated with anti-CLIC4 (Santa Cruz Biotechnology, TX; sc-135739, 1 μg/mL), anti-mitofusin 2 (Abcam, United Kingdom; ab205236, 3 μg/mL), anti-VDAC (Cell Signaling Technology, MA; 4661S, 1:1000, and Abcam, United Kingdom; ab14734, 1:500), anti-bestrophin1 (Alomone Labs, Israel; ABC-001, 1.6 μg/mL), anti-CFTR (Thermo Fisher Scientific, MA; PA5-121193, 5 μg/mL), and anti-CLIC3 (Proteintech, IL; 15971-1-AP, 1 μg/mL) overnight at 4°C. Secondary antibodies conjugated with anti-mouse Alexa-488 (Cell Signaling Technology, MA; 4408, 1:1000) and anti-rabbit Atto-647N (Sigma-Aldrich, MO; 40839, 1:1000) were added for 60 min at room temperature. To label nuclei, DAPI was added (1:10,000) to the wash solution. Coverslips were mounted with Mowiol 4-88 (Sigma-Aldrich, MO; 81381). Cells were imaged with Nikon A1R high-resolution confocal microscopy. Images were filtered by custom-built software, as described earlier (40).

Results

Single-channel currents in MERCs

Proteins from MERCs were reconstituted in planar bilayers by painting them directly onto the membranes (Fig. 1 A). We titrated the MERC proteins to 50 ng/μL to decrease the number of channels inserting at once. We further focused only on Cl⁻ channels and used NMDG-Cl in the recording solutions. Although other channels may exist, in the absence of cations or other anions, those channels are presumed not to be functional in our experiments. We obtained anion channel currents in 120 out of 155 independent experiments in MERCs using NMDG-Cl. In our experiments, usually several different channels were inserted, making it difficult to analyze single-channel kinetics (Fig. 1 B). However, in 89 experiments, 3 or fewer unique channels were observed. These channels were characterized according to their single-channel conductance and sensitivity to known chloride channel blockers.

As shown in Fig. 2, A and C, we have successfully recorded Cl⁻ currents in MERCs. While the currents in Fig. 2 A showed larger currents at negative holding potentials (Fig. 2 B), the channel represented in Fig. 2 C showed larger currents at positive holding potentials (Fig. 2 D). The current versus voltage curves were plotted for all the identified channels (Fig. 2, B and C are representative channels), and single-channel conductance were calculated. All the channels identified were plotted on frequency histograms, and we discovered that there are four distinct channel conductances at ∼50, ∼110, ∼170, and ∼210 pS (Fig. 2 E). The majority of the single-channel conductance was found to be ∼50 (33%) and ∼110 pS (20%). Open probability does not show any correlation with channel conductance. DIDS is a known “classical” anion channel and transporter blocker (41,42). Application of DIDS reduced open probability from 0.8 to 0.1, and it was reversible (Po: 0.7) by perfusing the bath solution (Fig. 2 F).

Figure 2.

Single-channel characterization of MERC membrane anion channels. (A) Representative traces of MERCs anion currents at different holding potentials. The solid line represents the closed state. (B) Current-voltage curve for MERC anion channels shown in (A). (C) Representative traces of other MERC anion currents at different holding potentials. The solid line represents the closed state. (D) Current-voltage curve for MERC anion channels shown in (C). (E) Frequency histogram of various conductances obtained for MERC anion channels. Corresponding open probability (Po) values are also indicated. (F) Representative traces of DIDS-sensitive anion currents at −40 mV. The addition of DIDS reduces the open probability, which was completely reversible after washing with the bath solutions. Frequency histograms to represent open probability are shown as insets. Recording buffers contained cis-trans: 150 mM NMDG-Cl (symmetrical) and 1 mM DTT.

Pharmacological characterization of MERCs Cl⁻ channels

After characterizing Cl⁻ channels in MERCs, we took the next steps to identify these channels. We incorporated commonly used and well-established Cl⁻ channel blockers: DIDS (50 μM), A9C (50 μM), NPPB (25 μM), and IAA-94 (30 μM) to characterize Cl⁻ channel currents. We calculated the open probability of channel conductance before and after the Cl⁻ channel blocker. We observed a very high block of Cl⁻ channels (70–90% reduction in open probability) with A9C (Fig. 3 A) and DIDS (Fig. 3 B). However, with NPPB and IAA-94, we only observed a 1 and 15% reduction (respectively) in the open probability (Po) of Cl⁻ channels. We also quantified the percentage of Cl⁻ currents sensitive to specific blockers. Among all the MERC Cl⁻ channels recorded in NMDG-Cl, 25% of total Cl⁻ currents were sensitive to DIDS (9 blocks out of 36 attempts), 33% of total Cl⁻ currents were sensitive to A9C (2 blocks out of 6 attempts), 21% of total Cl⁻ currents were sensitive to IAA-94 (8 blocks out of 37 attempts), and 12.5% of total Cl⁻ currents were sensitive to NPPB (1 partial block out of 8 attempts). There was a large proportion (29%) of Cl⁻ currents that were not sensitive to any of the 4 Cl⁻ channel blockers used. Although single-channel conductance and Cl⁻ channel blockers can provide some information on the identity of putative Cl⁻ channels in MERCs, it is not possible to establish the precise identity without western blots or immunocytochemistry.

Figure 3.

Pharmacological characterization of MERC membrane anion channels. Single-channel activity at +60 mV was obtained for MERCs and, after 30 s, several known pharmacological blockers were added. (A) A9C completely blocked the Cl⁻ current (n = 2/6). (B) DIDS, a known blocker for VDACs, ClC6, MaxiCl, and CLCC1, reduced the open probability of Cl⁻ currents (n = 9/36). (C) NPPB, a known blocker for ORCC and CaCC, had little effect on the open probability of Cl currents (n = 1/8). (D) IAA-94, a known chloride intracellular channel (CLIC1-6) substantially reduced the open probability of Cl currents (n = 8/37).

Biochemical characterization of MERCs Cl⁻ channels

Since we observed a significant block with Cl⁻ channel blockers, we further probed for their presence with well-characterized antibodies (Fig. 4, A and B). Since CLIC4 is reported to be present in MERCs (26), we probed our cardiac MERC preparations with anti-CLIC4 antibody. We noted that CLIC4, as well as GRP78 (a marker for MERCs), are present in cardiac MERCs. Further, GRP78 and CLIC4 were absent in ultrapure mitochondrial preparations (PM). Using these highly pure MERCs and mitochondrial preparations, we tested several putative Cl⁻ channel candidates. Although VDACs and CLIC3 were enriched in the ultrapure mitochondrion, we also found trace amounts in MERC preparations. VDACs have been shown to be present in MERCs earlier (3,24,43), but this is the first evidence of the existence of CLIC3 in MERCs. We also detected mitofusin 2 protein in mitochondria as well as MERCs. Mitofusins are known for their roles in mitochondrion-mitochondrion and mitochondrion-ER tethering (44,45). In our experiments, we did not detect bestrophin, CFTR, TMEM16A (ANO1), or CLIC1 in MERCs or mitochondria. Our results indicate that in addition to CLIC4, VDACs, and mitofusin 2, CLIC3 is present in cardiac MERCs. We also corroborated our results in MERCs isolated from the whole brain. As shown in Fig. S1, A and B, we detected CLIC4, VDACs, GRP78, and CLIC3 in MERCs, but the mitochondrial marker ATP5A was absent.

Figure 4.

Biochemical characterization of MERC membranes. The cellular fractions from the heart; WHL (whole organ lysate), cytoplasm, crude mitochondria (CM), ultrapure mitochondria (PM), and MERCs were lysed and probed with several different antibodies. (A) We obtained signals for CLIC4, GRP78, CLIC3, mitofusin 2, and VDAC in MERCs. Mitofusin 2, VDAC, and CLIC3 were present in ultrapure mitochondria. CLIC4 was absent in ultrapure mitochondria. GRP78 showed 2 bands at 72 and 78 kDa; the upper band was absent in ultrapure mitochondria. CFTR, TMEM16A, CLIC1, and bestrophin were absent in mitochondria or MERCs. (B) Quantification of the western blots from (A). The graphs depict the mean ± SEM for normalized expression of the proteins against the respective Ponceau S-stained blots; n = 3.

Localization of MERC Cl channels

We investigated the localization of ion channels identified by western blot using fluorescence confocal microscopy. As shown in Fig. 5, CLIC4 and mitofusin 2 colocalized in ACMs, corroborating (26) that CLIC4 is associated with MERCs. However, bestrophin and CFTR are localized to the plasma membrane of ACMs. Although the majority of CLIC3 and VDACs were present in the plasma membrane, a small fraction was localized to CLIC3, which is located in MERCs. Our combined results from biochemical and immunocytochemistry indicate that CLIC4, CLIC3, and VDACs are associated with cardiac MERCs. To further corroborate our findings, we also incorporated cardiac fibroblasts to test whether CLIC4, CLIC3, and VDACs localize to MERCs. We discovered that, similar to cardiomyocyte MERCs, CLIC3, CLIC4, and VDACs are present in fibroblast MERCs but bestrophin and CFTR are absent (Fig. S3).

Figure 5.

Localization of anion channels in adult cardiomyocytes. Representative images of isolated adult cardiomyocytes loaded with MitoTracker Red (magenta), fixed, permeabilized, and immunolabeled with anti-CLIC4 (red). Costaining (green) was performed with the MERC marker anti-mitofusin 2 (A) and chloride channel markers: anti-Bestrophin (B, BEST), anti-CFTR (C), anti-CLIC3 (D), and anti-VDAC (E). Nuclei were stained with DAPI (blue). The right panels show merged images at higher magnification with an intensity profile of colocalization. All experiments were independently repeated at least four times.

Discussion

MERCs, also known as MAMs or contact sites, were first described in rat liver (46), where they were proposed to be a site of mitochondrial protein biogenesis. In addition to protein synthesis, they are also implicated in lipid transport between organelles (47). In hepatocytes, it is proposed that one in four mitochondria are associated with the ER (47), whereas in striated muscle >97% of mitochondria have ER contact sites. At MERCs, mitochondria and the ER are located between ∼10 and ∼50 nm from each other, and the average length of MERCs ranges from 145 to 270 nm (47,48). The close association between the ER and mitochondria is due to the presence of tethers that link both organelles and keep these contact sites intact. Mitofusin 1 and 2 are mitochondrial shaping proteins that stabilize the interaction of mitochondrion to different organelles.

The in-depth proteomic profiles of MERCs derived from the rat heart revealed 1871 resident proteins (49), a number comparable with that of the previously reported MERC proteome from HEK293 cells, mouse liver, and brain (6,50,51). Comparative mass spectrometry across different tissue types and species revealed a core set of 170 conserved proteins. Notably, we found the presence of 34 ion channels and 92 transporters, emphasizing their functional role in maintaining ion homeostasis and metabolite exchange in MERCs (Tables S1 and S2). In the first study on the MERC proteome, the prominent ion channels discovered were VDAC1, VDAC2, VDAC3, and CLIC4 (52). Out of the 1212 proteins in the later study (6), the ion channels found were chloride channel 6 (ClC6), CLIC4, voltage-gated potassium channel subfamily A member 1 (K_v1.1), voltage-gated potassium channel subfamily A member 2 (K_v1.2), voltage-gated potassium channel subfamily A member 4 (K_v1.4), potassium/sodium hyperpolarization-activated cyclic nucleotide-gated channel 1 (HCN1), potassium/sodium hyperpolarization-activated cyclic nucleotide-gated channel 2 (HCN2), VDAC1, VDAC2, and regulatory subunits of Ca²⁺ channels (voltage-dependent calcium channel γ8 subunit [CACNG8], and voltage-dependent calcium channel subunit α2/δ3 [CACNA2D3]). In both studies, VDAC1, VDAC2, and CLIC4 were the prominent anion channels (6,52).

Recently, MERCs were implicated in Ca²⁺-mediated signaling as well as in Ca²⁺ flux. At these contact sites, IP3R and VDAC are postulated to form a Ca²⁺ tunnel (53). The Ca²⁺ tunnel is supported by GRP75 (glucose-regulated protein 75), which is homologous to heat shock protein 70. GRP75 tethers IP3R to VDAC, which regulates mitochondrial Ca²⁺ levels (54). In aging rats, disruption of MERC integrity, marked by a significant reduction of VDAC1, causes the disassembly of the IP3R-GRP75-VDAC1 tethering complex. This disruption impairs Ca²⁺ dynamics and destabilizes the ER-mitochondria organization network, potentially affecting ER-mitochondria interactions, lipid transport, and mitochondrial protein import (49). In addition to Ca²⁺, there is a substantial gradient for Cl⁻ ions between the ER and mitochondrion as well as the cytosol (22). Regulation of Cl⁻ in the cytosol and organelles is vital for pH and cell/organelle volume regulation. In the heart, CLIC4 was characterized in MERCs (26), but there is no evidence of the existence of functional Cl⁻ channels or “Cl⁻ tunnels” in MERCs. Active transport of Cl⁻ at the MERCs has not been reported, but diffusion, due to the existing concentration gradient of Cl⁻, will facilitate its transport from the mitochondrion to the ER. The transport of Cl⁻ can be described by Fick’s and Einstein’s diffusion theories. According to Fick’s laws, Cl⁻ will move from the mitochondrion or cytosol (high concentration) to the ER (low concentration), forming a concentration gradient that changes with time. Einstein’s diffusion equation (55) indicates that doubling the distance of MERCs will slow down the diffusion time of Cl⁻ by fourfold. In cellular membranes, the diffusion of ions is facilitated by ion channels and, to date, no functional anion channels have been reported in MERCs. However, it is obvious that the Cl⁻ channels are essential for these contact sites.

The major challenge in measuring the functional activity of ion channels in MERCs is the lack of direct approaches. There are several biochemical and microscopy-based approaches used to identify proteins in MERCs. A proximity-labeling-based method, Contact-ID (56), was recently used to identify proteins localized in MERCs. Using Contact-ID between ER and mitochondria, 115 unique proteins were identified in MERCs (56). The majority of the proteins identified were implicated in protein, lipid, and vesicle transport, autophagy, protein degradation, and apoptosis (57). In addition, three ion channels, namely chloride channel CLIC such as 1 (CLCC1), mitochondrial calcium uniporter protein, CLIC4, and VDACs 1–3 (56). CLCC1 is known to transport anions in planar bilayers, but the channel is proposed to present in ER and renamed as ER anion channel 1 (20). Similarly, ClC6 is present in the Golgi apparatus and is also predicted to be present in MERCs. CLCC1 (20), ClC6 (58), and VDACs (59) lack specific blockers but are blocked by DIDs.

MERCs are inaccessible to the traditional patch-clamp approach. We used a planar bilayer system in which purified MERCs were incorporated into lipid bilayers. Our approach, using recording solutions containing NMDG-Cl, will exclusively select for Cl⁻ channels. Surprisingly, there are several distinct Cl⁻ currents present in MERCs. Using a combination of pharmacological, biochemical, and immunocytochemical approaches, we narrowed down CLIC3, CLIC4, and VDACs as putative contributors to Cl⁻ currents in MERCs. VDAC is already implicated in Ca²⁺ flux at contact sites. CLIC4 has been shown to be present exclusively in MERCs and not in mitochondrial membranes, where CLIC4 is involved in modulating ER and mitochondrial Ca²⁺ homeostasis. Additionally, CLIC4 was implicated in protecting the heart from ischemia-reperfusion injury. CLIC3 is identified as a novel putative Cl⁻ channel, with its functional role in MERCs implicating both CLIC3 and CLIC4 as Cl⁻-conducting channels. Electrophysiological recordings from planar lipid bilayers incorporating the outer mitochondrial membrane revealed single-channel conductances of 600–750, 315, and 420 pS in symmetrical 100 mM KCl. Under these conditions, the majority of VDAC conductances were observed at 650 pS in the fully open state, but the most commonly observed conductance at physiological conditions is around 300 pS (60). We observed conductances at ∼170–210 pS under symmetrical 150 mM NMDG-Cl conditions, further supporting our claim for the presence of functional VDACs in MERCs. The difference in conductances reported in KCl versus NMDGCl could arise from the diffusion potentials of mobile ions in the channel. For KCl, the diffusion potential is similar because the mobilities of K⁺ and Cl⁻ are very close (7.62 × 10⁻⁴ and 7.92 × 10⁻⁴ cm² s⁻¹ V⁻¹, respectively) (61), but for NMDG-Cl there is hardly any mobility for cations as the membranes are impermeable to NMDG (62). The major limitation of using planar bilayers for recording activities from MERCs is the possibility of contamination from other mitochondrial or organellar membranes, as intricate subcellular organelles are difficult to fractionate.

In summary, we demonstrated that a combination of electrophysiological, biochemical, and immunocytochemical approaches can be successfully applied to study the functional presence of ion channels in MERCs. We have discovered that functional anion channels are present in MERCs (Fig. 6). Thus, we expect that this approach can be used to detect the presence of other ion channels in MERCs as well as any membrane contact site (e.g., ER-Golgi, ER-lysosome, ER-plasma membrane, mitochondria-lipid droplet, and extracellular vesicle-mitochondria) and intracellular organelles (39) in different cell types. These studies can also be carried out in physiological and stress conditions. Stress can disrupt ionic homeostasis, which can cause a change in the distance between the mitochondrion and the ER, therefore disrupting MERCs. These disruptions can result in a dynamic loss of intra-organellar transport of ions, lipids, and proteins. Hence, increasing our knowledge regarding the molecular aspects and functions of different ion channels and transporters in MERCs is the only way to provide future therapeutic targets for a wide range of pathological conditions.

Figure 6.

Anion channels of MERCs. VDAC, CLIC3, and CLIC4 are the putative anion channels associated with MERCs. Mitofusin 2 (MFN2), known for tethering different organelles to the mitochondrion is also present in MERCs.

Acknowledgments

S.K.S. was supported by The Ohio State University President’s Predoctoral Fellowship, The Ohio State University Department of Physiology and Cell Biology Margaret T. Nishikawara Merit Scholarship Endowment in Physiology, and The Ohio State University Graduate School’s Alumni Grants for Graduate Research and Scholarship Program. We thank Ms. Marykate Hill for editing and proofreading the manuscript. This work was supported by the National Heart, Lung, and Blood Institute (HL133050 and HL157453) and the American Heart Association–Transformational Project Award (965031) to H.S.

Author contributions

S.K.S., D.G., S.K.R., and H.S. designed and performed experiments, analyzed the data, and wrote and revised the manuscript. A.G., A.S., H.R.B., J.T., V.L.-C., D.P., J.D., and S.G.R. performed experiments and edited the article. H.S. conceptualized the article and acquired the funding.

Declaration of interests

The authors declare no competing interests.

Editor: Manu Ben Johny.