Protocol

Systematic Monitoring of Protein Complex Composition and Abundance by Blue-Native PAGE

ARC Centre of Excellence in Plant Energy Biology, University of Western Australia, Crawley, 6009, Australia

¹Corresponding author (hmillar{at}cyllene.uwa.edu.au)

INTRODUCTION

Many polypeptides do not perform their functions as single autonomous units in vivo. Instead, multiple polypeptides associate to form higher molecular mass structures. Blue-native polyacrylamide gel electrophoresis (BN-PAGE) allows a range of the major protein complexes involved in such protein-protein interactions to be visualized simultaneously and in a single experiment. When combined with a second dimension of sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), the BN/SDS-PAGE procedure can resolve the complexes according to their molecular weight, as well as the subunits within each complex, according to the molecular weights of the subunits. Similarly, used in conjunction with differential in-gel electrophoresis (DIGE), it can accurately quantify changes in protein complex abundance or subunit composition between different samples, or between different complexes within the same sample. The following basic protocol describes sample preparation and gel casting for the first (BN-PAGE) and second (SDS-PAGE) dimensions. Variants are presented with and without DIGE labeling, along with the additional steps required for the fluorescence DIGE technique.

RELATED INFORMATION

For details on the BN-PAGE and DIGE procedures described in this protocol, see Schägger and von Jagow (1991) and Perales et al. (2005), respectively. Additional downstream applications and alternative second and third dimensions for BN-PAGE are described elsewhere (Wittig et al. 2006).

MATERIALS

Reagents

Acrylamide (40%, w/v)

Ammonium persulfate (APS; 10%, w/v)

Bis-Tris (300 mM; 6X anode buffer)

Adjust the pH to 7.0 at 4°C with HCl.

BN gel buffer (6X)

Cathode buffer (5X)

Cathode buffer, second dimension (pH 8.25)

Coomassie brilliant blue G250 (5%, w/v)

DIGE labeling kit, 5-nmol (GE Healthcare)

Glycerol (100%)

Lysine (10 mM)

N,N'-Methylenebisacrylamide (2%, w/v)

Overlay buffer

Sample to be analyzed

SDS (Sodium dodecyl sulfate)

SDS equilibration solution

Solubilization buffer, DIGE

Solubilization buffer, standard

TEMED

Tricine gel buffer, with and without glycerol

Tris (0.2 M; second-dimension anode buffer)

Adjust the pH to 8.9 with HCl.

Equipment

Aluminum foil (optional for DIGE gels; see note prior to Step 22)

Analytical software (e.g. DeCyder, GE Healthcare)

Beakers, assorted

Blotting paper

Gel plates for slab gel system, low-fluorescence (optional; for DIGE only)

Gel staining boxes

Gradient maker, 25 mL per chamber

Ice

Imaging hardware (e.g., fluorescence scanner or camera system equipped with suitable light source and filters; optional; for DIGE only)

Various laser and charge-coupled device (CCD) camera-based systems are available from different companies.

Incubator preset to 37°C (optional; see Step 11)

Injection needle

Magnetic stirrer and stir bars

Microcentrifuge preset to 4°C

Peristaltic pump

Pipettes and tips

Power supply, programmable

Slab gel system, horizontal, including 1.0- and 1.5-mm gel spacers

METHOD

Gel Preparation for First Dimension of BN-PAGE

The first dimension of BN-PAGE consists of a separating gel (usually a gradient of 4.5%-16% acrylamide) topped by a stacking gel. The decreasing pore size created by the gradient ensures optimal resolution of protein complexes (Fig. 1) . The values presented here are suitable for separating respiratory chain components but might need to be optimized for other sample types. Casting the gel takes several hours and should be completed before commencing sample preparation.

View larger version (26K): [in this window] [in a new window]

Figure 1. Schematic overview of a typical BN gel, including a separated sample. As the acrylamide concentration in the separating gel increases, the pore size of the gel decreases. (HMW) High molecular weight band; (LMW) low molecular weight band.

1. Assemble the slab-gel casting device using the 1.5-mm spacers (Fig. 2 ).

View larger version (40K): [in this window] [in a new window]

Figure 2. Typical BN gel casting assembly. (Inset) Close-up of injection needle inserted from the bottom through a hole in the base of the gel stand, projecting between the two glass plates.

The exact methodology depends on the system used; refer to the manufacturer’s instructions.

2. From the top, inject ~5 mL of water in between the glass plates.
This acts as a cover fluid to ensure the top of the separating gel is even.

3. Cool the assembly and gradient maker to 4°C before casting the gel.

4. Prepare heavy and light gradient solutions:

Reagent Amount to add (per each 10 mL) Light (4.5% acrylamide) Heavy (16% acrylamide) 6X BN gel buffer 1.7 mL 1.7 mL 40% Acrylamide 1.1 mL 4.0 mL 2% Methylenebisacrylamide 0.7 mL 2.5 mL Glycerol --- 1.9 mL H2O to 10 mL to 10 mL

The final volume depends on the dimensions of the gel system; scale the amounts to add accordingly. About 70% of the overall gel volume is needed for the separating gel; enough space for a stacking gel should be left on top (Fig. 2). Prepare light and heavy solutions in a 10:9 ratio. This provides a small nongradient area at the top of the gel consisting of only the light solution.

5. Close the valves of the gradient mixer. Place the light solution in the first chamber of the gradient maker (i.e., the one connected to the outlet). Place the heavy solution in the second chamber. Add a small magnetic stir bar to each chamber. Cool the solutions to 4°C.

6. Insert the injection needle through the bottom of the casting stand to reach just between the two glass plates (Fig. 2, inset).
Some manufacturers provide slots or holes in the gel stand. If not, extra holes may be drilled.

7. Place the gradient caster on a magnetic stirrer. Connect it to the injection needle in the gel assembly via a peristaltic pump.
If no pump is available, gravity can be used to cast the gel by placing the gradient caster higher than the gel assembly.

8. Add 45 µL of 10% APS stock per 10 mL of gel solution to the first chamber. Add 4.5 µL of TEMED per 10 mL of gel solution. Mix thoroughly. Open the outlet valve of the gradient mixer.
When using a pump, the initial speed should be ~20% of the maximum to avoid mixing of the light solution with the overlaying water phase. Check for air bubbles in or around the needle tip and, if necessary, remove them by slightly tapping or tilting the gel assembly.

9. Shortly before the level in the first chamber drops to the level of the second chamber, add 33 µL of 10% APS stock per 10 mL of gel solution to the second chamber. Add 3.3 µL TEMED per 10 mL of gel solution. Mix thoroughly. Open the valve between the two chambers when the levels are even.
The heavy solution should now flow into the first chamber and mix with the light solution.

10. Slowly increase the pump speed to maximum in small increments over a 10-min period.
The whole casting procedure should take ~20-30 min. Monitor closely in the final stage to avoid air bubbles being pumped into the gel. If this happens, the gel is not usable and the casting process must be repeated.

11. Place the gel at room temperature or in a 37°C incubator. Make sure the assembly is level. Wait until a clear separation line appears between the water and the set gel (~1 h).
12. Discard the water. Dry the portion of the glass plates above the gel with blotting paper. Avoid touching the gel in the process.

13. Prepare the stacking gel solution (4% acrylamide):

Reagent Amount to add (per 10 mL) 6X BN gel buffer 1.7 mL 40% Acrylamide 0.9 mL 2% Methylenebisacrylamide 0.6 mL H2O to 10 mL 10% APS 43 µL TEMED 4.3 µL

The final volume depends on the dimensions of the gel system; scale the amounts to add accordingly.

14. Cast the stacking gel. Slightly warm the solution and the gel assembly to 30°C-40°C. Allow the gel to set.
Warming the assembly facilitates formation of geometric wells. Best results are obtained using a 10-tooth comb.

15. Assemble the gel system. Fill the reservoirs with 1X cathode and anode buffers (diluted from the 5X cathode and 6X anode buffer stocks, respectively). Chill to 4°C before running the gel.

Sample Preparation for First Dimension of BN-PAGE

This protocol is suitable for solubilizing respiratory protein complexes and supercomplexes from different organisms. Although it is a good starting point, conditions should be optimized for other systems. The choice of detergent is critical: Only very mild detergents (e.g., nonionic or zwitterionic) are suitable. Test different detergents and protein:detergent ratios before commencing a study. Historically, different buffers have been utilized for different detergents (Bis-Tris for n-dodecyl maltoside, imidazole for Triton X-100, and HEPES for digitonin), but the actual influence of the buffer on solubilization efficiency is not known. Generally, use a buffer system that interferes the least with the gel run and adjust the concentration of the chosen detergent to the optimum. A standard one-dimensional (1D) BN-PAGE can be performed with 100-250 µg, whereas ~250-800 µg of protein is needed for a Coomassie-stained two-dimensional BN/SDS-PAGE. Silver- or fluorescent dye-stained 1D BN gels require ~50-100 µg of protein.

16. Prepare the samples, as appropriate for the method of analysis:

For standard BN-PAGE:

i. Determine the protein concentration of the samples to calculate the sample volumes corresponding to the chosen protein load.

ii. Centrifuge cell, organelle, or membrane preparations and discard the supernatant to concentrate the samples.

iii. Solubilize the samples by adding 10 µL of standard solubilization buffer containing the detergent of choice for every 100 µg of protein.

iv. Incubate the samples for 20 min on ice.

v. Add 0.5 µL of 5% Coomassie brilliant blue G250 for every 10 µL of sample.
See Troubleshooting.

vi. Centrifuge the samples at 20,000g for 10 min to remove any unsolubilized material.
Proceed to Step 17.

For DIGE:

The simplest DIGE experiment requires a minimum of three replicate sample sets of a biological treatment and a control sample.

vii. Prepare CyDye DIGE Fluors from the DIGE labeling kit according to the manufacturer’s specifications.

viii. Determine the protein concentration of the samples.

ix. Aliquot a volume equivalent to 50 µg of protein for each of the six samples to be labeled.

x. Combine 25 µg of each of the samples to create a standard containing all of the proteins and protein complexes found in the three sample sets.

xi. Centrifuge cell, organelle, or membrane preparations and discard the supernatant to concentrate the samples.

xii. Solubilize each of the samples in 20 µL of DIGE solubilization buffer containing digitonin (pH 8.0). Solubilize the standard (to be used on all three gels) in 60 µL of the DIGE solubilization buffer.
Using these volumes, higher detergent/protein ratios than in the standard protocol are obtained. This is not a problem when digitonin is used for solubilization, but might create issues with other detergents.

xiii. Add 1 µL of Cy3 or Cy5 (400 nmol) to the experimental samples and controls. Add 3 µL of Cy2 to the standard.
Swap the Cy3 and Cy5 dyes between the samples according to the manufacturer’s specifications to prevent dye-derived artifacts.

xiv. Incubate samples for 30 min on ice.

xv. Terminate the labeling reaction by adding 1 µL of 10 mM lysine. Incubate the samples for 10 min on ice.

xvi. Mix control and experimental samples from the same set with one-third of the standard (from Step 16.x). Centrifuge at 20,000g for 10 min to remove any unsolubilized material.

xvii. Add 3 µL of 5% Coomassie brilliant blue G250 to the supernatant. Mix.
See Troubleshooting.
Continue with Step 17.

Running the First Dimension of the BN-PAGE Gel

The quality of BN-PAGE benefits from slow gel entry of the protein complexes during the initial portion of the run, probably because the range of molecular masses in native samples is much higher than those of denatured samples. Ideally, use a programmable power supply to define a power gradient for the first 2-3 h. After the initial period, limit the current to 15 mA with the voltage rising to a maximum of 500 V over the next 12-15 h. For delicate samples or protein loads >500 µg, use a maximum current of 10 mA.

17. Load the samples (from Step 16.vi or Step 16.xvii) on the gel.

18. Run the gel at 4°C, typically at 100 V max for 1 h, followed by 15 mA max for 12-15 h.
Run BN-DIGE gels in the dark to avoid photobleaching of the fluorophores.

19. Cut lanes from the first dimension BN-PAGE gels.
These can be used immediately for second dimensions or can be stored at 4°C for several days.

Lane Preparation for Second Dimension BN/SDS-PAGE

For subsequent SDS-PAGE, protein complexes must be denatured before transfer to the second dimension gel.

20. Incubate gel lane(s) (from Step 19) in SDS equilibration solution for 30 min.
Incubate BN-DIGE gels in the dark to avoid photobleaching of the fluorophores.

21. Wash the lane(s) in water for 30 sec to remove surface reductant that might inhibit acrylamide polymerization.

Casting the Second Dimension SDS-PAGE Gel

The BN gel can be slipped between the plates and placed on top of a second-dimension SDS-PAGE gel cast thicker than the BN dimension and then fixed in place with agarose. Alternatively, in the method presented below, the BN-PAGE gel lane is placed between glass plates and the second-dimension gel is cast around it. The SDS-PAGE gel must be thinner than the BN dimension so that the latter is squeezed and fixed in place during the casting. The procedure is the same for standard and DIGE samples, but DIGE gels should be exposed as little as possible to light to avoid photobleaching of the fluorophores: Cover the gel assembly with aluminum foil or store it in the dark between the casting steps.

22. Allow the water from the wash step to drip off the BN gel lane. Place the lane on a gel plate in the region usually occupied by the stacking gel. Place the strip slightly angled from the horizontal position, with the low molecular weight region of the BN lane slightly higher than the high molecular weight end (Fig. 3 ).

View larger version (23K): [in this window] [in a new window]

Figure 3. Schematic overview of a typical second-dimension tricine-buffered gel including a BN gel lane. Slightly angling the BN gel lane simplifies casting of the stacking gel. (HMW) High molecular weight band; (LMW) low molecular weight band.

Angling the strip reduces the chance of air bubbles being trapped beneath it when the stacking gel is cast. The acrylamide concentration is lower at the low molecular weight end and the gel widens more when squeezed.

23. Position 1-mm spacers on the plate. Cover with the second plate. Align the assembly and tighten the clamps. Blot off any liquid that might have been pressed out of the gel during this process.

24. Prepare the separating and spacer gels:

Reagent Amount to add (per each 10 mL) Separating (16% acrylamide) Spacer (10% acrylamide) 40% Acrylamide 4.0 mL 2.5 mL 2% Methylenebisacrylamide 2.5 mL 1.5 mL Tricine gel buffer w/glycerol 3.3 mL --- Tricine gel buffer w/o glycerol --- 3.3 mL H2O to 10 mL to 10 mL 10% APS 33 µL 33 µL TEMED 3.3 µL 4.6 µL

The final volume depends on the dimensions of the gel system; scale the amounts to add accordingly. The separating gel should account for approximately half of the gel volume, whereas the spacer gel should take up only ~20%. This leaves about one-third for the stacking gel incorporating the first dimension gel strip (Fig. 3).

25. Cast the separating gel.

26. Immediately after pouring the separating gel (i.e., before it solidifies), slowly pour the spacer gel with a pipette.
Tilt the gel assembly nearly horizontal to avoid mixing the two solutions.

27. Overlay the spacer gel with overlay buffer to create an even upper gel border. Allow the gels to polymerize.

28. Prepare the stacking gel (10% acrylamide):

Reagent Amount to add (per 10 mL) 6X BN gel buffer 3.3 mL 40% Acrylamide 2.5 mL 2% Methylenebisacrylamide 1.5 mL Glycerol 1.0 mL SDS 10 mg H2O to 10 mL 10% APS 83 µL TEMED 8.3 µL

The final volume depends on the dimensions of the gel system; scale the amounts to add accordingly.

29. Decant the overlay buffer. Cast the stacking gel around the first-dimension gel strip. Tilt the gel assembly slightly sideways to avoid collection of air bubbles beneath the gel strip. Allow the gel to polymerize.

30. Fully assemble the gel system. Fill with second-dimension anode and cathode buffers.

31. Run the gel at a constant 1.5 mA/cm of gel width (for 1.0-mm thick gels) or 2.3 mA/cm (for 1.5-mm second-dimension gels).
The run time depends on the dimensions of the gel system. These are usually two to three times longer for tricine-buffered gels than for glycine-buffered gels of the same dimensions.

Visualization

32. Visualize the gels as appropriate for the method used:

For standard BN/SDS-PAGE gels:

i. Stain the gel with any standard staining techniques (e.g., Coomassie, colloidal Coomassie, silver, fluorescent dyes).
Colloidal Coomassie staining (Fig. 4 , left) is the best compromise between cost, sensitivity, the ability to quantify spots on the gel, and compatibility with mass spectrometric protein identification.

View larger version (59K): [in this window] [in a new window]

Figure 4. (Left) Colloidal Coomassie-stained standard BN/SDS-PAGE loaded with 500 µg of Arabidopsis mitochondria isolated from a cell culture. (Right) Superimposed false-color spot intensity maps of a DIGE BN/SDS-PAGE. Samples are Arabidopsis mitochondria isolated from a cell culture. One sample was treated with a respiratory inhibitor, the other was not. Red spots are more intense in the treated samples, whereas green spots are more intense in control samples. Yellow spots are equally intense in both samples.

For DIGE gels:

ii. Scan or photograph the gels using appropriate excitation wavelength and emission filters for all three dyes.

iii. Upload gel images (i.e., spot intensity maps). Perform differential in-gel analysis (DIA) for each of three gel sets.

iv. Perform biological variance analysis.
This incorporates all three DIAs to generate statistical data on the significance and the average ratio of protein abundance between samples and controls.

v. Superimpose false-color spot intensity maps to create a visual overview of the changes observed in treatment vs. control samples (Fig. 4B, right).

TROUBLESHOOTING

Problem: Sample aggregation occurs.

[Steps 16.v and 16.xvii]

Solution: The presence of divalent cations can cause the Coomassie to aggregate, which can affect the gel detrimentally. Try switching to imidazole-buffered NaCl for the solubilization buffer.

DISCUSSION

Within the cellular environment, polypeptides often associate with each other to form higher molecular mass structures. These complexes can vary from several copies of the same polypeptide (i.e., homo-oligomers) to composites of structural proteins and enzymes performing discrete but related reactions within a metabolic pathway. The latter type of complex is also termed a metabolon, within which metabolic channeling increases the speed of the overall reaction, minimizes loss of intermediates into other pathways, and reduces unwanted nonenzymatic breakdown of unstable intermediates. The capacity of membranes to take up hydrophobic proteins might also be increased when they are assembled into protein complexes.

BN-PAGE has been used primarily to analyze the stable and highly abundant respiratory chain complexes and supercomplexes in mitochondria during the last decade. However, the technique has also been used to study other stable protein complexes. Although BN-PAGE provides a broad assessment of protein-protein interactions, only stable protein complexes are visualized; transient interactions cannot be investigated using this technique. Also, because of the enormous dynamic range in the abundance of protein complexes within cells, protein complexes present in low copy numbers usually are not observed using this technique. In proteomic studies, two-dimensional BN/SDS-PAGE may additionally serve to fill the gap between IEF/SDS-PAGE and 1D SDS-PAGE as it offers more resolution in the separation of hydrophobic proteins compared to 1D SDS-PAGE and can display proteins often under-represented in IEF/SDS-PAGE (Eubel et al. 2005).

REFERENCES

Eubel, H., Braun, H.P., and Millar, A.H. 2005. Blue-native PAGE in plants: A tool in analysis of protein-protein interactions. Plant Methods 1: 11.[Medline]

Perales, M., Eubel, H., Heinemeyer, J., Colaneri, A., Zabaleta, E., and Braun, H.P. 2005. Disruption of a nuclear gene encoding a mitochondrial gamma carbonic anhydrase reduces complex I and supercomplex I+III₂ levels and alters mitochondrial physiology in Arabidopsis. J. Mol. Biol. 350: 263–277.[Medline]

Schägger, H. and von Jagow, G. 1991. Blue native electrophoresis for isolation of membrane protein complexes in enzymatically active form. Anal. Biochem. 199: 223–231.[Medline]

Wittig, I., Braun, H.P., and Schägger, H. 2006. Blue native PAGE. Nat. Protoc 1: 418–428.[Medline]