In this study, we describe the analysis of MPCs by BN-PAGE. A 2D approach is used to first separate MPCs under native conditions, and then to further subdivide them into their individual components by a second dimension SDS-PAGE.
Samples are prepared from cell lysates. For the solubilization of many MPCs, an appropriate detergent is needed, which preserves the structure of the protein complexes. Here, we use 0.1% Triton X-100. However, the optimal detergent and its suitable concentration have to be determined empirically for every MPC. In case of Triton X-100, for example, it has been reported that low detergent concentrations allow the identification of a dimeric form of the F1F0-ATPase complex (Arnold, Pfeiffer et al. 1998). Higher Triton X-100 concentrations, however, lead to the dissociation of the dimer and to a corresponding increase of the monomeric F1F0-ATPase complex. This is in line with one of our former studies, were we show that the multivalent T-cell receptor complex (TCR) is preserved when extracted with low concentrations of Brij 96, whereas the usage of higher concentration or of another detergent called digitonin results in the extraction of monomeric TCR (Schamel, Arechaga et al. 2005). Commonly used detergents that can be tested include digitonin (0.5 to 1%), Triton X-100 (0.1 to 0.5%), Brij 96 (0.1 to 0.5%), or dodecylmaltoside (0.1 to 0.5%). These reagents are nonionic detergents, which tend to be best for MPC stability. Be aware that contact with SDS and other strong detergents should be avoided (Camacho-Carvajal, Wollscheid et al. 2004).
Dialysis of the lysates is required to achieve MPC separation in a BN-gel (Camacho-Carvajal, Wollscheid et al. 2004); (Heiss, Junkes et al. 2005). It seems that the adjustment of salt concentration or the removal of low molecular weight impurities is crucial for high resolution. It is noteworthy that also membrane preparations and MPCs, which have been immunopurified and later on eluted from the antibody, are suitable for BN-PAGE (Swamy, Siegers et al. 2006). In both cases, the samples do not have to be dialyzed for BN-PAGE separation, if membrane lysis or elution is carried out in BN-lysis buffer.
For protein separation by BN-PAGE, the dye Coomassie blue is needed, which binds unspecifically to proteins and covers them with negative charges. Thereby, Coomassie blue enables the electrophoretic mobility of proteins towards the cathode at neutral pH (Schägger and von Jagow 1991); (Schägger, Cramer et al. 1994). Furthermore, Coomassie blue prevents protein aggregation in the stacking gel during electrophoresis. For BN-PAGE, Coomassie G250 has to be used instead of Coomassie blue R250 or colloidal Coomassie blues.
Before running a BN-gel, it is necessary to ensure that the percentage of the gel fits to the expected size of the MPC of interest. Precast BN-gels with different gradients and suitable buffers are commercially available from Invitrogen (NativePAGE Novex Bis-Tris Gel System). But BN-gels can also be prepared using a gradient mixer together with a persistaltic pump. To guarantee an intact gradient, the liquid should flow constantly during pouring and bubbles should be avoided. We recommend the loading of different sample dilutions onto the gel because overloading can lead to protein precipitation during the electrophoresis process. In addition, BN-gels should be run at 4°C to prevent protein degradation and to keep the MPCs intact.
After BN-PAGE, visualization of MPCs can be achieved by Coomassie brilliant blue staining, silver staining or immunoblotting. Protein bands visualized by Coomassie or silver staining are suitable for further analysis by mass spectrometry (Camacho-Carvajal, Wollscheid et al. 2004). In case of immunoblotting, the optimal transfer conditions for the MPCs of interest have to be determined empirically. Be aware that Coomassie blue is also transferred during blotting of a BN-gel. Therefore, the gel will be colourless after the successful transfer, whereas the membrane will exert a blue colour. Further, it is important to mention that not every primary antibody, which works for detection after SDS-PAGE, is applicable to immunoblotting upon BN-PAGE. It can happen that antibodies do not recognize the MPC of interest because their epitope is hidden in the native conformation of the proteins. To overcome this problem, it is possible to denature the proteins within the BN-gel prior to the transfer by boiling the gel shortly in 1x SDS sample buffer.
In our example, we did not subject the BN-gel directly to detection of protein bands. Instead, we further divided the BN-PAGE-separated lysate by a second dimension SDS-PAGE. In the second dimension SDS-gel, monomeric proteins migrate within a hyperbolic diagonal due to the gradient gel in the first and the linear gel in the second dimension (Camacho-Carvajal, Wollscheid et al. 2004). This allows the easy identification of MPCs, since they are localized below this hyperbolic diagonal. Subcomponents of one distinct MPC are separated in a vertical line in the second dimension SDS-PAGE. Components that are constituents of several dinstinct MPCs can be identified on a horizontal line according to the size of the MPC. However, it has to be considered that several protein spots appearing in one vertical line could also be part of separate complexes that migrate at the same position in BN-PAGE. The final proof that they are present in the same MPC can be obtained by an antibody-based gel shift assay. In this assay, cellular lysate is incubated with an antibody against a protein represented by one of the identified spots prior to BN-PAGE. This results in a shift of all MPCs that contain this protein towards a higher molecular mass in the first dimension. Other proteins that are also a part of these MPCs will undergo this complex-specific shift and are therefore easy to identify in the second dimension SDS-gel.
Not only the composition of MPCs can be analyzed by BN-PAGE but also the determination of their stoichiometry is possible (Schamel and Reth 2000); (Schamel 2001), (Swamy, Minguet et al. 2007). For this purpose, a NAMOS assay (native antibody-based mobility-shift assay) can be performed. As in the antibody-based gel shift assay, the cellular lysates are incubated with monoclonal subunit-specific antibodies. This leads to the induction of electrophoretic immunoshifts in the BN-gels, which allow the inference from the extent of the shift on the stoichiometry of MPCs
In conlusion, BN-PAGE is suitable for the identification of MPCs and the determination of their size, composition, as well as relative abundance. Performed as a NAMOS assay, it also offers the possibility to determine the stoichiometry of a certain MPC. Given its general applicability, this technique is a very useful tool for the characterization of MPCs (Dekker, Müller et al. 1996); (Wittig and Schagger 2008); (Wagner, Rehling et al. 2009); (Wittig and Schägger 2009).