Proteome-Wide Quantitation by SILAC - LC-MS/MS in Proteomics: Methods and Applications - page 196

9. To account for systematic inaccuracies in the mixing ratio
the data often require normalization. The simplest way is
to calculate the median of the transformed data and divide
all transformed ratios with this value.
10. The final step prior to assessing the biological implications
of the data is to define a threshold for regarding a protein
as being significantly regulated. Under most circumstances
a change in protein expression >1.5-fold can be regarded
as significant since the average standard deviation of such
experiments is typically around 15%. For quantitation of
post translational modifications a more stringent criteria of
>twofold regulation is often required to account for the
larger variability of single peptide quantitation. Alterna-
tively sophisticated statistical test can be applied to evaluate
the probability of a regulation being significant. Different
statistical approaches have been suggested, but they all have
in common that they test for proteins being significant out-
liers from the large population of unregulated proteins ( 22 )
( see Fig. 11.3D ) .
4. Notes
1. In principle any amino acid can be used for labeling
although care should be taken when selecting which amino
acid to use. By far the most common SILAC amino acids is
a combination of lysine and arginine, since this ensures the
presence of at least one label in all tryptic peptides, except
for the C-terminal peptide. Alternatively, lysine labeling
alone in combination with endopeptidase Lys-C digestion
or arginine labeling in combination with trypsin digestion
can be employed. In the last case the number of peptides
amenable to quantitation is dramatically reduced. This is
however still a useful approach since most proteins will be
identified by a sufficiently large number of peptides to allow
accurate quantitation. In experiments targeting phospho-
rylations or other posttranslational modifications it is cru-
cial to label both arginine and lysine if tryptic digestion is
employed to ensure that all peptides can be quantified.
2. Methodologies for applying SILAC to tissue samples have
been reported, either by using SILAC labeled cells as a
common reference which is mixed in with each of the tis-
sue samples ( 16 ) or by using SILAC labeling of an entire
organism, exemplified by the 'SILAC mouse' ( 17 ) .
3. In some cell lines, high concentrations of arginine in the
growth media may result in its metabolic conversion into