Resolving Electronic Transitions in Synthetic Fluorescent Protein Chromophores by Magnetic Circular Dichroism

Article abstract of DOI:10.1002/cphc.201600313

The detailed electronic structures of fluorescent chromophores are important for their use in imaging of living cells. A series of green fluorescent protein chromophore derivatives is examined by magnetic circular dichroism (MCD) spectroscopy, which allows the resolution of more bands than plain absorption and fluorescence. Observed spectral patterns are rationalized with the aid of time-dependent density functional theory (TDDFT) computations and the sum-over-state (SOS) formalism, which also reveals a significant dependence of MCD intensities on chromophore conformation. The combination of organic and theoretical chemistry with spectroscopic techniques also appears useful in the rational design of fluorescence labels and understanding of the chromophore's properties. For example, the absorption threshold can be heavily affected by substitution on the phenyl ring but not much on the five-member ring, and methoxy groups can be used to further tune the electronic levels.

Full text of DOI:10.1002/cphc.201600313

DOI: 10.1002/cphc.201600313
Articles
Resolving Electronic Transitions in Synthetic Fluorescent
Protein Chromophores by Magnetic Circular Dichroism
[b]
[a]
[a]
[a]
[a]
ˇ
ˇ
ˇ
ˇ
ˇ
Petr Stepꢀnek, Thomas Y. Cowie, Martin Safarꢁk, Jaroslav Sebestꢁk, Radek Pohl, and
[a]
ˇ
Petr Bour*
The detailed electronic structures of fluorescent chromophores
are important for their use in imaging of living cells. A series of
green fluorescent protein chromophore derivatives is exam-
ined by magnetic circular dichroism (MCD) spectroscopy,
which allows the resolution of more bands than plain absorp-
tion and fluorescence. Observed spectral patterns are rational-
ized with the aid of time-dependent density functional theory
(TDDFT) computations and the sum-over-state (SOS) formalism,
which also reveals a significant dependence of MCD intensities
on chromophore conformation. The combination of organic
and theoretical chemistry with spectroscopic techniques also
appears useful in the rational design of fluorescence labels and
understanding of the chromophore’s properties. For example,
the absorption threshold can be heavily affected by substitu-
tion on the phenyl ring but not much on the five-member
ring, and methoxy groups can be used to further tune the
electronic levels.
1. Introduction
Fluorescent proteins (FPs) are fantastic tools to visualize the
cellular structure and processes in living organisms. They have
been extensively studied and have recently been used to di-
rectly influence metabolic pathways.[1] More than a decade
after discovery of the first green FP[2] the importance of the
structure to the chromophore has been recognized.[3] Since
then, numerous FP variants have been found, differing in both
the protein and chromophore components.[4] Although the
structure of the chromophore varies,[4,5] a five-member imida-
zolinone ring must be present as a core motif. The imidazoli-
none connected through the ꢀC= bridge to another residue
with conjugated electrons thus seems to be crucial for all fluo-
rescence applications.
In the present study we test how chemical modification can
influence spectral properties and the electronic structure of
a series of FP chromophore derivatives, and determine how re-
producible these changes are by advanced quantum chemical
computations. The fluorescence, absorption and magnetic cir-
cular dichroism (MCD) spectra of FP chromophore derivatives
were measured and compared with predictions based on the
time-dependent density functional theory (TDDFT). The ab-
sorption and MCD spectra are known to be closely related to
the fluorescent properties.[4,6]
MCD spectroscopy[7] has been useful in assigning and resolv-
ing the electronic transitions of a wide range of organic and in-
organic molecules. MCD is a differential absorption of the left-
and right-circularly polarized light in a permanent magnetic
field parallel to the light propagation. It can be both negative
and positive and provides additional information on the ab-
sorption and fluorescence properties. Planar molecules with
conjugated electrons do not exhibit natural circular dichroism
but may still have a large MCD signal. As of writing, MCD spec-
troscopy has not been applied to the examinations of FP chro-
mophores yet.
Modern quantum MCD theory appeared in the 1970s,[8] but
only the recent implementations within DFT[9] are applicable to
sizable systems. The present study thus also documents how
theoretical simulations significantly enhance the MCD potential
to provide insights into structural and chemical problems.[9f,10]
Several computational approaches have been applied to un-
derstand the FP chromophore’s excited electronic states in the
past.[11] Recently, the method for reduced virtual-space ap-
proach for approximating second-order coupled clusters (RVS-
CC2)[12] were applied to investigate changes in green FP chro-
mophore at different ionic states.[13] The importance of the po-
larization effect for excitation energies were emphasized in
Fluorescence properties are determined not only by the
chemical structure of the chromophore but also by the envi-
ronmental conditions of the protein including pH. Mutations
to the green FP have created mutant strains to provide varying
emission wavelengths, enhanced absorption cross-sections and
protein folds.[1,5] Due to the complexity in identifying specific
modifications it is currently difficult to design fluorescent
labels of specific properties by only using means of synthetic
and computational chemistry.
ˇ
ˇ
ˇ
ˇ
[a] T. Y. Cowie, Dr. M. Safarꢀk, Dr. J. Sebestꢀk, Dr. R. Pohl, Prof. P. Bour
Institute of Organic Chemistry and Biochemistry, AS CR
ˇ
Flemingovo nꢁmestꢀ 2, 166 10 Prague (Czech Republic)
E-mail: bour@uochb.cas.cz
ˇ
ˇ
[b] P. Stepꢁnek
NMR Research Group
Faculty of Science, University of Oulu
PO Box 3000, 90014 Oulu (Finland)
Supporting Information for this article can be found under http://
dx.doi.org/10.1002/cphc.201600313.
ChemPhysChem 2016, 17, 1 – 8
1
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Articles
a complex study using time-dependent density functional
theory (TDDFT), multiconfigurational perturbation theory
(CASPT2), and quantum Monte Carlo (QMC) methods.[14] The
polarization effect was confirmed by works which combined
TDDFT with polarizable embedding (PE) of the FP chromo-
phore.[15] The PE approximation was thoroughly tested and
found very suitable for fast estimations of the environmental
effects on the chromophore.[16] Effects of chemical modifica-
tions to the green FP chromophore were studied using TDDFT
and molecular dynamics (MD),[17] and intermolecular charge
transfer in yellow FP was estimated using TDDFT and CC2 ap-
proaches.[18] Ground- and excited-states vibrations and dielec-
tric response to photoexcitation were estimated at the com-
plete active space self-consistent field (CASSCF) level.[19] Exten-
sive benchmarking of the quantum chemical methods was
done on several simplified FP chromophores.[20]
has been successfully applied for fullerene derivatives.[10b,23] We
find it efficient with respect to the required computer time, al-
though efficient implementations of other procedures, such as
the complex polarization propagator (CPP) methods,[9b] have
also been reported and tested.[24] As shown below the SOS cal-
culations reproduce the main trends observed in the experi-
mental spectra.
2. Results and Discussion
Experimental and calculated absorption spectra of the chromo-
phore derivatives are plotted in Figure 2. For Ya, only the cal-
culated spectra are shown as the synthesis was not successful.
We use TDDFT to calculate excited electronic states of
a series of synthetic FP chromophore derivatives (Figure 1); the
method provides reasonably accurate results with acceptable
computational cost.[11,20a,b,21] The sum-over-state methodology
(SOS)[9 h,22] is used for simulations of MCD intensities, which
Figure 2. Experimental (black) and calculated (red, obtained as Boltzmann
averages of individual conformers; absorption lines in arbitrary units are in-
dicated as well) absorption spectra of model chromophores.
The measured wavenumber range was limited by a strong ab-
sorption of the chloroform solvent below ~240 nm. The spec-
tra confirm the stability of the electronic structure of the FP
chromophore skeleton; they are similar with a dominant high-
wavelength absorption band that is separated by about 70 nm
(~0.6 eV) from the others. The experimental spectral patterns
are consistent with previous observations on other FP chromo-
phore compounds.[11,17,20c,d,25] In the experiment, the broad lon-
gest-wavelength bands are often nearly split or exhibit fine
Figure 1. Investigated compounds, precursors (Xn) and variously substituted
native FP chromophores (Yn).
&
ChemPhysChem 2016, 17, 1 – 8
www.chemphyschem.org
2
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Articles
structure, most probably due to conformer equilibria and elec-
tronic–vibrational interactions.[19a,26]
Within the Y-series the least substituted Yd derivative exhib-
its the lowest-wavelength absorption threshold (~400 nm) and
the strongest experimental absorption at 355 nm. The me-
thoxy substitution in Ye has caused a red-shift of 18 nm and
with the dimethoxy derivative Yf the maximum has been shift-
ed by 6 nm further right to 379 nm. Replacement of the
methyl group by phenyl on the imidazolinone ring (Yd!
Yb,Yc) causes much larger shifts of about 50 nm from the
maximum. The calculations reveal that introduction of the
second benzene ring in Ya–Yb results in additional transitions
in the lower-wavelength region (240–350 nm) without a signifi-
cant change in the overall absorption. The Yd–Yf derivatives
have three transitions within 240-500 nm only. Computed ab-
sorption intensities appear mostly larger than the observed
ones, by ~20-200%, and the transitions wavelength are sys-
tematically higher, otherwise the computation reproduces the
main experimental trends. Note that accuracy of the experi-
mental intensity may vary as well because of limited solubility
and gradual evaporation of the chloroform solvent during the
relatively long MCD accumulation times.
In the precursor series (Xa–Xf) the same absorption pattern,
namely, a large relatively separate high-wavelength band, and
less intense transitions below 350 nm, is conserved. The nitro-
gen–oxygen exchange in the imidazolinone ring (Yd!Xd)
causes a large absorption increase (by about 50%) and a blue-
shift of the principle band from 355 to 360 nm (~0.2 eV) in ex-
periment, which is well reproduced by the calculations. Further
substitutions by the phenyl and methoxy groups cause similar
effects as for the nitrogen derivatives Ya–Yf.
Figure 3. Experimental (black) and calculated (red) MCD spectra of studied
chromophores, corresponding to the absorption in Figure 2. Experimental
intensities of Xa, Xb, Xc, Yb and Yc were multiplied by two.
Corresponding calculated and experimental MCD spectra are
plotted in Figure 3. As expected MCD intensities are more vari-
able than the absorption because they are more sensitive to
differences in the geometry and electronic structure.[9c,9 g,10b,27]
However, a general pattern can be observed for all com-
pounds, too. Typical is a negative MCD peak at the longest-
wavelength region (>300 nm), and stronger multiple bands
around 250 nm. Experimental MCD spectra of the lowest-
energy transition are hampered by a large noise caused by
small CD/absorption ratio. This highest-wavelength MCD signal
is in general more developed in the oxazolones Xa–Xf; the Xe
derivative provides the largest MCD intensity of
ꢀ0.25 Lcmꢀ1 molꢀ1 Tꢀ1 at around 356 nm.
applicable to a wider class of compounds than the more
common B3LYP.[29] However, for our compounds, none pro-
vides convincingly better agreement with the experiment than
the other. The B3LYP band wavelengths are slightly higher and
the M062X ones lower than in experiment. Another long-range
corrected functional, CAM-B3LYP,[30] provided similar results as
M062X, and the pure GGA B86[31] functional gave rather unreal-
istic spectra (Figure 2 SB). Such a limited accuracy of DFT is
consistent with previous works,[17,20b,21] and future improve-
ment may be difficult because of the complex interplay be-
tween the functional parameterization, solvent models, vibra-
tional effects, and so forth.[32]
The computations allow us to assign most of the observed
MCD spectral features to particular electronic transitions al-
though there are several inconsistencies explicable by the lim-
ited accuracy of the simulations. In Xd, however, the positive
experimental signal at 333 nm is not reproduced at all, and the
correspondence between the simulated and MCD intensities
within 250–300 nm for Yb and Yc is not obvious.
To investigate more limitations of the theoretical approach,
in Figure S1 MCD spectra are plotted as recalculated by alter-
nate SOS-gradient method.[9h] This, however, provided similar
intensities to the default SOS-length formula [Eq. (1) in the Ex-
perimental Section]. Larger discrepancies appeared between
the B3LYP[28] and M062X[29] functional (Figure S2A). The latter is
The dependence of the MCD spectra on the conformation is
documented for all stable conformers of Xc (listed in Table S1)
in Figure 4. Such sensitivity is rather surprising. For example, in
our previous study of aromatic aminoacids the effect of the
conformation on MCD intensities was rather limited.[33] For the
FP chromophores, however, relevant excitations involve elec-
tron rearrangement across the rotating bonds and proper con-
formational weighting is needed to reproduce the experimen-
tal data. The largest conformation-related changes occur
within 200–300 nm, while the highest-wavelength band
(~420 nm) is negative for all twelve conformers.
ChemPhysChem 2016, 17, 1 – 8
www.chemphyschem.org
3
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Articles
Figure 4. Calculated MCD spectra of stable Xc conformers.
Table 1. Energies of the lowest-energy absorption (calculated, EA,cal, and
experimental, EA,exp) and fluorescence bands (EF,cal, EF,exp), in eV.
EA,cal
EA,exp
EF,cal
EF,exp
ES,cal
ES,exp
EF,cal
EF,exp
Xa
Xb
Xc
Xd
Xe
Xf
3.14
2.97
2.81
3.57
3.35
3.18
3.40
3.20
3.11
3.75
3.44
3.31
2.67
2.59
2.53
3.07
3.01
2.83
3.02
–
2.68
3.15
2.92
2.81
Ya
Yb
Yc
Yd
Ye
Yf
3.09
2.98
2.96
3.38
3.27
3.25
–
2.47
2.42
2.41
2.82
2.82
2.80
–
3.15
3.05
3.49
3.33
3.27
2.67
2.64
3.15
2.84
2.79
cence maxima are compared to the computations in Table 1.
As expected, the experimental lowest-energy absorption and
fluorescence peak positions differ by the relaxation energy,
about 0.4–0.6 eV, which is well reproduced by the calculations.
The relaxation of the first excited singlet state leads to minor
changes in the chromophore geometries, which can be seen in
Figure S3. However, fluorescence intensity/efficiency signifi-
cantly varied among the derivatives, analysis of which goes
Figure 5. Experimental fluorescence (solid line) and absorption (dashed)
spectra, normalized to the maximum=1.
Most compounds also exhibited measurable fluorescence
and the spectra are plotted together with the normalized ab-
sorption in Figure 5. Energies of the absorption and fluores-
&
ChemPhysChem 2016, 17, 1 – 8
www.chemphyschem.org
4
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Articles
Table 2. Calculated wavelengths and principal orbital contributions to the lowest-energy transitions in lowest-energy cis and trans conformers (cf. angle
f1, Table S1) of Xd and Yd.
l [nm]
cis-Xd
l [nm]
cis-Yd
347
310
HOMO!LUMO (100%)
HOMOꢀ1!LUMO (94%)
367
302
289
HOMO!LUMO (98%)
HOMOꢀ1!LUMO (91%)
HOMOꢀ3!LUMO (93%)
trans-Xd
trans-Yd
352
314
HOMO!LUMO (100%)
HOMOꢀ1!LUMO (94%)
371
304
288
HOMO!LUMO (97%)
HOMOꢀ1!LUMO (91%)
HOMOꢀ3!LUMO (94%)
beyond the scope of the present study. The variations are con-
sistent with the strong dependence of the FP fluorescence on
the local environment of the chromophore.[4]
computations reproduced the main trends in the excitation
electronic energies, absorption, and MCD spectral intensities
that were observed experimentally.
The transition assignment is rather straightforward and anal-
ogous for all compounds. As an example the lowest-energy
transitions in Xd and Yd is summarized in Table 2. Also the
HOMO and LUMO orbitals are in different compounds remarka-
bly similar; in Figure 6 the orbitals are plotted for Xc–Xe only.
The lowest-energy transitions are thus almost exclusively
caused by the HOMO!LUMO excitation, with the second
lowest transition being predominantly HOMOꢀ1!LUMO, and
the character of the transitions is not significantly changed by
the conformation of the substituents. Perhaps this simplicity
and stability of the electronic structure contributes to the uni-
versal applicability of the fluorescent labels based on the FP
chromophores.
Experimental Section
Organic Synthesis
The synthesis was based on strategy of Kojima et al.[25] using oxa-
zolones precursors (Scheme 1). The precursors containing oxygen
in the five-membered ring were converted to FP chromophores
using condensation with methylamine. Purity of the products was
controlled using HPLC, mass spectroscopy and NMR. Synthetic de-
tails can be found in the Supporting Information.
Scheme 1. General synthesis scheme.
Spectra Measurement
MCD and absorption spectra were acquired on a Jasco J-815
(Japan) spectrometer equipped with 1.5 T permanent magnet. The
samples were dissolved in CHCl3 freshly purified by alumina and
measured in quartz cell of 1 mm optical path, at room temperature
(295 K), using a scanning speed of 5 nmminꢀ1, and concentrations
within 0.075–1.4 mgmLꢀ1 according to optimal absorption with re-
spect to MCD measurement. Fluorescence spectra were measured
on a Horiba Fluoromax-4 spectrometer for solutions in ethanol;
chloroform as a solvent provided negligible fluorescence intensities
only. Unlike the intensities, fluorescence peak positions measured
in ethanol and chloroform are expected to be similar.[34] Excitation
wavelengths listed in the Supporting Information were chosen to
maximize the signal.
Figure 6. Frontier orbitals of Xc–Xe.
3. Conclusions
The synthetic series of derivatives of the fluorescent protein
chromophore helped us to better understand the relation be-
tween the structure and electronic properties, important for ra-
tional design of fluorescent labels. MCD spectroscopy proved
useful both as a tool for chromophore discrimination and as-
signment of observed electronic transitions. Surprising was
also the dependence of the MCD pattern on the conformation,
possibly usable in future structural studies. The TDDFT/SOS
Spectra Modeling
Equilibrium geometries and excited electronic states were obtained
using the Gaussian 09 program.[35] The standard 6–311+ +G**
ChemPhysChem 2016, 17, 1 – 8
www.chemphyschem.org
5
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Articles
[6] M. F. Garcia-Parajo, G. M. J. Segers-Nolten, J. A. Veerman, J. Greve, N. F.
van Hulst, Proc. Natl. Acad. Sci. USA 2000, 97, 7237–7242.
[7] W. R. Mason, A Practical Guide to Magnetic Circular Dichroism Spectrosco-
py, Wiley-Interscience, New York, 2007.
[8] a) P. J. Stephens, J. Chem. Phys. 1970, 52, 3489–3516; b) P. J. Stephens,
Annu. Rev. Phys. Chem. 1974, 25, 201–232.
[9] a) H. Solheim, K. Ruud, S. Coriani, P. Norman, J. Phys. Chem. A 2008, 112,
9615–9618; b) H. Solheim, K. Ruud, S. Coriani, P. Norman, J. Chem. Phys.
2008, 128, 094103; c) T. Kjaergaard, S. Coriani, K. Ruud, Wiley Interdiscip.
Rev. Comput. Mol. Sci. 2012, 2, 443–455; d) M. Seth, M. Krykunov, T. Zie-
gler, J. Autschbach, A. Banerjee, J. Chem. Phys. 2008, 128, 144105; e) M.
Seth, M. Krykunov, T. Ziegler, J. Autschbach, J. Chem. Phys. 2008, 128,
234102; f) M. Seth, J. Autschbach, T. Ziegler, J. Chem. Theory Comput.
basis set and Gaussian version (CPCM) of the conductor-like sol-
vent model (COSMO)[36] were used, previously giving reasonable re-
sults for similar system.[33] Geometry optimizations were performed
at the B3LYP[28]/6–311+ +G** level of approximation. Excited
states were calculated using time-dependent DFT (TDDFT)[37] and
the B3LYP functional. In some cases M062X BP86 and CAM-B3LYP
functionals were tried as well. 400 lowest-energy excited states ob-
tained by Gaussian were used for sum-over-states (SOS)[9 h,22] simu-
lations of MCD intensities using our own software (“GUVCDE”). The
origin-independent SOS expression[22] was used by default where
the Faraday “B-term” determining the MCD intensities for an n!j
transition is equal to [Eq. (1)]
2007, 3, 434–447; g) M. Krykunov, M. Seth, T. Ziegler, J. Autschbach, J.
(
"
!#
ꢀ
ꢀ
ꢀ
ꢀ
ꢁ
ꢂ
P
ˇ
Ne
ˇ
ˇ
Chem. Phys. 2007, 127, 244102; h) P. Stepꢀnek, P. Bour, J. Comput. Chem.
2013, 34, 1531–1539; i) K. M. Lee, K. Yabana, G. F. Bertsch, J. Chem. Phys.
2011, 134, 144106.
X
X
X
mlk ꢁ rjl
k
i¼1 ri ꢁ ri n
1
2
1
2Ne
mkl ꢁ rln
mln ꢁ rkl
B ¼
þ
þ
ꢂ mnj ꢁ mjkþ
)
Ek0 n
El0n
Ek0 l
k¼n
l¼n
l¼k
"
!#
ꢀ
ꢀ
ꢀ
ꢁ
ꢂ
P
Ne
ꢀ
X
X
X
j
i¼1 ri ꢁ ri k
1
mjl ꢁ rlk
ˇ
[10] a) V. Andrushchenko, D. Padula, E. Zhivotova, S. Yamamoto, P. Bour,
þ
þ
ꢂ mnj ꢁ mkn
Ek0 j
Ek0 l
El0j
ˇ
2Ne
ˇ
Chirality 2014, 26, 655–662; b) P. Stepꢀnek, M. Straka, V. Andrushchen-
k¼j
l¼k
l¼j
ˇ
ko, P. Bour, J. Chem. Phys. 2013, 138, 151103; c) E. I. Solomon, K. M.
ð1Þ
Light, L. V. Liu, M. Srnec, S. D. Wong, Acc. Chem. Res. 2013, 46, 2725–
2739; d) N. Lin, H. Solheim, X. Zhao, F. Santoro, K. Ruud, J. Chem. Theory
ˇ
ˇ
Comput. 2013, 9, 1557–1567; e) P. Stepꢀnek, V. Andrushchenko, K.
where n denotes the ground state, j, k, and l are excited states
modeled by the TDDFT single excitations, i is used for the electron
index, Ne is the number of electrons, ri is position of an electron i,
mab ¼ hajmjbidenotes electric dipole matrix element, Eab =EaꢀEb is
the difference in excitation (not orbital) energies. The apostrophe
means that the energy in the denominator was replaced by a “de-
generacy-resistant” expression, as shown by Equation (2):
ˇ
Ruud, P. Bour, J. Phys. Chem. A 2012, 116, 778–783; f) M. Seth, T. Ziegler
in Advances in Inorganic Chemistry Vol. 62 (Eds.: R. VanEldik, J. Harley),
Elsevier Academic Press, Inc., San Diego, 2010, pp. 41–109; g) E. Her-
nandez-Marin, M. Seth, T. Ziegler, Inorg. Chem. 2010, 49, 6066–6076;
h) G. A. Peralta, M. Seth, H. Zhekova, T. Ziegler, Inorg. Chem. 2008, 47,
4185–4198; i) M. Seth, T. Ziegler, J. Autschbach, J. Chem. Phys. 2005,
122, 094112.
[11] A. V. Nemukhin, I. A. Topol, S. K. Burt, J. Chem. Theory Comput. 2006, 2,
292–299.
[12] R. Send, V. R. I. Kaila, D. Sundholm, J. Chem. Phys. 2011, 134, 214114.
[13] V. R. I. Kaila, R. Send, D. Sundholm, Phys. Chem. Chem. Phys. 2013, 15,
4491–4495.
1
Ea0 b
Eab
Ea2b þ G2
¼
ð2Þ
where G=0.02 hartree. Resultant absorption and MCD spectra
were simulated as a Boltzmann average of individual conformers,
using Lorenzian bandwidth of full width at half height of 0.2 eV. In
Figure 4 and S2B 15 nm width was used to emphasize individual
transitions. Fluorescence energies were computed as the vertical
excitation energies for the lowest-energy conformers optimized in
the first electronic excited state.
[14] C. Filippi, F. Buda, L. Guidoni, A. Sinicropi, J. Chem. Theory Comput.
2012, 8, 112–124.
[15] M. T. P. Beerepoot, A. H. Steindal, K. Ruud, J. M. H. Olsen, Comput. Theor.
Chem. 2014, 1040, 304–311.
[16] T. Schwabe, M. T. P. Beerepoot, J. M. H. Olsen, J. Kongsted, Phys. Chem.
Chem. Phys. 2015, 17, 2582–2588.
[17] M. T. P. Beerepoot, A. H. S. Steindal, J. Kongsted, B. O. Brandsdal, L. Fre-
diani, K. Ruud, J. M. H. Olsen, Phys. Chem. Chem. Phys. 2013, 15, 4735–
4743.
[18] M. T. P. Beerepoot, D. H. Friese, K. Ruud, Phys. Chem. Chem. Phys. 2014,
16, 5958–5964.
Acknowledgements
[19] a) R. Gnanasekaran, Chem. Phys. Lett. 2013, 587, 61–67; b) Y. Xu, R. Gna-
nasekaran, D. M. Leitner, Chem. Phys. Lett. 2013, 564, 78–82.
[20] a) R. Nifosꢁ, P. Amat, V. Tozzini, J. Comput. Chem. 2007, 28, 2366–2377;
b) E. Epifanovsky, I. Polyakov, B. Grigorenko, A. Nemukhin, A. I. Krylov, J.
Chem. Theory Comput. 2009, 5, 1895–1906; c) C. Filippi, M. Zaccheddu,
F. Buda, J. Chem. Theory Comput. 2009, 5, 2074–2087; d) H. M. List, J. M.
Olsen, T. R. Rinza, O. Christiansen, J. Kongsted, Int. J. Quantum Chem.
2012, 112, 789–800.
The work was supported by the Grant Agency of the Czech Re-
public (grant numbers 13–03978S and 16–059355). We thank Dr.
D. Padula for the help with the calculations.
Keywords: density functional calculations
·
fluorescence
protein chromophores · magnetic circular dichroism · organic
synthesis · spectral simulations
[21] M. A. L. Marques, X. Lꢃpez, D. Varsano, A. Castro, A. Rubio, Phys. Rev.
Lett. 2003, 90, 258101.
ˇ
ˇ
ˇ
[22] P. Stepꢀnek, P. Bour, J. Comput. Chem. 2015, 36, 723–730.
ˇ
ˇ
ˇ
ˇ
[23] P. Stepꢀnek, M. Straka, J. Sebestꢁk, P. Bour, Chem. Phys. Lett. 2016, 647,
117–121.
[1] C. Brieke, F. Rohrbach, A. Gottschalk, G. Mayer, A. Heckel, Angew. Chem.
Int. Ed. 2012, 51, 8446–8476; Angew. Chem. 2012, 124, 8572–8604.
[2] O. Shimomura, F. H. Johnson, Y. Saiga, J. Cell. Comp. Physiol. 1962, 59,
223–239.
[24] T. Fahleson, J. Kauczor, P. Norman, F. Santoro, R. Improta, S. Coriani, J.
Phys. Chem. A 2015, 119, 5476–5489.
[25] S. Kojima, H. Ohkawa, T. Hirano, S. Maki, H. Niwa, M. Ohashi, S. Inouye,
F. I. Tsuji, Tetrahedron Lett. 1998, 39, 5239–5242.
[26] D. Padula, D. Picconi, A. Lami, G. Pescitelli, F. Santoro, J. Phys. Chem. A
2013, 117, 3355–3368.
[3] O. Shimomura, FEBS Lett. 1979, 104, 220–222.
[4] D. M. Chudakov, M. V. Matz, S. Lukyanov, K. A. Lukyanov, Physiol. Rev.
2010, 90, 1103–1163.
[5] a) A. Follenius-Wund, M. Bourotte, M. Schmitt, F. Iyice, L. Hans, J. J. Bour-
guignon, J. Haiech, C. Pigault, Biophys. J. 2003, 85, 1839–1850; b) A. B.
Cubitt, R. Heim, S. R. Adams, A. E. Boyd, L. A. Gross, R. Y. Tsien, Trends
Biochem. Sci. 1995, 20, 448–455; c) B. P. Cormack, R. H. Valdivia, S.
Falkow, Gene 1996, 173, 33–38; d) S. J. Remington, Methods Enzymol.
2000, 305, 196–211.
[27] J. Michl, Tetrahedron 1984, 40, 3845–3934.
[28] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[29] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215–241.
[30] T. Yanai, D. Tew, N. C. Handy, Chem. Phys. Lett. 2004, 393, 51–57.
[31] J. P. Perdew, Phys. Rev. B 1986, 33, 8822–8824.
&
ChemPhysChem 2016, 17, 1 – 8
www.chemphyschem.org
6
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Articles
[32] A. D. Laurent, D. Jacquemin, Int. J. Quantum Chem. 2013, 113, 2019–
mann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dan-
nenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz,
J. Cioslowski, D. J. Fox, , Gaussian, Inc., Wallingford CT, 2009.
2039.
ˇ
ˇ
ˇ
[33] P. Stepꢀnek, P. Bour, Phys. Chem. Chem. Phys. 2014, 16, 20639–20649.
[34] J. Dong, K. M. Solntsev, L. M. Tolbert, J. Am. Chem. Soc. 2006, 128,
12038–12039.
[36] A. Klamt in The Encyclopedia of Computational Chemistry, Vol. 1 (Eds.:
P. R. Schleyer, N. L. Allinger, T. Clark, J. Gasteiger, P. A. Kollman, H. F.
[35] Gaussian 09 (Revision D.01), M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.
Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratch-
ian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M.
Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, T. Vreven, J. Montgomery, J. A., J. E. Peralta, F.
Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov,
R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Strat-
Schaefer III, P. R. Schreiner), John Wiley
pp. 604–615.
& Sons, Chichester, 1998,
[37] F. Furche, R. Ahlrichs, J. Chem. Phys. 2002, 117, 7433–7447.
Manuscript received: March 27, 2016
Accepted Article published: April 28, 2016
Final Article published: && &&, 2016
ChemPhysChem 2016, 17, 1 – 8
www.chemphyschem.org
7
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

ARTICLES
ˇ
ˇ
ˇ
ˇ
P. Stepꢁnek, T. Y. Cowie, M. Safarꢀk,
Better understanding: Magnetic circu-
lar dichroism for a series of fluorescent
protein chromophore derivatives pro-
vides specific spectral patterns, and in
connection with density functional
theory simulations it can be used for ra-
tional design of fluorescent probes.
ˇ
ˇ
J. Sebestꢀk, R. Pohl, P. Bour*
&& – &&
Resolving Electronic Transitions in
Synthetic Fluorescent Protein
Chromophores by Magnetic Circular
Dichroism
&
ChemPhysChem 2016, 17, 1 – 8
www.chemphyschem.org
8
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!