Fluorescent Sensors of Protein Kinases: From Basics to Biomedical Applications - Progress in Molecular Biology and Translational Science - page 226

two AFPs due to an intramolecular conformational change. A wide variety of
these genetically encoded biosensors have been developed (see Table 6.2 and
Figs. 6.3-6.6 ), but the basic structure of these kinase activity reporters (KARs)
is essentially the same: a kinase-specific substrate sequence bearing a consensus
phosphorylation site connected to a matching phosphoamino acid-binding
domain (PAABD) by a flexible linker and flanked by a pair of AFPs that
can transfer fluorescence resonance energy between each other. Following
phosphorylation of the substrate sequence by the kinase, the PAABD binds
the phosphorylated sequence, thereby promoting an intramolecular
conformational change which alters the distance and orientation of the AFPs.
Several factors have to be considered when aiming to design an optimally
responsive and selective genetically encoded FRET biosensor. Moreover,
the specificity and selectivity of a kinase/biosensor couple should always
be characterized in vitro and in cellulo .
First, needless to say, the size and sequence of the substrate have to be
tailored to the most suitable sequence to gain the highest level of selectivity.
The choice of the substrate sequence is essential for selectivity and may
either be identified through an in silico approach through database mining
from knowledge-based libraries, or simply through rational design of a se-
quence derived from a known substrate protein of a given kinase. The spec-
ificity of a kinase for a protein substrate is generally dependent on other
regions of the protein that “dock” onto the kinase to offer higher affin-
ity/recognition. Therefore, several designs include a docking domain dis-
tinct from the substrate sequence, so as to increase specificity for the
target kinase while also increasing the efficiency of phosphorylation of
the substrate by the kinase. This is well exemplified by MAPK protein ki-
nases (ERK, p38, and JNK), which have very similar phosphorylation sites,
yet distinct docking sites that can be made use of to improve substrate
targeting and selectivity. 81,82
Second, the choice of the PAABD is critical for an efficient and sensitive
biosensor. The sequence of the PAABD should present high affinity and ef-
ficient recognition of the phosphorylated substrate, as opposed to poor af-
finity for the unphosphorylated substrate, and should display a fully
reversible behavior between both forms. Several modular phosphobinding
domains (PBDs) have been well characterized, including Src-homology 2/3
(SH2 SH3) and phosphotyrosine-binding (PTB) domains, 14.3.3 proteins,
forkhead-associated (FHA) domains, WW domains, WD40 domains, and
the PBD of Plk1 (for reviews, see Refs. 83-89 ).