Site-specific analysis of protein hydration based on unnatural amino acid fluorescence

Article abstract of DOI:10.1021/jacs.5b01681

Hydration of proteins profoundly affects their functions. We describe a simple and general method for site-specific analysis of protein hydration based on the in vivo incorporation of fluorescent unnatural amino acids and their analysis by steady-state fl

Full text of DOI:10.1021/jacs.5b01681

Article
pubs.acs.org/JACS
Site-Specific Analysis of Protein Hydration Based on Unnatural
Amino Acid Fluorescence
Mariana Amaro,† Jan Brezovsky,‡,⌋ Silvia Kovacova,∥,⌋ Jan Sykora,† David Bednar,‡,⌋ Vaclav Nemec,∥,⌋
́
́
̌
́
́
́
̌
́
̌
Veronika Liskova,‡ Nagendra Prasad Kurumbang,‡,# Koen Beerens,‡ Radka Chaloupkova,‡
̌
́
́
Kamil Paruch,*,∥,⌋ Martin Hof,*,† and Jirí Damborsky*,‡,⌋
̌
́
†J. Heyrovsky Institute of Physical Chemistry of the ASCR, v. v. i., Academy of Sciences of the Czech Republic, Dolejskova 3, 182 23
Prague 8, Czech Republic
‡Loschmidt Laboratories, Department of Experimental Biology and Research Centre for Toxic Compounds in the Environment
RECETOX, Faculty of Science, Masaryk University, Kamenice 5/A13, 625 00 Brno, Czech Republic
∥Department of Chemistry, Faculty of Science, Masaryk University, Kamenice 5/A13, 625 00 Brno, Czech Republic
⌋International Clinical Research Center, St. Anne’s University Hospital Brno, Pekarska 53, 656 91 Brno, Czech Republic
S
* Supporting Information
ABSTRACT: Hydration of proteins profoundly affects their functions. We
describe a simple and general method for site-specific analysis of protein
hydration based on the in vivo incorporation of fluorescent unnatural amino
acids and their analysis by steady-state fluorescence spectroscopy. Using this
method, we investigate the hydration of functionally important regions of
dehalogenases. The experimental results are compared to findings from
molecular dynamics simulations.
molecular dynamics (MD) simulations and previous re-
ports.11,13
INTRODUCTION
■
Protein hydration is important in enzymatic catalysis1 since it
influences enzymes’ kinetics2 and enantioselectivity,3 protein
folding,4 ligand-binding, and DNA−protein interactions.5
Hydration has been studied using X-ray absorption,6 NMR
spectroscopy,7 neutron scattering,8 dielectric relaxation spec-
troscopy,9 two-dimensional infrared spectroscopy,10 and time-
resolved fluorescence spectroscopy.11 All these techniques
provide valuable information on the arrangement of water
molecules in the vicinity of specific protein moieties. However,
they require costly and highly specialized instrumentation. Here
we present a new technique for studying protein hydration
using very basic laboratory instruments. Our approach is based
on a recently developed method for the site-specific
incorporation of unnatural amino acids (UAAs) into protein
structures12 and the analysis of their properties using steady-
state fluorescence (SSF) spectroscopy. Additionally, we
developed a more economical synthesis of L-(7-hydroxycou-
marin-4-yl) ethylglycine, whose preparation had represented a
bottleneck for its wider use in protein labeling. Furthermore,
for the first time a quantitative analysis of the hydroxycoumarin
fluorescence within a specific protein region is performed,
followed by newly established data deconvolution that enables
us to estimate the level of protein hydration. We demonstrate
its effectiveness by characterizing the molecular environments
of UAAs in the tunnel mouths of two haloalkane dehalogenases
and comparing these experimental results to the output of
RESULTS AND DISCUSSION
■
To deliver an UAA into a specific site encoded by a nonsense
codon, the tRNA/aminoacyl-tRNAsynthetase pair needs to be
constructed. In this project we utilized the L-(7-hydroxycou-
marin-4-yl) ethylglycine,14 which contains the fluorophore 7-
hydroxy-4-methylcoumarin (7H4MC). The photophysics of
7H4MC is environment-sensitive.15,16 Three forms of 7H4MC,
neutral, anionic, complexed, exist in equilibrium (Supporting
Information Figure S1A), each with different excitation
maxima.17 In addition, a tautomeric form can be created in
the excited state via proton transfer between the spatially
separated carbonyl and hydroxyl groups (Figure 1A, Supporting
Information Figure S1B). Accordingly, the SSF spectra of all
four excited state forms are shifted in respect to each other and
thus their contributions to the overall signal can be separated.
As the equilibria (both in the ground and excited states) are
governed by the hydration state of the dye, information on the
contributions of the corresponding microenvironments to the
SSF spectra can be gained.
To demonstrate how the fluorescence of 7H4MC can
provide information on the extent of hydration of its vicinity,
Received: July 26, 2014
Revised: February 14, 2015
Published: March 27, 2015
© 2015 American Chemical Society
4988
DOI: 10.1021/jacs.5b01681
J. Am. Chem. Soc. 2015, 137, 4988−4992

Journal of the American Chemical Society
Article
Figure 1. (A) General reaction scheme of 7H4MC in the excited state. Neutral, anionic and complexed (for instance with an adjacent amino group)
forms of the fluorophore can exist and/or coexist. Proton transfer (reaction III) can additionally take place, creating a tautomeric form. This proton
transfer has been shown to be promoted by the presence of “structured water”.15 In bulk water, anion formation (reaction I) prevails because the
pKa* of the excited state is below 1.5.15 The numbers accompanying the rippled arrows correspond to the maximum of the emission spectra of each
particular 7H4MC form. Red reaction arrows highlight the dominant pathways in the excited state. (B) Dependence of the hydration parameter on
water content (w0) in AOT micelles. The graph illustrates the increase in the magnitude of the fluorescence emission of the anionic and tautomer
forms of 7H4MC (the sum of the decomposed spectrum areas emitted from the two forms) with the increasing amount of water added to the
reverse AOT micelles (w0). Imidazole was added in order to mimic −NH+− and −NH− functional groups present in the protein matrix, resulting in
the formation of the complex form. The areas were gained by the decomposition of the steady-state emission spectra. (C and D) Emission spectra of
UAA incorporated into DhaA:C176UAA and DbjA:G183UAA, respectively. The spectra were recorded at an excitation wavelength of 320 nm. The
black curve represents the recorded emission spectrum; the green, gray, blue and orange curves represent its decomposition into the neutral,
complexed, anionic and tautomeric components, respectively. The red curve stands for the result of the fit. (E) Hydration parameter upon thermal
denaturation. The temperature dependence of the hydration parameter (AA + AT) was measured for both herein investigated proteins. Tm values
given in the graph legend corresponds to the melting temperature, which is the temperature at which 50% of the protein becomes unfolded.
and specific interactions between the fluorophore and proton
donors or acceptors, we decomposed its SSF spectra in
docusate sodium (AOT) reverse micelles. At low water/
surfactant molar ratios (w0), water molecules associate strongly
with the polar heads of the surfactant molecules, forming
“structured water”. As w0 increases, “bulk” water is created
inside the reverse micelles. Studies on the photophysics of
7H4MC in reverse micelles15,16 have shown that in the absence
of water (w0 ≈ 0) the dye fluoresces mainly from its neutral
form. Stepwise addition of water causes an increasing
contribution from the tautomeric form, with the anionic form
also becoming important as w0 increases further. The relative
contributions of these particular 7H4MC forms can be
determined by decomposing the corresponding SSF spectra
(Supporting Information Figure S2). At w0 ≈ 0, only the
neutral form (emission maximum wavelength ∼380 nm) is
present (Supporting Information Table S1 and Figure S2A).
When w0 rises above 10, the tautomer band (∼480 nm) appears
and gradually becomes more intense, indicating that the dye is
surrounded by an increasing number of water molecules bound
to the detergent’s polar headgroups. Note that even at very low
w0 values, the complexed form (∼420 nm) can be detected if
proton donor or acceptor molecules are added to the AOT
reverse micelles (Supporting Information Table S2). Further
increases in the water content produce bulk water inside the
reverse micelles, sharply increasing the contribution of the
anionic band (∼450 nm; Supporting Information Figure S2C).
The anionic form is the dominant fluorescing species in bulk
water. We utilized this sensitivity of 7H4MC photophysics and
determined the sum of the contributions of the anionic and
tautomer forms to the total emission spectrum (quantified by
parameter AA + AT), whose formation is conditioned by the
presence of water. Indeed, AA + AT gradually rises with
increasing water content w0 for three different AOT systems
(Figure 1B, Supporting Information Tables S1 and S2). This
parameter can therefore be used as an indicator of the extent of
hydration.
Prior to testing this approach in proteins, the synthesis of the
UAA containing the 7H4MC fluorophore had to be optimized
since gram quantities of UAA were required for the cultivation
of recombinant Escherichia coli. We realized that the UAA is
poorly soluble in the mobile phase previously used for HPLC
purification (<400 mg/L).18 Therefore, we developed a new,
significantly more economical protocol that avoids the use of
4989
DOI: 10.1021/jacs.5b01681
J. Am. Chem. Soc. 2015, 137, 4988−4992

Journal of the American Chemical Society
Article
HPLC and utilizes ion exchange chromatography followed by
precipitation. The fluorophore was inserted at specific locations
within the sequences of two haloalkane dehalogenases
DhaA:C176UAA and DbjA:G183UAA. The selected enzymes
are functional variants derived from haloalkane dehalogenases
DhaA from Rhodococcus rhodochrous NCIMB130643319 and
DbjA from Bradyrhizobium japonicum USDA110.20 Circular
dichroism analysis (Supporting Information Figure S3) and
activity testing (Supporting Information Table S3) confirmed
that introduction of the UAA did not disrupt the structure and
the catalytic function of the enzymes. Although both enzymes
catalyze nucleophilic substitution via the same mechanism their
substrate specificity and enantioselectivity differ substantially.13
This is attributed to the different architectures of their active
sites as well as access tunnels. The fluorescent UAA was
inserted at the position 176 and 183 of DhaA and DbjA,
respectively, to probe the hydration of their tunnels (Figure
2A) which are known to be important for enzyme function.3,13
Table 1. Enantioselectivity and Its Thermodynamic
Components for the Kinetic Resolution of 2-Bromopentane
a
by Haloalkane Dehalogenases DhaA and DbjA
E value
ΔR‑SΔH⧧
TΔR‑SΔS⧧298K
ΔR‑SΔG⧧298K
enzyme
DhaA
298 K
[kJ mol−1
]
[kJ mol−1
]
[kJ mol−1
]
20
−15.37
−69.51
−8.22
−57.78
−7.15
b
DbjA
132
−11.73
a
E value−enzyme enantioselectivity, ΔR‑SΔG‡ − differential free
energy of activation between (R)- and (S)- enantiomer of 2-
bromopentane (ΔR‑SΔG‡ = ΔR‑SΔH‡ − TΔR‑SΔS‡) and its enthalpic
b
(ΔR‑SΔH‡ 298 K) and entropic (TΔR‑SΔS‡ 298 K) terms. Data from
Prokop et al.13
component of DbjA enantioselectivity are four times and seven
times greater in absolute values, respectively, than DhaA.
For the decomposition of the SSF spectrum into its separate
components, the excitation wavelength was set to 320 nm in
order to predominantly excite the neutral form of the
fluorophore. The neutral form then populates the complexed,
anionic and tautomeric forms (Figure 1A). As shown in Figure
1C, the fluorescence emission of DhaA:C176UAA originates
from the neutral, complexed and anionic forms. The tautomer
contributes only marginally to the emission spectrum.
Conversely, significant tautomer emission was observed for
DbjA:G183UAA (Figure 1D). The formation of the tautomer
in DbjA:G183UAA indicates that the chromophore is exposed
to “structured water”, that is, water molecules with residence
times longer than the fluorescence time scale. The presence of
the tautomer was also confirmed by time-resolved fluorescence
spectroscopy (Supporting Information Figure S4). The decay
curve contained a negative component due to the population of
the tautomer from the excited neutral species. Such particular
feature of the fluorescence decay was not observed for
DhaA:C176UAA. Because the tautomeric and anionic species
are only formed in aqueous environments, their summed
contributions to the overall signal reflect the hydration level
within the dye’s microenvironment. The anionic and
tautomeric forms account only for 50% of the emission signal
of DhaA:C176UAA. However, 70% of the emission signal of
DbjA:G183UAA can be attributed to these forms (Figure 1E,
Supporting Information Table S4). This indicates that the
microenvironment surrounding the UAA in DhaA:C176UAA is
much less extensively hydrated than in DbjA:G183UAA. To
further validate the approach, we thermally denaturated the
enzymes. The UAA originally buried within the protein interior
becomes exposed to more hydrated microenvironment upon
the denaturation process. As expected, for both enzymes the
increasing temperature causes an increase in the contribution of
the anion at the expense of the others forms, leading to the
growth of AA + AT parameter (Figure 1E, Supporting
Information Table S4).
Figure 2. Local environment and hydration of the UAAs in
DhaA:C176UAA and DbjA:G183UAA. (A) Positions of the UAAs
(indicated by the red surface patches) and the tunnel mouths
(indicated by black arrows) on the surface of the enzymes. (B) Local
interactions of UAAs with hydrogen acceptors and amino groups
(purple sticks). The buried (UAA_3) and exposed (UAA_4)
conformations of the UAA in DbjA:G183UAA are indicated by
green and yellow sticks, respectively. The conformation of the UAA in
DhaA:C176UAA is indicated by the orange stick model. (C)
Hydration of UAAs. The blue surface represents enzyme regions
that are occupied by water molecules for at least 40% of the total
simulation time.
Specifically, DhaA possesses a narrower tunnel and at the same
time exhibits significantly lower enantioselectivity to β-
bromoalkanes than DbjA.13 Recently, we have demonstrated
that hydration and dynamics of the tunnel mouth are important
determinants of DbjA enantioselectivity toward β-bromoal-
kanes.3 Thermodynamic analysis revealed that enantiodiscrimi-
nation of β-bromoalkanes by DbjA and DhaA is differently
influenced by individual thermodynamic contributions−differ-
ential activation enthalpy (ΔR‑SΔH⧧) and entropy (ΔR‑SΔS⧧).
While the resolution of 2-bromopentane by DbjA is driven
almost equally by both enthalpic and entropic terms (where the
entropic term represents 83% of the enthalpic term) the DhaA
enantioselectivity is clearly dominated by enthalpy (Table 1). In
both cases, the preferred (R)-enantiomer is favored by the
enthalpy while the nonpreferred (S)-enantiomer is favored by
the entropy. Interestingly, the enthalpic and the entropic
Replicated 200 ns MD simulations were performed for both
labeled and wild type structures of studied enzymes. The
periodic-boundary NPT simulations were carried out in
AMBER12 (University of California, San Francisco, 2012)
using ff10 force field (Supporting Information Table S5 and
Figure S5).21−23 The level of hydration within the tunnel
mouth of the wild type enzymes was about 1.7-times higher in
DbjA than in DhaA (Table 2). The MD simulations show that
the UAA is in one dominant conformation in the tunnel of
DhaA:C176UAA (Supporting Information Figure S6). Con-
versely, the wider tunnel mouth of DbjA:G183UAA allowed the
UAA to adopt two equally relevant stable conformations, one
4990
DOI: 10.1021/jacs.5b01681
J. Am. Chem. Soc. 2015, 137, 4988−4992

Journal of the American Chemical Society
Article
Table 2. Hydration of Haloalkane Dehalogenases DhaA, DbjA, DhaA:C176UAA, and DbjA:G183UAA Revealed by
b
Fluorescence Spectroscopy and Molecular Dynamics Simulations
parameter
DhaA:C176UAA DbjA:G183UAA
DhaA
DbjA
overall contributions of anionic and tautomeric forms to the emission UAA less
UAA more
hydrated
less hydrated tunnel
more hydrated tunnel
a
a
SSF spectra
hydrated
mouth
mouth
number of water molecules within 5 Å
9
2
21 5 or 28
4
16
4
27
4
a
b
The enzymes have been characterized previously.11 The methodology used was different from the SSF method. Experimental parameters and
results are presented in normal text; parameters and results obtained from MD simulations are presented in italics.
buried into the tunnel (UAA_3) and one more exposed
(UAA_4; Supporting Information Table S6 and Figure S6).
The simulations suggested that the UAA microenvironment is
substantially less hydrated in DhaA:C176UAA than in both
variants of DbjA:G183UAA (Table 2, Figure 2C), which is
consistent with our experimental results. Additionally, the
simulation indicated that the fluorophore often forms hydrogen
bonds with adjacent amino acid residues, Thr148 in
DhaA:C176UAA and Glu146 in DbjA:G183UAA , and with
amino acids whose side chains contain amino groups, His272
for DhaA:C176UAA and Arg179 or His139 for DbjA:-
G183UAA (Figure 2B, Supporting Information Table S7).
This may facilitate the formation of the complexed form
detected in our experiments. Finally, the simulations indicated
that water molecules in the microenvironment of the buried
conformation of DbjA:G183UAA have residence times of up to
60 ns, which is much longer than the fluorescence time scale
(Supporting Information Figure S7). This explains the strong
contribution of the tautomeric form for this mutant. Such
extremely long residence times are not found in structures
without UAA (Supporting Information Figure S7). This
suggests that the “structured water” detected in
DbjA:G183UAA is induced by the UAA itself locking several
water molecules within the active site pocket.
*hof@jh-inst.cas.cz
*jiri@chemi.muni.cz
Present Address
#Great Lakes Bioenergy Research Center, College of Engineer-
ing, University of Wisconsin-Madison, Wisconsin 53706-1567,
United States.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Prof. Peter G. Schultz and Dr. Jeremy Mills from The Scripps
Research Institute (USA) are greatly acknowledged for
providing genetic material and valuable advices in UAA
technology. Financial support from the Czech Science
Foundation via grants P208/12/G016 (M.H. and J.S.) and
P207/12/0775 (R.Ch.) and the Ministry of Education of the
Czech Republic (LO1214; CZ.1.05/1.1.00/02.0123) is ac-
knowledged. Moreover, M.H. acknowledges the Praemium
Academie Award from Academy of Sciences of the Czech
Republic. The work of J.B. and K.B. was supported by Program
of “Employment of Best Young Scientists for International
Cooperation Empowerment” (CZ1.07/2.3.00/30.0037) with
cofinancing from the European Social Fund and the state
budget of the Czech Republic. MetaCentrum is acknowledged
for providing access to their computing facilities, supported by
the Ministry of Education of the Czech Republic
(LM2010005). CERIT-SC is acknowledged for providing
access to their computing facilities, under the program Center
CERIT scientific Cloud (CZ.1.05/3.2.00/08.0144).
CONCLUSIONS
■
Recent findings emphasizing the importance of dynamics and
hydration in enzymatic catalysis and rational protein design2,3
have created a strong demand for methods that provide site-
specific information on these factors. In our approach, site-
specificity is guaranteed by using UAA. SSF spectroscopic
analysis of the hydroxycoumarin probe incorporated into the
structure of the enzymes is a strikingly simple and universally
applicable experimental procedure. The photophysics of the
UAA provide qualitative information on the extent of hydration
as demonstrated for two HLDs with already characterized
hydration levels.11 Although MD simulations show that
incorporation of the chromophore can influence the residential
times of water molecules, the conclusions on the hydration
levels are valid. Given the ongoing development of UAA
technology, this method could potentially be used to analyze
hydration at specific sites in a wide range of proteins.
REFERENCES
■
(1) Fogarty, A. C.; Duboue-Dijon, E.; Sterpone, F.; Hynes, J. T.;
Laage, D. Chem. Soc. Rev. 2013, 42, 5672.
(2) Privett, H. K.; Kiss, G.; Lee, T. M.; Blomberg, R.; Chica, R. A.;
Thomas, L. M.; Hilvert, D.; Houk, K. N.; Mayo, S. L. Proc. Natl. Acad.
Sci. U. S. A. 2012, 109, 3790.
(3) Sykora, J.; Brezovsky, J.; Koudelakova, T.; Lahoda, M.; Fortova,
A.; Chernovets, T.; Chaloupkova, R.; Stepankova, V.; Prokop, Z.; Kuta
Smatanova, I.; Hof, M.; Damborsky, J. Nat. Chem. Biol. 2014, 10, 428.
(4) Sekhar, A.; Vallurupalli, P.; Kay, L. E. Proc. Natl. Acad. Sci. U. S. A.
2013, 110, 11391.
(5) Levy, Y.; Onuchic, J. N. Annu. Rev. Biophys. Biomol. Struct. 2006,
35, 389.
(6) Grossman, M.; Born, B.; Heyden, M.; Tworowski, D.; Fields, G.
B.; Sagi, I.; Havenith, M. Nat. Struct. Mol. Biol. 2011, 18, 1102.
(7) Nucci, N. V.; Pometun, M. S.; Wand, A. J. Nat. Struct. Mol. Biol.
2011, 18, 245.
(8) Russo, D.; Hura, G.; Head-Gordon, T. Biophys. J. 2004, 86, 1852.
(9) Oleinikova, A.; Sasisanker, P.; Weingartner, H. J. Phys. Chem. B
2004, 108, 8467.
(10) King, J. T.; Kubarych, K. J. J. Am. Chem. Soc. 2012, 134, 18705.
(11) Jesenska, A.; Sykora, J.; Olzynska, A.; Brezovsky, J.; Zdrahal, Z.;
Damborsky, J.; Hof, M. J. Am. Chem. Soc. 2009, 131, 494.
(12) Summerer, D.; Chen, S.; Wu, N.; Deiters, A.; Chin, J. W.;
Schultz, P. G. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 9785.
ASSOCIATED CONTENT
* Supporting Information
Experimental procedures, synthesis of UAA, results of emission
spectra deconvolution, CD spectra, and results from MD
simulations. This material is available free of charge via the
Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Authors
*paruch@chemi.muni.cz
■
4991
DOI: 10.1021/jacs.5b01681
J. Am. Chem. Soc. 2015, 137, 4988−4992

Journal of the American Chemical Society
Article
(13) Prokop, Z.; Sato, Y.; Brezovsky, J.; Mozga, T.; Chaloupkova, R.;
Koudelakova, T.; Jerabek, P.; Stepankova, V.; Natsume, R.; van
Leeuwen, J. G. E.; Janssen, D. B.; Florian, J.; Nagata, Y.; Senda, T.;
Damborsky, J. Angew. Chem., Int. Ed. 2010, 49, 6111.
(14) Schultz, P. G.; Mills, J. H.; Lee, H. S.; Liu, C. C.; Wang, J. Y.
ChemBioChem. 2009, 10, 2162.
(15) Choudhury, S. D.; Nath, S.; Pal, H. J. Phys. Chem. B 2008, 112,
7748.
(16) Choudhury, S. D.; Pal, H. J. Phys. Chem. B 2009, 113, 6736.
(17) Moriya, T. Bull. Chem. Soc. Jpn. 1983, 56, 6.
(18) Wang, J. Y.; Xie, J. M.; Schultz, P. G. J. Am. Chem. Soc. 2006,
128, 8738.
(19) Kulakova, A. N.; Larkin, M. J.; Kulakov, L. A. Microbiology 1997,
143, 109.
(20) Sato, Y.; Monincova, M.; Chaloupkova, R.; Prokop, Z.;
Ohtsubo, Y.; Minamisawa, K.; Tsuda, M.; Damborsky, J.; Nagata, Y.
Appl. Environ. Microb. 2005, 71, 4372.
(21) Hornak, V.; Abel, R.; Okur, A.; Strockbine, B.; Roitberg, A.;
Simmerling, C. Proteins 2006, 65, 712.
(22) Joung, I. S.; Cheatham, T. E. J. Phys. Chem. B 2008, 112, 9020.
(23) Joung, I. S.; Cheatham, T. E. J. Phys. Chem. B 2009, 113, 13279.
4992
DOI: 10.1021/jacs.5b01681
J. Am. Chem. Soc. 2015, 137, 4988−4992