90
dyes rather than FPs as the signaling component of the sensor. As
there is much greater control over the precise location of synthetic
fluorophores, synthetic dyes can theoretically allow for even small
conformational changes to produce a measurable FRET signal.
While a large conformational change for a given protein is always
desirable for sensor construction, it is often a necessity when using
FPs, meaning that synthetic dyes are potentially applicable across a
much larger range of proteins, rather than being restricted to those
with large distance-based conformational changes.
Sensors that use synthetic dyes are either intensity-based sensors (non-FRET), which require multiple synthetic components,
or must be developed through extensive high-throughput screening [3, 4]. For example, in addition to the use of two synthetic
dyes, the Snifit-type sensor design requires the development and
synthesis of a tethered competitive ligand that can occupy the
binding active site, which necessarily introduces an extra design
phase of engineering and screening [3]. On the other hand, the
EOS-type sensor developed by Namiki et al. only requires a single
synthetic component (a dye) [4], but still requires exhaustive
screening of different residues to find a location that gives a strong
signal. In addition, EOS sensors do not produce ratiometric output and are therefore not quantitative.
It is possible to improve on one or more of these design components when creating a synthetic dye-based sensor. Specifically, it
has been shown that it is possible to create FRET sensors that are
a combination of one FP and one synthetic dye [5], which is an
improvement over both single fluorophore and two dye sensors as
it is ratiometric and requires one fewer site-specific modification.
Additionally, rather than using brute force high-throughput
screening of all residue locations to identify a dye labeling site, we
have expedited sensor development through the use of computational screening, which can reduce the number of possible dye
labeling sites to a subset with a higher likelihood of yielding a functional sensor. We have used these computational techniques in tandem with synthetic dyes (via thiol-maleimide labeling of residues)
to develop optical sensors with large dynamic ranges.
2 Materials
Whenever possible, prepare all stock solutions and buffers in ultrapure water (MilliQ). For reagents that have poor solubility in water,
dissolve them with the smallest possible proportion of organic solvent (i.e., 5% DMSO would be preferable to 10% DMSO) as some
proteins may have poor stability in organic solvent. All solutions
and reagents should be prepared as fresh as possible as some reagents
will have short lifetimes when in solution, even at −20 °C. All
buffer solutions should be filtered through a membrane filter
(pore size 0.45 μm or smaller) after preparation.
Joshua A. Mitchell et al.
dyes rather than FPs as the signaling component of the sensor. As
there is much greater control over the precise location of synthetic
fluorophores, synthetic dyes can theoretically allow for even small
conformational changes to produce a measurable FRET signal.
While a large conformational change for a given protein is always
desirable for sensor construction, it is often a necessity when using
FPs, meaning that synthetic dyes are potentially applicable across a
much larger range of proteins, rather than being restricted to those
with large distance-based conformational changes.
Sensors that use synthetic dyes are either intensity-based sensors (non-FRET), which require multiple synthetic components,
or must be developed through extensive high-throughput screening [3, 4]. For example, in addition to the use of two synthetic
dyes, the Snifit-type sensor design requires the development and
synthesis of a tethered competitive ligand that can occupy the
binding active site, which necessarily introduces an extra design
phase of engineering and screening [3]. On the other hand, the
EOS-type sensor developed by Namiki et al. only requires a single
synthetic component (a dye) [4], but still requires exhaustive
screening of different residues to find a location that gives a strong
signal. In addition, EOS sensors do not produce ratiometric output and are therefore not quantitative.
It is possible to improve on one or more of these design components when creating a synthetic dye-based sensor. Specifically, it
has been shown that it is possible to create FRET sensors that are
a combination of one FP and one synthetic dye [5], which is an
improvement over both single fluorophore and two dye sensors as
it is ratiometric and requires one fewer site-specific modification.
Additionally, rather than using brute force high-throughput
screening of all residue locations to identify a dye labeling site, we
have expedited sensor development through the use of computational screening, which can reduce the number of possible dye
labeling sites to a subset with a higher likelihood of yielding a functional sensor. We have used these computational techniques in tandem with synthetic dyes (via thiol-maleimide labeling of residues)
to develop optical sensors with large dynamic ranges.
2 Materials
Whenever possible, prepare all stock solutions and buffers in ultrapure water (MilliQ). For reagents that have poor solubility in water,
dissolve them with the smallest possible proportion of organic solvent (i.e., 5% DMSO would be preferable to 10% DMSO) as some
proteins may have poor stability in organic solvent. All solutions
and reagents should be prepared as fresh as possible as some reagents
will have short lifetimes when in solution, even at −20 °C. All
buffer solutions should be filtered through a membrane filter
(pore size 0.45 μm or smaller) after preparation.
Joshua A. Mitchell et al.