Mitochondria are tiny organelles, less than a micrometre across, found inside almost all eukaryotic cells. Their main function is to act as the ‘power plant’ of the cell, generating adenosine triphosphate or ATP, which is the source of chemical energy for cellular processes. Beyond generating ATP, mitochondria perform many other functions: they contribute to various signalling pathways; they influence cellular differentiation; and they are involved in processes related to cell death.

Mitochondria are quite distinctive in appearance—they are enclosed by two membranes, a porous outer one and a largely impermeable inner membrane. Most mitochondrial functions involve proteins that control the movement of various molecules and ions across the inner membrane. One particularly important ion that must pass through this membrane is calcium; once inside the mitochondria, these calcium ions regulate cell survival and the generation of ATP.

Although several calcium import mechanisms exist, the best-studied pathway involves a pore-forming protein complex called the mitochondrial calcium uniporter. This ion channel has an exquisite selectivity, allowing only calcium into mitochondria even when other ions outnumber it a million-fold. Previously, researchers had identified several genes that are required for the formation of the uniporter, but it had not been established which of these encodes the central pore through which the calcium ions pass. Now, Chaudhuri et al. have shown that one of these—a gene called mitochondrial calcium uniporter (MCU)—codes for the protein subunit that creates the pore.

Prior studies used optical methods or purified proteins to study genes encoding the uniporter complex, producing controversial results regarding pore identity. This study uses a much more direct assay, namely electrophysiology performed on mitochondrial inner membranes. To access the inner membrane, the authors stripped off the outer membrane from whole mitochondria, and made them expand. By using a technique called voltage-clamping, Chaudhuri et al. were able to precisely measure calcium ion movement through intact or mutated channels. This technique controls confounding factors and minimizes the effect of contaminants that can plague interpretation of data acquired by other methods. They showed that blocking the expression of the MCU gene reduced the flow of calcium ions through the uniporter, whereas increasing MCU expression increased calcium transport.

One unique feature of the mitochondrial calcium uniporter is that it can be blocked by a dye called ruthenium red. Chaudhuri et al. used this property to confirm that the MCU gene encodes the pore-forming subunit of the channel complex—they identified a single point mutation in MCU that did not affect the channel’s ability to transport calcium ions, but did abolish its sensitivity to ruthenium red. Together, these results show that the MCU gene encodes the pore of the mitochondrial calcium uniporter, and should lead to further research into the physiology and structure of this channel.

Since the initial demonstration that mitochondria take up substantial amounts of cytoplasmic Ca²⁺ (Deluca and Engstrom, 1961), detailed studies have revealed that this uptake can sculpt the cytoplasmic Ca²⁺ transient (Wheeler et al., 2012), enhance ATP synthesis (Balaban, 2009), and trigger cell death (Zoratti and Szabo, 1995). Of several pathways for Ca²⁺ entry, a uniporter found in the inner membrane possesses the largest capacity for uptake and was shown to be a highly Ca²⁺-selective ion channel (Kirichok et al., 2004). However, despite this substantial progress, the identities of the genes encoding the functional uniporter were largely unknown until only recently. In the past several years, investigators from several laboratories have identified mitochondrial calcium uptake 1 (MICU1, formerly Cbara1) (Perocchi et al., 2010), mitochondrial calcium uniporter (MCU, formerly Ccdc109a) (Baughman et al., 2011; De Stefani et al., 2011), and mitochondrial calcium uniporter regulator 1 (MCUR1, formerly Ccdc90a) (Mallilankaraman et al., 2012) as potentially integral components of this channel.

These discoveries have spurred an explosion of research into mitochondrial Ca²⁺ transport, yet an outstanding question in this field is which, if any, of these several genes forms the pore subunit of the channel. So far, the discovery of each of these genes occurred via Ca²⁺ imaging assays following RNAi-mediated knockdown. Although extremely useful for screening, this technique is ill-suited for separating modulatory effects from altered expression of the channel-forming subunit itself, and each of these genes was found to inhibit Ca²⁺ uptake via this assay. This difficulty arises because Ca²⁺ imaging measures Ca²⁺ entry, which depends critically on (i) the number of open portals, (ii) the voltage gradient, and (iii) the Ca²⁺ concentration gradient driving flux, and none of these is controlled during imaging experiments. In fact, changes in Ca²⁺ buffers or efflux pathways can alter matrix Ca²⁺, and changes in pH or other divalents can alter indicator fluorescence, without reflecting a true difference in uniporter activity. Rather, the ideal method for assaying electrogenic ion transport is voltage-clamping, which allows precise control of the voltage and concentration gradient and thus accurate measurement of the changes in channel density.

Initial attempts to apply voltage-clamping focused on MCU, which possesses two transmembrane domains and highly-conserved acidic residues in a putative pore domain (Baughman et al., 2011; De Stefani et al., 2011; Bick et al., 2012). However, although MCU peptides reconstituted into lipid bilayers showed some activity in single-channel recordings (De Stefani et al., 2011), these experiments failed to recapitulate the typical single-channel behavior of the uniporter, most importantly its characteristic conductance and long-lasting openings (99% open probability) (Kirichok et al., 2004). Although such a discrepancy may be explained by the incorporation of the channel into bilayers in the absence of regulatory subunits, lipid bilayer reconstitution is also notorious for false positive findings. Since it detects the activity of single molecules, channel-like behavior can be measured from trace contaminants after even crystallography-grade purification (Accardi et al., 2004). Thus, the identity of the pore-forming subunit remains to be definitively established.

To overcome the limitations of single-channel recordings, we chose to voltage-clamp whole mitoplasts (mitochondria stripped of their outer membrane and expanded osmotically to allow access to the inner membrane) (Figure 1A). This allows electrophysiological interrogation of all the channels assembled in their native milieu in individual mitochondria. Moreover, this technique allows us to express mutated channels and examine mitochondrial Ca²⁺ currents (I_MiCa) for altered key features. Using this method, we show here that MCU does recapitulate key features of I_MiCa, and show that a single point-mutation can generate resistance to pharmacologic inhibition.

Figure 1

Download asset Open asset

Stable transfection of HEK-293T cells with a short-hairpin RNA targeting MCU produced a substantial reduction in the protein when assayed by Western blot (Figure 1B) or quantitative real-time polymerase chain reaction (17 ± 5% transcripts remaining compared to shGFP). We isolated mitoplasts from these cells using the Kirichok protocol (Fedorenko et al., 2012; Fieni et al., 2012; Figure 1A). As expected from Ca²⁺-imaging experiments (Baughman et al., 2011), mitoplasts from control cells showed robust I_MiCa during voltage ramps from −160 mV to +80 mV (Figure 1C). Because I_MiCa features a half-saturation value (K_0.5) of 20 mM [Ca²⁺]_bath, we maximized current by recording in a 100 mM Ca²⁺ gluconate bath solution (Kirichok et al., 2004). Utilizing high external Ca²⁺ allows us to conclude that changes observed after modifying MCU expression is due to altered channel levels rather than modulation of K_0.5, which might be set by accessory subunits. Other critical features of I_MiCa replicated in HEK-293T cells include its strong inward-rectification and high-affinity blockade by ruthenium red (RuR, 87 ± 2% inhibition in 100 nM RuR, Figure 1C,E).

Compared to the control condition, I_MiCa in mitoplasts from shMCU-expressing cells was markedly smaller (Figure 1D). The RuR-sensitive component of total Ca²⁺ current was reduced by 78 ± 14% (p<0.001, Figure 1E), with no significant difference in the RuR-insensitive residual component, suggesting that the knockdown was specific to I_MiCa and not a generalized reduction in membrane conductance. Moreover, differences were not due to alterations in mitochondrial structure, as mitoplast capacitance, a surrogate for inner membrane surface area (100µm²/pF), was consistent across all conditions tested here (shGFP: 0.48 ± 0.10 pF, shMCU: 0.34 ± 0.09 pF, p>0.05).

Next, we examined if overexpression of wild-type or mutant human MCU proteins substantially changed I_MiCa. We focused on an MCU mutated to encode an alanine at a highly-conserved serine in the putative pore-forming loop (S259A) (Baughman et al., 2011; Bick et al., 2012). The wild-type and mutant MCU proteins were tagged with a carboxyl-terminal FLAG epitope, and localized appropriately to mitochondria (Figure 2A,B). To confirm targeting to the inner membrane, we treated HEK-293T mitochondrial fractions with increasing concentrations of digitonin, and assayed for enzymatic digestion with proteinase K (Baughman et al., 2011; Figure 2C,D). Since the FLAG-tagged proteins were protected from digestion to a comparable extent as peptidyl-prolyl cis-trans isomerase F (PPIF, a matrix protein), whereas proteins of the outer- and inter-membrane spaces were not, the MCU constructs are appropriately targeted with their carboxyl-termini facing the mitochondrial matrix. Turning to our whole-mitoplast recordings (Figure 2E,G), we found that overexpression of wild-type MCU-Flag produced a robust increase in I_MiCa of approximately 3.4-fold compared to endogenous HEK-293T currents (compare to Figure 1C,E). This enhanced I_MiCa retained its sensitivity to RuR (Figure 2E,G).

Figure 2

Download asset Open asset

Finally, we studied the S259A-MCU mutant to see if it disrupted key features of I_MiCa. In Ca²⁺-imaging experiments, this mutant had preserved Ca²⁺ uptake but diminished sensitivity to uniporter inhibitors (Baughman et al., 2011). At baseline, mitoplasts from HEK-293T cells transfected with the S259A mutant had capacitances similar to mitoplasts after wild-type MCU overexpression (MCU-Flag: 0.32 ± 0.06 pF, S259A: 0.29 ± 0.10 pF, p>0.05). Moreover, S259A overexpression mirrored the increase in I_MiCa seen after wild-type MCU transfection, confirming a fully-functional channel (Figure 2F,G). However, this variant displayed markedly decreased sensitivity to RuR, with minimal inhibition at 100 nM. Since overexpression occurred on a background of endogenous channels, this mutant appears to act in a dominant-negative fashion. In particular, the S295A RuR-inhibited fraction (148 ± 33 pA/pF, Figure 2G) was much less than the RuR-inhibited fraction in endogenous I_MiCa (372 ± 42 pA/pF, Figure 1F). This suggests that, since the channel is an oligomer of multiple MCU subunits (Baughman et al., 2011), mutant S259A-MCU can incorporate with endogenous, wild-type subunits to form a Ca²⁺-conducting channel, which nonetheless has drastically-reduced RuR sensitivity.

In our study, we use the most direct assay available to measure mitochondrial Ca²⁺ currents—whole-mitoplast electrophysiology (Kirichok et al., 2004). By isolating mitoplasts and recording directly from all the channels embedded in the inner membrane, we measure the ensemble current and thus avoid confounding minor contaminants. This method, moreover, is the same technique that originally demonstrated that the mitochondrial Ca²⁺ uniporter was an ion channel, allowing us to make a direct comparison to the characteristic features of this current (I_MiCa), and to determine if the genetically-modified channels could alter them.

We focused on MCU, the leading candidate for the pore-forming subunit, as it possesses two transmembrane domains and has highly-conserved acidic residues in a putative pore-like domain. First, we show that reducing MCU transcripts via knockdown, or enhancing them through overexpression, produce parallel changes in I_MiCa. Second, we show that a central feature of I_MiCa—its exquisite sensitivity to blockade with ruthenium red—can be abolished via a single point mutation in the pore domain. MCU subunits carrying this mutation act in a dominant-negative fashion, producing functional channels that conduct Ca²⁺ ions but are largely insensitive to ruthenium red. We conclude that the functional uniporter pore at the mitochondrial inner membrane is formed by MCU multimers. We believe these experiments put to rest the outstanding question of the pore identity, firmly establishing that MCU produces I_MiCa.

1. 1. Accardi A
   2. Kolmakova-Partensky L
   3. Williams C
   4. Miller C

   (2004) Ionic currents mediated by a prokaryotic homologue of CLC Cl⁻ channels

   J Gen Physiol 123:109–119.

   https://doi.org/10.1085/jgp.200308935

   • Google Scholar
2. 1. Balaban RS

   (2009) The role of Ca(2+) signaling in the coordination of mitochondrial ATP production with cardiac work

   Biochim Biophys Acta 1787:1334–1341.

   https://doi.org/10.1016/j.bbabio.2009.05.011

   • Google Scholar
3. 1. Baughman JM
   2. Perocchi F
   3. Girgis HS
   4. Plovanich M
   5. Belcher-Timme CA
   6. Sancak Y

   et al. (2011) Integrative genomics identifies MCU as an essential component of the mitochondrial calcium uniporter

   Nature 476:341–345.

   https://doi.org/10.1038/nature10234

   • Google Scholar
4. 1. Bick AG
   2. Calvo SE
   3. Mootha VK

   (2012) Evolutionary diversity of the mitochondrial calcium uniporter

   Science 336:886.

   https://doi.org/10.1126/science.1214977

   • Google Scholar
5. 1. De Stefani D
   2. Raffaello A
   3. Teardo E
   4. Szabo I
   5. Rizzuto R

   (2011) A forty-kilodalton protein of the inner membrane is the mitochondrial calcium uniporter

   Nature 476:336–340.

   https://doi.org/10.1038/nature10230

   • Google Scholar
6. 1. Deluca HF
   2. Engstrom GW

   (1961) Calcium uptake by rat kidney mitochondria

   Proc Natl Acad Sci USA 47:1744–1750.

   https://doi.org/10.1073/pnas.47.11.1744

   • Google Scholar
7. 1. Fedorenko A
   2. Lishko PV
   3. Kirichok Y

   (2012) Mechanism of fatty-acid-dependent UCP1 uncoupling in brown fat mitochondria

   Cell 151:400–413.

   https://doi.org/10.1016/j.cell.2012.09.010

   • Google Scholar
8. 1. Fieni F
   2. Bae Lee S
   3. Jan YN
   4. Kirichok Y

   (2012) Activity of the mitochondrial calcium uniporter varies greatly between tissues

   Nat Commun 3:1317.

   https://doi.org/10.1038/ncomms2325

   • Google Scholar
9. 1. Kirichok Y
   2. Krapivinsky G
   3. Clapham DE

   (2004) The mitochondrial calcium uniporter is a highly selective ion channel

   Nature 427:360–364.

   https://doi.org/10.1038/nature02246

   • Google Scholar
10. 1. Mallilankaraman K
    2. Cardenas C
    3. Doonan PJ
    4. Chandramoorthy HC
    5. Irrinki KM
    6. Golenar T

    et al. (2012) MCUR1 is an essential component of mitochondrial Ca(2+) uptake that regulates cellular metabolism

    Nat Cell Biol 14:1336–1343.

    https://doi.org/10.1038/ncb2669

    • Google Scholar
11. 1. Perocchi F
    2. Gohil VM
    3. Girgis HS
    4. Bao XR
    5. McCombs JE
    6. Palmer AE

    et al. (2010) MICU1 encodes a mitochondrial EF hand protein required for Ca(2+) uptake

    Nature 467:291–296.

    https://doi.org/10.1038/nature09358

    • Google Scholar
12. 1. Sancak Y
    2. Bar-Peled L
    3. Zoncu R
    4. Markhard AL
    5. Nada S
    6. Sabatini DM

    (2010) Ragulator-Rag complex targets mTORC1 to the lysosomal surface and is necessary for its activation by amino acids

    Cell 141:290–303.

    https://doi.org/10.1016/j.cell.2010.02.024

    • Google Scholar
13. 1. Wheeler DG
    2. Groth RD
    3. Ma H
    4. Barrett CF
    5. Owen SF
    6. Safa P

    et al. (2012) Ca(V)1 and Ca(V)2 channels engage distinct modes of Ca(2+) signaling to control CREB-dependent gene expression

    Cell 149:1112–1124.

    https://doi.org/10.1016/j.cell.2012.03.041

    • Google Scholar
14. 1. Zoratti M
    2. Szabo I

    (1995) The mitochondrial permeability transition

    Biochim Biophys Acta 1241:139–176.

    https://doi.org/10.1016/0304-4157(95)00003-A

    • Google Scholar

Author details

1. Dipayan Chaudhuri

   1. Cardiovascular Research Center, Massachusetts General Hospital, Boston, United States
   2. Department of Cardiology, Howard Hughes Medical Institute, Boston Children’s Hospital, Boston, United States

   Contribution

   DC, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

2. Yasemin Sancak

   1. Departments of Molecular Biology and Medicine, Massachusetts General Hospital, Boston, United States
   2. Department of Systems Biology, Harvard Medical School, Boston, United States

   Contribution

   YS, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

3. Vamsi K Mootha

   1. Departments of Molecular Biology and Medicine, Massachusetts General Hospital, Boston, United States
   2. Department of Systems Biology, Harvard Medical School, Boston, United States

   Contribution

   VKM, Conception and design, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

4. David E Clapham

   1. Department of Cardiology, Howard Hughes Medical Institute, Boston Children’s Hospital, Boston, United States
   2. Department of Neurobiology, Harvard Medical School, Boston, United States

   Contribution

   DEC, Conception and design, Drafting or revising the article

   For correspondence

   dclapham@enders.tch.harvard.edu

   Competing interests

   The authors declare that no competing interests exist.

• 1,832

  Page views

• 319

  Downloads

• 146

  Citations

Article citation count generated by polling the highest count across the following sources: Scopus, Crossref, PubMed Central.