Chemistry Reference
In-Depth Information
In each of the examples discussed below, the labeling positions were chosen so
that FRET ef
ciency is high in the folded state, with D and A held very close to each
other, whereas, in the unfolded state, they are further away. For each protein
detected, the FRET ef
ciency E is calculated as a ratio between the number of
photons detected from the acceptor versus the number of photons detected from
both donor and acceptor channels. When E is high (close to 1), the distance R
between D and A is small, indicating a folded conformation of the protein. When
the protein is unfolded, the protein is in a more expanded conformation with
increased R, and decreased E.
In the
first study of protein folding at the single-molecule level, Hochstrasser and
collaborators took advantage of the unique properties of GCN4, a transcription factor
fromyeast [79]. The protein is a cooperatively folded coiled-coil dimer with a disul
de
bond between the C termini of identical subunits. This arrangements allows for a
simple labeling scheme: a donor
fluorophore (rhodamine 6G) was attached to the N
terminus of one sample, and an acceptor (Texas Red) was attached to the N terminus
of the other. These weremixed, and allowed to fold, forming disul
de bonds between
the C termini. After puri
cation, a sample containing heterodimers labeled with one
D and one A remained, and was used for single-molecule experiments.
In our
first study of protein folding with Chymotrypsin Inhibitor 2 [80], amino acid
positions 1 and 40 were chosen since they were fairly close in the folded structure, but
were well separated in the polypeptide chain. They were also not inside any major
secondary structural elements in the folded protein. The protein was synthesized in
two pieces using solid phase synthesis, and one piece was labeled
first. The protein
was then assembled in refolding conditions, and spliced together using a ligation
reaction. Then, the other label was attached, and several puri
cation steps were
required to obtain the
final sample.
Some subsequent protein folding papers have used a less demanding scheme for
obtaining dual-labeled samples. Proteins, including CspTm [81] and CI2 [80], are
expressed as two-cysteine mutants. One
fluorophore is used to label the protein, and
the singly-labeled protein is separated from unlabeled and dual-labeled protein. The
other
fluorophore is added, and dual-labeled proteins were then separated from
single-labeled proteins. The problem with this scheme is that, even though the
protein is known to have one donor and one acceptor, these
uorophores may be at
either position. Careful controls are then required to ensure that any effects seen are
not due to the presence of two different labeled populations.
Our group introduced two other ways to site-speci
cally label a protein at two sites.
In Jager et al. [28], the procedure of labeling is the same as that described in the
previous paragraph, with one critical change: in the
first labeling step, one of the
labeling sites is hidden using a protein that binds to the CI2. This prevents labeling at
that site, and, upon puri
cation, allows the remaining site to be labeled site
speci
cally with the other
fluorophore. This method may be employed in many
situations, as long as there is a protein that binds to and blocks a region of the protein
of interest. In Jager et al. [29], a short glutamine tag was added to the N terminus and
enzyme ligation was used to speci
cally tag CI2 at that position. This allowed site-
speci
c labeling in two positions
