Materials and Reagents

Pipette tips
Cell culture dishes (Corning, catalog number: 430293 )
Pasteur pipette
15 ml tube
HEK 293 cells (Incubation: 37 °C, 5.0% CO₂) (ATCC, catalog number: CRL-1573 )
DMEM (high glucose, no glutamine) (store at 4 °C) (Thermo Fisher Scientific, Gibco^TM, catalog number: 11960051 )
Trypsin
Premium fetal bovine serum (store at -20 °C) (Atlanta Biologicals, catalog number: S11150 )
L-glutamine (store at -20 °C) (Thermo Fisher Scientific, Gibco^TM, catalog number: 25030081 )
HEPES (store at RT) (Sigma-Aldrich, catalog number: H3375-1KG )
Potassium chloride (KCl) (store at RT) (Fisher Scientific, catalog number: P217-500 )
Potassium phosphate dibasic (K₂HPO₄) (store at RT) (Fisher Scientific, catalog number: P290-500 )
Magnesium chloride hexahydrate (MgCl₂·6H₂O) (store at RT) (Fisher Scientific, catalog number: M33-500 )
Potassium hydroxide (KOH) (store at RT) (for pH to 7.4) (Fisher Scientific, catalog number: P250-500 )
Succinate (store at RT) (Sigma-Aldrich, catalog number: S3674-100G )
Calcium Green 5N - hexapotassium salt (CG-5N) (Thermo Fisher Scientific, Invitrogen^TM, catalog number: C3737 )
Note: Store at -20 °C as powder; 4 °C dissolved in ddH₂O. Recommended: dissolve to a concentration of 0.5 μM for stock solution.
Digitonin (Sigma-Aldrich, catalog number: D141 )
Note: Store at -20 °C as powder. Dissolve in ddH₂O immediately before use. Recommended: dissolve to a concentration of 30 mM for stock solution.
Ruthedium 360 (Ru360) (Santa Cruz Biotechnology, catalog number: sc-222265 )
Note: Store at -20 °C as powder; 4 °C dissolved in ddH₂O.
Calcium Chloride Dihydrate (Fisher Scientific, catalog number: C79-500 )
Recommended: Dissolve in water to a concentration of 10 mM for stock solution.
Growth media (see Recipes)
Wash buffer (WB) (see Recipes)
Recording buffer (RB) (see Recipes)

Equipment

Pipettes
Incubator (Fisher Scientific, Isotemp^TM, catalog number: 13-255-26 )
Centrifuge (Thermo Fisher Scientific, model: Heraeus^TM Labofuge^TM 400 , catalog number: 75008164)
Quartz cuvette (Fisher Scientific, Fisherbrand, catalog number: 14-958-128 )
Small stir bar (Hach, catalog number: 2095349 )
Thermo Cell Holder with Stirrer (Hitatchi, model: 251-0148 )
F-2500 Fluorescence spectrophotometer (Hitachi, model: 251-0090 )

Software

IGOR Pro
Microsoft Excel

Procedure

Cell (HEK-293) preparation
1. Carefully remove growth media from a confluent 10-cm dish (i.e., with a Pasteur pipette. See Note 3 for more on cell density). Add 10 ml WB (Wash Buffer) warmed to 37 °C to the plate with 10 ml pipet, pipetting up and down to gently lift cells from the bottom of the plate, and transfer WB-cell slurry to a 15 ml tube. These preparations are to be carried out on the bench at room temperature. If cells are strongly adherent and not easily dislodged by gentle up-and-down pipetting, add 0.7 ml of Trypsin (or just enough to evenly cover cells) warmed to 37 °C after washing cells once with 10 ml WB and let stand in a 37 °C incubator for 2-3 min. Remove cells from the incubator. Cells should now be easily displaced by gentle up-and-down pipetting with 10 ml WB. Transfer WB-cell slurry to a 15 ml tube.
2. Gently pellet cells by centrifugation at room temperature at 1,000 x g for 5 min. Discard supernatant.
3. Wash pellet by re-suspending cells in 10 ml WB, spin for 5 min at 1,000 x g, and discard supernatant.
4. Resuspend cells in 2.5 ml recording buffer (RB), also warmed to 37 °C. Transfer 2.0 ml of this slurry to the quartz cuvette with a small stir bar. (The remaining 0.5 ml can be used separately to analyze protein content via Western Blot analysis. See Note 5 for more on Western Blots)
  Note: Gently flick cuvette to remove bubbles before the start of the experiment, as these can create unwanted noise in the data.
Flux experiment
1. Place cuvette with RB-cell slurry into the cell-holder/stirrer affixed to the spectrophotometer. Turn on a stirrer and stir cells slowly (see Note 1).
2. Set spectrophotometer to these parameters:
  1. Excitation wavelength: 506 nm
  2. Emission wavelength: 531 nm
  3. Em/Ex slit width: 2.5/2.5 nm
  4. Time scan
    1. 600 sec (or longer)
    2. 0 sec delay
  5. PMT Voltage: 700 Volts
  6. Response: 0.04 sec
  7. Data Mode: Fluorescence
3. Start recording time scan. See Figure 1 below.
  
  Figure 1. Starting the Recording. Left: cartoon of quartz cuvette with 2 ml recording buffer suspending wild-type (WT) HEK cells, loaded into the stirrer on the fluorescence spectrophotometer (small stir bar not shown). Right: spectrophotometric trace about 45 sec after starting the recording.
4. Add CG-5N to a final concentration of 0.25 nM (i.e., 1 μl of 0.5 μM stock). CG-5N is a membrane impermeable calcium-sensitive fluorophore (Kd = 14 μM). The fluorescence signal will go up with the addition of the fluorophore as trace amounts of calcium are present in the RB. See Figure 2 below.
  
  Figure 2. Adding CG-5N. Adding CG-5N to a concentration of 0.25 nM (left) will increase the fluorescence signal in the trace profile (right).
5. Add digitonin to a final concentration of 30 μM (i.e., 2 μl of 30 mM stock). Digitonin permeabilizes the plasma membrane by extracting cholesterol, leaving those intracellular membranes lacking cholesterol (e.g., the mitochondrial inner membrane) intact. This brings mitochondria in direct contact with the buffer solution and susceptible to influence by experimental reagents (the outer mitochondrial membrane is already in a steady-state with the cytoplasm with regards to small molecules like the reagents used in this protocol). CG-5N (membrane-impermeable) is afterward reporting ‘extra-mitochondrial’ calcium levels. A typical dip in fluorescence is almost always observed upon the addition of digitonin, as can be seen in Figure 3 below.
  
  Figure 3. Adding digitonin. Adding digitonin to a concentration of 30 μM (left) will cause the spectrophotometric trace profile to drop (right).
6. Add CaCl₂ to a final concentration of 10 μM (i.e., 2 μl of 10mM stock). The fluorescence signal will go up as calcium binds to CG-5N. If the cells are harboring activated mitochondria containing functional MCU (i.e., WT-HEK cells), a precipitous declination in fluorescence will immediately follow, signifying MCU-mediated calcium uptake (see Figure 4).
  Note: If cells lack MCU, the fluorescence signal in the trace profile will increase but then will immediately flatten out, showing that the observed declination in the trace profile of WT cells after the addition of calcium is due to MCU-mediated calcium uptake AND NOT by other intra-cellular calcium uptake pathways such as SERCA (Sarco/Endoplasmic Calcium-ATPase) (Note 6).
  
  Figure 4. Adding calcium. Adding calcium to a final concentration of 10 μM (left) will produce a sharp increase in the trace profile (right). The following drop in fluorescence signal is indicative of MCU-mediated calcium flux.
7. After some time has passed, typically between 30 sec to 1 min, add Ru360 to a final concentration of 0.5 μM. Ru360 (i.e., 0.5 μl of 2 mM stock) is a potent inhibitor of MCU (Kd = 340 pM) and will cause the spectrophotometric trace to flatten as the steady-state exchange between the CG-5N-calcium-bound and CG-5N-calcium unbound species returns due to the lack of calcium mobility away from CG-5N. This step testifies to the dependence of MCU on the removal of calcium from the buffer solution observed in the previous step. See Figure 5 below; here, Ru360 is added to the cuvette after a second shot of 10 μM calcium.
  
  Figure 5. Ru360. Adding Ru360 to a final concentration of 0.5 μM (left) will cause the spectrophotometric trace profile to flatten (right).

Data analysis

For reference data, please refer to Figure 4–figure supplement 2C, and Figure 4–figure supplement 3A in our paper ‘Dual functions of a small regulatory subunit in the mitochondrial calcium uniporter complex’ by following the link provided below:
http://cdn.elifesciences.org/elife-articles/15545/figures-pdf/elife15545-figures.pdf?_ga=1.67863441.1487892113.1470405893
Raw data from the spectrophotometric traces were uploaded into IGOR Pro and the slope of the a.u./time trace, starting immediately following the addition of calcium and for the following 10 sec, was calculated from each graph. The average of 3 slopes from cells expressing WT-hMCU and WT-hEMRE was calculated in Microsoft Excel along with the concomitant standard error. These values were used to normalize all following experimental data sets against.

Notes

Stir bar speed
The speed of the stir bar can be an important but overlooked factor in determining the quality of the fluorescence trace profile. Spinning too fast can result in cell damage (cells will clump together in the cuvette), while spinning too slowly may limit the speed of reagent mixing within the cuvette, which can be seen in the spectrophotometric data primarily as enhanced noise and hyperbolic-like transitions from one steady-state to another upon the addition of a new reagent to the cuvette. In general, we’ve found that starting the stir bar in its slowest setting first and then slowly ramping up speed is the best way to find the optimal speed leading to intact cells and quick rates of reagent mixing. In our hands, this ideal speed is, qualitatively, on the slow end within the range of possible stir-bar speeds.
Digitonin
We have found that digitonin from Sigma-Aldrich works particularly well for this experiment. Also, make digitonin stock fresh before each experiment, as this reagent tends to crash out of solution between experiments (within an hour).
Cell density
While the number of HEK cells used in an experiment is completely up to the experimenter, we have found the ideal number to be 2.0 x 10⁷, or 1 fully confluent 10 cm dish.
Cell types
We have performed this experiment exclusively with HEK 293 cells, and cannot say how it might work using other cell types.
Western Blot Analysis
Recording Buffer (RB) does not interfere Western Blot analysis. Simply spin the remaining 0.5 ml of cell slurry down, discard sup (RB), and lyse cells with ice cold lysis buffer (we found RIPA buffer works well for this). Lysing cells with about 50 μl RIPA buffer yields protein concentrations in a range appropriate for the comfortable loading of between 10 μg and 50 μg of protein in a 15 μl or 50 μl per well gel. A high-speed spin step at Step A4 is also required after lysis to pellet cell derbies and to keep the lysate from becoming ‘goopy’ and unmanageable during loading; keep the supernatant. Follow instructions for your lysis buffer of choice (add protease inhibitors, work on ice, etc.) and carefully quantify protein concentration after the high-speed spin with your method of choice. We always run a loading control gel on which we detect Actin (we load 10 μg protein for this blot). Other gels should be run on which to look for the protein of interest (typically MCU or one of its regulatory partners, but this obviously depends on the experiment and what’s desirable to detect). In general, expect to spend time optimizing western blots to see high-quality (high signal to noise) data.
A clear difference between the trace profiles of wild-type cells and MCU-knock out cells (cells in which MCU has been deleted from the genome) is that in the latter, the slope, after the addition of calcium, is nearly zero, while in the former, the slope is clearly negative. The profile’s negative slope after calcium addition is therefore indicative of MCU-mediated calcium uptake.

Figure 6. Spectrophotometric Profile of MCU-knockout Cells. Trace profile is flat after the addition of Ca²⁺ when MCU has been deleted from the genome.

Recipes

Growth media
90% (v/v) DMEM
4.5 g/L D-glucose, L-glutamate, sodium pyruvate
10% (v/v) premium fetal bovine serum
2 mM L-glutamine
Store at 4 °C
Notes:
1. Prepare media under sterile conditions, i.e., in a laminar flow hood.
2. Filter media through a 0.22 μm filter after preparation.
Wash buffer (WB)
20 mM HEPES
125 mM KCl
2 mM K₂HPO₄
1 mM MgCl₂
Adjust pH to 7.4 with KOH
Store at 4 °C
Recording buffer (RB)
WB (see above)
5 mM succinate
Store at 4 °C

Acknowledgments

This work was carried out in Dr. Christopher Miller’s lab (HHMI; Brandeis University) and was supervised by Dr. Ming-Feng Tsai.
The above protocol is a classic protocol which has been used for decades in the field of mitochondrial calcium handling. Our thanks to those original thinkers who gave it life and to those who have refined it over the years. The author declares that there are no conflicts of interest or competing interests.

References

Glancy, B. and Balaban, R. S. (2012). Role of mitochondrial Ca²⁺ in the regulation of cellular energetics. Biochemistry 51(14): 2959-2973.
Pacher, P. and Hajnoczky, G. (2001). Propagation of the apoptotic signal by mitochondrial waves. EMBO J 20(15): 4107-4121.
Wheeler, D., Groth, R., Ma, H., Barrett, C., Owen, S., Safa, P., and Tsien, R. (2012). Ca_V1 and Ca_V2 channels engage distinct modes of Ca²⁺ signaling to control CREB-dependent gene expression. Cell 149(5): 1112-1124.
Zoratti, M. and Szabo, I. (1995). The mitochondrial permeability transition. Biochim Biophys Acta 1241(2): 139-176.