Mass Spectrometry in Systems Biology - Systems Biology: Networks, Models, and Applications - page 17

2. Enrichment of amino acids with 13 C [46].
3. Incorporation of only a select number of isotopically enriched
amino acids in cell culture, SILAC [47].
Sometimes, manipulation of biological samples is not possible. In such
situations, chemical labeling is an alternative. There are a number of
possible chemistries for stable isotope labeling of peptides/proteins
(for review see ref. [48]). Developed by Aebersold and colleagues, ICAT
(isotope-coded affinity tag reagents) is one of the most successful
strategies for chemical labeling of peptides [49]. ICAT reagent contains
three units: an affinity tag (usually biotin), a linker (that is isotope-
coded), and a thiol-specific reactive group. A variety of ICAT reagents
have been developed over the years (for review see ref. [50]). ICAT
strategies have a particular advantage, which is that it simplifies the
protein digestion mixture by affinity enrichment of a subpopulation
of peptides. However, because most of the ICAT reagent targets only
cysteine residues, its applicability is limited to cysteine-containing
peptides. As such, identification of posttranslational modified regions
of proteins is reduced to only those present in cysteine-containing
peptides [51].
One of the most stringent requirements for quantitative analyses is
that the standard compound is to be added at the exact starting point
of the experiment. Therefore, requirements for quantitative analysis of
a whole class of biological samples, in particular human tissue, empha-
size the need for quantitative proteomics without stable isotope
labeling. Proof of principle experiments have been performed for pro-
filing of protein complexes [52] and quantification relative to a native
reference peptide (from the protein of interest) of site-specific phospho-
rylation [53]. Western blotted quantities of nitrated proteins paralleled
quantities derived from site-specific nitration relative to native refer-
ence peptide [54]. Absolute amounts (mole of modifications/mole of
protein) can be obtained by constructing a response curve with exter-
nal standards. Alternatively, by spiking the protein solution with a
native peptide (isotopically labeled) that is concomitantly processed by
digestion, one could obtain absolute quantification of proteins levels
while accounting for digestion efficiency [55]. Absolute quantitation
can also be obtained by addition of an isotope label internal standard
of the peptide of interest (already processed by proteolysis), followed
by a single reaction monitoring experiment (one of the most sensitive
MS routines for quantification) that detects both the analyte and its
labeled standard (a method known as AQUA [56]). Practical consider-
ations for protein profiling when performing large-scale analysis of
peptides by MudPIT led Liu et al. [57] to statistically evaluate the sam-
pling of peptide ions by data-dependent acquisition (as defined above).
Their model showed that spectral sampling is an accurate estimate of
relative protein abundance over two orders of magnitude.