Stable Isotope Labeling by Amino Acids in Cell Culture (SILAC)

Standard proteomic approaches tend to produce large lists of proteins, some of which are specific to the purified complex or organelle of interest but many of which are contaminants that bind non-specifically. In order to quickly identify "real" hits and rule out contaminants,our laboratory maps protein interactomes using a quantitative mass spectrometric method based on differential metabolic labeling of proteins in vivo. This technique, called SILAC (Stable Isotope Labeling by Amino acids in Culture) was originally used primarily with cultured cells but has recently been extended to complex organisms, including C. elegans, drosophila and mice. It can be applied to the analysis of both endogenous and tagged proteins, although certain caveats apply under both conditions (e.g. antibody cross-reactivity, overexpression artefacts).

The main advantage of SILAC for labeling-based quantitative proteomics is that the control and experimental samples are differentially labeled upstream of processing steps that can introduce variability between them, such as SDS-PAGE separation, digestion and MS analysis.



Isotopic amino acids introduce mass shifts into peptides that can be detected by MS analysis:

In the example shown here, two different isotopes of arginine can be used in combination with normal (light) arginine: one in which 6 carbons are substituted with heavy carbon (R6) and one in which 6 carbons and 4 nitrogens are substituted with heavy carbon/nitrogen (R10). They result in mass shifts of 6 Da and 10 Da, respectively, in peptides containing a single arginine. For cultured cells, isotopic arginine is usually combined with isotopic lysine, to ensure that all peptides generated by trypsin digestion are quantifiable, as in theory they all end in an arginine or a lysine.

SILAC-based quantitative IP experiments can be carried out on a tagged protein (e.g. GFP-tagged, as shown above) oran endogenous protein. The experimental design is similar, with the proteins combined after IP and processed together at every step afterwards to eliminate variability that can be introduced downstream by sample handling and separate MS analysis.

For our purposes, expressing the 3 mammalian PP1 isoforms as GFP-tagged proteinsoffers us the opportunity to compare their dynamic localizations within live cells and the ability to recover the fusion proteins from cell lysates or purified structures and analyze the proteins with which they associate. This type of approach identifies both targeting subunits (proteins that interact directly with PP1) and other proteins that are in larger complexes with PP1 but do not interact with the phosphatase directly.

Three populations of a cell line (or three different cell lines, as shown here) can be metabolically labeled with either normal arginine or lysine (12C6-Arg/12C6-Lys, also referred to as R0K0), carbon-substituted arginine and hydrogen-substituted lysine (13C6-Arg and D4-Lys, also referred to as R6K4) and or carbon- plus nitrogen-substituted arginine and lysine (13C615N4-Arg and 13C615N4-Lys, also referred to as R10K8) respectively, for at least five cell doublings. This ‘triple encoding’ procedure allows three cell states to be measured in one experiment. In the case shown here, we can compare complexes that co-purify with PP1 vs. the control (GFP alone), but also complexes that co-purify specifically with one of these two PP1 isoforms.

By fractionating cells prior to analysis, we improve our coverage and can also compare both isoform-specific and compartment-specific PP1 complexes. Having previously demonstrated this for nuclear vs. cytoplasmic extracts, we recently extended our interactome analyses to nucleolar extracts. This necessitated optimization of the extraction of nucleolar protein complexes, as all previously published extraction methods were relatively inefficient (~50% total protein extracted from purified nucleoli). The basic steps in our interactome analyses of PP1 isoforms, as summarized above, include fractionation of cultured mammalian cells stably expressing either GFP alone or a GFP-PP1 fusion protein (or preparation of whole cell extracts), immunoprecipitation of these GFP proteins usingthe high affinity GFP_Trap_A reagent,elution and separation of the proteins by 1-dimensional polyacrylamide gel electrophoresis and LC/MS separation, identification of the proteins by peptide mass fingerprinting and mass spectrometry and bioinformatic analyses (cross-reference of protein informatics with genomic databases). Identified proteins are then further characterized by expression in mammalian and bacterial cells, generation of antibodies, etc.


Endogenous or tagged? This is a commonly asked question in interactome analyses, and there is no easy answer. A pulldown of an endogenous protein, while more physiological, has the caveat that the antibody may not be specific to only that protein. Cross-reactivity will lead to pulldown of the cross-reacting protein(s) and its interactors, which complicates analysis of the dataset. A pulldown of a tagged, exogenously expressed protein is an attractive option, however it is essential to first demonstrate that the tagged protein is full-length and functional, and accurately represents the pool of endogenous protein (see checklist below).In the example shown in the bottom right panel, an endogenous protein was pulled down using an antibody, however the depletion was not very efficient and only a subset of known interactors was identified. In cells expressing the same level of GFP-tagged as endogenous protein, a reasonably efficient depletion of the GFP-tagged protein gave similar results. However, when a more efficient affinity reagent was used to deplete all of the GFP-tagged protein, its full set of known interactors was identified. This highlights the important point that in any MS-based IP experiment, the identification of interactors depends to a very large part on the actual amount of protein pulled down.

SILAC Protocols:

For details on sourcing reagents and preparing the media, download our SILAC Reagent Protocol.

For details on covalently coupling antibodies to Protein G sepharose, download our Covalent Coupling Protocol.

For details on cell fractionation and preparation of whole cell vs. nuclear and cytoplasmic lysates, download our Cell Fractionation Protocol.

For details on trypsin digestion of gel slices for MS analysis, download our In Gel Digestion Protocol.

Related publications:

Trinkle-Mulcahy L .Resolving protein interactions and complexes by affinity purification followed by label-based quantitative mass spectrometry.Proteomics.12:1623-38, 2012.

Chamousset, D., De Wever, V., Moorhead, G., Chen, Y., Boisvert, F.M., Lamond, A.I. and Trinkle-Mulcahy, L. RRP1B targets PP1 to mammalian cell nucleoli and is associated with pre-60S ribosomal subunits. Mol. Biol. Cell 21:4212-26, 2010.

Schmitz MH, Held M, Janssens V, Hutchins JR, Hudecz O, Ivanova E, Goris J, Trinkle-Mulcahy L, Lamond AI, Poser I, Hyman AA, Mechtler K, Peters JM, Gerlich DW .Live-cell imaging RNAi screen identifies PP2A-B55alpha and importin-beta1 as key mitotic exit regulators in human cells. Nat Cell Biol 12:886-93, 2010.

Chamousset D, Mamane S, Boisvert FM, Trinkle-Mulcahy L .Efficient extraction of nucleolar proteins for interactome analyses .Proteomics 10:3045-50, 2010.

Trinkle-Mulcahy L., Boulon S., Lam Y.W., Urcia R., Boisvert F.M., Vandermoere F., Morrice N.A., Swift S., Rothbauer U., Leonhardt H. and Lamond A.I. Identifying specific protein interaction partners using quantitative mass spectrometry and bead proteomes. J Cell Biol. 183:223-39, 2008.

Trinkle-Mulcahy, L., Andersen, J., Lam, Y.W., Moorhead, G., Mann, M. and Lamond, A.I. Repo-Man recruits PP1γ to chromatin and is essential for cell viability. J. Cell Biol. 172:679-92, 2006.