Molecular Activation Mechanism and Structural Dynamics of Orange Carotenoid Protein

: Like most photosynthetic organisms, cyanobacteria are vulnerable to ﬂuctuations in light intensity, which can damage their photosynthetic machinery. To protect against this, they use a photoprotective mechanism called non-photochemical quenching (NPQ), where excess absorbed photo-energy is dissipated as heat. In cyanobacteria, light activation of Orange Carotenoid Protein (OCP) is the critical ﬁrst step in the NPQ response. OCP is also the only known photosensitive protein, which uses carotenoid for its activation. We summarize the current knowledge on the light induced reactions of OCP; the different mechanisms of activation that have been proposed; photocycle kinetics and characteristics; and the reported structural intermediates. We discuss the possible interpretations of reported experimental results, and we formulate important open questions and directions for future work, to reveal the molecular and structural basis of photosensing


Introduction
Photosensitive proteins play a crucial role in the life and development of various organisms, from animals to unicellular bacteria, where they regulate a vast range or processes, such as circadian rhythms, organism development, phototaxis and many others [1,2].Photosynthesis is another process relying on different photosensors and photoreceptors [3].More than half of the total photosynthesis activity on earth occurs in single-celled organisms, especially algae and cyanobacteria.Therefore, cyanobacteria are excellent model organisms to study the efficiency and regulation of light energy harnessing and storage.During their evolution, photosynthetic organisms have evolved various photoprotective strategies to cope with fluctuating light intensity and quality, as well as other environmental stresses, by adjusting their photosynthetic machinery.Adaptation to rapid environmental changes is crucial for maintaining optimal photosynthetic efficiency, and is ultimately key to their survival.Most photosynthetic organisms protect their photosynthetic apparatus against rapidly increasing light intensities using non-photochemical quenching (NPQ): a mechanism by which excess photo-energy is dissipated as heat [4,5].This response is rapid (minutes or less) and does not involve changes in gene expression, but instead requires a rapid functional switch of the photosynthetic machinery from a "light-harvesting" into a "heat-dissipating" state.In cyanobacteria, this is controlled by a unique photosensitive protein, Orange Carotenoid Protein (OCP) [6].OCP is the only known photoreceptor protein which uses a carotenoid for its activation.Because of its unique structure and importance, it is vital for the scientific community to fully understand how this carotenoidbased light-activation process is activated, and subsequently controls NPQ in cyanobacteria.These goals require an understanding of the initial activation mechanism, the intermediate and final structural states, and the kinetics of the OCP photocycle.However, following important spectroscopy and structural studies over the last decade, many of the fundamental questions still need to be answered.Therefore, there is a strong need to summarize
states, and the kinetics of the OCP photocycle.However, following important spectroscopy and structural studies over the last decade, many of the fundamental questions still need to be answered.Therefore, there is a strong need to summarize and discuss all the progress and evidence from structural and time-resolved spectroscopic studies, and give directions for future work.This is the aim of this review.

Structure and Function of OCP
OCP is a water-soluble protein composed of two domains: an all-helical N-terminal domain (NTD, residues 15-165-annotation is based on PDB ID code 5TV0) that is unique for cyanobacteria; and a C-terminal domain (CTD, residues 190-317), which is an α/β-fold domain of the ubiquitous Nuclear Transport Factor-2 (NTF2) superfamily, also found in many other organisms (Figure 1).The structure of the C-terminal domain resembles blue-light responsive BLUF and LOV domains, which also contain a core β-sheet surrounded by helices [6].The two domains are joined together by: (1) an N-terminal extension (NTE, residues [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15], packing against the central β-sheet of the CTD; (2) a flexible linker between the two domains running on the molecular surface opposite the NTE (residues 166-189) [7]; (3) salt bridge between Arg155 and Glu244; (4) by keto-carotenoid buried inside the two domains (Figure 1a,b).To protect themselves against fluctuating light intensities, cyanobacteria can adjust energy flows through their photosynthetic antenna complexes, the phycobilisomes (PBSs).OCP plays a crucial role in this process.Absorption of blue-green light by the keto-carotenoid induces conformational changes in the carotenoid and the protein, converting stable inactive OCP O (the so-called orange form) into unstable but active OCP R (red form), which is also followed by drastic changes in the OCP absorption.It is known that only OCP R can bind to PBSs and induce NPQ [8,9].OCP R could potentially quench To protect themselves against fluctuating light intensities, cyanobacteria can adjust energy flows through their photosynthetic antenna complexes, the phycobilisomes (PBSs).OCP plays a crucial role in this process.Absorption of blue-green light by the ketocarotenoid induces conformational changes in the carotenoid and the protein, converting stable inactive OCP O (the so-called orange form) into unstable but active OCP R (red form), which is also followed by drastic changes in the OCP absorption.It is known that only OCP R can bind to PBSs and induce NPQ [8,9].OCP R could potentially quench excitation energy either directly or indirectly (see the proposed mechanisms [10][11][12] and the recent review [13]).

Role of Individual Amino Acids in OCP Light Activation
For OCP to be photoactive, a keto-carotenoid should be present in the middle of its structure (Figure 1).The OCP has been shown to bind and be active with various keto-carotenoids (3-hydroxyl-echinenone, hECN; echinenone, ECN; and canthaxanthin, CAN) [14][15][16].All of these are composed of a conjugated chain of 11 carbon-carbon double bonds, which are in all trans configuration in OCP O , and a keto-group.The keto-group is located in CTD in the β1-ring of the keto-carotenoid.It is involved in a hydrogen bond formation between the keto-carotenoid and the protein moiety.The latter is supplied by the hydroxyl group of Tyr201 and the amino group of Trp288, both located in CTD (Figure 2).The other ring of the carotenoid (β2) is nestled within a group of conserved aromatic residues (Trp41, Tyr44, Trp110) in the NTD (Figure 2) [7].Multiple mutagenesis studies narrowed down the list of amino acids playing a crucial role in OCP function (Figure 2): If Tyr201 or Trp288 gets mutated, the OCP R is not accumulated; therefore, OCP activity is considered to be lost [17].In addition, these amino acids are responsible for carotenoid affinity and specificity [18].

Role of Individual Amino Acids in OCP Light Activation
For OCP to be photoactive, a keto-carotenoid should be present in the middle of its structure (Figure 1).The OCP has been shown to bind and be active with various keto-carotenoids (3-hydroxyl-echinenone, hECN; echinenone, ECN; and canthaxanthin, CAN) [14][15][16].All of these are composed of a conjugated chain of 11 carbon-carbon double bonds, which are in all trans configuration in OCP O , and a keto-group.The keto-group is located in CTD in the β1-ring of the keto-carotenoid.It is involved in a hydrogen bond formation between the keto-carotenoid and the protein moiety.The latter is supplied by the hydroxyl group of Tyr201 and the amino group of Trp288, both located in CTD (Figure 2).The other ring of the carotenoid (β2) is nestled within a group of conserved aromatic residues (Trp41, Tyr44, Trp110) in the NTD (Figure 2) [7].Multiple mutagenesis studies narrowed down the list of amino acids playing a crucial role in OCP function (Figure 2): If Tyr201 or Trp288 gets mutated, the OCP R is not accumulated; therefore, OCP activity is considered to be lost [17].In addition, these amino acids are responsible for carotenoid affinity and specificity [18].If Tyr44 is replaced by Ser, OCP R is also not accumulated at any temperature indicating Tyr44, an essential role in OCP light activation process.On the other hand, the replacement of Trp110 by Ser affects the activation energy, necessary for the OCP O to OCP R conversion.Replacement of Tyr44 and Trp110 with Phe does not affect the OCP R accumulation, demonstrating the critical function of the aromatic rings of the side chains in the OCP photocycle [19].Other site-directed mutagenesis studies demonstrated that salt bridge between Arg155 and Glu244 stabilizes the interaction between the N-and C-terminal domains: its disruption increased the rate of photoactivity and the stability of the photoactivated OCP R [18].Changing Glu244 to Leu had no effect on OCP binding to PBS.By contrast, substitution of Arg155 with a neutral or a negatively charged amino acid largely decreased OCP binding to the PBs, whereas substitution with a lysine slightly perturbed the interaction.Based on that finding, it was suggested that the surface of the N-terminal domain, containing the Arg155, is the place where OCP binds/interacts with the PBS.
The specific roles or functional associations of different amino acids in the light activation process, or in NPQ, indicate that light activation and photoprotection are not If Tyr44 is replaced by Ser, OCP R is also not accumulated at any temperature indicating Tyr44, an essential role in OCP light activation process.On the other hand, the replacement of Trp110 by Ser affects the activation energy, necessary for the OCP O to OCP R conversion.Replacement of Tyr44 and Trp110 with Phe does not affect the OCP R accumulation, demonstrating the critical function of the aromatic rings of the side chains in the OCP photocycle [19].Other site-directed mutagenesis studies demonstrated that salt bridge between Arg155 and Glu244 stabilizes the interaction between the N-and C-terminal domains: its disruption increased the rate of photoactivity and the stability of the photoactivated OCP R [18].Changing Glu244 to Leu had no effect on OCP binding to PBS.By contrast, substitution of Arg155 with a neutral or a negatively charged amino acid largely decreased OCP binding to the PBs, whereas substitution with a lysine slightly perturbed the interaction.Based on that finding, it was suggested that the surface of the N-terminal domain, containing the Arg155, is the place where OCP binds/interacts with the PBS.
The specific roles or functional associations of different amino acids in the light activation process, or in NPQ, indicate that light activation and photoprotection are not interconnected processes, although some amino acids could play a dual role.Since many amino acids have a direct effect on light activation, the process must be driven by a cascade of protein-conformational changes, initiated by carotenoid light-absorption.This cascade of events will be discussed in the next section.Further discussion of the possible quenching mechanisms of PSBs by OCP has been presented previously [13].

Global Structural Changes
The molecular weight (MW) of OCP O was originally measured to be 34.6 kDa by matrix-assisted laser desorption/ionization mass-spectrometry (MALDI-MS) and electrospray techniques, and was later confirmed by the resolved X-ray structure [7,20], while the MW of OCP R is still under debate.Solution studies using small angle X-ray scattering (SAXS), and dynamic light scattering, demonstrated a large difference in the global structure of OCP O and OCP R , with OCP R showing a ~30% and ~50% increase in the radius of gyration and maximal dimension, respectively [21].In all studies, size exclusion chromatography (SEC) of OCP R gives an earlier eluting peak than for OCP O , indicating a larger hydrodynamic volume for OCP R [21,22].Leverenz and co-workers predicted globular mass of 43 kDa, while Gupta and co-workers found it to be 63 kDa.The discrepancy is intriguing and seems to arise from different illumination protocols.Different illumination schemes were used in these two studies, suggesting that the resolved MW might correspond to MW of different intermediates from the OCP O -OCP R photocycle.But more importantly, both studies observed an increase in various measurements of the OCP R size, as compared to OCP O .
Far-UV CD measurements reveal global changes in secondary structure of OCP upon light illumination [23,24].Comparison of far-UV CD measurements in OCP O and OCP R demonstrate only minor changes in CD signal, indicating that the secondary structure of OCP R is hardly different from OCP O [21,22].Using several algorithms to calculate the predicted secondary structure of OCP O and OCP R , from the experimental far-UV CD curves, Gupta et al. interpreted minor changes as loss of only a small portion of the protein's α-helical structure (2% α-helical content = ~6 residues) after illumination [21].Mass spectrometry foot printing revealed that salt bridging between R155 and E244 weakens or completely disrupts during OCP O to OCP R transition.All these results point towards CTD and NTD separation [21,22,25].
Keto-carotenoids are nearly symmetric, and thus, normally inactive as seen by VIS CD [26].However, the binding of a keto-carotenoid to the protein introduces asymmetry into the chromophore, giving rise to a VIS CD signal.The VIS CD spectra of the OCP R are substantially lower than OCP O , suggesting that CTD from NTD separation upon light activation causes lowering of keto-carotenoid asymmetry, and less protein-chromophore interactions [27].There have so far been no successful attempts to obtain crystal structures of the fully active OCP R form to unambiguously resolve the change in protein-chromophore interactions after light activation.However, it was discovered that NTD as a truncated fragment can bind keto-carotenoid, giving rise to red carotenoid protein (RCP) [15,22].Comparisons between the crystal structures of OCP and RCP has led to a recent proposal that carotenoid undergoes a 12-Å translocation relative to the protein framework, in forming photoactivated OCP R [15].However, it is important to learn whether the 12-Å carotenoid translocation occurs in a native OCP R , which steps of structural events trigger such a substantial change in keto-carotenoid coordinates, and, in particular, how it affects the protein structure and protein binding interface.
To summarize, OCP R represents a molten-globule state, the intermediate of protein folding (Figure 3), which has a secondary structure similar to the native form.This state is characterized by CTD and NTD domain separation, caused by modifications of carotenoidprotein interactions, and as a result, salt bridge disruption and detachment of the αA helix [21,28].

Changes in the Secondary and Tertiary Structure
Two structural approaches were used to resolve changes in the secondary and tertiary structure of OCP, upon illumination.Hydrogen/deuterium Exchange Mass Spectrometry (HDX-MS), together with X-ray protein foot printing (XF-MS), were used to resolve solution phase protein conformation and differences [21].This study was later complimented, with the first result demonstrating structural changes on an atomistic level, using X-ray crystallography in illuminated OCP crystals [16].X-ray protein foot printing (XF-MS), and HDX-MS techniques, demonstrate that activation of the OCP results in disruption of the minor interface, the interaction between the N-terminal α-helix and the solvent-exposed β-sheet of the CTD, via the unfolding of the αA (NTE): XF-MS showed a significant increase in accessibility in OCP R relative to OCP O at the minor interface; NTD residues A8-R9 and P13-P22 on the loop joining αA to αB by, commensurate with the corresponding residues V301-A302-I303 on the surface of the β-sheet facing helix αA.High water mobility was also observed.Two clusters of highly mobile waters are centered on the domain interface and close to β2-ring, playing a role in domain separation (Figure 4a).
Furthermore, a substantial decrease in solvent accessibility upon illumination was observed at residue clusters, which are in the carotenoid vicinity in NTD (W41-Y42-Y44, M47, M74-P76-M83, and M117), concomitantly with an increase in solvent accessibility in CTD residue clusters surrounding carotenoid in CTD (P276-W277-F278-M284 and F290), supporting the hypothesis of carotenoid translocation from CTD to NTD during light activation [15].However, surprisingly, residues involved in H-bonding with the β1-ring of keto-carotenoid, Trp-201 and Tyr-288, were absolutely conserved in solvent accessibility, suggesting that H-bond is still present in the illuminated state.Furthermore, residue Phe-240, which projects toward the β1-ring of carotenoid, and Phe-241, which projects toward the core of CTD, also showed no modification.These observations raise an

Changes in the Secondary and Tertiary Structure
Two structural approaches were used to resolve changes in the secondary and tertiary structure of OCP, upon illumination.Hydrogen/deuterium Exchange Mass Spectrometry (HDX-MS), together with X-ray protein foot printing (XF-MS), were used to resolve solution phase protein conformation and differences [21].This study was later complimented, with the first result demonstrating structural changes on an atomistic level, using X-ray crystallography in illuminated OCP crystals [16].X-ray protein foot printing (XF-MS), and HDX-MS techniques, demonstrate that activation of the OCP results in disruption of the minor interface, the interaction between the N-terminal α-helix and the solvent-exposed β-sheet of the CTD, via the unfolding of the αA (NTE): XF-MS showed a significant increase in accessibility in OCP R relative to OCP O at the minor interface; NTD residues A8-R9 and P13-P22 on the loop joining αA to αB by, commensurate with the corresponding residues V301-A302-I303 on the surface of the β-sheet facing helix αA.High water mobility was also observed.Two clusters of highly mobile waters are centered on the domain interface and close to β2-ring, playing a role in domain separation (Figure 4a).
Furthermore, a substantial decrease in solvent accessibility upon illumination was observed at residue clusters, which are in the carotenoid vicinity in NTD (W41-Y42-Y44, M47, M74-P76-M83, and M117), concomitantly with an increase in solvent accessibility in CTD residue clusters surrounding carotenoid in CTD (P276-W277-F278-M284 and F290), supporting the hypothesis of carotenoid translocation from CTD to NTD during light activation [15].However, surprisingly, residues involved in H-bonding with the β1-ring of keto-carotenoid, Trp-201 and Tyr-288, were absolutely conserved in solvent accessibility, suggesting that H-bond is still present in the illuminated state.Furthermore, residue Phe-240, which projects toward the β1-ring of carotenoid, and Phe-241, which projects toward the core of CTD, also showed no modification.These observations raise an important question: the XF-MS and HDX-MS results suggest that even if the translocation (partially) occurs during OCP O to OCP R transition, it should not be substantial, since the hydrogen bond between β1, Tyr-288 and Trp-201 appears to remain and would not be ruptured in OCP R .An alternative explanation is that the illuminated state in the study is not the final OCP R state, suggesting that H-bond rupture is not the primary photoactivation event, but occurs as a result of domain separation in later stages.
Physchem 2022, 2, FOR PEER REVIEW 6 not the final OCP R state, suggesting that H-bond rupture is not the primary photoactivation event, but occurs as a result of domain separation in later stages.Recently, Bandara and co-workers demonstrated for the first time that OCP could be light-activated in the crystalline state, by showing Fourier difference maps between dark and light states (the crystals were illuminated for 10-15 min at cryo temperatures 100-180 K).While structural differences were demonstrated, the crystals did not show UV/VIS spectroscopic differences upon illumination.This important result provided further opportunities for more in-depth studies of OCP using crystallography.The authors observed that difference densities were mostly localized in CTD, suggesting that CTD is the primary domain responsible for structural rearrangements.In contrast to the small-amplitude movement in NTD (<0.8 Å), CTD underwent significant atomic displacements (as large as 2 Å) in a direction away from NTD and the dimer interface (Figure 4b).Three helices (αK, αL and αM) and the central β-sheet of CTD move together like a rigid body, away from the NTD.As a result of that movement, hydrogen bonds from Recently, Bandara and co-workers demonstrated for the first time that OCP could be light-activated in the crystalline state, by showing Fourier difference maps between dark and light states (the crystals were illuminated for 10-15 min at cryo temperatures 100-180 K).While structural differences were demonstrated, the crystals did not show UV/VIS spectroscopic differences upon illumination.This important result provided further opportunities for more in-depth studies of OCP using crystallography.The authors observed that difference densities were mostly localized in CTD, suggesting that CTD is the primary domain responsible for structural rearrangements.In contrast to the smallamplitude movement in NTD (<0.8 Å), CTD underwent significant atomic displacements (as large as 2 Å) in a direction away from NTD and the dimer interface (Figure 4b).Three helices (αK, αL and αM) and the central β-sheet of CTD move together like a rigid body, away from the NTD.As a result of that movement, hydrogen bonds from the 4-keto group to Tyr201 and Trp288 become longer by 0.2-0.4Å.Both the raw and SVD-filtered difference maps show evenly spaced positive and negative difference densities associated with carotenoid, suggesting a small shift roughly along the conjugated plane of the polyene chain.The authors also observed movements associated with water molecules at the dimer interface, where a network of hydrogen bonds and the Arg155-Glu244 salt bridge are altered.Based on the elongation of the H-bond by 0.2-0.4Å, the authors proposed that the H-bond eventually gets ruptured, possibly due to the shift in keto-enol equilibrium; however, it could be a secondary effect resulting from domain separation, in agreement with observations from Gupta et al. [21].
Both structural studies showed a separation of CTD from NTD, which was observed together with the presence of highly-mobile water molecules at the dimer interface [16,21].This interface also played a role in domain separation and the modification of the Arg155-Glu244 salt bridge.In the work of Bandara and co-workers, this process is not accompanied by unfolding of αA [16], unlike reported by Gupta et al. [21].Since large scale protein and/or cofactor movements would be restrained by crystal lattice, the OCP R state in crystalline likely resembles an intermediate state of the OCP O -OCP R photocycle in solution.All in all, there are some discrepancies between the results of structural studies of OCP activation [16,21]; a full structural mechanism and determination of intermediates is still lacking.

OCP O -OCP R -OCP O Photocycle
The structural studies reported above aimed to unravel the structural differences between OCP O and OCP R , but did not provide dynamics and kinetics information for these changes.Time-resolved spectroscopy has been an established technique in studying kinetics of the photoactivation and relaxation processes within photosensitive proteins [29][30][31][32].One of the main technical difficulties in applying time-resolved spectroscopy techniques for OCP research is the very low OCP O -OCP R conversion efficiency (1-3%) [8,33], and quite long OCP R -OCP O relaxation time (minutes).To overcome these issues, either relatively long excitation pulses (from several ps to ns) or continues illumination should be used, in combination with efficient sample refreshment using a Lissajous sample scanner, or flow-through cuvette.

Flash Photolysis
Maksimov and co-workers reported the first study of the OCP O -OCP R -OCP O photocycle with subsecond resolution [33].They performed nanosecond flash photolysis to monitor hECN absorption changes, as well as protein structural changes.The latter was probed by time-resolved Trp autofluorescence measurements, and time-resolved fluorescence measurement of Nile Red label, the hydrophobicity reporter dye.The OCP O to OCP R transition, as seen by 550 nm OD changes, was single exponential and its rate was recorded to be fluxdepended, while OCP R -OCP O relaxation in darkness was biexponential.The characteristic relaxation times were 13 s and 66 s with an amplitude ratio of 3:1 in the region between 20 and 45 • C, and of 3:2 at lower temperatures (5-17 • C) [33].Changes in OCP O -OCP R -OCP O kinetics, as monitored by OD change recorded at 550 nm OD, followed the changes in Trp fluorescence yield [33].This correlation suggested the principal role of carotenoid-protein interactions or, at least, carotenoid-Trp interactions in the formation of OCP photocycle.From these measurements it is not possible to determine which Trp contributes to these NR fluorescence quenching and recovery; therefore, the OCP O -OCP R -OCP O photocycle also involves some hydrophobic areas getting exposed, then concealed.Recently, the same authors resolved two additional timesteps and assigned them to actual intermediates in the OCP photocycle [34].The first one is an intermediate OCP O -OCP R state >>200 ns (below instrument resolution).This intermediate is assigned to a protein conformation close to OCP O , but carotenoid is in its final red state, according to hECH absorption measurements; however, no Trp autofluorescence could be reported for this state, due to instrument limitations, which could confirm the protein-conformation assignment.The authors also resolved a recovery step of 3 s, when NTD and CTD are properly oriented against each other (likely to be assisted by Fluorescence Recovery Protein in vivo), as concluded from the combination of hECN absorption changes and Trp autofluorescence transients.The authors suggest that this step is a prerequisite for the hydrogen bond formation between Tyr-201 and Trp-288, and the keto-group of carotenoid [34].The knowledge of the OCP O -OCP R -OCP O photocycle, based on the flash photolysis studies [33,34], is summarized in Figure 5.
Physchem 2022, 2, FOR PEER REVIEW changes in Trp fluorescence during OCP O -OCP R -OCP O followed NR fluorescence quenching and recovery; therefore, the OCP O -OCP R -OCP O photocycle also involves some hydrophobic areas getting exposed, then concealed.Recently, the same authors resolved two additional timesteps and assigned them to actual intermediates in the OCP photo cycle [34].The first one is an intermediate OCP O -OCP R state >>200 ns (below instrumen resolution).This intermediate is assigned to a protein conformation close to OCP O , bu carotenoid is in its final red state, according to hECH absorption measurements; howev er, no Trp autofluorescence could be reported for this state, due to instrument limitations which could confirm the protein-conformation assignment.The authors also resolved a recovery step of 3 s, when NTD and CTD are properly oriented against each other (likely to be assisted by Fluorescence Recovery Protein in vivo), as concluded from the combi nation of hECN absorption changes and Trp autofluorescence transients.The author suggest that this step is a prerequisite for the hydrogen bond formation between Tyr-201 and Trp-288, and the keto-group of carotenoid [34].The knowledge of the OCP O -OCP R -OCP O photocycle, based on the flash photolysis studies [33,34], is summa rized in Figure 5.

Intermediates Resolved by Transient Absorption Spectroscopy and FTIR
Konold and co-workers, using femtosecond to microsecond visible, and infrared transient absorption spectroscopy for the first time, were able to identify ultrafast struc tural intermediates (in ps-µs range) in the OCP O -OCP R pathway, which had not been observed before [35].
The first structurally defined intermediate rises in 23 ps and is characterized by a replacement of 1685 cm −1 bleach, which is present at 5 ps and assigned to C=O stretch o β1-ring, by a smaller stretch mode at 1680 cm −1 (Figures 5 and 6a).This 5 cm −1 shift and loss in intensity were assigned to the loss of a hydrogen bond between keto-group, and Trp-288 and Tyr-201.This assignment is based on previous findings suggesting that OCP light activation is triggered by H-bond rupture between the keto-group of β1 ring and the two amino acids [16], and on the assumption that C=O stretch in OCP is substantially red shifted than in the solution, due to strong bonding [35,37].The next state arises in 0.

Intermediates Resolved by Transient Absorption Spectroscopy and FTIR
Konold and co-workers, using femtosecond to microsecond visible, and infrared transient absorption spectroscopy for the first time, were able to identify ultrafast structural intermediates (in ps-µs range) in the OCP O -OCP R pathway, which had not been observed before [35].
The first structurally defined intermediate rises in 23 ps and is characterized by a replacement of 1685 cm −1 bleach, which is present at 5 ps and assigned to C=O stretch of β1-ring, by a smaller stretch mode at 1680 cm −1 (Figures 5 and 6a).This 5 cm −1 shift and loss in intensity were assigned to the loss of a hydrogen bond between keto-group, and Trp-288 and Tyr-201.This assignment is based on previous findings suggesting that OCP light activation is triggered by H-bond rupture between the keto-group of β1 ring and the two amino acids [16], and on the assumption that C=O stretch in OCP is substantially red shifted than in the solution, due to strong bonding [35,37].The next state arises in 0.5-1.1 µs and manifests itself with a broad and strong bleach centered around 1655 cm −1 with no downshift upon H/D exchange, with shoulders near 1680 and 1635 cm −1 , and nearly zero amplitude elsewhere (Figure 6a).Due to the spectra, and the fact that the band is still present in ∆NTE mutant, it was assigned to amide I band, arising from the loss of nonsolvent-exposed α-helix.The authors assign it to the NTD.In 10 µs a formation of the red carotenoid state is observed, which is characteristic for OCP R , but it is not accompanied by any changes in the IR region; therefore, this state affects only Car conformation and not the protein environment.Based on the proposed hypothesis of the carotenoid 12 Å translocation into NTD as a result of OCP R conversion, Konold and co-workers suggest that this OCP "red" state can be explained by carotenoid translocation into NTD.

22, 2, FOR PEER REVIEW
downshift upon H/D exchange, with shoulders near 1680 and 1635 cm −1 , and nearly amplitude elsewhere (Figure 6a).Due to the spectra, and the fact that the band i present in ΔNTE mutant, it was assigned to amide I band, arising from the loss of solvent-exposed α-helix.The authors assign it to the NTD.In 10 µs a formation of th carotenoid state is observed, which is characteristic for OCP R , but it is not accompa by any changes in the IR region; therefore, this state affects only Car conformation not the protein environment.Based on the proposed hypothesis of the carotenoid translocation into NTD as a result of OCP R conversion, Konold and co-workers su that this OCP "red" state can be explained by carotenoid translocation into NTD.Recently, details of the OCP O -OCP R -OCP O photocycle were studied by time-resolved FTIR Difference spectroscopy study, under continuous illumination Because of continuous illumination, intrinsic lifetimes cannot be assessed directly, bu measurements do resolve the change in population accumulation rates at given il nation power densities.The authors found two different conformational states d OCP O to OCP R conversion: The first process dominates in the beginning, after the o and involves partial reorganization of nonsolvent-exposed α-helix, having a vibr band at 1651 cm −1 (Figure 6b), which is virtually identical to the 1655 cm −1 contributi the 0.5-1.1 µs-step resolved by Konold et al. [35].In both cases the process was assi to unfolding of the α-helix in NTD.On the same timescales, a positive band at ~1619 of unknown identity was resolved, which, based on the analysis of difference sp recorded in H2O and D2O, was tentatively assigned to movement of water molec which may play a key role in the photoactivation mechanism, according to XF HDX-MS and crystallographic results.The second step becomes dominant later, m festing itself with 1644 cm −1 bleach, and involves a solvent-exposed α-helix.Therefo was assigned to NTE.

Open Questions
The summarized knowledge, based on different spectroscopic studies rega Recently, details of the OCP O -OCP R -OCP O photocycle were studied by the timeresolved FTIR Difference spectroscopy study, under continuous illumination [36].Because of continuous illumination, intrinsic lifetimes cannot be assessed directly, but the measurements do resolve the change in population accumulation rates at given illumination power densities.The authors found two different conformational states during OCP O to OCP R conversion: The first process dominates in the beginning, after the onset, and involves partial reorganization of nonsolvent-exposed α-helix, having a vibration band at 1651 cm −1 (Figure 6b), which is virtually identical to the 1655 cm −1 contributing to the 0.5-1.1 µs-step resolved by Konold et al. [35].In both cases the process was assigned to unfolding of the α-helix in NTD.On the same timescales, a positive band at ~1619 cm −1 of unknown identity was resolved, which, based on the analysis of difference spectra recorded in H 2 O and D 2 O, was tentatively assigned to movement of water molecules, which may play a key role in the photoactivation mechanism, according to XF-MS, HDX-MS and crystallographic results.The second step becomes dominant later, manifesting itself with 1644 cm −1 bleach, and involves a solvent-exposed α-helix.Therefore, it was assigned to NTE.

Open Questions
The summarized knowledge, based on different spectroscopic studies regarding OCP O -OCP R -OCP O photocycle, is presented on Figure 5.The key questions currently discussed in the literature regarding OCP photocycle are: (1) the time when the OCP R photoproduct is formed and; (2) what mechanism is causing its red shift.Konold et al. observed the photoproduct formation in 10 µs after excitation, but it is not accompanied by any detectable changes in the IR region; therefore, they assign it to carotenoid translocation into NTD [35].Maksimov and colleagues, however, report the red state of carotenoid to be formed >>200 ns, and assign it to the hydrogen bond rupture between Tyr-201, Trp-288 and the keto-group of β1 ring, while protein conformation is suggested to be close to OCP O [34].
Konold et al., on the other hand, suggest that this H-bond rupture occurs in 23 ps and is manifested by a loss of 1685 cm −1 band and its 5 cm −1 downshift, while no changes in visible spectrum seem to be associated with this state [35].The presence of the 23 ps time constant that shows spectroscopic differences at 1680 cm −1 character, raises questions and requires additional investigation, to confirm its assignment to C=O stretch and H-bond rupture of the keto-group in the β1 ring, since IR studies on isolated ECH in solution report different absorption frequencies for C=O stretch [38,39].The assignment of certain IR frequencies to specific structural rearrangements is challenging, particularly with large protein structures.Therefore, it will be necessary to investigate and exclude the possibility of other pigment-protein interactions contributing to the 23 ps state, for example, a light activation mechanism located in the β2 ring, according to recent findings [40] or other proposed trigger mechanisms (see Section 6).
Both IR TA and FTIR studies observe bleach of 1651-1655 cm −1 upon OCP light activation, which is assigned to (partial) unfolding of nonsolvent-exposed alfa-helix, most likely in NTD.However, far-UV CD, demonstrated loss of only a small portion of the protein's α-helical structure (2% α-helical content = ~6 residues) after illumination, which was assigned to NTE based on HDX-MS and XF-MS results, while NTD stayed mainly unchanged, except for minor changes of amino acids in the carotenoid vicinity [21].Due to this variation in IR and HDX/XF-MS signatures, it is very inviting to consider whether the bleach of 1651-1655 cm −1 upon OCP light activation is actually a signature of alfa-helix unfolding, or some other change in the secondary or tertiary alfa-helix structure in NTD or CTD.

The Mechanism and Trigger of Light Activation
Most research groups have concluded that OCP photoactivation is initiated by H-bond rupture [15,16,18,19,35,41].This proposed mechanism is partially based on multiple mutagenesis studies, demonstrating that upon Tyr-201 or Trp-288 mutation in the mutated OCPs, although still binding keto-carotenoids, OCP R cannot be accumulated [16,17].Partially, it is also based on the difference maps obtained from dark and illuminated crystals, where the distance between the 4-keto group to Tyr201 and Trp288, becomes longer by 0.2-0.4Å upon illumination.
There are still two important questions which need to be answered: (1) What is the driving force behind this H-bond rupture?(2) Is H-bond rupture the first light activation reaction?Bandara and co-workers hypothesized that keto-carotenoid illumination results in a shift in the keto-enol equilibrium, which weakens or disrupts the hydrogen bonds between the β1 ring and protein moiety (Figure 7).The proposed enolization mechanism implies sp2→sp3 orbital transition at the C6 position of the carotenoid, which was not directly observed in the difference maps.However, crystal structures were obtained after 10-15 min of illumination; therefore, transient signals, if present, could not be resolved.Another proposed mechanism of H-bond rupture concerns photo-induced s-isomerization of the C6−C7 single bond, which would flip the orientation of the β1-ring by 90 • , going from an s-trans to an s-cis conformation.Such photo-induced isomerization mechanisms, which rupture or rearrange hydrogen-bonding networks, are commonly in effect in photoreceptor proteins, such as microbial rhodopsins, photoactive yellow protein, and phytochromes [42][43][44].Support for such a mechanism in OCP came from X-ray structure of the isolated N-terminal domain (also known as red carotenoid protein, RCP) where keto-carotenoid was present in a C6−C7 s-cis conformation [15], and from resonant Raman spectroscopy, where such a conformation would be one of the structures consistent with OCP R Raman spectra [39].DFT calculations suggest that there are three additional metastable configurations of the terminal ring, which retain hydrogen bonding to Trp and Tyr, with varying C6-C7 dihedral angles (50 • , 80 • , 130 • ) between the polyene chain and the β1-ring [33].The energy barrier for a β1-ring rotation from 130 • to 50 • is identical to the value of activation energy (1.7 kcal/mol at high temperatures), estimated from Arrhenius plots during OCP O -OCP R photoconversion [33].The change of torsional angle of the β ring in lutein was recently suggested to be responsible for alternation of the carotenoid energetic landscape, to trigger quenching mechanism in plants [45].In contradiction to these conclusions, Raman optical activity experiments from Konold and co-workers essentially demonstrated identical anisotropy for OCP O and OCP R states (0.23 and 0.20, respectively), showing that the angle (Θ) between electronic and vibrational transition dipoles states is around the same in both states.It is therefore likely that carotenoid should stay in C6−C7 trans conformation.Based on femtosecond transient absorption data, Konold and co-workers suggest that a specific carotenoid state, typically denoted as the S* state, is the key state leading to disruption of hydrogen bonds [35].The S* state of long (having more than 11 conjugated C=C bonds) carotenoids is either related to structurally distorted form of the optically forbidden state S1 [35], or to a hot carotenoid ground state that dissipates its energy to vibrational modes of the amino acids near carotenoid [46].Two recent time-resolved spectroscopic studies also proposed the existence of the S* state as a precursor of the first OCP photoproduct [47,48].In addition, both works also observed carotenoid cation formation.The increase of the S* formation yield seems to go in parallel with cation increase upon UV excitation [47].However, Nizi ński and co-workers demonstrate the oppositive phenomenon, where OCP photoproduct and cation formation are two decoupled phenomena, upon VIS light excitation [48].Therefore, it is likely that the S* and cation formation play different roles in OCP function.
Physchem 2022, 2, FOR PEER REVIEW 11 identical to the value of activation energy (1.7 kcal/mol at high temperatures), estimated from Arrhenius plots during OCP O -OCP R photoconversion [33].The change of torsional angle of the β ring in lutein was recently suggested to be responsible for alternation of the carotenoid energetic landscape, to trigger quenching mechanism in plants [45].In contradiction to these conclusions, Raman optical activity experiments from Konold and co-workers essentially demonstrated identical anisotropy for OCP O and OCP R states (0.23 and 0.20, respectively), showing that the angle (Θ) between electronic and vibrational transition dipoles states is around the same in both states.It is therefore likely that carotenoid should stay in C6−C7 trans conformation.Based on femtosecond transient absorption data, Konold and co-workers suggest that a specific carotenoid state, typically denoted as the S* state, is the key state leading to disruption of hydrogen bonds [35].The S* state of long (having more than 11 conjugated C=C bonds) carotenoids is either related to structurally distorted form of the optically forbidden state S1 [35], or to a hot carotenoid ground state that dissipates its energy to vibrational modes of the amino acids near carotenoid [46].Two recent time-resolved spectroscopic studies also proposed the existence of the S* state as a precursor of the first OCP photoproduct [47,48].In addition, both works also observed carotenoid cation formation.The increase of the S* formation yield seems to go in parallel with cation increase upon UV excitation [47].However, Niziński and co-workers demonstrate the oppositive phenomenon, where OCP photoproduct and cation formation are two decoupled phenomena, upon VIS light excitation [48].Therefore, it is likely that the S* and cation formation play different roles in OCP function.HDX/XF-MS study suggests that H-bond rupture is not the primary event of OCP light activation and, therefore, S*-unrelated, but occurs after domain separation [21].In  HDX/XF-MS study suggests that H-bond rupture is not the primary event of OCP light activation and, therefore, S*-unrelated, but occurs after domain separation [21].In line with that, by studying fluorescence line shapes, quantum yields, and system−bath couplings in hECN and CAN in OCP O , Gurchiek et al. suggested that the fluorescence spectra and yields and, therefore, S2 potential surface, are controlled by friction and are modulated by specific interactions of the β2 rings with Leu37 in the NTD of OCP O .It would suggest that the mechanism of OCP photoactivation is β1-insenstative, and involves out-of-plane motions of the β2 ring.

OCP Excited State Dynamics
A range of time-resolved spectroscopic studies have been performed over the last 15 years to describe the energetic landscape of OCP in its two conformations (OCP O and OCP R ), and in RCP [49][50][51][52][53].The structural mechanism of activation is expected to be initiated via the S 1 /ICT (2 1 A g − ) dark state that is created in 150 fs from the optically allowed S 2 (1 1 B u + ), and has a 3.5 ps lifetime [49] (Figure 8).

OCP Excited State Dynamics
A range of time-resolved spectroscopic studies have been performed over the l years to describe the energetic landscape of OCP in its two conformations (OCP OCP R ), and in RCP [49][50][51][52][53],.The structural mechanism of activation is expected to b tiated via the S1/ICT (2 1 Ag − ) dark state that is created in 150 fs from the optically all S2 (1 1 Bu + ), and has a 3.5 ps lifetime [49] (Figure 8).The lowest excited state (S1 (2 1 Ag − )) lifetime of hECN, which in carbonyl carote is usually dramatically changed by the solvent polarity [54,55], was reported to be a identical both in polar, and non-polar solvents, indicating that it exhibits the we polarity-induced effects observed so far among carbonyl carotenoids [52].In an study, the absence of any polarity-induced shortening of the S1 was observed in thaxanthin (CAN) and echinenone (ECH), another two keto-carotenoids, which can to OCP [56].On the other hand, the conformational change and hydrogen bondin the carbonyl group of keto-carotenoid to OCP, results in dramatic shortening of OCP O to 3.3 ps, instead of 6.5 ps observed for hECN in solution [52].This shorten due to the change in β1 orientation and, as a result, in effective conjugation leng OCP O it takes the s-trans configuration, while s-cis is the lowest energy configurat solution [57].The first time-resolved study on OCP claimed the S1 state of hECN in to decay with two time constants of 0.9 and 3.3 ps [52].The multiexponential character of S1 state, however, is currently unresolved and is important.

Intramolecular Charge Transfer
It has been reported by several groups that the excited state absorption of the S and ICT → Sn transitions contribute to broad spectral features in the 550-700 nm [8,35,52].The intramolecular charge transfer (ICT) state is a characteristic feature o tain carbonyl carotenoids, and is readily observed in transient absorption spectra separate band appearing only in polar solvents [54,55].However, keto-carotenoids long (N > 11) conjugation, and with a conjugated keto-group located at the termina such as hECN, ECN or CAN, typically do not exhibit such a band [56].Multiple st however, demonstrate the presence of the ICT band upon binding these caroteno OCP, indicating that the OCP binding pocket activates the ICT state [49][50][51].Berer co-workers were the first to assign 0.6 ps lifetime to ICT, due to the broad band in th to 800 nm region, which becomes more pronounced in OCP R .The second componen ps) was assigned to S1 excited state, resolved both in OCP O  The lowest excited state (S 1 (2 1 A g − )) lifetime of hECN, which in carbonyl carotenoids is usually dramatically changed by the solvent polarity [54,55], was reported to be almost identical both in polar, and non-polar solvents, indicating that it exhibits the weakest polarity-induced effects observed so far among carbonyl carotenoids [52].In another study, the absence of any polarity-induced shortening of the S1 was observed in canthaxanthin (CAN) and echinenone (ECH), another two keto-carotenoids, which can bind to OCP [56].On the other hand, the conformational change and hydrogen bonding via the carbonyl group of keto-carotenoid to OCP, results in dramatic shortening of S1 in OCP O to 3.3 ps, instead of 6.5 ps observed for hECN in solution [52].This shortening is due to the change in β1 orientation and, as a result, in effective conjugation length: in OCP O it takes the s-trans configuration, while s-cis is the lowest energy configuration in solution [57].The first time-resolved study on OCP claimed the S1 state of hECN in OCP to decay with two time constants of 0.9 and 3.3 ps [52].The multiexponential decay character of S1 state, however, is currently unresolved and is important.

Intramolecular Charge Transfer
It has been reported by several groups that the excited state absorption of the S 1 → S n and ICT → S n transitions contribute to broad spectral features in the 550-700 nm range [8,35,52].The intramolecular charge transfer (ICT) state is a characteristic feature of certain carbonyl carotenoids, and is readily observed in transient absorption spectra as a separate band appearing only in polar solvents [54,55].However, keto-carotenoids with long (N > 11) conjugation, and with a conjugated keto-group located at the terminal ring, such as hECN, ECN or CAN, typically do not exhibit such a band [56].Multiple studies, however, demonstrate the presence of the ICT band upon binding these carotenoids to OCP, indicating that the OCP binding pocket activates the ICT state [49][50][51].Berera and co-workers were the first to assign 0.6 ps lifetime to ICT, due to the broad band in the 670 to 800 nm region, which becomes more pronounced in OCP R .The second component (3.5 ps) was assigned to S1 excited state, resolved both in OCP O and OCP R [49].
To exclude any OCP O -OCP R conversion during the measurements, the same authors later performed transient absorption studies on OCP O and OCP R , locked in their activation stages via freezing at 77 K.The results again demonstrated that both OCP forms show two intrinsic decay rates of the S1/ICT state of hECN.Moreover, the decay lifetimes seem to be almost identical in both OCPs (2.0 and 7.5 ps in OCP O and 2.3 and 7.6 ps in OCP R ) [51].
The predominantly ICT character has a lifetime of 2 ps and becomes more dominant in OCP R , and another state with mainly S1 character has a lifetime of around 7.5 ps (Figure 8 model 1).While the spectroscopic features related to the ICT were reported to be even more pronounced in OCP R [49,51] or RCP [50], no significant shortening of the overall lifetime in OCP R versus OCP O was observed [58].

Heterogeneity
Besides the role of the ICT state in the multiexponential character of S1, carotenoid heterogeneity once bound to OCP is another ongoing discussion in the literature.Slight differences in spectral and dynamical responses, obtained after excitation of OCP O at 495 and 530 nm, were reported in 2005, where S1 state of hECN in OCP decayed with two time constants of 0.9 and 3.3 ps [52].These lifetimes were originally explained by the presence of two populations of hECN in OCP, which differ in hydrogen bonding (Figure 8 model 2).In another study, the same authors suggested that the presence of two decay rates is associated with the coexistence of two forms of OCP, based on RT transient absorption studies, performed in VIS-NIR, where 0.9 ps is assigned to S1 of OCP R and 3.3 ps to S1 of OCP O [59].By deconvoluting the absorption spectra of OCP O and OCP R , and separating S0-S2 and S2-S1 contributions, the authors estimated S1 energy of hECN in OCP O and in OCP R to be 14,700 cm −1 and 14,000 cm −1 , respectively, indicating that significant changes in the vicinity of the conjugated carbonyl group occur upon activation [59].However, transient absorption data measured on OCP binding zeaxanthin, which is incapable of OCP O -OCP R conversion [17], demonstrated that even for this OCP, which is locked in OCP O state, two spectroscopically distinct zeaxanthins were identified [50].The presence of two subpopulations within the same class of OCP was concluded from 77 K transient absorption studies, where one population has more enhance ICT character than the other [53].Until now, the possible origin of this OCP heterogeneity has been unclear.For example, based on Resonance Raman spectra and DFT calculations, Maksimov et al. suggested that there are additional metastable configurations of the terminal ring, which retain hydrogen bonding to Trp and Tyr with varying C6-C7 dihedral angles (50 • , 80 • , 130 • ) between the polyene chain and the β1-ring [33].

RCP
The S1 lifetime of hECN in RCP (RCP-hECH) is 5.5 ps, which is markedly longer than in OCP O (3.3 ps), but close to the 6 ps obtained for hECN in organic solvent.This suggests that the binding of hECN to the C-terminal domain in OCP is primarily responsible for tuning its excited state properties [26].In contrast to that, RCP-CAN has a very similar S1 lifetime to its OCP O state.Furthermore, while RCP-CAN has an active ICT state [50], RCP-hECN does not [26].In addition, the presence of ICT in RCP-CAN hardly affects its ultrafast kinetics, compared to OCP O with the same bound carotenoid [57].These observations seem to indicate that the energy landscape of keto-carotenoid in OCP has a carotenoid-specificity, and both domains (CTD, NTD) have important roles in providing this specificity.

Summary and Open Questions
While some authors assign the multiexponential character of the S1 excited state in OCP O and OCP R to OCP population heterogeneity [50,52,53], and others to the different contribution of ICT to the excited state kinetics [49][50][51], none of the mechanisms can relate these changes to physiological phenomena.For example, presence of ICT in RCP-CAN hardly affects its ultrafast kinetics, compared to OCP O with the same carotenoid bound [50].Furthermore, while RCP-CAN has an active ICT state, RCP-hECH [26] has not, but both serve as excellent quenchers of PBS fluorescence [15,22].
A further consideration is that changes in the S1 excited state lifetime in OCP R , or in OCP O with different H-bond conformations and/or β1 ring torsions, is suggested by some studies not to contribute to a more efficient excited state quenching: The excited-state decay of carotenoid S1 state is three orders of magnitude faster, compared to the excited state decay of bilin in phycobilisome.Therefore, any carotenoid would serve as an excellent quencher for a bilin, with essentially identical quenching efficiency.
These observations seem to suggest that excited state heterogeneity and ICT state are a consequence of the fine-tuning of S1 energy, through ring rotations and the conformation changes of keto-carotenoid bound to OCP.Due to the absence of unambiguous evidence that OCP quenches PBS energy directly, it raises an important question: Could it be that these changes in OCP excited state properties reflect structural changes, which are important for OCP binding to promote indirect quenching?
Another question which requires more in-depth study of the reasons and consequences of carotenoid-specificity on OCP excited state properties is: Does it affect light activation and/or antenna quenching and, if it does, which domains, amino acids, and bonds have an important role in these processes?

General Conclusions and Future Prospects
Over the last 15 years, since OCP structure was first resolved (which at the time was a protein of unknown function), many advances in our understanding of its role in cyanobacterial photoprotection have been achieved.
Overall, upon light activation, NTD and CTD are separated and carotenoid is translocated into NTD [16,21].Various amino acids (e.g., Tyr201, Trp288 or Trp41) play an active role in the light activation process [17][18][19]; however, the stage at which they contribute to the photocycle is still unclear.Different photocycle intermediates were observed spectroscopically, from a picosecond to a second timescale [33][34][35][36].There are still structural data missing to confirm any of the proposed models of the OCP photocycle; therefore, interpretations are provisional and based on spectroscopic evidence only.Different hypothesis about the trigger of light activation have been also put forward [15,16,18,19,35,41,45].Most suggest that the H-bond rupture is the primary photoevent [15,16,18,19,35,41]; however, of all the hypotheses proposed, high-resolution structural data is urgently needed to resolve the triggering mechanism.
To summarize, many questions regarding the mechanism of OCP light activation remain unanswered, or are under debate.We believe that the current key topics regarding OCP light activationare: 1.
Trigger mechanism: Based on either structural or spectroscopic studies, proposed candidates for the trigger of OCP light activation are (1) β1-ring flip, due to photoisomerization; (2) enolization of β1-ring and protein moiety; (3) H-bond rupture due to a "hot" excited state; (4) out-of-plane motions in β2-ring.None of these mechanisms were exclusively supported in the literature, and there many results pointing in favor of different models.Unambiguously resolving the trigger and location of OCP light activation is essential to understand, and potentially controlling the OCP activation mechanism.The combination of ultrafast time-resolved spectroscopy, and structural time-resolved techniques, should be exploited to investigate this point.

2.
Intermediates: To fully understand the process and potential gain control of OCP light activation, we need to resolve all OCP intermediate states, and their lifetimes.For example, there are spectroscopic results suggesting that OCP reaches its red state in less than 0.2 µs, before major structural reorganizations (which seem to be mainly localized in NTD), while other results suggest that this process is 50 times slower, and happens after NTD reorganizations.It is also not clear when H-bond rupture occurs, what its spectroscopic signature is, and its role within light activation.
Other intermediates and lifetimes of OCP O -OCP R -OCP O photocycle still need to be either resolved or confirmed, based on spectroscopic characters and more advanced structural insights.

3.
Carotenoid/OCP specificity: Spectroscopic studies suggest keto-carotenoid specificity, affecting energetic landscape and excited state kinetics of OCP and RCP.Furthermore, genomic data are beginning to hint of a wider role for OCP homologs beyond

Figure 1 .
Figure 1.Structure of Orange Carotenoid Protein.(a) Crystal structure of Synechocystis sp.PCC6803 OCP (PDB ID code 5TV0) consisting of two domains, NTD and CTD, NTE and the loop as described in the main text introduction.(b) Amino acid residues within 4.5 Å of the carotenoid are shown by sticks.

Figure 1 .
Figure 1.Structure of Orange Carotenoid Protein.(a) Crystal structure of Synechocystis sp.PCC6803 OCP (PDB ID code 5TV0) consisting of two domains, NTD and CTD, NTE and the loop as described in the main text introduction.(b) Amino acid residues within 4.5 Å of the carotenoid are shown by sticks.

Figure 2 .
Figure 2. Tentative role of individual amino acids in OCP light activation in a close Car vicinity based on mutagenesis studies [17-19].

Figure 2 .
Figure 2. Tentative role of individual amino acids in OCP light activation in a close Car vicinity based on mutagenesis studies [17-19].

Figure 3 .
Figure 3. Current knowledge on OCP R structure.(a) Crystal structure of Synechocystis sp.PCC6803 OCP O (PDB ID code 5TV0) consisting of two domains, NTD and CTD, NTE and the loop as shown in Figure 1.The salt bridge R155-E244 plays an important role in connecting the two domains.(b)The model of OCP R by Leverenz et al.[15,22]: NTD and CTD are partially or entirely separated, promoted by salt bridge disruption, and keto-carotenoid is translocated into NTD[21].(c) Comparison of OCP O (5TV0, coloring according to Figure1) and RCP (amino acids and the keto-carotenoid are colored with cyan and red, respectively).(d) Superposition of NTD and RCP demonstrates 12 Å translocation of keto-carotenoid inside the protein network.This translocation is considered to be present in OCP R[15].

Figure 3 .
Figure 3. Current knowledge on OCP R structure.(a) Crystal structure of Synechocystis sp.PCC6803 OCP O (PDB ID code 5TV0) consisting of two domains, NTD and CTD, NTE and the loop as shown in Figure 1.The salt bridge R155-E244 plays an important role in connecting the two domains.(b) The model of OCP R by Leverenz et al. [15,22]: NTD and CTD are partially or entirely separated, promoted by salt bridge disruption, and keto-carotenoid is translocated into NTD [21].(c) Comparison of OCP O (5TV0, coloring according to Figure 1) and RCP (amino acids and the keto-carotenoid are colored with cyan and red, respectively).(d) Superposition of NTD and RCP demonstrates 12 Å translocation of keto-carotenoid inside the protein network.This translocation is considered to be present in OCP R [15].

Figure 4 .
Figure 4. Summary of light-induced structural changes in OCP resolved by HDX-MS and XF-MS [21], and X-ray crystallography [16].(a) Structural changes as resolved by HDX-MS and XF-MS [21].The amino acids which showed an increase and decrease in solvent accessibility are colored in blue and red, respectively.Amino acids which showed no change in solvent accessibility are colored in grey.(b) Structural changes as resolved by T-scan X-ray crystallography [16].Amino acids showing movement in illuminated crystals are colored in red.(c) Summary of the differences of the two studies.

Figure 4 .
Figure 4. Summary of light-induced structural changes in OCP resolved by HDX-MS and XF-MS [21], and X-ray crystallography [16].(a) Structural changes as resolved by HDX-MS and XF-MS [21].The amino acids which showed an increase and decrease in solvent accessibility are colored in blue and red, respectively.Amino acids which showed no change in solvent accessibility are colored in grey.(b) Structural changes as resolved by T-scan X-ray crystallography [16].Amino acids showing movement in illuminated crystals are colored in red.(c) Summary of the differences of the two studies.
dynamics and assign these signals to actual structural intermediates.Time-resolved Trp measurements in the ns range demonstrated an increase in the average lifetime and amplitude of OCP R , as compared to OCP O .The difference between the two fluorescence traces can be fit monoexponentially with 5 ns lifetime, indicating that only one type of fluorescent Trp interaction plays a role during OCP O -OCP R -OCP O conversion.It is accompanied with a 3 nm blue shift of the Trp spectral band in the OCP R state, indicating changes in the local Trp environment.These changes in Trp fluorescence during OCP O -OCP R -OCP O followed Physchem 2022, 2 242

Figure 6 .
Figure 6.IR time-resolved signals of OCP photocycle [35,36].(a) Evolution-associated diffe spectra determined from global analysis of IR transient absorption data, where ps pulses were to initiate the OCP photocycle.Reprinted with permission from [35].Copyright© 2018 Am Chemical Society.(b) FTIR difference spectra recorded during continuous illumination at dif times after light onset.Reprinted with permission from [36].Copyright© 2019 American Che Society.

Figure 6 .
Figure 6.IR time-resolved signals of OCP photocycle [35,36].(a) Evolution-associated difference spectra determined from global analysis of IR transient absorption data, where ps pulses were used to initiate the OCP photocycle.Reprinted with permission from [35].Copyright© 2018 American Chemical Society.(b) FTIR difference spectra recorded during continuous illumination at different times after light onset.Reprinted with permission from [36].Copyright© 2019 American Chemical Society.
line with that, by studying fluorescence line shapes, quantum yields, and system−bath couplings in hECN and CAN in OCP O , Gurchiek et al. suggested that the fluorescence spectra and yields and, therefore, S2 potential surface, are controlled by friction and are modulated by specific interactions of the β2 rings with Leu37 in the NTD of OCP O .It would suggest that the mechanism of OCP photoactivation is β1-insenstative, and involves out-of-plane motions of the β2 ring.

Figure 8 .
Figure 8. Two schematic models of the excited state dynamics of carotenoid in OCP.Th model assigns the multiexponential character of the S1 excited state in OCP to the differen tribution of ICT in excited state kinetics[49][50][51], while model two shows OCP population h geneity[50,52,53].The origin and contribution of S* to OCP photocycle is discussed in Sectio [50,52,53]49].Two schematic models of the excited state dynamics of carotenoid in OCP.The first model assigns the multiexponential character of the S1 excited state in OCP to the different contribution of ICT in excited state kinetics[49][50][51], while model two shows OCP population heterogeneity[50,52,53].The origin and contribution of S* to OCP photocycle is discussed in Section 6.