Biology Reference
In-Depth Information
once in a given proteome and are termed
proteotypic peptides
.
For closely related proteins or splice isoforms, fi nding proteo-
typic peptides with favorable LC and MS properties might be
challenging. Hence, sometimes statements can be only made
for a protein or closely related protein group, but not for splice
isoforms of the protein or individual proteins, respectively.
Using trypsin, or any other sequence specifi c endoprote-
ase, the surrounding protein sequence should be checked for
rugged ends, which are sequences of repeating recognition
amino acids, e.g., KK, KR, RK, or RR in the case of trypsin. If
possible, peptides with rugged ends should be avoided as both
the rugged end and canonically cleaved peptide may coexist
and are hard to predict a priori. Rugged end peptides are not
part of typical peptide libraries (mentioned in Subheading
2.2.3
).
However, when unavoidable, rugged end proteotypic peptides
can be used for SRM-MS and transitions are obtained as
described above in Subheadings 2.1.1/2.1.2/2.1.3/2.1.4. It
is advisable to monitor the canonical and rugged end peptide.
2. Interferences
SRM transitions are defi ned by Q1/Q3 pairs, which represent
the
m/z
of the precursor ion and the targeted fragment ion,
respectively. Unlike shotgun MS/MS where whole spectra are
recorded, SRM-MS heavily relies on the specifi city of these two
fi lters as the electron multiplier detector only records the amount
of ions present following the “fi ltration.” The window, or pore
size, of these fi lters is typically
m/z
0.7; tighter windows might
increase specifi city, but decrease signal intensity. Peptides to be
monitored could theoretically share transitions with other pep-
tides. Practically, in whole cell lysates of complex proteomes,
shared transitions are observable, especially for short peptides
and fragment ions with low sequence information [
24
]. The
most important denominator is the retention time, as these
shared transitions only play a role if they coelute during the peak
of the targeted peptide. There is a computational tool predicting
transition interferences called SRM Collider [
25
], which bases its
calculations on all theoretically possible fragment ions of a given
proteome (without any PTM) as well as all peptides ever detected
in PeptideAtlas [
26
] and predicts retention time using SSR-Calc
[
27
]. The output of SRM Collider should be taken into consid-
eration, e.g., if a certain peptide has a long list of shared transi-
tions, its best to choose another peptide. But, if the list of shared
transitions is limited or nonexistent, there is no guarantee that
the targeted peptide will be free of interferences. After all due
diligence, only the actual LC-SRM-MS experiment will deter-
mine if the targeted peptide can be measured and quantifi ed.
There are two ways to detect interferences in complex
samples. Generally, a low complexity sample is compared to a
