From FRET Imaging to Practical Methodology for Kinase Activity Sensing in Living Cells - Progress in Molecular Biology and Translational Science - page 186

The identification and/or optimization of the different components of a
KAR are reviewed in this section. A range of KARs have been designed on
this model, which allow activity monitoring of many tyrosine and serine/
threonine kinases.
6.1. Substrate peptide identification
Many methodological approaches have proven to be useful to identify and
characterize that a peptide sequence acts as a specific substrate for a kinase of
interest.
The first approach identifies a specific substrate in silico by using databases
of known substrate sequences of kinases such as UniProtKB/Swiss-Prot,
which provide reliable protein sequences associated with a very high level
of annotation and a high level of integration with additional databases. How-
ever, peptides selected by some knowledge-based libraries (such as UniProt,
for instance) can have multiple phospho-acceptor sites. Other databases such
as KinasePhos can predict phosphorylation sites within given proteins and
provide information on the exact positioning of phosphorylation sites with
a link to the corresponding catalytic protein kinases involved. 4
Another approach relies on the use of the kinase target sequence from a
protein known to be phosphorylated by the kinase of interest. Since many
kinases have multiple isoforms, substrate sequence could be modified to be
more specific toward a protein kinase isoform. The protein kinase C family,
with its 10 members, is a good example. 100,101 The first genetically encoded
PKC (protein kinase C)-FRET-based reporter (CKAR, C-kinase activity
reporter) is an effective reporter for all PKC isoforms, 102,103 but each
isoform has its own activity signature. 104 Kajimoto et al. have designed a
new genetically encoded reporter based on the first CKAR but with an
ultraspecific substrate to measure only PKC d activity in different cell
compartments. In order to make CKAR more selective for PKC d , they
selected 11 known substrates of PKC d threonine as phospho-acceptor
residues, and kept isoleucine at the position þ 3 to facilitate binding of
the PAABD (see Section 6.2 ). Sensors with candidate substrate sequences
were characterized in vitro and in cellulo (see Section 6.6 ) for specificity
and selectivity. In spite of being time consuming, this kind of approach
can provide highly specific and selective KARs.
The last approach enables the identification of protein kinase substrates
through large-scale analysis using high-density peptide microarrays from
various biological samples. The use of high-throughput technologies has