The role of the correlated motion(s) of the chromophore in photoswitching of green and red forms of the photoconvertible fluorescent protein mSAASoti.

Wild-type SAASoti and its monomeric variant mSAASoti can undergo phototransformations, including reversible photoswitching of the green form to a nonfluorescent state and irreversible green-to-red photoconversion. In this study, we extend the photochemistry of mSAASoti variants to enable reversible photoswitching of the red form. This result is achieved by rational and site-saturated mutagenesis of the M163 and F177 residues. In the case of mSAASoti it is M163T substitution that leads to the fastest switching and the most photostable variant, and reversible photoswitching can be observed for both green and red forms when expressed in eukaryotic cells. We obtained a 13-fold increase in the switching efficiency with the maximum switching contrast of the green form and the appearance of comparable switching of the red form for the C21N/M163T mSAASoti variant. The crystal structure of the C21N mSAASoti in its green on-state was obtained for the first time at 3.0 Å resolution, and it is in good agreement with previously calculated 3D-model. Dynamic network analysis reveals that efficient photoswitching occurs if motions of the 66H residue and phenyl fragment of chromophore are correlated and these moieties belong to the same community.

Multifunctionality of arginine residues in the active sites of non-canonical d-amino acid transaminases.

Structure-function relationships are key to understanding enzyme mechanisms, controlling enzyme activities, and designing biocatalysts. Here, we investigate the functions of arginine residues in the active sites of pyridoxal-5'-phosphate (PLP)-dependent non-canonical d-amino acid transaminases, focusing on the analysis of a transaminase from Haliscomenobacter hydrossis. Our results show that the tandem of arginine residues R28* and R90, which form the conserved R-[RK] motif in non-canonical d-amino acid transaminases, not only facilitates effective substrate binding but also regulates the catalytic properties of PLP. Non-covalent interactions between residues R28*, R90, and Y147 strengthen the hydrogen bond between Y147 and PLP, thereby maintaining the reactivity of the cofactor. Next, the R90 residue contributes to the stability of the holoenzyme. Finally, the R90I substitution induces structural changes that lead to substrate promiscuity, as evidenced by the effective binding of substrates with and without the α-carboxylate group. This study sheds light on the structural determinants of the activity of non-canonical d-amino acid transaminases. Understanding the structural basis of the active site plasticity in the non-canonical transaminase from H. hydrossis, which is characterized by effective conversion of d-amino acids and α-keto acids, may help to tailor it for industrial applications.

Structural framework for the understanding spectroscopic and functional signatures of the cyanobacterial Orange Carotenoid Protein families.

The Orange Carotenoid Protein (OCP) is a unique photoreceptor crucial for cyanobacterial photoprotection. Best studied Synechocystis sp. PCC 6803 OCP belongs to the large OCP1 family. Downregulated by the Fluorescence Recovery Protein (FRP) in low-light, high-light-activated OCP1 binds to the phycobilisomes and performs non-photochemical quenching. Recently discovered families OCP2 and OCP3 remain structurally and functionally underexplored, and no systematic comparative studies have ever been conducted. Here we present two first crystal structures of OCP2 from morphoecophysiologically different cyanobacteria and provide their comprehensive structural, spectroscopic and functional comparison with OCP1, the recently described OCP3 and all-OCP ancestor. Structures enable correlation of spectroscopic signatures with the effective number of hydrogen and discovered here chalcogen bonds anchoring the ketocarotenoid in OCP, as well as with the rotation of the echinenone's ß-ionone ring in the CTD. Structural data also helped rationalize the observed differences in OCP/FRP and OCP/phycobilisome functional interactions. These data are expected to foster OCP research and applications in optogenetics, targeted carotenoid delivery and cyanobacterial biomass engineering.

Expanded Substrate Specificity in D-Amino Acid Transaminases: A Case Study of Transaminase from Blastococcus saxobsidens.

Enzymes with expanded substrate specificity are good starting points for the design of biocatalysts for target reactions. However, the structural basis of the expanded substrate specificity is still elusive, especially in the superfamily of pyridoxal-5'-phosphate-dependent transaminases, which are characterized by a conserved organization of both the active site and functional dimer. Here, we analyze the structure-function relationships in a non-canonical D-amino acid transaminase from Blastococcus saxobsidens, which is active towards D-amino acids and primary (R)-amines. A detailed study of the enzyme includes a kinetic analysis of its substrate scope and a structural analysis of the holoenzyme and its complex with phenylhydrazine-a reversible inhibitor and analogue of (R)-1-phenylethylamine-a benchmark substrate of (R)-selective amine transaminases. We suggest that the features of the active site of transaminase from B. saxobsidens, such as the flexibility of the R34 and R96 residues, the lack of bulky residues in the ß-turn at the entrance to the active site, and the short O-pocket loop, facilitate the binding of substrates with and without α-carboxylate groups. The proposed structural determinants of the expanded substrate specificity can be used for the design of transaminases for the stereoselective amination of keto compounds.

Can Activation of Acetylcholinesterase by ß-Amyloid Peptide Decrease the Effectiveness of Cholinesterase Inhibitors?

A central event in the pathogenesis of Alzheimer's disease (AD) is the accumulation of senile plaques composed of aggregated amyloid-ß (Aß) peptides. The main class of drugs currently used for the treatment of AD are the acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) inhibitors. In this study, it has been shown that Aß augmented AChE activity in vitro, maximum activation of 548 ± 5% was achieved following 48 h of incubation with 10 µM of Aß1-40, leading to a 7.7-fold increase in catalytic efficiency. The observed non-competitive type of AChE activation by Aß1-40 was associated with increased Vmax and unchanged Km. Although BChE activity also increased following incubation with Aß1-40, this was less efficiently achieved as compared with AChE. Ex vivo electrophysiological experiments showed that 10 µM of Aß1-40 significantly decreased the effect of the AChE inhibitor huperzine A on the synaptic potential parameters.

Crystal structure reveals canonical recognition of the phosphorylated cytoplasmic loop of human alpha7 nicotinic acetylcholine receptor by 14-3-3 protein.

Nicotinic acetylcholine receptors (nAChRs) are ligand-gated ion channels composed of five homologous subunits. The homopentameric α7-nAChR, abundantly expressed in the brain, is involved in the regulation of the neuronal plasticity and memory and undergoes phosphorylation by protein kinase A (PKA). Here, we extracted native α7-nAChR from murine brain, validated its assembly by cryo-EM and showed that phosphorylation by PKA in vitro enables its interaction with the abundant human brain protein 14-3-3ζ. Bioinformatic analysis narrowed the putative 14-3-3-binding site down to the fragment of the intracellular loop (ICL) containing Ser365 (Q361RRCSLASVEMS372), known to be phosphorylated in vivo. We reconstructed the 14-3-3ζ/ICL peptide complex and determined its structure by X-ray crystallography, which confirmed the Ser365 phosphorylation-dependent canonical recognition of the ICL by 14-3-3. A common mechanism of nAChRs' regulation by ICL phosphorylation and 14-3-3 binding that potentially affects nAChR activity, stoichiometry, and surface expression is suggested.

In search for structural targets for engineering d-amino acid transaminase: modulation of pH optimum and substrate specificity.

The development of biocatalysts requires reorganization of the enzyme's active site to facilitate the productive binding of the target substrate and improve turnover number at desired conditions. Pyridoxal-5'-phosphate (PLP) - dependent transaminases are highly efficient biocatalysts for asymmetric amination of ketones and keto acids. However, transaminases, being stereoselective enzymes, have a narrow substrate specificity due to the ordered structure of the active site and work only in neutral-alkaline media. Here, we investigated the d-amino acid transaminase from Aminobacterium colombiense, with the active site organized differently from that of the canonical d-amino acid transaminase from Bacillus sp. YM-1. Using a combination of site-directed mutagenesis, kinetic analysis, molecular modeling, and structural analysis we determined the active site residues responsible for substrate binding, substrate differentiation, thermostability of a functional dimer, and affecting the pH optimum. We demonstrated that the high specificity toward d-glutamate/α-ketoglutarate is due to the interactions of a γ-carboxylate group with K237 residue, while binding of other substrates stems from the effectiveness of their accommodation in the active site optimized for d-glutamate/α-ketoglutarate binding. Furthermore, we showed that the K237A substitution shifts the catalytic activity optimum to acidic pH. Our findings are useful for achieving target substrate specificity and demonstrate the potential for developing and optimizing transaminases for various applications.

YTnC2, an improved genetically encoded green calcium indicator based on toadfish troponin C.

Genetically encoded calcium indicators based on truncated troponin C are attractive probes for calcium imaging due to their relatively small molecular size and twofold reduced calcium ion buffering. However, the best-suited members of this family, YTnC and cNTnC, suffer from low molecular brightness, limited dynamic range, and/or poor sensitivity to calcium transients in neurons. To overcome these limitations, we developed an enhanced version of YTnC, named YTnC2. Compared with YTnC, YTnC2 had 5.7-fold higher molecular brightness and 6.4-fold increased dynamic range in vitro. YTnC2 was successfully used to reveal calcium transients in the cytosol and in the lumen of mitochondria of both mammalian cells and cultured neurons. Finally, we obtained and analyzed the crystal structure of the fluorescent domain of the YTnC2 mutant.

Mechanism of D-Cycloserine Inhibition of D-Amino Acid Transaminase from Haliscomenobacter hydrossis.

D-cycloserine inhibits pyridoxal-5'-phosphate (PLP)-dependent enzymes. Inhibition effect depend on organization of the active site and mechanism of the catalyzed reaction. D-cycloserine interacts with the PLP form of the enzyme similarly to the substrate (amino acid), and this interaction is predominantly reversible. Several products of the interaction of PLP with D-cycloserine are known. For some enzymes formation of a stable aromatic product - hydroxyisoxazole-pyridoxamine-5'-phosphate at certain pH - leads to irreversible inhibition. The aim of this work was to study the mechanism of D-cycloserine inhibition of the PLP-dependent D-amino acid transaminase from Haliscomenobacter hydrossis. Spectral methods revealed several products of interaction of D-cycloserine with PLP in the active site of transaminase: oxime between PLP and ß-aminooxy-D-alanine, ketimine between pyridoxamine-5'-phosphate and cyclic form of D-cycloserine, and pyridoxamine-5'-phosphate. Formation of hydroxyisoxazole-pyridoxamine-5'-phosphate was not observed. 3D structure of the complex with D-cycloserine was obtained using X-ray diffraction analysis. In the active site of transaminase, a ketimine adduct between pyridoxamine-5'-phosphate and D-cycloserine in the cyclic form was found. Ketimine occupied two positions interacting with different active site residues via hydrogen bonds. Using kinetic and spectral methods we have shown that D-cycloserine inhibition is reversible, and activity of the inhibited transaminase from H. hydrossis could be restored by adding excess of keto substrate or excess of cofactor. The obtained results confirm reversibility of the inhibition by D-cycloserine and interconversion of various adducts of D-cycloserine and PLP.

FNCaMP, ratiometric green calcium indicator based on mNeonGreen protein.

Neurobiologists widely use green genetically encoded calcium indicators (GECIs) for visualization of neuronal activity. Among them, ratiometric GECIs allow imaging of both active and non-active neuronal populations. However, they are not popular, since their properties are inferior to intensiometric GCaMP series of GECIs. The most characterized and developed ratiometric green GECI is FGCaMP7. However, the dynamic range and sensitivity of its large Stock's shift green (LSS-Green) form is significantly lower than its Green form and its molecular design is not optimal. To address these drawbacks, we engineered a ratiometric green calcium indicator, called FNCaMP, which is based on bright mNeonGreen protein and calmodulin from A. niger and has optimal NTnC-like design. We compared the properties of the FNCaMP and FGCaMP7 indicators in vitro, in mammalian cells, and in neuronal cultures. Finally, we obtained and analyzed X-ray structure of the FNCaMP indicator.

Friunavirus Phage-Encoded Depolymerases Specific to Different Capsular Types of Acinetobacter baumannii.

Acinetobacter baumannii is a critical priority nosocomial pathogen that produces a variety of capsular polysaccharides (CPSs), the primary receptors for specific depolymerase-carrying phages. In this study, the tailspike depolymerases (TSDs) encoded in genomes of six novel Friunaviruses, APK09, APK14, APK16, APK86, APK127v, APK128, and one previously described Friunavirus phage, APK37.1, were characterized. For all TSDs, the mechanism of specific cleavage of corresponding A. baumannii capsular polysaccharides (CPSs) was established. The structures of oligosaccharide fragments derived from K9, K14, K16, K37/K3-v1, K86, K127, and K128 CPSs degradation by the recombinant depolymerases have been determined. The crystal structures of three of the studied TSDs were obtained. A significant reduction in mortality of Galleria mellonella larvae infected with A. baumannii of K9 capsular type was shown in the example of recombinant TSD APK09_gp48. The data obtained will provide a better understanding of the interaction of phage-bacterial host systems and will contribute to the formation of principles of rational usage of lytic phages and phage-derived enzymes as antibacterial agents.

Combined Structural and Computational Study of the mRubyFT Fluorescent Timer Locked in Its Blue Form.

The mRubyFT is a monomeric genetically encoded fluorescent timer based on the mRuby2 fluorescent protein, which is characterized by the complete maturation of the blue form with the subsequent conversion to the red one. It has higher brightness in mammalian cells and higher photostability compared with other fluorescent timers. A high-resolution structure is a known characteristic of the mRubyFT with the red form chromophore, but structural details of its blue form remain obscure. In order to obtain insight into this, we obtained an S148I variant of the mRubyFT (mRubyFTS148I) with the blocked over time blue form of the chromophore. X-ray data at a 1.8 Å resolution allowed us to propose a chromophore conformation and its interactions with the neighboring residues. The imidazolidinone moiety of the chromophore is completely matured, being a conjugated π-system. The methine bridge is not oxidized in the blue form bringing flexibility to the phenolic moiety that manifests itself in poor electron density. Integration of these data with the results of molecular dynamic simulation disclosed that the OH group of the phenolic moiety forms a hydrogen bond with the side chain of the T163 residue. A detailed comparison of mRubyFTS148I with other available structures of the blue form of fluorescent proteins, Blue102 and mTagBFP, revealed a number of characteristic differences. Molecular dynamic simulations with the combined quantum mechanic/molecular mechanic potentials demonstrated that the blue form exists in two protonation states, anion and zwitterion, both sharing enolate tautomeric forms of the C=C-O- fragment. These two forms have similar excitation energies, as evaluated by calculations. Finally, excited state molecular dynamic simulations showed that excitation of the chromophore in both protonation states leads to the same anionic fluorescent state. The data obtained shed light on the structural features and spectral properties of the blue form of the mRubyFT timer.

To the Understanding of Catalysis by D-Amino Acid Transaminases: A Case Study of the Enzyme from Aminobacterium colombiense.

Pyridoxal-5'-phosphate (PLP)-dependent transaminases are highly efficient biocatalysts for stereoselective amination. D-amino acid transaminases can catalyze stereoselective transamination producing optically pure D-amino acids. The knowledge of substrate binding mode and substrate differentiation mechanism in D-amino acid transaminases comes down to the analysis of the transaminase from Bacillus subtilis. However, at least two groups of D-amino acid transaminases differing in the active site organization are known today. Here, we present a detailed study of D-amino acid transaminase from the gram-negative bacterium Aminobacterium colombiense with a substrate binding mode different from that for the transaminase from B. subtilis. We study the enzyme using kinetic analysis, molecular modeling, and structural analysis of holoenzyme and its complex with D-glutamate. We compare the multipoint binding of D-glutamate with the binding of other substrates, D-aspartate and D-ornithine. QM/MM MD simulation reveals that the substrate can act as a base and its proton can be transferred from the amino group to the α-carboxylate group. This process occurs simultaneously with the nucleophilic attack of the PLP carbon atom by the nitrogen atom of the substrate forming gem-diamine at the transimination step. This explains the absence of the catalytic activity toward (R)-amines that lack an α-carboxylate group. The obtained results clarify another substrate binding mode in D-amino acid transaminases and underpinned the substrate activation mechanism.

Crystal Structure of Inhibitor-Bound Bacterial Oligopeptidase B in the Closed State: Similarity and Difference between Protozoan and Bacterial Enzymes.

The crystal structure of bacterial oligopeptidase B from Serratia proteamaculans (SpOpB) in complex with a chloromethyl ketone inhibitor was determined at 2.2 Å resolution. SpOpB was crystallized in a closed (catalytically active) conformation. A single inhibitor molecule bound simultaneously to the catalytic residues S532 and H652 mimicked a tetrahedral intermediate of the catalytic reaction. A comparative analysis of the obtained structure and the structure of OpB from Trypanosoma brucei (TbOpB) in a closed conformation showed that in both enzymes, the stabilization of the D-loop (carrying the catalytic D) in a position favorable for the formation of a tetrahedral complex occurs due to interaction with the neighboring loop from the ß-propeller. However, the modes of interdomain interactions were significantly different for bacterial and protozoan OpBs. Instead of a salt bridge (as in TbOpB), in SpOpB, a pair of polar residues following the catalytic D617 and a pair of neighboring arginine residues from the ß-propeller domain formed complementary oppositely charged surfaces. Bioinformatics analysis and structural modeling show that all bacterial OpBs can be divided into two large groups according to these two modes of D-loop stabilization in closed conformations.

Blue-to-Red TagFT, mTagFT, mTsFT, and Green-to-FarRed mNeptusFT2 Proteins, Genetically Encoded True and Tandem Fluorescent Timers.

True genetically encoded monomeric fluorescent timers (tFTs) change their fluorescent color as a result of the complete transition of the blue form into the red form over time. Tandem FTs (tdFTs) change their color as a consequence of the fast and slow independent maturation of two forms with different colors. However, tFTs are limited to derivatives of the mCherry and mRuby red fluorescent proteins and have low brightness and photostability. The number of tdFTs is also limited, and there are no blue-to-red or green-to-far-red tdFTs. tFTs and tdFTs have not previously been directly compared. Here, we engineered novel blue-to-red tFTs, called TagFT and mTagFT, which were derived from the TagRFP protein. The main spectral and timing characteristics of the TagFT and mTagFT timers were determined in vitro. The brightnesses and photoconversions of the TagFT and mTagFT tFTs were characterized in live mammalian cells. The engineered split version of the TagFT timer matured in mammalian cells at 37 °C and allowed the detection of interactions between two proteins. The TagFT timer under the control of the minimal arc promoter, successfully visualized immediate-early gene induction in neuronal cultures. We also developed and optimized green-to-far-red and blue-to-red tdFTs, named mNeptusFT and mTsFT, which were based on mNeptune-sfGFP and mTagBFP2-mScarlet fusion proteins, respectively. We developed the FucciFT2 system based on the TagFT-hCdt1-100/mNeptusFT2-hGeminin combination, which could visualize the transitions between the G1 and S/G2/M phases of the cell cycle with better resolution than the conventional Fucci system because of the fluorescent color changes of the timers over time in different phases of the cell cycle. Finally, we determined the X-ray crystal structure of the mTagFT timer and analyzed it using directed mutagenesis.

Extracellular Fe(III) reductase structure reveals a modular organization enabling S-layer insertion and electron transfer to insoluble substrates.

The thermophilic anaerobic Gram-positive bacterium Carboxydothermus ferrireducens utilizes insoluble Fe(III) oxides as electron acceptors in respiratory processes using an extracellular 11-heme cytochrome c OmhA as a terminal reductase. OmhA is able to transfer electrons to soluble and insoluble Fe(III) compounds, substrates of multiheme oxidoreductases, and soluble electron shuttles. The crystal structure of OmhA at 2.5 Å resolution shows that it consists of two functionally distinct parts: the cytochrome с electron transfer and the S-layer binding domains. Nonaheme C-terminal subdomain of the cytochrome с domain is structurally similar to the extracellular multiheme cytochrome OcwA from the metal-reducing Gram-positive bacterium "Thermincola potens." S-layer binding domain of OmhA is responsible for interaction with the S-layer that surrounds the Carboxydothermus ferrireducens cell envelope. The structural foundations enabling the embedding of extracellular multiheme cytochromes to the S-layer of a Gram-positive-type cell wall and putative electron transfer pathways to insoluble minerals are discussed.

Structure-Functional Examination of Novel Ribonucleoside Hydrolase C (RihC) from Limosilactobacillus reuteri LR1.

Ribonucleoside hydrolase C (RihC, EC 3.2.2.1, 3.2.2.2, 3.2.2.3, 3.2.2.7, 3.2.2.8) belongs to the family of ribonucleoside hydrolases Rih and catalyzes the cleavage of ribonucleosides to nitrogenous bases and ribose. RihC is one of the enzymes that are synthesized by lactobacilli in response to the presence of Klebsiella. To characterize this protein from Limosilactobacillus reuteri LR1, we cloned and expressed it. The activity of the enzyme was studied towards a wide range of substrates, including ribonucleosides, deoxyribonucleosides as well as an arabinoside. It was shown that the enzyme is active only with ribonucleosides and arabinoside, with the best substrate being uridine. The thermal stability of this enzyme was studied, and its crystal structure was obtained, which demonstrated the tetrameric architecture of the enzyme and allowed to shed light on a correlation between its structure and enzymatic activity. Comprehensive comparisons of all known RihC structures, both existing crystal structures and computed model structures from various species, were made, allowing for the identification of structural motifs important for enzyme functioning.

cNTnC and fYTnC2, Genetically Encoded Green Calcium Indicators Based on Troponin C from Fast Animals.

NTnC-like green fluorescent genetically encoded calcium indicators (GECIs) with two calcium ion binding sites were constructed using the insertion of truncated troponin C (TnC) from Opsanus tau into green fluorescent proteins (GFPs). These GECIs are small proteins containing the N- and C-termini of GFP; they exert a limited effect on the cellular free calcium ion concentration; and in contrast to calmodulin-based calcium indicators they lack undesired interactions with intracellular proteins in neurons. The available TnC-based NTnC or YTnC GECIs had either an inverted response and high brightness but a limited dynamic range or a positive response and fast kinetics in neurons but lower brightness and an enhanced but still limited dF/F dynamic range. Here, we solved the crystal structure of NTnC at 2.5 Å resolution. Based on this structure, we developed positive NTnC2 and inverted iNTnC2 GECIs with a large dF/F dynamic range in vitro but very slow rise and decay kinetics in neurons. To overcome their slow responsiveness, we swapped TnC from O. tau in NTnC2 with truncated troponin C proteins from the muscles of fast animals, namely, the falcon, hummingbird, cheetah, bat, rattlesnake, and ant, and then optimized the resulting constructs using directed molecular evolution. Characterization of the engineered variants using purified proteins, mammalian cells, and neuronal cultures revealed cNTnC GECI with truncated TnC from Calypte anna (hummingbird) to have the largest dF/F fluorescence response and fast dissociation kinetics in neuronal cultures. In addition, based on the insertion of truncated TnCs from fast animals into YTnC2, we developed fYTnC2 GECI with TnC from Falco peregrinus (falcon). The purified proteins cNTnC and fYTnC2 had 8- and 6-fold higher molecular brightness and 7- and 6-fold larger dF/F responses to the increase in Ca2+ ion concentration than YTnC, respectively. cNTnC GECI was also 4-fold more photostable than YTnC and fYTnC2 GECIs. Finally, we assessed the developed GECIs in primary mouse neuronal cultures stimulated with an external electric field; in these conditions, cNTnC had a 2.4-fold higher dF/F fluorescence response than YTnC and fYTnC2 and was the same or slightly slower (1.4-fold) than fYTnC2 and YTnC in the rise and decay half-times, respectively.

Silkworm carotenoprotein as an efficient carotenoid extractor, solubilizer and transporter.

Found in many organisms, water-soluble carotenoproteins are prospective antioxidant nanocarriers for biomedical applications. Yet, the toolkit of characterized carotenoproteins is rather limited: such proteins are either too specific binders of only few different carotenoids, or their ability to transfer carotenoids to various acceptor systems is unknown. Here, by focusing on a recently characterized recombinant ~27-kDa Carotenoid-Binding Protein from Bombyx mori (BmCBP) [Slonimskiy et al., International Journal of Biological Macromolecules 214 (2022): 664-671], we analyze its carotenoid-binding repertoire and potential as a carotenoid delivery module. We show that BmCBP forms productive complexes with both hydroxyl- and ketocarotenoids - lutein, zeaxanthin, astaxanthin, canthaxanthin and a smaller antioxidant, aporhodoxanthinone, but not with ß-carotene or retinal, which defines its broad ligand specificity toward xanthophylls valuable to human health. Moreover, the His-tagged BmCBP apoform is capable of cost-efficient and scalable enrichment of xanthophylls from various crude methanolic herbal extracts. Upon carotenoid binding, BmCBP remains monomeric and shows a remarkable ability to dynamically shuttle carotenoids to biological membrane models and to unrelated carotenoproteins, which in particular makes from the cyanobacterial Orange Carotenoid Protein a blue-light controlled photoswitch. Furthermore, administration of BmCBP loaded by zeaxanthin stimulates fibroblast growth, which is attractive for cell- and tissue-based assays.

Structural basis for the carotenoid binding and transport function of a START domain.

STARD3, a steroidogenic acute regulatory lipid transfer protein, was identified as a key xanthophyll-binding protein in the human retina. STARD3 and its homologs in invertebrates are known to bind and transport carotenoids, but this lacks structural elucidation. Here, we report high-resolution crystal structures of the apo- and zeaxanthin (ZEA)-bound carotenoid-binding protein from silkworm Bombyx mori (BmCBP). Having a STARD3-like fold, BmCBP features novel elements, including the Ω1-loop that, in the apoform, is uniquely fixed on the α4-helix by an R173-D279 salt bridge. We exploit absorbance, Raman and dichroism spectroscopy, and calorimetry to describe how ZEA and BmCBP mutually affect each other in the complex. We identify key carotenoid-binding residues, confirm their roles by ZEA-binding capacity and X-ray structures of BmCBP mutants, and also demonstrate that markedly different carotenoid-binding capacities of BmCBP and human STARD3 stem from differences in the structural organization of their carotenoid-binding cavity.