![Chromophore-binding pocket of BV-and PCB-binding phytochromes: Agp1 PGP in Pr state, PDB 5I5L [18] and Cph1 in Pr state, PDB 2VEA [19].](https://www.researchgate.net/publication/365936768/figure/fig1/AS:11431281104222898@1669976864814/Chromophore-binding-pocket-of-BV-and-PCB-binding-phytochromes-Agp1-PGP-in-Pr-state-PDB_Q320.jpg)
Available via license: CC BY 4.0
Content may be subject to copyright.
Citation: Sadeghi, M.; Balke, J.;
Rafaluk-Mohr, T.; Alexiev, U. Long-
Distance Protonation-Conformation
Coupling in Phytochrome Species.
Molecules 2022,27, 8395. https://
doi.org/10.3390/molecules27238395
Academic Editor: Carlos Alemán
Received: 31 October 2022
Accepted: 27 November 2022
Published: 1 December 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
molecules
Article
Long-Distance Protonation-Conformation Coupling in
Phytochrome Species
Maryam Sadeghi, Jens Balke , Timm Rafaluk-Mohr and Ulrike Alexiev *
Department of Physics, Freie Universität Berlin, Arnimallee 14, 14195 Berlin, Germany
*Correspondence: ulrike.alexiev@fu-berlin.de
Abstract:
Phytochromes are biological red/far-red light sensors found in many organisms. The
connection between photoconversion and the cellular output signal involves light-mediated global
structural changes in the interaction between the photosensory module (PAS-GAF-PHY, PGP) and
the C-terminal transmitter (output) module. We recently showed a direct correlation of chromophore
deprotonation with pH-dependent conformational changes in the various domains of the proto-
typical phytochrome Cph1 PGP. These results suggested that the transient phycocyanobilin (PCB)
chromophore deprotonation is closely associated with a higher protein mobility both in proximal and
distal protein sites, implying a causal relationship that might be important for the global large-scale
protein rearrangements. Here, we investigate the prototypical biliverdin (BV)-binding phytochrome
Agp1. The structural changes at various positions in Agp1 PGP were investigated as a function of pH
using picosecond time-resolved fluorescence anisotropy and site-directed fluorescence labeling of
cysteine variants of Agp1 PGP. We show that the direct correlation of chromophore deprotonation
with pH-dependent conformational changes does not occur in Agp1. Together with the absence of
long-range effects between the PHY domain and chromophore pK
a
, in contrast to the findings in
Cph1, our results imply phytochrome species-specific correlations between transient chromophore
deprotonation and intramolecular signal transduction.
Keywords:
phytochrome; Agp1; Cph1; biliverdin; chromophore protonation; conformational coupling
1. Introduction
Phytochromes are photoreceptor proteins and belong to the class of bilin-containing
proteins. Phytochromes control many developmental processes in plants such as seed
germination, de-etiolation, or flowering [
1
]. The discovery of the bacterial phytochrome
Cph1 [
2
] indicates the prokaryotic origin of these photoreceptor proteins, which include
prototypical and bathy phytochromes. Prototypical phytochromes, including Cph1 from
the cyanobacterium Synechocystis 6803 [
2
] and Agp1 from the soil bacterium Agrobacterium
tumefaciens [
3
], act as photochemical switches that interconvert between stable red (Pr)-
and metastable far-red (Pfr)-absorbing states. This interconversion is induced by photoi-
somerization of the bilin chromophore after light activation [
4
,
5
]. The bilin chromophore
deprotonates transiently during the Pr to Pfr photoconversion in association with exten-
sive global structural changes required for signal transmission [
6
]. Bathy phytochromes,
including Agp2 from Agrobacterium tumefaciens, interconvert between a stable Pfr state and
a metastable Pr state [7].
Despite numerous studies on the structure and function of phytochromes, fundamental
questions regarding the light-activation mechanism and its coupling to phytochrome
function remain. For example, it is not fully understood how the structural changes in Pr
and Pfr are triggered by the light-induced isomerization of the chromophore, although
structural differences between Pr and Pfr have been identified in bacteriophytochrome [
6
].
Recent results on ultrafast proton-coupled isomerization in the phototransformation of
phytochrome demonstrate how proton-coupled dynamics in the excited state of Pfr lead
Molecules 2022,27, 8395. https://doi.org/10.3390/molecules27238395 https://www.mdpi.com/journal/molecules
Molecules 2022,27, 8395 2 of 15
to a restructured hydrogen-bond environment in early Lumi-F, which is thought to be
a trigger for downstream protein structural changes [
4
,
5
,
8
]. Moreover, several studies on
the phytochrome have revealed that proton translocation plays a crucial role in the coupling
of chromophore and protein conformational changes [9,10].
In addition to structural and ultrafast spectroscopic studies, time-resolved fluorescence
anisotropy measurements in the picosecond to nanoseconds time range enable a variety
of experiments in which fluorescent dyes can be used to monitor the conformational
dynamics of proteins at a specific site [
11
–
15
]. In this respect, the fluorescence labeling of
phytochromes [
13
,
16
] provides an opportunity to study local conformational changes in
aqueous solution that would be otherwise hidden in crystal structure studies.
Using this fluorescence-based approach, we test here the hypothesis that chromophore
deprotonation is coupled to conformational changes in the different protein domains, i.e.,
whether long-range conformational changes exists and whether a long-range H-bonding
network between the chromophore-binding pocket and the PHY domain controls the chro-
mophore pK
a
in Agp1, as shown for Cph1 [
13
]. The canonical (prototypical) phytochrome
Agp1 differs from Cph1 in the bound bilin chromophore. While Cph1 and plant phy-
tochromes bind phycocyanobilin (PCB), Agp1 binds biliverdin (BV). The domain structure
of plant phytochromes, Cph1, Agp1 and some other phytochromes are shown in Figure 1
for comparison. The structure of the chromophore-binding pocket of the two phytochromes,
Cph1 and Agp1 with the PCB and BV chromophore, respectively, is shown in Figure 2.
Among others, the conserved tyrosines Y166 and Y253 in Agp1 and Y176 and Y263 in Cph1,
as well as the conserved salt bridge between D197 (GAF) and R462 (PHY) in Agp1 and
D207 and R472 in Cph1, whose cleavage is thought to be essential to interconvert from Pr
to Pfr, are indicated. Via this strictly conserved aspartic acid in the chromophore-binding
pocket, the chromophore connects to the PHY-tongue.
Molecules 2022, 27, x FOR PEER REVIEW 2 of 15
changes in Pr and Pfr are triggered by the light-induced isomerization of the chromo-
phore, although structural differences between Pr and Pfr have been identified in bacteri-
ophytochrome [6]. Recent results on ultrafast proton-coupled isomerization in the photo-
transformation of phytochrome demonstrate how proton-coupled dynamics in the excited
state of Pfr lead to a restructured hydrogen-bond environment in early Lumi-F, which is
thought to be a trigger for downstream protein structural changes [4,5,8]. Moreover, sev-
eral studies on the phytochrome have revealed that proton translocation plays a crucial
role in the coupling of chromophore and protein conformational changes [9,10].
In addition to structural and ultrafast spectroscopic studies, time-resolved fluores-
cence anisotropy measurements in the picosecond to nanoseconds time range enable a
variety of experiments in which fluorescent dyes can be used to monitor the conforma-
tional dynamics of proteins at a specific site [11–15]. In this respect, the fluorescence label-
ing of phytochromes [13,16] provides an opportunity to study local conformational
changes in aqueous solution that would be otherwise hidden in crystal structure studies.
Using this fluorescence-based approach, we test here the hypothesis that chromo-
phore deprotonation is coupled to conformational changes in the different protein do-
mains, i.e., whether long-range conformational changes exists and whether a long-range
H-bonding network between the chromophore-binding pocket and the PHY domain con-
trols the chromophore pKa in Agp1, as shown for Cph1 [13]. The canonical (prototypical)
phytochrome Agp1 differs from Cph1 in the bound bilin chromophore. While Cph1 and
plant phytochromes bind phycocyanobilin (PCB), Agp1 binds biliverdin (BV). The do-
main structure of plant phytochromes, Cph1, Agp1 and some other phytochromes are
shown in Figure 1 for comparison. The structure of the chromophore-binding pocket of
the two phytochromes, Cph1 and Agp1 with the PCB and BV chromophore, respectively,
is shown in Figure 2. Among others, the conserved tyrosines Y166 and Y253 in Agp1 and
Y176 and Y263 in Cph1, as well as the conserved salt bridge between D197 (GAF) and
R462 (PHY) in Agp1 and D207 and R472 in Cph1, whose cleavage is thought to be essential
to interconvert from Pr to Pfr, are indicated. Via this strictly conserved aspartic acid in the
chromophore-binding pocket, the chromophore connects to the PHY-tongue.
Figure 1. Protein domains in different phytochromes. The different photosensory domains PAS
(blue), GAF (magenta), PHY (green), and the output domain (histidine kinase (HK) or similar (STK-
HK: HK related domain; GGDEF-EAL: c-di-GMP turnover domain) (yellow) are shown according
to the N- to C-terminal phytochrome sequence. Almost all phytochromes hold only a single chro-
mophore-binding site at a conserved cysteine residue either at the PAS domain in the case of Agp1
and Agp2 or GAF domain in the case of plant phytochromes, Cph1, and Cph2.
Figure 1.
Protein domains in different phytochromes. The different photosensory domains PAS (blue),
GAF (magenta), PHY (green), and the output domain (histidine kinase (HK) or similar (STK-HK: HK
related domain; GGDEF-EAL: c-di-GMP turnover domain) (yellow) are shown according to the N-
to C-terminal phytochrome sequence. Almost all phytochromes hold only a single chromophore-
binding site at a conserved cysteine residue either at the PAS domain in the case of Agp1 and Agp2
or GAF domain in the case of plant phytochromes, Cph1, and Cph2.
By using chromophore titrations and picosecond time-resolved fluorescence anisotropy
measurements to observe structural dynamics in Agp1 at the site of fluorescence-sensor
attachment, we answer the following questions: Does the modification in the PHY domain
of Agp1 affect chromophore deprotonation, similar to what was observed in Cph1? Is
Molecules 2022,27, 8395 3 of 15
the long-distance protonation-conformation coupling, as observed for Cph1 between the
chromophore protonation state and the distant PHY domain, also present in Agp1 and
correlated to protonation heterogeneity in the chromophore-binding pocket [13,17]?
Molecules 2022, 27, x FOR PEER REVIEW 3 of 15
By using chromophore titrations and picosecond time-resolved fluorescence anisot-
ropy measurements to observe structural dynamics in Agp1 at the site of fluorescence-
sensor attachment, we answer the following questions: Does the modification in the PHY
domain of Agp1 affect chromophore deprotonation, similar to what was observed in
Cph1? Is the long-distance protonation-conformation coupling, as observed for Cph1 be-
tween the chromophore protonation state and the distant PHY domain, also present in
Agp1 and correlated to protonation heterogeneity in the chromophore-binding pocket
[13,17]?
We show that local conformational dynamics in the different photosensory domains
and its coupling to the chromophore behaves differently in PCB-binding Cph1 and BV-
binding Agp1 and is, thus, connected to a different long-range H-bonding network. This
indicates phytochrome species-specific correlations between transient chromophore
deprotonation and the photo-induced large global protein rearrangements and intramo-
lecular signal transduction.
Figure 2. Chromophore-binding pocket of BV- and PCB-binding phytochromes: Agp1 PGP in Pr
state, PDB 5I5L [18] and Cph1 in Pr state, PDB 2VEA [19].
2. Results
2.1. Protein Variants, Their Spectroscopic Characterization and Fluorescein Labeling of the
Photosensory Module
Agp1-PGP (wild type, WT), Agp1-PGP-C279S/C295S, Agp1-PGP-C279S, Agp1-PGP-
C295S, and Agp1-PGP-C279S/C295S/V364C (Figure 3A), were labeled with the pH-indi-
cator dye 5-iodoacetamidofluorescein (5-IAF) to about 80–110%, yielding WT-AF, C279-
AF, C295-AF, and V364C-AF. The corresponding UV–Vis absorbance spectra of the IAF-
labeled and unlabeled Agp1-PGP samples are shown in Figure 3B and were used for cal-
culating the labeling stoichiometry (LS) according to Materials and Methods. To test
whether the introduced single labeling site (C279, C205, or V364C) is the only accessible
labeling site for the fluorophore IAF in the respective Agp1 variant, we also performed
IAF-labeling for the variant Agp1-PGP-C279S/C295S, which only contains the cysteine in
position 20 for covalent binding of the BV chromophore.
Figure 2.
Chromophore-binding pocket of BV- and PCB-binding phytochromes: Agp1 PGP in Pr
state, PDB 5I5L [18] and Cph1 in Pr state, PDB 2VEA [19].
We show that local conformational dynamics in the different photosensory domains
and its coupling to the chromophore behaves differently in PCB-binding Cph1 and BV-
binding Agp1 and is, thus, connected to a different long-range H-bonding network. This
indicates phytochrome species-specific correlations between transient chromophore depro-
tonation and the photo-induced large global protein rearrangements and intramolecular
signal transduction.
2. Results
2.1. Protein Variants, Their Spectroscopic Characterization and Fluorescein Labeling of the
Photosensory Module
Agp1-PGP (wild type, WT), Agp1-PGP-C279S/C295S, Agp1-PGP-C279S, Agp1-PGP-
C295S, and Agp1-PGP-C279S/C295S/V364C (Figure 3A), were labeled with the pH-
indicator dye 5-iodoacetamidofluorescein (5-IAF) to about 80–110%, yielding WT-AF,
C279-AF, C295-AF, and V364C-AF. The corresponding UV–Vis absorbance spectra of the
IAF-labeled and unlabeled Agp1-PGP samples are shown in Figure 3B and were used for
calculating the labeling stoichiometry (LS) according to Materials and Methods. To test
whether the introduced single labeling site (C279, C205, or V364C) is the only accessible
labeling site for the fluorophore IAF in the respective Agp1 variant, we also performed
IAF-labeling for the variant Agp1-PGP-C279S/C295S, which only contains the cysteine in
position 20 for covalent binding of the BV chromophore.
Indeed, IAF-labeling of Agp1-PGP-C279S/C295S resulted in only residual labeling
of the protein, probably due to minor amounts of apoprotein in the sample. We further
tested whether this base mutant C279S/C295S changes the photochromicity of Agp1. The
UV
−
Vis absorption spectrum of Agp1-PGP C279S/C295S in Pr is identical to that of WT
showing the same maximum absorption wavelength (
λmax
) of 702 nm. The similarity also
holds true for the formation of the Pfr state after illumination, as shown in Figure 4. This
indicates that replacing the two cysteines in the GAF domain by serine does not affect the
chromophore-binding pocket.
Molecules 2022,27, 8395 4 of 15
Molecules 2022, 27, x FOR PEER REVIEW 4 of 15
Figure 3. Agp1 PGP structure, single accessible cysteine mutants, and fluorescein-labeling. (A)
Structural model of Agp1 PGP (PDB 5I5L) with accessible single cysteines for fluorescence labeling
in the indicated in red (C279 and C295 as cysteines in the native sequence, and V364C as a cysteine
introduced in the PHY domain), the PCB chromophore attached to Cys20 in black. To introduce
single accessible cysteine, the following mutants were constructed: C297S/C295, C295S/C279, and
C279S/C295S/V364C. The PAS, GAF, and PHY domains are colored in blue, pink, and green, respec-
tively. The figure was generated with VMD. (B) Pr absorption spectrum of Agp1 PGP and IAF-
labeled variants at pH 7.2, normalized to the chromophore peak at 702 nm; WT in black, WT-AF in
orange, V364C-AF in red, C295-AF in blue, and C297-AF in green. Labeling was performed as de-
scribed in Material and Methods and the labeling stoichiometry (LS) was calculated according to
Eq. 1, which yielded LS of 110%, 98%, 93%, and 80% for WT-AF, V364C-AF, C295-AF, and C297-AF,
respectively. In WT, the two cysteines C295 and C279 are partly accessible as evidenced by the LS
below 200%. The maximum absorption wavelength of the fluorescein peak in WT is indicated. Con-
ditions: 300 mM NaCl, 50 mM Tris, pH 7.2 at 20 °C .
Indeed, IAF-labeling of Agp1-PGP-C279S/C295S resulted in only residual labeling of
the protein, probably due to minor amounts of apoprotein in the sample. We further tested
whether this base mutant C279S/C295S changes the photochromicity of Agp1. The
UV−Vis absorption spectrum of Agp1-PGP C279S/C295S in Pr is identical to that of WT
showing the same maximum absorption wavelength (λmax) of 702 nm. The similarity also
holds true for the formation of the Pfr state after illumination, as shown in Figure 4. This
indicates that replacing the two cysteines in the GAF domain by serine does not affect the
chromophore-binding pocket.
Figure 4. Characterization of the accessible cysteine-less base variant Agp1 PGP C279S/C295S. The
UV–Vis absorption spectra in the Pr state (in red), after illumination in the Pr/Pfr state (in blue), and
the difference spectra (in grey) are shown. The maximum absorbance wavelength of the Pr state
Figure 3.
Agp1 PGP structure, single accessible cysteine mutants, and fluorescein-labeling. (
A
)
Structural model of Agp1 PGP (PDB 5I5L) with accessible single cysteines for fluorescence labeling
in the indicated in red (C279 and C295 as cysteines in the native sequence, and V364C as a cysteine
introduced in the PHY domain), the PCB chromophore attached to Cys20 in black. To introduce
single accessible cysteine, the following mutants were constructed: C297S/C295, C295S/C279,
and C279S/C295S/V364C. The PAS, GAF, and PHY domains are colored in blue, pink, and green,
respectively. The figure was generated with VMD. (
B
) Pr absorption spectrum of Agp1 PGP and
IAF-labeled variants at pH 7.2, normalized to the chromophore peak at 702 nm; WT in black, WT-AF
in orange, V364C-AF in red, C295-AF in blue, and C297-AF in green. Labeling was performed as
described in Section 4and the labeling stoichiometry (LS) was calculated according to Equation (1),
which yielded LS of 110%, 98%, 93%, and 80% for WT-AF, V364C-AF, C295-AF, and C297-AF,
respectively. In WT, the two cysteines C295 and C279 are partly accessible as evidenced by the
LS below 200%. The maximum absorption wavelength of the fluorescein peak in WT is indicated.
Conditions: 300 mM NaCl, 50 mM Tris, pH 7.2 at 20 ◦C.
Molecules 2022, 27, x FOR PEER REVIEW 4 of 15
Figure 3. Agp1 PGP structure, single accessible cysteine mutants, and fluorescein-labeling. (A)
Structural model of Agp1 PGP (PDB 5I5L) with accessible single cysteines for fluorescence labeling
in the indicated in red (C279 and C295 as cysteines in the native sequence, and V364C as a cysteine
introduced in the PHY domain), the PCB chromophore attached to Cys20 in black. To introduce
single accessible cysteine, the following mutants were constructed: C297S/C295, C295S/C279, and
C279S/C295S/V364C. The PAS, GAF, and PHY domains are colored in blue, pink, and green, respec-
tively. The figure was generated with VMD. (B) Pr absorption spectrum of Agp1 PGP and IAF-
labeled variants at pH 7.2, normalized to the chromophore peak at 702 nm; WT in black, WT-AF in
orange, V364C-AF in red, C295-AF in blue, and C297-AF in green. Labeling was performed as de-
scribed in Material and Methods and the labeling stoichiometry (LS) was calculated according to
Eq. 1, which yielded LS of 110%, 98%, 93%, and 80% for WT-AF, V364C-AF, C295-AF, and C297-AF,
respectively. In WT, the two cysteines C295 and C279 are partly accessible as evidenced by the LS
below 200%. The maximum absorption wavelength of the fluorescein peak in WT is indicated. Con-
ditions: 300 mM NaCl, 50 mM Tris, pH 7.2 at 20 °C .
Indeed, IAF-labeling of Agp1-PGP-C279S/C295S resulted in only residual labeling of
the protein, probably due to minor amounts of apoprotein in the sample. We further tested
whether this base mutant C279S/C295S changes the photochromicity of Agp1. The
UV−Vis absorption spectrum of Agp1-PGP C279S/C295S in Pr is identical to that of WT
showing the same maximum absorption wavelength (λmax) of 702 nm. The similarity also
holds true for the formation of the Pfr state after illumination, as shown in Figure 4. This
indicates that replacing the two cysteines in the GAF domain by serine does not affect the
chromophore-binding pocket.
Figure 4. Characterization of the accessible cysteine-less base variant Agp1 PGP C279S/C295S. The
UV–Vis absorption spectra in the Pr state (in red), after illumination in the Pr/Pfr state (in blue), and
the difference spectra (in grey) are shown. The maximum absorbance wavelength of the Pr state
Figure 4.
Characterization of the accessible cysteine-less base variant Agp1 PGP C279S/C295S. The
UV–Vis absorption spectra in the Pr state (in red), after illumination in the Pr/Pfr state (in blue), and
the difference spectra (in grey) are shown. The maximum absorbance wavelength of the Pr state peak
is given. (
A
) Agp1 PGP WT. (
B
) Agp1 PGP C279S/C295S. Conditions: 300 mM NaCl, 50 mM Tris,
pH 7.8 at 20 ◦C.
2.2. pH-Dependence of UV–Vis Absorption of Agp1-PGP and Its Variants
In recent studies [
13
,
17
], we determined the PCB chromophore pK
a
in Cph1-PGP and
fluorescein-labeled Cph1 variants. As observed for Cph1 [
13
], fluorescein-labeling at sites
distal to the bilin chromophore affected chromophore pK
a
substantially. Such a long-range
effect suggests the existence of a long-range H-bond network from the PHY domain to the
chromophore-binding pocket [
13
]. Here, we investigate the pH-dependence of the Agp1-
Molecules 2022,27, 8395 5 of 15
PGP absorption spectrum with and without labeling of the protein with fluorescein to test
whether fluorescein-labeling of the different Agp1 domains affects the chromophore pKa.
Absorption spectra of Agp1-PGP and its variants were measured at different pH
(Figure 5). Titration curves from the absorbance changes at nine distinct wavelengths were
used to calculate the pK
a
values in the Pr state. In agreement with earlier studies of Agp1,
the pH titration of Agp1-PGP WT yields only one pK
a
of 10.70
±
0.02. This pK
a
-value is
slightly lower than the value of 11.1 determined earlier for Agp1 [
20
], probably due to slight
variations in the protein purification or BV assembling protocols. In contrast to our findings
in the canonical phytochrome Cph1 [
13
], we found only minor changes in the chromophore
pK
a
between Agp1-PGP WT and its variants (Table 1). The maximum change in chro-
mophore pK
a
was observed between Agp1-PGP WT and Agp1-PGP C279S/C295S/V364C
with pK
a
=
−
0.2. The corresponding labeling site in Cph1-PGP (C371) yielded a down-shift
in chromophore pKaof pKa=−0.8 [13].
Table 1.
Comparison of pK
a
values of biliverdin chromophore deprotonation (pK
DPC
) of the different
AF-labeled and unlabeled Agp1-PGP variants from Figures 5and 6. The standard error is given (
±SD
).
Sample Chromophore pKDPC pKDPC a
Agp1 PGP WT 10.70 ±0.02 -
Agp1 PGP C295S 10.70 ±0.01 0 ±0.03
Agp1 PGP C279S 10.40 ±0.02 −0.3 ±0.03
Agp1 PGP C279S/C295S 10.51 ±0.02 −0.19 ±0.04
Agp1 PGP C279S/C295S/V364C 10.50 ±0.02 −0.20 ±0.04
Agp1 PGP WT-AF 10.42 ±0.02 −0.28 ±0.04
Agp1 PGP C295S/C279-AF 10.49 ±0.02 −0.21 ±0.04
Agp1 PGP C279S/C295-AF 10.45 ±0.03 −0.25 ±0.05
Agp1 PGP C279S/C295S/V364C-AF 10.60 ±0.03 −0.09 ±0.05
aKDPC values were calculated for the different constructs with respect to PGP WT. The standard error is given.
Next, we measured the absorption spectra of fluorescein-labeled Agp1-PGP and Agp1-
PGP variants at different pH values (Figure 6). Again, only a slight drop in the pK
a
values was determined compared to the unlabeled WT (Table 1). For V364C-AF with the
labeling position in the chromophore distant PHY domain, we calculated a change in the
chromophore pK
a
of only
−
0.1 pH units. In addition, we analyzed the pH-dependent
absorbance band of bound fluorescein and determined the respective pK
a
values of the
fluorescein at the different sites (Figure 6, bottom panel). The pK
a
values of the bound
fluorescein are upshifted compared to the unbound fluorophore (pK
a
= 6.5 [
21
]), indicating
a similar negative surface potential at the labeling sites. The
λmax
values of the fluorescein
peak in the different variants, however, indicate different polarities at the labeling sites.
The polarity was determined according to Alexiev et al. [
22
,
23
]. The fluorescein label in
position 364 (Figure 3) located in the
β
-sheets in the chromophore distant PHY domain
experiences the most hydrophilic environment with a
λmax
= 498 nm, while position 297
in the GAF domain close to the chromophore-binding pocket (Figure 3) shows the most
hydrophobic environment (λmax = 505 nm).
Molecules 2022,27, 8395 6 of 15
Molecules 2022, 27, x FOR PEER REVIEW 6 of 15
Figure 5. pH-dependent Pr absorption spectra of Agp1-PGP and its variants (top panel). (A) C295S,
(B) C279S, (C) C279S/C295S, (D) C279S/C295S/V364C. The direction of the pH titration is indicated
by the black arrow. The bottom panel shows the respective titration data at nine wavelengths 678 (
■), 683.5 (●), 689 (▲), 694.5 (▼), 700 (⧫), 705.5 (◄), 711 (►), 716.5 (⬣), and 722 nm (★). Selected fit
curves at 678, 700, and 711 nm are marked in red. The pH-dependence was fitted by the Hender-
son−Hasselbalch equation (Equation (2)) using Origin Pro 2019 (Origin Lab). The fit results are sum-
marized in Table 1. The absorption of the deprotonated chromophore at high pH was estimated to
20% of the maximum absorption of the respective wavelength. Conditions: 300 mM NaCl, 50 mM
Tris/HCl buffer for pH 6.2–9.5, or 100 mM Na2CO3/NaHCO3 buffer for pH 9.5–10.8, at 20 °C .
Next, we measured the absorption spectra of fluorescein-labeled Agp1-PGP and
Agp1-PGP variants at different pH values (Figure 6). Again, only a slight drop in the pKa
values was determined compared to the unlabeled WT (Table 1). For V364C-AF with the
labeling position in the chromophore distant PHY domain, we calculated a change in the
chromophore pKa of only −0.1 pH units. In addition, we analyzed the pH-dependent ab-
sorbance band of bound fluorescein and determined the respective pKa values of the flu-
orescein at the different sites (Figure 6, bottom panel). The pKa values of the bound fluo-
rescein are upshifted compared to the unbound fluorophore (pKa = 6.5 [21]), indicating a
similar negative surface potential at the labeling sites. The λmax values of the fluorescein
peak in the different variants, however, indicate different polarities at the labeling sites.
The polarity was determined according to Alexiev et al. [22,23]. The fluorescein label in
position 364 (Figure 3) located in the β-sheets in the chromophore distant PHY domain
experiences the most hydrophilic environment with a λmax = 498 nm, while position 297 in
the GAF domain close to the chromophore-binding pocket (Figure 3) shows the most hy-
drophobic environment (λmax = 505 nm).
Taken together, we observe a robust WT-like chromophore pKa in the Agp1 variants
investigated here, and upon modification of these variants with fluorescein in the GAF
and PHY domain. This result is in contrast to our observation in Cph1 where a similar
modification with fluorescein in the PHY domain at position C371 alters the chromophore
pKa by about −0.8 pH units, indicating a (direct or indirect) long-range H-bond network
from the chromophore to the PHY domain in Cph1 [13]. Thus, we conclude from our re-
sults here that a similar long-range interaction from the PHY domain β-sheets to the chro-
mophore is absent in Agp1. However, it is important to note that the chromophore pKa of
Agp1 is affected when the H-bond network within the chromophore-binding pocket is
disrupted as shown previously for the mutant D1976A, which is part of the conserved
Figure 5.
pH-dependent Pr absorption spectra of Agp1-PGP and its variants (top panel). (
A
)
C295S, (
B
) C279S, (
C
) C279S/C295S, (
D
) C279S/C295S/V364C. The direction of the pH titration
is indicated by the black arrow. The bottom panel shows the respective titration data at nine
wavelengths 678 (
), 683.5 (
•
), 689 (
N
), 694.5 (
H
), 700 (
␄
), 705.5 (
), 711 (
), 716.5 (
), and 722 nm
(
F
). Selected fit curves at 678, 700, and 711 nm are marked in red. The pH-dependence was fitted
by the Henderson
−
Hasselbalch equation (Equation (2)) using Origin Pro 2019 (Origin Lab). The fit
results are summarized in Table 1. The absorption of the deprotonated chromophore at high pH was
estimated to 20% of the maximum absorption of the respective wavelength. Conditions: 300 mM
NaCl, 50 mM Tris/HCl buffer for pH 6.2–9.5, or 100 mM Na
2
CO
3
/NaHCO
3
buffer for pH 9.5–10.8,
at 20 ◦C.
Taken together, we observe a robust WT-like chromophore pK
a
in the Agp1 variants
investigated here, and upon modification of these variants with fluorescein in the GAF and
PHY domain. This result is in contrast to our observation in Cph1 where a similar modifi-
cation with fluorescein in the PHY domain at position C371 alters the chromophore pK
a
by
about
−
0.8 pH units, indicating a (direct or indirect) long-range H-bond network from the
chromophore to the PHY domain in Cph1 [
13
]. Thus, we conclude from our results here
that a similar long-range interaction from the PHY domain
β
-sheets to the chromophore
is absent in Agp1. However, it is important to note that the chromophore pK
a
of Agp1 is
affected when the H-bond network within the chromophore-binding pocket is disrupted
as shown previously for the mutant D1976A, which is part of the conserved Asp-Arg salt
bridge (D197-R462 in Agp1, D207-R472 in Cph1) that connects the chromophore-binding
pocket with the tongue (Figure 2). A drastic downshift in chromophore pK
a
to about 7.6
was observed for this mutant [
20
]. This salt bridge and the hydrogen-bonding interaction of
the aspartate were shown in simulations to affect the stabilization of the early/late Lumi-R
intermediates featuring a more disordered chromophore-binding pocket [8].
Molecules 2022,27, 8395 7 of 15
Molecules 2022, 27, x FOR PEER REVIEW 7 of 15
Asp-Arg salt bridge (D197-R462 in Agp1, D207-R472 in Cph1) that connects the chromo-
phore-binding pocket with the tongue (Figure 2). A drastic downshift in chromophore pKa
to about 7.6 was observed for this mutant [20]. This salt bridge and the hydrogen-bonding
interaction of the aspartate were shown in simulations to affect the stabilization of the
early/late Lumi-R intermediates featuring a more disordered chromophore-binding
pocket [8].
Figure 6. pH-dependent Pr absorption spectra of IAF-labeled Agp1-PGP and its variants (top panel).
(A) WT-AF, (B) C295S/C279-AF, (C) C279S/C295-AF, (D) C279S/C295S/V364C-AF. The direction of
the pH titration is indicated by the black arrow. The middle panel shows the respective titration
data at nine wavelengths 678 (■), 683.5 (●), 689 (▲), 694.5 (▼), 700 (⧫), 705.5 (◄), 711 (►), 716.5 (⬣),
and 722 nm (★). Selected fit curves at 678, 700, and 711 nm are marked in red. The bottom panel
shows the respective titration data at the absorption maxima of the pH-sensitive absorption band of
fluorescein with 502 nm (A), 505 nm (B), 501 nm (C), and 498 nm (D). The pH-dependence was fitted
by the Henderson−Hasselbalch equation (Equation (2)) using Origin Pro 2019 (Origin Lab). The fit
results for the middle panels are summarized in Table 1, the pKa values for bound fluorescein are
indicated in the figures. The absorption of the deprotonated chromophore at high pH was estimated
to 20% of the maximum absorption of the respective wavelength. Conditions: 300 mM NaCl, 50 mM
Tris/HCl buffer for pH 6.2–9.5, or 100 mM Na2CO3/NaHCO3 buffer for pH 9.5–10.8, at 20 °C .
2.3. pH-Dependence of Conformational Dynamics and Structural Constraints in Cph1
Constructs Using Time-Resolved Fluorescence Anisotropy
In prokaryotic phytochromes, transient bilin chromophore deprotonation after pho-
toactivation is associated with conformational changes that culminate in de-/activation of
the histidine kinase transmitter [24]. In Cph1, we found a direct correlation between chro-
mophore deprotonation and pH-dependent conformational dynamics and flexibility of
Cph1-PGP in its different domains in equilibrium experiments of the Pr-state [13].
We hypothesize that the absent long-range interaction between chromophore pKa
and PHY domain in Agp1 compared to Cph1 [13], as observed by the fluorescein-labeling
experiments, also leads to an absent correlation between chromophore pKa and pH de-
pendence of conformational dynamics and protein flexibility in the Pr state.
Figure 6.
pH-dependent Pr absorption spectra of IAF-labeled Agp1-PGP and its variants (top panel).
(
A
) WT-AF, (
B
) C295S/C279-AF, (
C
) C279S/C295-AF, (
D
) C279S/C295S/V364C-AF. The direction
of the pH titration is indicated by the black arrow. The middle panel shows the respective titration
data at nine wavelengths 678 (
), 683.5 (
•
), 689 (
N
), 694.5 (
H
), 700 (
␄
), 705.5 (
), 711 (
), 716.5 (
),
and 722 nm (
F
). Selected fit curves at 678, 700, and 711 nm are marked in red. The bottom panel
shows the respective titration data at the absorption maxima of the pH-sensitive absorption band of
fluorescein with 502 nm (
A
), 505 nm (
B
), 501 nm (
C
), and 498 nm (
D
). The pH-dependence was fitted
by the Henderson
−
Hasselbalch equation (Equation (2)) using Origin Pro 2019 (Origin Lab). The fit
results for the middle panels are summarized in Table 1, the pK
a
values for bound fluorescein are
indicated in the figures. The absorption of the deprotonated chromophore at high pH was estimated
to 20% of the maximum absorption of the respective wavelength. Conditions: 300 mM NaCl, 50 mM
Tris/HCl buffer for pH 6.2–9.5, or 100 mM Na2CO3/NaHCO3buffer for pH 9.5–10.8, at 20 ◦C.
2.3. pH-Dependence of Conformational Dynamics and Structural Constraints in Cph1 Constructs
Using Time-Resolved Fluorescence Anisotropy
In prokaryotic phytochromes, transient bilin chromophore deprotonation after pho-
toactivation is associated with conformational changes that culminate in de-/activation
of the histidine kinase transmitter [
24
]. In Cph1, we found a direct correlation between
chromophore deprotonation and pH-dependent conformational dynamics and flexibility
of Cph1-PGP in its different domains in equilibrium experiments of the Pr-state [13].
We hypothesize that the absent long-range interaction between chromophore pK
a
and
PHY domain in Agp1 compared to Cph1 [
13
], as observed by the fluorescein-labeling exper-
iments, also leads to an absent correlation between chromophore pK
a
and pH dependence
of conformational dynamics and protein flexibility in the Pr state.
Thus, we measured the time-resolved fluorescence depolarization of Agp1-bound flu-
orescein to detect the pico-nanoseconds structural dynamics of the protein. The anisotropy
decay curve of the bound fluorophore provides information on global and local protein
dynamics as well as on the protein structure and conformational changes [11,12,25].
Figure 7shows the anisotropy curves at five to eight different pH values between
pH 6.3 and pH 11 for Agp1-WT-AF and the fluorescein-labeled variants. In WT-AF, the
fluorescein reporter groups are located at the two native cysteines, one in the
β
-sheet
region in the GAF domain (C279) and the other in the long helix part (C295) close to the
Molecules 2022,27, 8395 8 of 15
chromophore-binding pocket (Figure 3). Data analysis revealed relatively small changes
in the rotational correlation times for the
β
-sheets with 0.9
±
0.09 ns as a function of pH
(Table 2), whereas the amplitudes of the anisotropy decay components and, in particular,
the steric restriction, change significantly with pH (Table 2). Titration curves from the
anisotropy amplitude changes were used to calculate the pK
a
-value (Figure 7B), which
was found to be 9.0
±
0.2 in WT-AF. Next, we directly observed the pH-dependent GAF
β
-sheet mobility in C295S/C279-AF, i.e., the conformational space/steric restriction of
β
-strand movement, and found a similar pK
a
= 9.2
±
0.1 (Figure 7D). In comparison to the
GAF
β
-sheet mobility, the long helix (spine) in the GAF domain exhibits a different pH
dependence of its mobility as revealed in C279S/C295-AF. A faster rotational correlation
time of this segment (0.5 ns at pH 7.5) correlates with a reduced pH-dependent change in
conformational space and an increase in the corresponding pK
a
to 9.7
±
0.2 (Figure 7F).
When investigating the conformational dynamics in the PHY domain
β
-sheet using V364C-
AF, a slower rotational correlation time for the
β
-sheets with 1.4 ns at pH 7.5 was found.
The change in pH-dependent conformational space was smaller than for the GAF domain
β-sheets, but a similar pKa-value of 9.0 ±0.2 was found (Figure 7H, Table 2).
Molecules 2022, 27, x FOR PEER REVIEW 8 of 15
Thus, we measured the time-resolved fluorescence depolarization of Agp1-bound
fluorescein to detect the pico-nanoseconds structural dynamics of the protein. The anisot-
ropy decay curve of the bound fluorophore provides information on global and local pro-
tein dynamics as well as on the protein structure and conformational changes [11,12,25].
Figure 7 shows the anisotropy curves at five to eight different pH values between pH
6.3 and pH 11 for Agp1-WT-AF and the fluorescein-labeled variants. In WT-AF, the fluo-
rescein reporter groups are located at the two native cysteines, one in the β-sheet region
in the GAF domain (C279) and the other in the long helix part (C295) close to the chromo-
phore-binding pocket (Figure 3). Data analysis revealed relatively small changes in the
rotational correlation times for the β-sheets with 0.9 ± 0.09 ns as a function of pH (Table
2), whereas the amplitudes of the anisotropy decay components and, in particular, the
steric restriction, change significantly with pH (Table 2). Titration curves from the anisot-
ropy amplitude changes were used to calculate the pKa-value (Figure 7B), which was
found to be 9.0 ± 0.2 in WT-AF. Next, we directly observed the pH-dependent GAF β-
sheet mobility in C295S/C279-AF, i.e., the conformational space/steric restriction of β-
strand movement, and found a similar pKa = 9.2 ± 0.1 (Figure 7D). In comparison to the
GAF β-sheet mobility, the long helix (spine) in the GAF domain exhibits a different pH
dependence of its mobility as revealed in C279S/C295-AF. A faster rotational correlation
time of this segment (0.5 ns at pH 7.5) correlates with a reduced pH-dependent change in
conformational space and an increase in the corresponding pKa to 9.7 ± 0.2 (Figure 7F).
When investigating the conformational dynamics in the PHY domain β-sheet using
V364C-AF, a slower rotational correlation time for the β-sheets with 1.4 ns at pH 7.5 was
found. The change in pH-dependent conformational space was smaller than for the GAF
domain β-sheets, but a similar pKa-value of 9.0 ± 0.2 was found (Figure 7H, Table 2).
Figure 7. pH-dependent time-resolved fluorescence anisotropy data for Agp1 variants. (A,C,E,G)
show the anisotropy decay curves (grey) and the respective fits (colored) at the different pH.
(B,D,F,H) show the titrations curves of β-sheet flexibility expressed as the relative mobility β′2 =
β2/(β2 + β3) (red data points), and of the steric restriction of the respective protein segment by the
surrounding protein constituents expressed by β3 (black data points). The pKa values of the pH-
Figure 7.
pH-dependent time-resolved fluorescence anisotropy data for Agp1 variants. (
A
,
C
,
E
,
G
)
show the anisotropy decay curves (grey) and the respective fits (colored) at the different pH. (
B
,
D
,
F
,
H
)
show the titrations curves of
β
-sheet flexibility expressed as the relative mobility
β02
=
β2
/(
β2
+
β3
)
(red data points), and of the steric restriction of the respective protein segment by the surrounding
protein constituents expressed by
β3
(black data points). The pK
a
values of the pH-dependent
conformational changes were obtained by global fitting of steric restriction and
β02
using Equation (2).
The fit curves are shown. The respective pK
a
values are given. (
A
,
B
) Agp1 PGP WT-AF, (
C
,
D
) Agp1 PGP
C295S/C279-AF, (
E
,
F
) Agp1 PGP C279S/C295-AF, and (
G
,
H
) Agp1 PGP C279S/C295S/V364C-AF. The
chromophore pK
a
values are depicted alongside with the structural models at the right. The anisotropy
fit values are summarized in Tables 2and 3. Conditions: 300 mM NaCl, 50 mM Tris, at 19 ◦C.
Molecules 2022,27, 8395 9 of 15
Table 2.
Time-resolved anisotropy fit results of Agp1-PGP-AF variants. WT-AF (A), C279-AF (B),
C295-AF (C), V364C-AF (D). r
0
is the initial anisotropy and the amplitudes
β1
and
β2
indicate the
degree of depolarization of the anisotropy decay components with the correlation times
φ1
and
φ2
,
respectively. The tumbling of the whole protein is characterized by
φ3,
the corresponding steric
restriction by β3. The reduced χ2(χred 2) is given as a measure of the goodness of the fit.
(A) Agp1 PGP WT-AF
pH r0φ1(ns) φ2(ns) φ3(ns) aβ1β2β3χred2
7.4 0.35 0.10 1.05 30.0 0.052 0.065 0.232 0.95
8.0 0.35 0.08 0.82 30.0 0.063 0.066 0.220 0.93
8.6 0.34 0.08 0.82 30.0 0.058 0.071 0.211 0.93
9.0 0.34 0.11 0.91 30.0 0.066 0.066 0.207 0.92
9.5 0.34 0.07 0.87 30.0 0.062 0.076 0.202 1.00
10.2 0.33 0.13 0.92 30.0 0.059 0.079 0.192 0.93
(B) Agp1 PGP C279-AF
pH r0φ1(ns) φ2(ns) φ3(ns) aβ1β2β3χred2
7 0.34 0.10 1.10 30 0.048 0.078 0.214 0.96
7.5 0.34 0.31 2.25 30 0.070 0.071 0.199 1.04
9.1 0.34 0.23 1.17 30 0.078 0.082 0.180 0.87
9.5 0.34 0.17 1.10 30 0.065 0.098 0.174 1.06
10.3 0.34 0.18 1.17 30 0.077 0.097 0.161 1.02
10.8 0.34 0.27 1.5 30 0.088 0.093 0.160 0.96
(C) Agp1 PGP C295-AF
pH r0φ1(ns) φ2(ns) φ3(ns) aβ1β2β3χred2
7.5 0.34 0.04 0.50 30 0.054 0.055 0.231 1.01
7.8 0.34 0.04 0.41 30 0.049 0.056 0.235 0.96
8.0 0.34 0.04 0.50 30 0.056 0.054 0.229 0.95
8.6 0.34 0.05 0.51 30 0.049 0.056 0.235 0.95
9.2 0.34 0.04 0.52 30 0.058 0.058 0.224 1.01
10.0 0.34 0.06 0.60 30 0.060 0.060 0.2196 0.96
10.5 0.34 0.07 0.70 30 0.060 0.063 0.2162 0.98
11.0 0.34 0.05 0.60 30 0.061 0.066 0.2135 0.94
(D) Agp1 PGP V364C-AF
pH r0φ1(ns) φ2(ns) φ3(ns) aβ1β2β3χred2
6.3 0.34 0.15 1.40 30 0.049 0.086 0.204 0.96
7.5 0.34 0.15 1.40 30 0.059 0.085 0.199 0.94
8.9 0.34 0.15 1.35 30 0.065 0.088 0.187 1.01
9.5 0.34 0.15 0.60 30 0.076 0.089 0.174 0.97
10.0 0.34 0.15 1.70 30 0.080 0.088 0.171 0.98
aφ3(ns) is kept constant at 30 ns for fitting.
Table 3.
Comparison of pK
a
values of biliverdin chromophore deprotonation (pK
DPC
) of the different
AF-labeled Agp1-PGP variants and the pK
a
values of protein conformational changes in the different
domains. The standard error is given (±SD).
Sample Chromophore pKDPC pKaof Conformational
Change
Agp1 PGP WT-AF
(labeling in GAF) 10.42 ±0.02 9.0 ±0.2
Agp1 PGP C295S/C279-AF
(labeling in β-sheet of GAF) 10.49 ±0.02 9.2 ±0.1
Agp1 PGP C279S/C295-AF
(labeling in long helix of GAF) 10.45 ±0.03 9.7 ±0.2
Agp1 PGP C279S/C295S/V364C-AF
(labeling in β-sheet of PHY) 10.60 ±0.03 9.0 ±0.2
Molecules 2022,27, 8395 10 of 15
The respective pK
a
values of chromophore deprotonation (pK
DPC
) are depicted along-
side the structural models on the right side of Figure 7. The comparison between the pK
a
values of chromophore deprotonation and the pK
a
value of the different domain mobilities
shows no correlation (Table 3, Figure 8), as hypothesized above.
Molecules 2022, 27, x FOR PEER REVIEW 9 of 15
dependent conformational changes were obtained by global fitting of steric restriction and β′2 using
Equation (2). The fit curves are shown. The respective pKa values are given. (A,B) Agp1 PGP WT-
AF, (C,D) Agp1 PGP C295S/C279-AF, (E,F) Agp1 PGP C279S/C295-AF, and (G,H) Agp1 PGP
C279S/C295S/V364C-AF. The chromophore pKa values are depicted alongside with the structural
models at the right. The anisotropy fit values are summarized in Tables 2 and 3. Conditions: 300
mM NaCl, 50 mM Tris, at 19 °C.
The respective pKa values of chromophore deprotonation (pKDPC) are depicted along-
side the structural models on the right side of Figure 7. The comparison between the pKa
values of chromophore deprotonation and the pKa value of the different domain mobili-
ties shows no correlation (Table 3, Figure 8), as hypothesized above.
Figure 8. Comparison between chromophore pKa and pH-dependent conformational change in the
PHY and GAF domain for Cph1 and Agp1. Data from Figure 7, Table 3 and reference [13].
Table 2. Time-resolved anisotropy fit results of Agp1-PGP-AF variants. WT-AF (A), C279-AF (B),
C295-AF (C), V364C-AF (D). r0 is the initial anisotropy and the amplitudes β1 and β2 indicate the
degree of depolarization of the anisotropy decay components with the correlation times ϕ1 and ϕ2,
respectively. The tumbling of the whole protein is characterized by ϕ3, the corresponding steric re-
striction by β3. The reduced χ2 (χred2) is given as a measure of the goodness of the fit.
(A) Agp1 PGP WT-AF
pH
r0
ϕ1 (ns)
ϕ2 (ns)
ϕ3 (ns) a
β1
β2
β3
χred2
7.4
0.35
0.10
1.05
30.0
0.052
0.065
0.232
0.95
8.0
0.35
0.08
0.82
30.0
0.063
0.066
0.220
0.93
8.6
0.34
0.08
0.82
30.0
0.058
0.071
0.211
0.93
9.0
0.34
0.11
0.91
30.0
0.066
0.066
0.207
0.92
9.5
0.34
0.07
0.87
30.0
0.062
0.076
0.202
1.00
10.2
0.33
0.13
0.92
30.0
0.059
0.079
0.192
0.93
(B) Agp1 PGP C279-AF
pH
r0
ϕ1 (ns)
ϕ2 (ns)
ϕ3 (ns) a
β1
β2
β3
χred2
7
0.34
0.10
1.10
30
0.048
0.078
0.214
0.96
7.5
0.34
0.31
2.25
30
0.070
0.071
0.199
1.04
9.1
0.34
0.23
1.17
30
0.078
0.082
0.180
0.87
9.5
0.34
0.17
1.10
30
0.065
0.098
0.174
1.06
10.3
0.34
0.18
1.17
30
0.077
0.097
0.161
1.02
10.8
0.34
0.27
1.5
30
0.088
0.093
0.160
0.96
Figure 8.
Comparison between chromophore pK
a
and pH-dependent conformational change in the
PHY and GAF domain for Cph1 and Agp1. Data from Figure 7, Table 3and reference [13].
3. Discussion
A number of fundamental questions regarding the mechanism of phytochrome light
activation and subsequent signal transfer remain unanswered. Although structural differ-
ences between Pr and Pfr have been identified in bacteriophytochrome [
6
], it is not fully
understood how these changes are triggered by the light-induced isomerization of the
chromophore. Also, the mechanism of signal transmission from the chromophore-binding
pocket to the transmitter region is still unclear. While a recent femtosecond X-ray laser
study [
4
] focused on ultrafast collective changes in the chromophore-binding pocket, in-
cluding protein backbone and water movements around the chromophore, a theoretical
study [
26
] shed new light on the activation mechanism leading to the large structural
change enabling signal transduction. The latter study indicates a large structural relaxation
in solution compared to the crystal structure and an internal re-organization of the PHY do-
main upon light-activation. This internal re-organization, i.e., rotation of the PHY domain,
is thought to be a consequence of the tongue re-folding from the
β
-sheet to the
α
-helical
conformation in Pfr after chromophore isomerization, leading to a new conformation of
the long helix through constraint of the H-bonds to the adjacent GAF and PHY domains,
and thus to the large-scale conformational change enabling transmitter activation. Tongue
re-folding was experimentally shown to occur with the last transition between the inter-
mediate Meta-R and the Pfr state in Deinococcus radiodurans phytochrome (DrBphP) [
27
].
Together with the new understanding of the phytochrome activation pathways in the study
by Macaluso et al. [
26
], this study underlines the importance of solution experiments that
provide structural information.
An open question also concerns the de-/protonation of the chromophore during the
photoconversion from Pr to Pfr. While theoreticians focus on the chromophore protonation
in the Pr and Pfr ground or excited states [
28
,
29
], it is generally accepted that proton (s)
are released and subsequently taken up by the protein upon the transition from Pr to
Pfr, as shown in spectroscopic experiments [
9
,
10
]. However, the exact correlation with
intermediate states, and whether tongue re-folding is causally associated with protonation
dynamics remains elusive.
Molecules 2022,27, 8395 11 of 15
The pH titration of phytochromes indicates that chromophore absorption is highly
pH-dependent in the Pr-state as found for many bacteriophytochrome species in the
range between pH 6 and 10 [
10
,
20
], also recently shown in detail for Cph1 [
17
]. This pH-
dependent difference in absorption can be explained by chromophore deprotonation, since
deprotonation of the bilin chromophore nitrogen results in lowering of the chromophore
extinction coefficient [
30
,
31
]. Employing such equilibrium titrations in fluorescently labeled
Cph1 in position 371 we obtained an unexpected result. We found that both labeling with
fluorescein and the mutation C371S, located in the chromophore-distant PHY domain, lead
to a substantial downshift of the chromophore pK
a
. Our findings were rationalized in
our previous publication as follows [
13
]. We assumed a conformational change, induced
by the mutation/labeling with fluorescein, that propagates through the arrangement of
hydrogen bonds [
32
] from the PHY domain
β
-sheet to the chromophore-binding pocket.
This hypothesis was supported by the correlation of the pH dependence of the local
conformational change in the distant PHY domain with the chromophore pKa[13].
Here, we investigated whether a similar long-range rearrangement of hydrogen bonds
exist and would lead to a correlation of chromophore pK
a
and pH-dependent conforma-
tional flexibility in the distant PHY domain of another canonical phytochrome, namely
Agp1 which, however, binds the biliverdin chromophore in the PAS domain, most distant
to the PHY domain (Figure 1). We constructed the mutant Agp1-PGP V346C to test this
hypothesis. To our surprise, neither a mutation-dependent change in chromophore pK
a
nor a correlation between chromophore pK
a
and pH-dependent domain mobility, was
observed for Agp1 (Figure 8). This result is also in line with the absence of intersubunit
distance changes as found for Agp1 using PELDOR measurements [33].
We conclude the following: (i) Agp1, which binds BV to position C20 (PAS) instead
of PCB to position C259 (GAF) in Cph1, behaves differently compared to Cph1 when
the analogous position to C371 in the PHY domain is mutated (V346C) or labeled with
fluorescein. No long-distance effects on the chromophore deprotonation were observed
in Agp1. (ii) The absence of these long-distance effects correlates with an absence of the
coupling between pH-dependent conformational changes in the distant PHY domain and
the chromophore pK
a
. (iii) While the properties within the chromophore-binding pocket
and the activation mechanisms seem to be highly conserved [
8
], activation pathways from
the chromophore-binding pocket to the PHY domain may vary and were suggested to
allow phytochrome to exploit redundant routes for its activation [26].
The latter is supported by experimental results where substitution of the conserved
tyrosine Y263 to phenylalanine resulted in a decoupling between protein conformation (Pfr)
and chromophore state (Pr) in DrBphP [
34
]. Similarly, the Y263S mutation in the canonical
Cph1 phytochrome resulted in a block of the Pr to Pfr photoconversion as evidenced by
time-resolved absorption experiments [35].
Regarding the mechanism of signal transmission from the chromophore-binding
pocket to the transmitter region, we now provide further experimental evidence that
various mechanisms may exist that couple chromophore deprotonation to the mobility of
the PHY domain which could translate into the global conformational change. We speculate
that the long-range coupling observed in the canonical phytochrome Cph1 harboring the
PCB chromophore and the absence of this coupling in the canonical Agp1 with the BV
chromophore might be connected to the two different chromophores, and thus probably also
to the protonation heterogeneity observed in the Pr state of the PCB-binding Cph1 [
13
,
17
]
but not in the BV-binding Agp1.
4. Materials and Methods
4.1. Phytochrome Mutagenese, Expression and Purification
The photosensory module of wild-type Agp1 (residues 1–504 with a C-terminal His6-
tag, Agp1-PGP), and the constructs Agp1-PGP-C279S/C295S, Agp1-PGP-C279S, Agp1-
PGP-C295S, and Agp1-PGP-C279S/C295S/V364C were expressed in E. coli, the cells lysed,
and the soluble fraction purified following established methods [
19
]. The apoprotein was
Molecules 2022,27, 8395 12 of 15
expressed and assembled with BV according to established protocols [
36
]. The plasmid
agro1-M15-9N, based on the original plasmid [
18
] was kindly provided by Prof. Hildebrand
(TU Berlin). This plasmid and the plasmid without deletion of the first 9 N-terminal amino
acids were used for the mutagenesis. Site-directed mutagenesis of Agp1-PGP was carried
out according to the Quik Change
TM
mutagenesis protocol (Agilent Technologies, Santa
Clara, CA, USA). The required enzymes were purchased from New England Biolabs,
Ipswich, MA, USA.
4.2. Phytochrome Labeling with IAF
The photosensory module Agp1-PGP, Agp1-PGP-C279S/C295S, Agp1-PGP-C279S,
Agp1-PGP-C295S, and Agp1-PGP-C279S/C295S/V364C were labeled with a 10-fold molar
excess of 5-iodacetamidfluorescein (IAF, Invitrogen Molecular Probes, Waltham, MA, USA)
and 2-fold molar excess of DTT in 50 mM Tris pH 7.8, 150 mM NaCl for 8 h at room
temperature in the dark. Gel filtration (Sephadex G-25 fine, GE Healthcare, Chicago,
IL, USA) was used to remove unlabeled dye (adapted from [
16
,
25
]). The molar labeling
stoichiometry was determined by
cLabel
cProtein
=∆ALabel
εLabel εProtein,Pr
AProtein,Pr (1)
The absorbance A
Protein,Pr
was measured at
λmax
= 702 nm. The corresponding extinc-
tion coefficient is
εProtein,Pr
= 90,000 M
−1
cm
−1
[
37
]. The measurements were performed
under green light to avoid photoconversion of Pr. The absorbance difference between the
labeled and unlabeled samples,
∆
A
L
, at
λmax
(500 nm) is the absorbance of the fluorescein
label. The corresponding extinction coefficient of IAF at
λmax
is
εLabel
= 67,000 M
−1
cm
−1
at pH 7.0 and
εLabel
= 77,000 M
−1
cm
−1
at pH 7.8 (Invitrogen Molecular Probes). Covalent
binding of IAF to Agp1-PGP and its variants, and removal of excess unlabeled dye, was
verified by Tricine-SDS-PAGE using the fluorescence band of IAF-labeled Agp1-PGP.
4.3. UV–Vis Spectroscopy, pH-Titration and pKaDetermination
UV–Vis absorbance spectra were measured with a Shimadzu UV2450 spectrometer
(Shimadzu, Kyoto, Japan). pH titration was carried out at 20
◦
C in 50 mM Tris-citrate
buffer including 15 mM NaCl. The pH was adjusted with small aliquots of NaOH and
HCl in the pH range from 6.5 to 10.3. The Pr state was generated by saturating irradiation
using a 735 nm LED (Conrad Electronics, Hirschau, Germany). The experiments were
performed with aliquots in parallel, with different aliquots used in different pH ranges.
The pH was measured before each experiment with a microelectrode (Metrohm, Herisau,
Switzerland) in a measurement volume of 60
µ
L. The following controls were carried out:
The samples were tested at the extreme pH values to identify the pH range in which the
protein was stable for the time of the measurement. For pH values below pH 6 and above
pH 11 immediate aggregation was identified by UV–Vis absorption. Also, before each
TCSPC experiment, the absorption spectrum was checked. Only low intensity background
light was used during the measurements to avoid photoconversion of Pr. The absorbance
values were scaled to the extinction coefficient
ε700nm
of Pr at pH 7.8 (90,000 M
−1
cm
−1
)
for presentation. The pH-titration curves were generated from the respective absorbance
values at nine wavelengths from 678–722 nm. A global fit with the Henderson
−
Hasselbalch
equation (Equation (2)) was performed
∆A(pH)=∆Amax /1+10(pK−p H)(2)
4.4. Time-Resolved Fluorescence Spectroscopy
The fluorescence anisotropy decay measurements were performed as described
in [
14
,
25
,
38
]. Briefly, the sample was measured using a picosecond Ti:sapphire laser system
(Millenia vs and Tsunami, Spectra Physics, Milpitas, CA, USA) and a microchannel plate
Molecules 2022,27, 8395 13 of 15
detector (model #R3809U, Hamamatsu, Shizuoka, Japan) in a time-correlated single-photon
counting (TCSPC) setup with TCSPC card SPC-830 (Becker & Hickl, Berlin, Germany).
Fluorescence excitation was at 488 nm and emission was detected using a 515 nm long-pass
filter (GG515). The fluorescence lifetime traces were recorded in 1024 time channels with
an instrument response function (IRF) of 30–40 ps. The instrument response function (IRF)
of the TCSPC setup was determined at the corresponding wavelengths with a colloidal
silica solution as the scattering material (LUDOX, Grace). Fluorescence decay traces were
fitted using a sum of exponentials with an amplitude
αi
and a decay constant
τi
after
deconvolution with the IRF
I(t) =
n
∑
i=1
αie−t/τi(3)
Fluorescence anisotropy time traces r(t) were calculated by measuring parallel
Ik(t)
and perpendicular
I⊥(t)
polarized fluorescence intensity compared to the excitation
polarization using
r(t) = Ik(t)−I⊥(t)
Ik(t) + 2I⊥(t)(4)
Anisotropy fit data were acquired using the following model function in the software
Globals (Laboratory for Fluorescence Dynamics, University of California, Irvine, CA, USA)
r(t) =
3
∑
i=1
βie−t
φi(5)
with the initial anisotropy (at t= 0: r
0
= (
β1
+
β2
+
β3
). The rotational correlation times
φ1
and
φ2
describe the rotational motion of the label and the protein segment to which the labeled
is attached.
φ3
describes the rotational diffusion of the whole system. The amplitudes
β1
and
β2
indicate the degree of depolarization of the anisotropy decay components with
the correlation times
φ1
and
φ2
, respectively. If the rotational correlation time of the last
decay component is much slower than the lifetime of the fluorescent probe, the anisotropy
decays virtually to a constant end value r
∞
in the ns time range of the measurements that
is determined by the lifetime of the fluorescent probe. The value of r
∞
, or the amplitude
of the slowest anisotropy decay component, in our case
β3
, represents a measure of the
degree of steric hindrance by the protein surface.
The conformational space of the protein segment is expressed as relative mobility using
β0
2=β2/(β2+β3)(6)
Author Contributions:
U.A. conceived and supervised the project. M.S., J.B. and U.A. analyzed and
interpreted the experimental data. U.A., J.B. and T.R.-M. wrote the manuscript. All authors discussed
the results and commented on the manuscript. All authors have read and agreed to the published
version of the manuscript.
Funding: This work was supported by DFG (SFB1078 Project-ID 221545957), project A2 to U.A.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement:
The data presented in this study are openly available in FigShare at
10.6084/m9.figshare.21647348.
Acknowledgments:
The providing of the initial vector for Agp1 and discussion by AG Scheerer,
Charite Berlin, is gratefully acknowledged.
Conflicts of Interest: The authors declare no conflict of interest.
Sample Availability: Samples of the compounds are available from the authors.
Molecules 2022,27, 8395 14 of 15
References
1.
Macháˇcková, I.; Kendrick, R.E.; Kronenberg, G.H.M. Photomorphogenesis in plants, 2nd Edition. Biol. Plant.
1994
,36, 564.
[CrossRef]
2.
Hughes, J.; Lamparter, T. Prokaryotes and Phytochrome. The Connection to Chromophores and Signaling1. Plant Physiol.
1999
,
121, 1059–1068. [CrossRef] [PubMed]
3.
Lamparter, T.; Michael, N.; Mittmann, F.; Esteban, B. Phytochrome from Agrobacterium tumefaciens has unusual spectral
properties and reveals an N-terminal chromophore attachment site. Proc. Natl. Acad. Sci. USA
2002
,99, 11628–11633. [CrossRef]
4.
Claesson, E.; Wahlgren, W.Y.; Takala, H.; Pandey, S.; Castillon, L.; Kuznetsova, V.; Henry, L.; Panman, M.; Carrillo, M.; Kubel, J.;
et al. The primary structural photoresponse of phytochrome proteins captured by a femtosecond X-ray laser. eLife
2020
,9, e53514.
[CrossRef] [PubMed]
5.
Yang, Y.; Stensitzki, T.; Sauthof, L.; Schmidt, A.; Piwowarski, P.; Velazquez Escobar, F.; Michael, N.; Nguyen, A.D.; Szczepek,
M.; Brünig, F.N.; et al. Ultrafast proton-coupled isomerization in the phototransformation of phytochrome. Nat. Chem. 2022,14,
823–830. [CrossRef]
6.
Takala, H.; Bjorling, A.; Berntsson, O.; Lehtivuori, H.; Niebling, S.; Hoernke, M.; Kosheleva, I.; Henning, R.; Menzel, A.; Ihalainen,
J.A.; et al. Signal amplification and transduction in phytochrome photosensors. Nature 2014,509, 245–248. [CrossRef]
7.
Krieger, A.; Molina, I.; Oberpichler, I.; Michael, N.; Lamparter, T. Spectral properties of phytochrome Agp2 from Agrobacterium
tumefaciens are specifically modified by a compound of the cell extract. J. Photochem. Photobiol. B 2008,93, 16–22. [CrossRef]
8.
Salvadori, G.; Macaluso, V.; Pellicci, G.; Cupellini, L.; Granucci, G.; Mennucci, B. Protein control of photochemistry and transient
intermediates in phytochromes. Nat. Commun. 2022,13, 6838. [CrossRef] [PubMed]
9.
Borucki, B.; von Stetten, D.; Seibeck, S.; Lamparter, T.; Michael, N.; Mroginski, M.A.; Otto, H.; Murgida, D.H.; Heyn, M.P.;
Hildebrandt, P. Light-induced proton release of phytochrome is coupled to the transient deprotonation of the tetrapyrrole
chromophore. J. Biol. Chem. 2005,280, 34358–34364. [CrossRef] [PubMed]
10.
van Thor, J.J.; Borucki, B.; Crielaard, W.; Otto, H.; Lamparter, T.; Hughes, J.; Hellingwerf, K.J.; Heyn, M.P. Light-induced proton
release and proton uptake reactions in the cyanobacterial phytochrome Cph1. Biochemistry 2001,40, 11460–11471. [CrossRef]
11.
Alexiev, U.; Farrens, D.L. Fluorescence spectroscopy of rhodopsins: Insights and approaches. Biochim. Biophys. Acta Bioenerg.
2014,1837, 694–709. [CrossRef] [PubMed]
12.
Kirchberg, K.; Kim, T.Y.; Moller, M.; Skegro, D.; Dasara Raju, G.; Granzin, J.; Buldt, G.; Schlesinger, R.; Alexiev, U. Conformational
dynamics of helix 8 in the GPCR rhodopsin controls arrestin activation in the desensitization process. Proc. Natl. Acad. Sci. USA
2011,108, 18690–18695. [CrossRef] [PubMed]
13.
Sadeghi, M.; Balke, J.; Schneider, C.; Nagano, S.; Stellmacher, J.; Lochnit, G.; Lang, C.; Weise, C.; Hughes, J.; Alexiev, U. Transient
Deprotonation of the Chromophore Affects Protein Dynamics Proximal and Distal to the Linear Tetrapyrrole Chromophore in
Phytochrome Cph1. Biochemistry 2020,59, 1051–1062. [CrossRef]
14.
Volz, P.; Krause, N.; Balke, J.; Schneider, C.; Walter, M.; Schneider, F.; Schlesinger, R.; Alexiev, U. Light and pH-induced Changes
in Structure and Accessibility of Transmembrane Helix B and Its Immediate Environment in Channelrhodopsin-2. J. Biol. Chem.
2016,291, 17382–17393. [CrossRef]
15.
Winkler, K.; Winter, A.; Rueckert, C.; Uchanska-Ziegler, B.; Alexiev, U. Natural MHC class I polymorphism controls the pathway
of peptide dissociation from HLA-B27 complexes. Biophys. J. 2007,93, 2743–2755. [CrossRef]
16.
Richter, C.; Schneider, C.; Quick, M.T.; Volz, P.; Mahrwald, R.; Hughes, J.; Dick, B.; Alexiev, U.; Ernsting, N.P. Dual-fluorescence
pH probe for bio-labelling. Phys. Chem. Chem. Phys. 2015,17, 30590–30597. [CrossRef] [PubMed]
17.
Escobar, F.V.; Lang, C.; Takiden, A.; Schneider, C.; Balke, J.; Hughes, J.; Alexiev, U.; Hildebrandt, P.; Mroginski, M.A. Protonation-
Dependent Structural Heterogeneity in the Chromophore Binding Site of Cyanobacterial Phytochrome Cph1. J. Phys. Chem. B
2017,121, 47–57. [CrossRef]
18.
Nagano, S.; Scheerer, P.; Zubow, K.; Michael, N.; Inomata, K.; Lamparter, T.; Krauß, N. The Crystal Structures of the N-terminal
Photosensory Core Module of Agrobacterium Phytochrome Agp1 as Parallel and Anti-parallel Dimers. J. Biol. Chem.
2016
,291,
20674–20691. [CrossRef]
19.
Essen, L.O.; Mailliet, J.; Hughes, J. The structure of a complete phytochrome sensory module in the Pr ground state. Proc. Natl.
Acad. Sci. USA 2008,105, 14709–14714. [CrossRef]
20.
von Stetten, D.; Seibeck, S.; Michael, N.; Scheerer, P.; Mroginski, M.A.; Murgida, D.H.; Krauss, N.; Heyn, M.P.; Hildebrandt, P.;
Borucki, B.; et al. Highly Conserved Residues Asp-197 and His-250 in Agp1 Phytochrome Control the Proton Affinity of the
Chromophore and Pfr Formation. J. Biol. Chem. 2007,282, 2116–2123. [CrossRef]
21.
Möller, M.; Alexiev, U. Surface Charge Changes upon Formation of the Signaling State in Visual Rhodopsin. Photochem. Photobiol.
2009,85, 501–508. [CrossRef] [PubMed]
22.
Alexiev, U.; Marti, T.; Heyn, M.P.; Khorana, H.G.; Scherrer, P. Surface charge of bacteriorhodopsin detected with covalently bound
pH indicators at selected extracellular and cytoplasmic sites. Biochemistry 1994,33, 298–306. [CrossRef] [PubMed]
23.
Kirchberg, K.; Michel, H.; Alexiev, U. Exploring the entrance of proton pathways in cytochrome c oxidase from Paracoccus
denitrificans: Surface charge, buffer capacity and redox-dependent polarity changes at the internal surface. Biochim. Biophys. Acta
2013,1827, 276–284. [CrossRef] [PubMed]
24.
Björling, A.; Berntsson, O.; Lehtivuori, H.; Takala, H.; Hughes, A.J.; Panman, M.; Hoernke, M.; Niebling, S.; Henry, L.; Henning,
R.; et al. Structural photoactivation of a full-length bacterial phytochrome. Sci. Adv. 2016,2, e1600920. [CrossRef] [PubMed]
Molecules 2022,27, 8395 15 of 15
25.
Alexiev, U.; Rimke, I.; Pohlmann, T. Elucidation of the nature of the conformational changes of the EF-interhelical loop in bacteri-
orhodopsin and of the helix VIII on the cytoplasmic surface of bovine rhodopsin: A time-resolved fluorescence depolarization
study. J. Mol. Biol. 2003,328, 705–719. [CrossRef] [PubMed]
26.
Macaluso, V.; Salvadori, G.; Cupellini, L.; Mennucci, B. The structural changes in the signaling mechanism of bacteriophy-
tochromes in solution revealed by a multiscale computational investigation. Chem. Sci. 2021,12, 5555–5565. [CrossRef]
27.
Ihalainen, J.A.; Gustavsson, E.; Schroeder, L.; Donnini, S.; Lehtivuori, H.; Isaksson, L.; Thoing, C.; Modi, V.; Berntsson, O.; Stucki-
Buchli, B.; et al. Chromophore-Protein Interplay during the Phytochrome Photocycle Revealed by Step-Scan FTIR Spectroscopy. J.
Am. Chem. Soc. 2018,140, 12396–12404. [CrossRef]
28.
Borg, O.A.; Durbeej, B. Relative ground and excited-state pKa values of phytochromobilin in the photoactivation of phytochrome:
A computational study. J. Phys. Chem. B 2007,111, 11554–11565. [CrossRef]
29.
Modi, V.; Donnini, S.; Groenhof, G.; Morozov, D. Protonation of the Biliverdin IX alpha Chromophore in the Red and Far-Red
Photoactive States of a Bacteriophytochrome. J. Phys. Chem. B 2019,123, 2325–2334. [CrossRef]
30.
Margulies, L.; Stockburger, M. Spectroscopic studies on model compounds of the phytochrome chromophore. Protonation and
deprotonation of biliverdin dimethyl ester. J. Am. Chem. Soc. 1979,101, 743–744. [CrossRef]
31.
Mizutani, Y.; Tokutomi, S.; Aoyagi, K.; Horitsu, K.; Kitagawa, T. Resonance Raman study on intact pea phytochrome and its model
compounds: Evidence for proton migration during the phototransformation. Biochemistry
1991
,30, 10693–10700. [CrossRef]
[PubMed]
32. Oliveira, A.S.F.; Campos, S.R.R.; Baptista, A.M.; Soares, C.M. Coupling between protonation and conformation in cytochrome c
oxidase: Insights from constant-pH MD simulations. Biochim. Biophys. Acta 2016,1857, 759–771. [CrossRef] [PubMed]
33.
Kacprzak, S.; Njimona, I.; Renz, A.; Feng, J.; Reijerse, E.; Lubitz, W.; Krauss, N.; Scheerer, P.; Nagano, S.; Lamparter, T.; et al.
Intersubunit distances in full-length, dimeric, bacterial phytochrome Agp1, as measured by pulsed electron-electron double
resonance (PELDOR) between different spin label positions, remain unchanged upon photoconversion. J. Biol. Chem.
2017
,292,
7598–7606. [CrossRef] [PubMed]
34.
Takala, H.; Lehtivuori, H.K.; Berntsson, O.; Hughes, A.; Nanekar, R.; Niebling, S.; Panman, M.; Henry, L.; Menzel, A.; Westenhoff,
S.; et al. On the (un)coupling of the chromophore, tongue interactions, and overall conformation in a bacterial phytochrome. J.
Biol. Chem. 2018,293, 8161–8172. [CrossRef]
35.
Nagano, S.; Sadeghi, M.; Balke, J.; Fleck, M.; Heckmann, N.; Psakis, G.; Alexiev, U. Improved fluorescent phytochromes for in situ
imaging. Sci. Rep. 2022,12, 5587. [CrossRef]
36.
Lamparter, T.; Esteban, B.; Hughes, J. Phytochrome Cph1 from the cyanobacterium Synechocystis PCC6803—Purification,
assembly, and quaternary structure. Eur. J. Biochem. 2001,268, 4720–4730. [CrossRef]
37.
Lamparter, T.; Michael, N.; Caspani, O.; Miyata, T.; Shirai, K.; Inomata, K. Biliverdin Binds Covalently to Agrobacterium
Phytochrome Agp1 via Its Ring A Vinyl Side Chain. J. Biol. Chem. 2003,278, 33786–33792. [CrossRef]
38.
Kim, T.Y.; Winkler, K.; Alexiev, U. Picosecond multidimensional fluorescence spectroscopy: A tool to measure real-time protein
dynamics during function. Photochem. Photobiol. 2007,83, 378–384. [CrossRef]
