A small fluorophore reporter of protein conformation and redox state

Article abstract of DOI:10.1039/c1cc10896d

A new thiol-specific reagent introduces a small bis(methylamino) terephthalic acid fluorophore into proteins. The noninvasive probe with distinct spectroscopic properties offers many advantages for protein labeling, purification, and mechanistic work prom

Full text of DOI:10.1039/c1cc10896d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 5714–5716
www.rsc.org/chemcomm
COMMUNICATION
A small fluorophore reporter of protein conformation and redox statew
Graham J. Pound, Alexandre A. Pletnev, Xiaomin Fang and Ekaterina V. Pletneva*
Received 15th February 2011, Accepted 14th March 2011
DOI: 10.1039/c1cc10896d
A new thiol-specific reagent introduces a small bis(methylamino)-
terephthalic acid fluorophore into proteins. The noninvasive probe
with distinct spectroscopic properties offers many advantages for
protein labeling, purification, and mechanistic work promising to
serve as a powerful tool in studies of protein folding and heme
redox reactions.
The small size, ease of purification of labeled proteins, and
distinct spectroscopic properties of the Atpt fluorescent probe
offer a number of advantages for protein modification and
mechanistic work. We show that this minimally invasive label
provides multiple options for monitoring protein conformational
dynamics through fluorescence resonance energy transfer
(FRET) interactions with two different intrinsic chromophores,
both as a donor (D) and an acceptor (A), as well as the probe
quenching in folded proteins. Furthermore, owing to uncommon
difference in the spectral overlap of the dye emission with
absorption of the reduced and oxidized hemes (Fig. S1, ESIw),
the small probe also acts as a convenient fluorescent reporter
of heme redox-state changes.
There is a strong interest in site-specific protein labeling
reagents that introduce small fluorophore probes.1,2 In studies
of protein folding and electron transfer (ET), small proteins,
which are amenable to both theory and experiments, are the
prime subjects of detailed mechanistic investigations. For
them, interprobe distances derived from large chemical dyes
and even larger fluorescent proteins become ambiguous.
Moreover, perturbation of both the protein structure and
the delicate nature of functional conformational changes upon
dye labeling is a valid concern.3,4
Scheme S1 (ESIw) outlines a developed synthetic route to 1
starting with commercially available 3,5-dinitro-p-toluic acid.
Reaction of 1 with N-acetyl-L-cysteine yielded a model compound,
Atpt-Cys. Atpt-Cys shows bright blue fluorescence with lmax
at 459 nm in an aqueous solution, shifted to the red compared
to its 2,6-bis(methylamino)benzoate analog.11 The fluorescence
Anthranilic (o-aminobenzoic) acid, Ant, which carries a
single aromatic ring, is one of the smallest fluorophores
available.5 An intramolecular NHÁ Á ÁOQC hydrogen bond
between its amino and carboxyl groups is important for the
Ant photophysical properties. This probe and its N-methyl
derivative are popular for labeling of nucleotides6 and
peptides7,8 but in the absence of efficient labeling reagents
their applications to proteins so far have been rare and rather
specialized.9,10 Two N-methyl anthraniloyl-derived thiol-labeling
reagents (N-[2-[(bromoacetyl)amino]ethyl]-2-(methylamino)-
benzamide10 and 2-[(iodoacetyl)-methyl-amino]-6-(methylamino)-
benzoic acid methyl ester)11 have been reported but they have
not found widespread use in protein chemistry possibly owing
to thermal degradation10 and limited solubility in aqueous
solutions. Analysis of their structures and reported properties,
as well as our previous work with dye-labeled proteins,4
suggested that an additional ionizable group at the aromatic
ring could not only improve the reagent’s solubility in aqueous
solutions but also allow for easy chromatographic separation
of labeled and unlabeled proteins. This seemingly simple
strategy led us to a new water-soluble thiol-reactive reagent,
Atpt iodoacetamide, 1 (Fig. 1a), based on the Ant analog,
2,6-bis(methylamino)terephthalic acid.
Fig. 1 (a) Atpt iodoacetamide 1 and structure of yeast iso-1 cyt c
showing the Atpt labeling site. Indicated is the Ca(Residue66)–Fe
distance from the crystal structure, 1YCC.12 (b) Distributions of
fluorescence decay rates, P(k) and D–A distances, P(r), from regular-
ization analysis4 of FRET kinetics for folded Atpt66-cyt c. The
average lifetime t0 = 8.0 ns of the unquenched probe was used to
calculate P(r). (c) Refolding of Atpt66-cyt c from 2.0 M to 0.2 M
GuHCl at pH 7.0. Inset: steady-state fluorescence spectra at pH 7.0 for
the protein folded in a 100 mM sodium phosphate buffer (red) and
unfolded in 5.9 M GuHCl (blue).
Department of Chemistry, Dartmouth College, Hanover, NH 03755,
USA. E-mail: ekaterina.pletneva@dartmouth.edu;
Fax: +1 603 646 3946; Tel: +1 603 646 0933
w Electronic supplementary information (ESI) available: Atpt iodo-
acetamide synthesis (Scheme S1), experimental details, Fig. S1–S5. See
DOI: 10.1039/c1cc10896d
c
5714 Chem. Commun., 2011, 47, 5714–5716
This journal is The Royal Society of Chemistry 2011

View Article Online
decay of Atpt-Cys is biexponential in a phosphate buffer at pH
7.4 (with very similar components t1 = 6.1 Æ 0.2 ns (23%) and
t2 = 8.6 Æ 0.1 ns (77%), F = 0.28) and monoexponential in
100% ethanol (t = 9.3 Æ 0.1 ns, F = 0.51). Such a solvent
dependence and two-component decay in aqueous solutions
have been previously reported for other anthraniloyl probes.10
We have labeled with Atpt cysteine mutants of two
representative proteins, a positively charged heme protein
cytochrome c (cyt c, Fig. 1),12 and a negatively charged D1
domain of cytoskeletal protein vinculin (Fig. 2).13 The labeled
and unlabeled proteins elute at different concentrations of
salt on cation- and anion-exchange columns (Fig. S2, ESIw)
providing an easy way to purify reaction products, critical for
quantitative FRET work. The proteins were labeled under
nondenaturing conditions and the yields of purified labeled
proteins were above 50%. For both proteins, labeling at
solvent-exposed sites did not cause perturbations in the
protein secondary structure, stability (Fig. S3, ESIw) and,
based on results below, tertiary structure.
accounts for the length of the probe linker4 and conformational
freedom of the C-terminal Trp258.13 Moreover, Atpt lifetime
measurements in Atpt66-cyt c gave a very similar fluorophore–
heme distance to that in the protein derivative labeled with
dansyl at the same site.4 Guanidine hydrochloride enhances
Atpt and Trp fluorescence in cyt c (Fig. 1) and vinculin
(Fig. 2), respectively, reflecting increases in the Atpt66–heme
and Trp258–Atpt121 distances upon protein unfolding.
Prominent fluorescence signals with Atpt-labeled proteins
enable high signal-to-noise kinetics measurements of Atpt–heme
and Trp–Atpt approach during refolding.
We find that Atpt fluorescence is considerably quenched in
the vinculin D1 folded state (Fig. 2). Previous studies have
found drastic reduction of the Ant quantum yield when the
probe is coupled to the Na-amino group of Trp compared to
other amino acids.14 High Trp concentrations do increase the
rate of the Atpt-Cys fluorescence decay (Fig. S5, ESIw),
suggesting that Trp could indeed act as a dynamic quencher
of Atpt fluorescence when the Atpt–Trp pair comes into
contact. However, the Trp-to-Atpt distance of 20 A in vinculin
is too long to support contact dynamic quenching between
these sites. Instead, distinct differences in the steady-state
spectra and yet very similar Atpt lifetimes for folded and
unfolded Atpt-labeled vinculin D1 (Fig. 2b) suggest that the
mechanism of quenching is static. Furthermore, the Atpt
quenching is not a unique property of the site 121, as we have
also observed a similar decrease in the Atpt fluorescence signal
with folded vinculin D1 labeled at site 85.
Absorption and fluorescence spectra of Atpt overlap with
Trp emission and heme absorption spectra, respectively
(Fig. S4, ESIw), allowing FRET measurements of distances
in these proteins. Analyses of Atpt quenching in cyt c (Fig. 1b)
and Trp quenching in vinculin (Fig. 2b) yielded Atpt66–heme
and Trp258–Atpt121 distances of 21 and 20 A, respectively.
These distances are consistent with crystal structures, if one
A couple of scenarios could explain such quenching.
Previous studies have indicated that only one of the two Ant
rotamers contributes to the observed fluorescence spectrum.5
Thus, similar to cis–trans prolyl imide bond isomerization,
protein folding may change the ratio of the two Atpt rotamers,
affecting the steady-state fluorescence signal but not necessarily
the probe lifetime or anisotropy. Alternatively, the close
interactions between Atpt and another group15 in the folded
protein could lead to rapid quenching not detectable by the
fluorescence lifetime measurements. Future studies with a
variety of model peptides will distinguish between these two
possibilities. However, because the Atpt signal is so intense
and sensitive to the protein folded state, it is already apparent
that it provides another powerful option for monitoring
protein folding.
Interestingly, the Atpt-labeled cyt c shows distinct fluorescence
spectra in the ferrous and ferric states (Fig. 3), allowing for
differentiation of the heme redox state by Atpt fluorescence in
the labeled protein. Furthermore, the progress of heme redox
reactions in Atpt-labeled cyt c can be conveniently monitored
by fluorescence. Importantly, the observed kinetics match
those monitored by traditional absorption methods.16 Fluorescent
redox-state sensors have advanced studies of copper metallo-
proteins,17,18 however, applications of FRET to heme redox
processes so far have been limited to the Cy5-labeled cyt
c550.17 Modification with the bulky red and near-IR fluores-
cent dyes, such as Cy5, is known to perturb the protein
structure, stability, and binding interactions,3,19,20 properties
that may critically affect metalloprotein redox behavior.21,22
Such perturbations are less likely with much smaller dyes that
typically have blue or green emission. However, prominent
Fig. 2 (a) Structure of vinculin D1 showing the Atpt labeling site and
Trp258. Indicated is the Ca(Residue121)–Ca(Residue258) distance
from the crystal structure, 1RKC.13 (b) Steady-state fluorescence
spectra at pH 7.4 for Atpt121–vinculin D1, folded in a 100 mM
HEPES buffer (red) and unfolded in 6.3 M GuHCl (blue). The
spectrum of unlabeled folded protein is shown in a dashed line.
(c) Refolding of Atpt121–vinculin D1 from 4.2 M to 0.4 M GuHCl
at pH 7.4 monitored by decrease in Atpt (top) and Trp (bottom)
fluorescence. Global fitting of the kinetics data yielded k1 = 4.7 sÀ1
and k2 = 0.3 sÀ1
.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 5714–5716 5715

View Article Online
intrinsic as well as extrinsic chromophores and the probe
quenching in folded proteins. With the bright Atpt fluorescence,
such studies may be potentially expanded to single-molecule
work. Differences in overlap of the Atpt emission with absorption
spectra of reduced and oxidized hemes afford new applications
of fluorescently-labeled proteins for probing heme redox
reactions.
We thank Dr Jay R. Winkler for recognizing the potential
of Ant derivatives in protein folding research and Professor
Tina Izard for the gift of vinculin D1 plasmid. This study was
supported by
(E.V.P).
a NSF CAREER Award, CHE-0953693
Notes and references
1 A. A. Ensign, I. Jo, I. Yildirim, T. D. Krauss and K. L. Bren, Proc.
Natl. Acad. Sci. U. S. A., 2008, 105, 10779–10784.
2 J. W. Taraska, M. C. Puljung and W. N. Zagotta, Proc. Natl.
Acad. Sci. U. S. A., 2009, 106, 16227–16232.
3+
3 D. S. Kudryashov and E. Reisler, Biophys. J., 2003, 85, 2466–2475.
4 E. V. Pletneva, H. B. Gray and J. R. Winkler, J. Mol. Biol., 2005,
345, 855–867.
5 C. A. Southern, D. H. Levy, G. M. Florio, A. Longarte and
T. S. Zwier, J. Phys. Chem. A, 2003, 107, 4032–4040.
6 A. Leskovar and J. Reinstein, Arch. Biochem. Biophys., 2008, 473,
16–24.
Fig. 3 Kinetics of cyt c oxidation by 0.73 mM Co(phen)3
in a
50mM sodium phosphate buffer at pH 7.0 monitored by (a) Atpt66
fluorescence (kbim = 1.1 Â 103
M
À1sÀ1) and (b) heme absorption
(kbim = 1.0 Â 103
M
À1sÀ1) changes. Inset: steady-state fluorescence
spectra for folded ferrous (solid line) and ferric (dashed line) Atpt66-cyt c.
7 J. Ø. Duus, M. Meldal and J. R. Winkler, J. Phys. Chem. B, 1998,
102, 6413–6418.
8 A. R. Mezo, R. P. Cheng and B. Imperiali, J. Am. Chem. Soc.,
2001, 123, 3885–3891.
9 N. Hagag, E. R. Birnbaum and D. W. Darnall, Biochemistry, 1983,
22, 2420–2427.
10 D. J. Pouchnik, L. E. Laverman, F. Janiak-Spens, D. M. Jameson,
G. D. Reinhart and C. R. Cremo, Anal. Biochem., 1996, 235,
26–35.
11 G. S. Reddy, H.-Y. Chen and I.-J. Chang, J. Chin. Chem. Soc.,
2006, 53, 1303–1308.
12 G. V. Louie and G. D. Brayer, J. Mol. Biol., 1990, 214, 527–555.
13 T. Izard, G. Evans, R. A. Borgon, C. L. Rush, G. Bricogne and
P. R. Bois, Nature, 2004, 427, 171–175.
absorption spectra of both ferrous and ferric hemes in this
spectral region cause small differences in their overlap with
emission of many popular small dyes (Fig. S1, ESIw), which
makes it difficult to discriminate the two oxidation states by
fluorescence. The broad Atpt emission spectrum and its
favorable position of lmax at 459 nm allow for the overlap
with both the Soret and Q absorption bands of cyt c amplifying
differences in the integrals J(l) for the reduced and oxidized
hemes (Fig. S1 and S4, ESIw).
The kinetic results obtained with a small noninvasive Atpt
dye suggest a valuable approach for studying redox reactions of
heme proteins, particularly those involving two or more heme
centers. ET in cyt c oxidase, bc1 complex, and between cyt c and
cyt c peroxidase are just a few examples21 where overlapping
heme spectra complicate measurements of the heme redox
changes by absorption techniques. Replacement of the heme
iron by zinc in one of the heme centers for photoinduced redox
reactions is one strategy to solve this problem.23 However, the
metal reconstitution procedure requires protein treatment
with hydrogen fluoride, which may lead to large protein losses,
and is not easily applicable to multiheme24–26 proteins. A well-
established thiol labeling procedure, which is effective even
for more challenging membrane proteins,27 provides a straight-
forward alternative to protein modification for studies of
the thermal redox reactions with widely-available fluorescence
instrumentation. Furthermore, because of the FRET distance
dependence, a strategic placement of Atpt in multiheme proteins
could enable changes in the redox and potentially also the
ligation state of a particular heme center to be measured.
In summary, Atpt iodoacetamide is a new reagent for
highly-efficient and minimally invasive fluorescent labeling of
proteins. The photophysical properties of the small Atpt probe
provide multiple options for monitoring protein folding and
conformational changes through FRET interactions with
14 A. S. Ito, R. D. Turchiello, I. Y. Hirata, M. H. Cezari, M. Meldal
and L. Juliano, Biospectroscopy, 1998, 4, 395–402.
15 H. Chen, S. S. Ahsan, M. B. Santiago-Berrios, H. D. Abruna and
W. W. Webb, J. Am. Chem. Soc., 2010, 132, 7244–7245.
16 A. M. Carver, M. De, H. Bayraktar, S. Rana, V. M. Rotello and
M. J. Knapp, J. Am. Chem. Soc., 2009, 131, 3798–3799.
17 S. Kuznetsova, G. Zauner, R. Schmauder, O. Mayboroda,
A. Deelder, T. J. Aartsma and G. W. Canters, Anal. Biochem.,
2006, 350, 52–60.
18 J. M. Salverda, A. V. Patil, G. Mizzon, S. Kuznetsova, G. Zauner,
N. Akkilic, G. W. Canters, J. J. Davis, H. A. Heering and
T. J. Aartsma, Angew. Chem., Int. Ed., 2010, 49, 5776–5779.
19 H. Wang, K. Chen, G. Niu and X. Chen, Mol. Pharmaceutics,
2009, 6, 285–294.
20 B. Schuler, E. A. Lipman and W. A. Eaton, Nature, 2002, 419,
743–747.
21 Biological Inorganic Chemistry: Structure and Reactivity, ed.
I. Bertini, H. B. Gray, E. I. Stiefel and J. S. Valentine, University
Science Books, Sausalito, CA, 2007.
22 F. A. Armstrong, Curr. Opin. Chem. Biol., 2005, 9, 110–117.
23 S. A. Kang, P. J. Marjavaara and B. R. Crane, J. Am. Chem. Soc.,
2004, 126, 10836–10837.
24 G. Su Pulcu, B. L. Elmore, D. M. Arciero, A. B. Hooper and
S. J. Elliott, J. Am. Chem. Soc., 2007, 129, 1838–1839.
25 O. Farver, P. M. Kroneck, W. G. Zumft and I. Pecht, Proc. Natl.
Acad. Sci. U. S. A., 2003, 100, 7622–7625.
26 C. M. Paquete and R. O. Louro, Dalton Trans., 2010, 39,
4259–4266.
27 R. B. Spruijt, C. J. Wolfs, J. W. Verver and M. A. Hemminga,
Biochemistry, 1996, 35, 10383–10391.
c
5716 Chem. Commun., 2011, 47, 5714–5716
This journal is The Royal Society of Chemistry 2011