Fluorescent Protein-Based Biosensors and Their Clinical Applications - Progress in Molecular Biology and Translational Science - page 338

CrkL phosphorylation by BCR-ABL triggers a structural change in CrkL,
leading to an increase in FRET efficiency due to the close proximity of CFP
and YFP; conversely, inhibition of BCR-ABL kinase activity by inhibitors
results in a decrease in FRET ( Fig. 8.5B ). The final version of the biosensor,
named “Pickles” for phosphorylation indicator of CrkL en substrate,
required fine-tuning beyond merely adding CFP and YFP to either end
of CrkL. As in the case of development of other FRET biosensors,
certain modifications, including truncation of CrkL, circular permutations
of CFP, and monomeric mutation in of YFP, were required for its
development and clinical application, and the detailed procedures are
described in our previously published research paper. 5
The final product of Pickles (version 2.31; Pickles_2.31) displayed an in-
crease in FRET efficiency of approximately 80-100% upon phosphoryla-
tion by BCR-ABL; this increase was specific for BCR-ABL (and its
cellular counterpart c-Abl) over other nonreceptor-type tyrosine kinases
tested. Because our biosensor was designed based on bioimaging techniques,
a faithful time-dependent decrease in FRET efficiency was observed in the
drug-treated cells compared to the control cells using time-lapse microscopy
in tumor cells ( Fig. 8.5C ). In addition, when we tested its dose-dependent
response to IM, Pickles could detect the drug effect at lower concentrations
and in a wider measureable range compared to other existing techniques, for
example, Western blotting ( Fig. 8.5D ). Moreover, we noticed that the time
required for evaluation is 12 h; BCR-ABL activity reached a nadir 3-6 h
after drug treatment ( Fig. 8.5C ).
7.2. Special features of Pickles
As described above, GFP-based FRET technology was originally used to
observe the spatiotemporal activation of proteins in living cells. Therefore,
the most prominent feature of our Pickles system is that we can perform
evaluation of drug responsiveness at the single-cell level using FRET tech-
nology. In fact, a drug-resistant cell population of 1% or less can be distin-
guished from other drug-sensitive cells. In our system, in which a researcher
manually observes a few hundreds of cells with a microscope, 1% is the
highest sensitivity. To improve sensitivity, we attempted similar analyses
using a combination of immunofluorescence with an antibody specific for
phosphorylated CrkL and a flow cytometer, which can simultaneously an-
alyze significantly higher cell numbers; however, unexpectedly, the detect-
ability of resistant cells was the same as that in our system. This highlights the