Evaluation of Data-Dependent and -Independent Mass Spectrometric Workflows for Sensitive Quantification of Proteins and Phosphorylation Sites

In recent years, directed and, particularly,
targeted mass spectrometric
workflows have gained momentum as alternative techniques to conventional
data-dependent acquisition (DDA) LC–MS/MS approaches. By focusing
on specific peptide species, these methods allow hypothesis-driven
analysis of selected proteins of interest, and they have been shown
to be suited to monitor low-abundance proteins within complex mixtures.
Despite their growing popularity, no study has systematically evaluated
these various MS strategies in terms of quantification, detection,
and identification limits when they are applied to complex samples.
Here, we systematically compared the performance of conventional DDA,
directed, and various targeted MS approaches on two different instruments,
namely, a hybrid linear ion trap–Orbitrap and a triple quadrupole
instrument. We assessed the limits of identification, quantification,
and detection for each method by analyzing a dilution series of 20
unmodified and 10 phosphorylated synthetic heavy-labeled reference
peptides, respectively, covering 6 orders of magnitude in peptide
concentration with and without a complex human cell digest background.
We found that all methods performed similarly in the absence of background
proteins; however, when analyzing whole-cell lysates, targeted methods
were at least 5–10 times more sensitive than that of the directed
or DDA method. In particular, higher stage fragmentation (MS3) of
the neutral loss peak using a linear ion trap increased the dynamic
quantification range of some phosphopeptides up to 100-fold. We illustrate
the power of this targeted MS3 approach for phosphopeptide monitoring
by successfully quantifying nine phosphorylation sites of the kinetochore
and spindle assembly checkpoint component Mad1 over different cell
cycle states from nonenriched pull-down samples.