PROTOCOLQuantitative proteomics using stable isotope labeling with ami no acids in cell cultureH CHarsha1-fHenrik Molina3,4 & Akhilesh Pandey3-61Institute of Bioinformatics, International Technology Park, Bangalore 560066, Karnataka, India. 2Manipal University, Manipal 576104,Karnataka, India. 3McKusick- Nathans Institute of Genetic Medicine, JohnsHopkins University, Baltimore, Maryland 21205, USA. 4Department of Biological Chemistry, Johns Hopkins University, Baltimore, Maryla nd 21205, USA. 5Departme nt of Pathology, Joh nsHopk ins Un iversity, Baltimore, Maryla nd 21205, USA. 6Departme nt of On cology, Joh nsHopki ns University, Baltimore, Maryland 21205, USA. Correspondenceshould be addressedto A.P.(pandeyjhmi.edu).Publishedonline 6 March2008; doi:10.1038/nprot.2008.2Stable isotope labeling with amino acids in cell culture (SILAC) is a simple in vivo labeling strategy for mass spectrometry-based quantitative proteomics. It relies on the metabolic incorporation of nonradioactive heavy isotopic forms of amino acids into cellular proteins, which can be readily distinguished in a mass spectrometer. As the samples are mixed before processing in the SILAC methodology, the sample handling errors are also minimized. Here we present protocols for using SILACin the following types of experiments: (i) studying inducible protein complexes, (ii) identi?cation of Tyr kinase substrates, (iii) differential membrane proteomics and (iv) studying temporal dynamics using SILAC5-plexing. Although the overall time is largely dependent on the rate of cell growth and various sample processing steps employed, atypical SILACexperiment from start to ?nish, including data analysis, should take anywhere between 20 and 25 d.NATUREPROTOCOLS VOL.3NO.3 | 2008 | #NATUREPROTOCOLS VOL.3NO.3 | 2008 | #PUPG gsbup pnraN 0802 ?INTRODUCTIONA number of massspectrometry(MS)-basedquantitativeproteomic methods are in usetoday with each having its unique advantages and disadvantagesSomeof the commonly used methods include SILAC1, ICAT2, iTRAQ3 and 18O labeling4. In contrast to others, SILAC is an in vivo labeling strategy in which the proteome is labeled as the cells grow in culture. Further, in the SILAC method, the samplesare mixed at the very beg inning before a ny puri?cati on or process in gstepswhich con siderablyreducessample handling errors.Although SILAC offers greatadvantages,there are a few limitations. SILAC can be usedonly in cell culture where the cells are metabolically activeand thus cannot be usedfor tissuesamples.The other major issue is the limited availability of the repertoire of heavy forms of ami no acids, which restrictsthe n umber of cellular statesthat can be directly compared. NeverthelessSILAC isasgood or better than most existing protein/peptide labeling methodswith the ability to compare up to ?ve states in a single experiment (see REAGENT SETUPa nd Table 1 for choosi ng ami no acids for multiplex in g). While carry ing out 3-, 4- or 5-state experime nts, deuterated am ino acidsareused, which might affect separati onby reversedphasechromatography5.Here, we provide protocols for designing and implementing severaltypesof SILAC-basedexperiments.A continuously updated website(http:/www.silac.org) is a resourcefor additi on al literature, reagents,recipesand applications pertaining to this method.Experimental designAll SILAC-basedexperiments essentiallycomprise the following steps(Fig. 1):1. Adapt ing cellsto media containing heavy am ino acids,2. Cell man ipulati on and biochemical process in gor en rich- ment and3. MS and data analysis.Adapt ing the cellsiso neof the crucial steps in an y SILAC-based experiment. The successof a SILAC experiment largely relies on ensuring heavy amino acid incorporation in the cell population.Almost all previous studies that in volve SILAC haveused dialyzed serum to grow the cells to avoid free amino acids that are potentially present in the serum. Although this has worked well for severalcell li nesthat have bee nusedi n the previous studies, we havealsolearnedthat therearecellsthat might not grow well in this serum. To overcomethis problem, wehavetestedtheextent of light ami no acid i ncorporati on whe n thecellslabeledwith heavyam ino acidsaregrown in complete serum. In approximately ?ve passages, we observe nearcomplete i ncorporati on of heavyam ino acidswith anegligible background from light amino acidsin the presenceof completeserum we haveused (I nvitroge n, cat.26140-079)(Fig. 2). If the serum is from some other manufacturer, we recommend carrying out a similar experiment before using it to prepare the medium for SILAC experiments.TABLE1 | List of recommendedamino acid combinations while using trypsin.Number of statesbeing comparedHeavy amino acidsMass difference (Da)213C6-Lys613C6-Arg63D4-Lys413C5-Arg613C6,15N2-Lys813C6, 15N4-Arg104D4-Lys413C5-Arg613C6,15N2-Lys81