Bioluminescence Imaging of Potassium Ion Using a Sensory Luciferin and an Engineered Luciferase

Bioluminescent indicators are power tools for studying dynamic biological processes. In this study, we present the generation of novel bioluminescent indicators by modifying the luciferin molecule with an analyte-binding moiety. Specifically, we have successfully developed the first bioluminescent indicator for potassium ions (K+), which are critical electrolytes in biological systems. Our approach involved the design and synthesis of a K+-binding luciferin named potassiorin. Additionally, we engineered a luciferase enzyme called BRIPO (bioluminescent red indicator for potassium) to work synergistically with potassiorin, resulting in optimized K+-dependent bioluminescence responses. Through extensive validation in cell lines, primary neurons, and live mice, we demonstrated the efficacy of this new tool for detecting K+. Our research demonstrates an innovative concept of incorporating sensory moieties into luciferins to modulate luciferase activity. This approach has great potential for developing a wide range of bioluminescent indicators, advancing bioluminescence imaging (BLI), and enabling the study of various analytes in biological systems.


■ INTRODUCTION
Fluorescent indicators have revolutionized our understanding of cellular processes by enabling real-time visualization of dynamic events in living systems. 1,2These indicators have become indispensable tools for researchers in various fields, facilitating significant discoveries and advancements in our understanding of life processes.−5 BLI operates through a biochemical reaction involving the oxidation of a substrate (luciferin) by an enzyme (luciferase), allowing photon emission without external light excitation.This unique characteristic not only eliminates concerns of autofluorescence and phototoxicity but also enables deeper tissue imaging. 5Consequently, BLI is particularly attractive for studying biological processes in thick tissue and live animals. 6,7espite its potential, the progress of BLI in visualizing biological activities is impeded by the limited availability and undesirable properties of current bioluminescent indicators.
Firefly luciferase (FLuc) is a widely used bioluminescent label.−11 These modified substrates, known as caged luciferins, are designed to undergo uncaging reactions in the presence of specific molecules or enzymes, enabling the detection of biological activities (Figure 1a).However, these indicators rely on the availability of ATP since FLuc consumes ATP during the bioluminescence process. 4,12Therefore, there is a concern that they may disrupt cell physiology since ATP serves as both a vital energy currency and a crucial signaling molecule in living systems. 13n this context, luciferases derived from marine organisms have emerged as promising candidates for indicator development. 14,15These luciferases utilize coelenterazine (CTZ) as their natural luciferin and do not require ATP for their activity. 4mong them, the NanoLuc system derived from the deep-sea shrimp Oplophorus gracilirostris has gained significant attention. 16This system offers additional advantages such as a small protein size, high enzyme stability, and a remarkable >150-fold increase in luminescence compared to traditional luciferases.In the presence of the synthetic luciferin furimazine, NanoLuc emits intense blue photons at around 450 nm.−21 For indicator development, NanoLuc is frequently utilized as a RET donor.−26 However, these RET-based indicators face challenges in achieving a wide dynamic range and are better suited for in vitro assays rather than in vivo BLI applications due to the strong attenuation of NanoLuc's blue emission by mammalian tissue. 4,5nother promising approach involves directly incorporating sensory domains into the structure of NanoLuc or its derived luciferases (Figure 1b). 21,27,28This results in the modulation of luciferase activity through structural changes upon analyte binding.In certain cases, the luciferases are further fused with red-emitting FPs to achieve red-shifted emission for better tissue penetration.−30 However, this approach necessitates the identification of appropriate protein-based sensory domains for  A one-site binding model was used to fit the data and derive the apparent dissociation constants (K d ).BL, bioluminescence.specific analytes of interest, and the engineering process can be tedious with unpredictable outcomes.
In this study, we aimed to expand the strategies for generating bioluminescent indicators.Specifically, we explored the method of modifying luciferins with sensory moieties (Figure 1c).We successfully designed and synthesized a luciferin called potassiorin, which selectively binds to potassium ions (K + ), an essential electrolyte in living systems. 31Additionally, we engineered a luciferase named BRIPO (Bioluminescent Red Indicator for Potassium) to work in conjunction with potassiorin, producing bioluminescence signals responsive to the physiological concentrations of K + .To our knowledge, this development represents the first bioluminescent indicator for K + , expanding the capability of monitoring K + in living systems.Our indicator was thoroughly tested in diverse settings, consistently demonstrating the capability in real-time monitoring of K + dynamics.Overall, our study not only presents a valuable bioluminescent indicator for studying K + physiology but also introduces a powerful approach to designing bioluminescent indicators.

Design and Synthesis of Potassiorin.
In a previous study, we introduced a NanoLuc variant called teLuc, which emits teal bioluminescence at around 500 nm when combined with the synthetic luciferin DTZ (Figure 2a). 18Notably, DTZ could be readily synthesized from commercially available reagents in just two steps with a good yield.Due to the red-shifted emission of teLuc compared to NanoLuc, we successfully developed BREP (Figure 2b), 28 a fusion construct between teLuc and a red FP (RFP) mScarlet-I, 32 emitting approximately 60% of its total emission above 600 nm.BREP enables deep-tissue photon penetration and has emerged as one of the most powerful luciferases for in vivo BLI.Building upon these findings, we selected DTZ and BREP as the foundations for our current study.
K + is the most abundant intracellular cation, with a high concentration of 140−150 mM within cells. 31It plays a critical role in generating functional activity in muscle cells, neurons, and cardiac tissue. 31To address the limited methods available for tracking K + in living systems, we aimed to develop a novel bioluminescent K + indicator by incorporating a K + -binding moiety into DTZ.Specifically, we selected a crown ether called 1-aza-18-crown-6, known for its ability to form a complex with K + , 33,34 to derivatize DTZ.
At the beginning of this project, to overcome the challenge of not having a cocrystal structure of NanoLuc and its substrate, we utilized a previously generated docking structure of CTZ in NanoLuc (Figure 2c). 18From this model, we deduced that installing the 1-aza-18-crown-6 moiety through C6 or C8 of the imidazopyrazinone core of the substrate would likely result in the complete loss of bioluminescence activity due to their buried positions and the inability of the putative substrate-binding pocket to accommodate the size of the K + -binding moiety.However, we identified the aromatic ring at the C2 position of imidazopyrazinone as a promising site for installing the K +binding moiety, as it extends outside of the putative substratebinding pocket of the luciferase enzyme (Figure 2c).Notably, this rationale, which was derived from the docking model, is consistent with a recently available cocrystal structure of NanoLuc and its inactive substrate analog. 35e designed a DTZ analog called potassiorin, which incorporates the 1-aza-18-crown-6 moiety extended through the C2 position (Figure 2a).The synthesis of potassiorin involved multiple steps.Briefly, starting from commercially available 4-benzyloxybenzyl alcohol (compound 1 in Figure S1), we synthesized 3-(4-(2-azidoethoxy)phenyl)-1,1-diethoxypropan-2-one (6) in five steps with an overall yield of 9.3% (Figure S1a).Simultaneously, we prepared N-(4-ethynylphenyl)aza-18crown-6 (10) in three steps from N-phenyldiethanolamine (7), with an overall yield of 32.6%.The subsequent Cu(I)-catalyzed azide−alkyne cycloaddition (CuAAC) reaction between compounds 6 and 10 produced 11 in a 77% yield (Figure S1b).Finally, we employed our previously established procedure to synthesize 5-diphenylpyrazin-2-amine (12), which was then subjected to an acid-catalyzed cyclization reaction with 11, resulting in the final product, potassiorin, with a 10% yield (Figure S1b).
Initial Characterization of Potassiorin with BREP.After synthesizing potassiorin, we assessed the compound using purified BREP protein.We measured the emission spectra of BREP and potassiorin in the absence and presence of 150 mM K + ions (Figure 2d).The results indicated that the presence of 150 mM K + led to a reduction in bioluminescence by approximately 30%.In contrast, the bioluminescence of BREP and DTZ (as a negative control) showed only marginal changes in response to K + (Figure 2e).Furthermore, we examined the bioluminescence of BREP and potassiorin in the presence of various concentrations of K + or Na + .Through our analysis, we determined the apparent affinities for K + and Na + (i.e., the concentrations to cause 50% of the overall bioluminescence change) to be 108 and 185 mM, respectively (Figure 2f).These findings suggest that potassiorin exhibits K + -dependent bioluminescence, although further improvements are necessary to enhance the dynamic range and selectivity toward K + over Na + .
Engineering BREP into BRIPO for Enhanced Responsiveness.To enhance the dynamic range and selectivity, we employed random mutagenesis on BREP using error-prone PCRs.We screened the resulting gene library and selected clones exhibiting large K + -dependent bioluminescence changes.These clones were subjected to counter-selection to ensure minimal bioluminescence changes in response to Na + .We conducted seven rounds of random mutagenesis and screening but observed only marginal improvement (Figure S2).To further enhance performance, we turned to multisite-directed mutagenesis.By utilizing the docking model, we identified and simultaneously mutated three amino acids (W233, S260, and V261) near the C2 position of the luciferin.Through screening this library, we successfully identified a mutant that exhibited a remarkably improved response magnitude and selectivity.This mutant, named BRIPO, contains a total of six mutations from the original BREP sequence (Figures 3a and S3).Notably, the S260R and V261W mutations obtained during the final step of engineering played a crucial role in enhancing the performance (Figure S2a).The cocrystal structure of NanoLuc and its substrate analog reaffirmed that these mutations are close to the C2 position of the luciferin substrate (Figure S2b), suggesting their ability to influence the interactions between potassiorin and the enzyme.
In the presence of BRIPO and potassiorin, 150 mM K + resulted in a remarkably 6-fold decrease in bioluminescence intensity (Figure 3b).Conversely, when BRIPO was combined with DTZ, the responsiveness to K + was minimal (Figure 3c).Further titration experiments involving varying concentrations of K + and Na + revealed that the apparent affinities of the BRIPO and potassiorin pair for K + and Na + have been altered to 26 and 67 mM, respectively (Figure 3d,e).Since the intracellular concentration of Na + is approximately 10 times lower than that of K + , 31 the BRIPO-potassiorin system should exhibit sufficient specificity to accurately sense intracellular K + levels over Na + .
To validate the specificity of the BRIPO-potassiorin system, we conducted additional tests using a range of cations at physiologically relevant or higher concentrations (Figure 3f).The results confirmed the system's selectivity toward K + .Notably, high concentrations of Cu 2+ led to bioluminescence quenching (Figure S4a), which is consistent with a previously documented mechanism involving the Cu 2+ -mediated oxidation of the luciferin. 36However, these high Cu 2+ concentrations are unlikely to be physiologically relevant. 37oreover, we conducted additional investigations into the potassiorin concentration dependency of BRIPO bioluminescence in the presence or absence of 150 mM K + (Figure S4b).Our findings indicate that the presence of K + enhances the binding of potassiorin to the enzyme, resulting in a lower Michaelis constant (K M ).However, this slightly enhanced affinity does not lead to increased bioluminescence.Instead, we observed a nearly 5-fold reduction in maximal photon production rates (V max ) and a small decrease in quantum yields (Table S1).These results support that a K + -dependent reduction in enzyme catalysis is the primary factor contributing to the observed bioluminescence turn-off response.
Imaging Intracellular K + Dynamics in Cultured Cell Lines.To investigate the ability of BRIPO and potassiorin to visualize K + dynamics in live mammalian cells, we expressed BRIPO in HEK 293T cells and imaged the cells in a low K + buffer supplemented with potassiorin.Inducing K + efflux with a combination of nigericin (a K + ionophore), bumetanide (an inhibitor of Na + /K + /2Cl − cotransporter), and ouabain (an inhibitor of Na + , K + -ATPase pump) 38 resulted in an approximately 30% increase in bioluminescence (Figure 4a−c and Movie S1).Furthermore, we replicated these experiments using HEK 293T cells in a high K + (200 mM) buffer, facilitating the movement of K + from the extracellular space to the intracellular space, resulting in an approximately 8% decrease in bioluminescence (Figure S5).In addition, we expressed BRIPO in a HEK 293T cell line stably expressing a mouse leak K + channel (mTrek) and a few other ion channels. 39We next used arachidonic acid to stimulate the mTrek channel. 40allowing K + efflux to the extracellular low K + space.This led to a nearly 50% increase in bioluminescence (Figure 4d−f and Movie S2).For all three cases, control experiments with DTZ did not show much change in bioluminescence (Figures 4 and S5).These findings support the effectiveness of the BRIPO-potassiorin system for selectively monitoring K + dynamics in live mammalian cells.
Imaging K + Dynamics in Primary Neurons and Live Mice.Using the BRIPO-potassiorin system, we imaged K + dynamics in primary mouse neurons.We transduced the neurons with adeno-associated viruses (AAVs) and imaged them in a low K + buffer with potassiorin.Glutamate was used to activate the cells, causing membrane depolarization followed by repolarization due to K + channel activation and K + efflux. 41round 40% of the examined neurons, which are likely to express glutamate receptors, exhibited robust bioluminescence increases (Figure 5a,b).In contrast, almost no cells in the control experiments using DTZ showed obvious changes in bioluminescence.
To further validate the effectiveness of the BRIPO-potassiorin system for in vivo imaging, we performed BLI in live mouse brains.We administered AAVs carrying the BRIPO gene into the hippocampal and cortical regions of mice.After 3 weeks of gene expression, we injected potassiorin and conducted timelapse imaging on anesthetized mice placed in a dark box.Further delivery of glutamate resulted in notable increases in bioluminescence in all five mice injected with potassiorin, indicating the detection of K + dynamics (Figure 5c,d and  Movie S3).In contrast, control experiments using DTZ showed minimal changes in bioluminescence, which can be attributed to animal movement or alterations in blood flow.These findings provide strong evidence for the efficacy of the BRIPOpotassiorin system for in vivo imaging applications.

■ DISCUSSION
In this study, we introduce a novel bioluminescence imaging method for studying K + dynamics.We successfully developed potassiorin, a luciferin responsive to K + , and engineered the luciferase enzyme BRIPO to work in synergy with potassiorin.The BRIPO-potassiorin system demonstrated robust K +dependent bioluminescence quenching in purified proteins, cell lines, primary neurons, and live mice, validating the effectiveness of this new system.
Interestingly, the presence of K + was observed to only minimally enhance the affinity of potassium for the enzyme and slightly reduce the bioluminescence quantum yield, but it resulted in a notable decrease in photon production rate.This observation suggests that K + binding to potassiorin allosterically modulates the enzyme activity, potentially through interactions involving the mutated residues.In the protein engineering process, the introduction of the S260R and V261W mutations was identified as crucial for augmenting the response magnitude and selectivity of the BRIPO-potassiorin system.Hypothetically, the positively charged guanidinium group in R260 could potentially interact with the crown ether ring of potassiorin, while W261 may engage in a cation-π interaction when K + binds to the crown ether ring.
Importantly, the K + -induced turn-off response provides a practical advantage: the system exhibits a bioluminescence turnon response when cells are activated physiologically since under these conditions, K + efflux occurs through K + channels, leading to hyperpolarization of the cell membrane.In addition, our method overcomes the limitations of traditional K + -detection techniques like ion-selective electrodes, 42 flame photometry, 43 and fluorescent indicators, 33,34,38,41,44,45 which face challenges in tissue-and organism-level studies.By enabling visualization and study of K + dynamics from live cells to animals, our approach opens new avenues for understanding the role of K + in physiology and disease.
Additionally, this study introduces a novel approach for generating bioluminescent indicators by modifying the luciferin molecule with an analyte-binding moiety.We identified that extending the luciferin molecule through the C2 aromatic ring of imidazopyrazinone is a viable option, as NanoLuc-derived luciferases display tolerance to such structural modifications while remaining sensitive to the modulation caused by analyte binding.The effectiveness of this strategy was successfully demonstrated through our development of the indicator for K + .Furthermore, our preliminary studies have led to the development of a prototype bioluminescent Na + indicator by substituting the 18-member crown ether moiety in potassiorin with a 15-member ring. 46In upcoming studies, we plan to further enhance the sensory luciferin by incorporating ligands with higher affinity to K + and Na + , 47,48 aiming to improve sensitivity and enable the detection of extracellular K + as well as intracellular and extracellular Na + .
Prior research has utilized the caged luciferin strategy (Figure 1a) to successfully create indicators for metal ions such as Cu 2+ and Fe 2+ , 11,49,50 contingent upon the catalytic capability of these transition metals in catalyzing decaging reactions.In these cases, the bioluminescence intensity is influenced by the kinetics of decaging and luciferin clearance, which may not accurately reflect the spatiotemporal dynamics of the metals.In contrast, our approach hinges on binding and dissociation kinetics, which often occurs more rapidly than decaging reactions and luciferin clearance.Moreover, our approach can be extended to metal ions that are not amenable to decaging reactions, as well as broader targets including anions and nonionic entities.Through the integration of various sensory moieties into the luciferin structure, our methodology has the potential to significantly broaden the spectrum of bioluminescent indicators.
Overall, this research sets the stage for future advancements in bioluminescent sensors, allowing for the creation of versatile indicators that can be adapted to monitor various ions, molecules, and molecular interactions.The flexibility in sensor design opens up new avenues for broad applications BLI, which will enhance our understanding of biological systems and drive forward biomedical research.

■ METHODS
General Methods and Information.DNA oligos were purchased from either Integrated DNA Technologies or Eurofins Genomics.Restriction enzymes and Phusion High-Fidelity DNA polymerase were purchased from Thermo Fisher.Taq DNA polymerase was purchased from New England Biolabs.DNA sequencing was performed by Eurofins Genomics.All animal experiments were conducted following the guidelines and approval (Protocol #4196) of the University of Virginia Institutional Animal Care and Use Committee.BALB/cJ mice (#000651) were obtained from the Jackson Laboratory and housed in a temperature-controlled room (∼23 °C) with a 12 h light-dark cycle and approximately 50% humidity.At approximately 6 weeks of age, the mice were randomly assigned to experimental groups, ensuring a balance of both female and male animals.All 1 H and 13 C NMR spectra were collected on a Bruker Avance DRX 600 NMR Spectrometer at the UVA Biomolecular Magnetic Resonance Facility.Chemical shifts (δ) are given with the internal standards: 1 H (7.26 ppm) and 13 C (77.0 ppm) for CDCl 3 ; 1 H (5.32) for CD 2 Cl 2 , and 13 C (49.00 ppm) for CD 3 OD.Splitting patterns are reported as s (singlet), d (doublet), t (triplet), dd (doublet of doublets), and m (multiplet).Coupling constants (J) are reported in Hz.Synthetic schemes and compound numbering information are shown in Figure S1.NMR spectra for key compounds are presented in Figures S6 and S7.ChatGPT was utilized to paraphrase sentences and correct grammatical errors in this manuscript.

Synthesis of N-(4-Ethynylphenyl)aza-18-crown-6 (10).
Compound 10 was synthesized from N-phenyldiethanolamine (7) through several steps.First, 8 was obtained as a white solid according to the literature. 54Then, 9 was prepared from 8 using a published procedure. 55In the next step, compound 10, which was reported previously, 56 was obtained from 9 using a revised procedure.Briefly, 9 (600 mg, 1.63 mmol, 1 equiv) was dissolved in 10 mL dry methanol and stirred with dry K 2 CO 3 (899 mg, 6.52 mmol, 4 equiv) in a 100 mL round-bottom flask at room temperature.Then, 588 uL of dimethyl (1diazo-2-oxopropyl) phosphonate (753 mg, 3.9 mmol, 2.4 equiv) was added.The reaction mixture was stirred overnight at room temperature and monitored by thin-layer chromatography (TLC).After the reaction neared completion, 50 mL ddH 2 O was added to the reaction mixture.Then, the resulting mixture was subjected to three extractions with 50 mL of ethyl acetate each time.The organic layers were combined, dried over anhydrous Na 2 SO 4 , filtered, and concentrated under vacuum.The resulting residue was purified by silica column chromatography using an elution solvent mixture of ethyl acetate and hexane (3:10, gradually shifting to pure ethyl acetate).This yielded compound 10 as a white solid (473 mg, 1.3 mmol, 80% yield).11).Compound 10 (180 mg, 0.585 mmol) and compound 6 (236 mg, 0.585 mmol) were suspended in a mixture of ddH 2 O (4 mL) and tert-butyl alcohol (4 mL).Then, sodium ascorbate (12.8 mg, 0.0585 mmol, freshly prepared as a 5 mL solution in ddH 2 O) was added, followed by copper(II) sulfate pentahydrate (1 mg, 0.00585 mmol, predissolved in 5 mL ddH 2 O).The resulting heterogeneous mixture was vigorously stirred overnight until it cleared, and TLC analysis confirmed the complete consumption of the reactants.The reaction mixture was subsequently diluted with 20 mL of ddH 2 O and extracted three times with 20 mL of ethyl acetate.The organic layers were combined, washed with 20 mL of brine, dried over Na 2 SO 4 , filtered, and concentrated under vacuum.The resulting residue was purified by silica column chromatography using an elution solvent mixture of ethyl acetate and hexane (3:10, gradually shifting to pure ethyl acetate).This yielded 300 mg (77% yield) of pure product as a sticky light-yellow oil.Synthesis of Potassiorin (13).3,5-Diphenylpyrazin-2-amine (12) was prepared from commercially available 3,5-dibromopyrazin-2-amine according to a previously described procedure. 18Next, a solution of compound 12 (25 mg, 0.1 mmol, 1 equiv) and compound 11 (134 mg, 0.2 mmol, 2 equiv) in 5 mL of degassed 1,4-dioxane was prepared.Then, 0.5 mL of 6 N HCl (30 equiv) was added to the solution.The resulting mixture was stirred at 80 °C in a sealed pressure tube (MilliporeSigma, Cat.#Z568767) for 12 h.Afterward, the reaction was cooled down to room temperature, and the solvent was removed under vacuum.The residue was dissolved in a 1 mL solution of methanol and water (1:1, v/v).The resulting mixture was filtered through a 0.22 μm poly(tetrafluoroethylene) (PTFE) membrane filter and further purified with a Waters Prep 150 liquid chromatography coupled with an SQ Detector 2 mass spectrometer.An XBridge BEH Amide/Phenyl OBD Prep Column (130 Å, 5 μm, 30 mm × 150 mm) was used along with a gradient elution of acetonitrile and water (1:99 to 90:10) at a flow rate of 20 mL/min.The fractions containing the desired product were combined and subjected to lyophilization, resulting in the potassiorin compound as an orange powder (8 mg, 0.01 mmol, 10% yield). 1   Library Construction and Screening.To create libraries with random mutations, the BREP gene was amplified from our previously described pcDNA3-BREP plasmid (Addgene, Cat.#172337) 28 using Taq DNA polymerase under a previously established error-prone condition. 57The resulting mutated genes were then subcloned into a pBAD/His B plasmid using Gibson assembly. 58Escherichia coli DH10B competent cells were transformed by electroporation and plated on 2xYT agar supplemented with 100 μg/mL ampicillin and 0.2% (w/v) L- arabinose.After overnight incubation at 37 °C, approximately 200 μL of 25 μM potassiorin was sprayed onto the colonies on each plate.BLI was performed using a UVP BioSpectrum dark box, a Computar Motorized ZOOM lens (M6Z1212MP3), and a Teledyne Photometrics Evolve 16 EMCCD camera.Colonies displaying strong bioluminescence were selected and cultured individually in wells of 96-well plates containing 1 mL of 2xYT media supplemented with 100 μg/mL ampicillin and 0.2% (w/v) L-arabinose.After shaking at 37 °C for 20 h, bacterial cells were pelleted by centrifugation and lysed using 500 μL of Thermo Fisher Bacterial Protein Extraction Reagent (B-PER).In the initial screening stage, the bioluminescence of E. coli lysates was measured in the presence of potassiorin under two KCl concentrations: 0 mM and 150 mM.For this, 30 μL of each cell lysate was diluted with 50 μL of MOPS buffer (10 mM, pH 7.4) containing either 0 mM or 240 mM KCl, resulting in final KCl concentrations of 0 mM and 150 mM, respectively.Meanwhile, potassiorin (5 mM) dissolved in a premade stock solution (ethanol/1,2-propanediol = 1:1 (v/v), supplemented with 0.88 mg/mL L-ascorbic acid) was diluted to 100 μM using the MOPS buffer containing no KCl or 150 mM KCl.A 20 μL aliquot of the potassiorin solution was dispensed into each well of a microplate using an automated dispenser on a CLARIOstar Microplate Reader (BMG Labtech).After a 1 s shake, the bioluminescence spectra ranging from 450 to 700 nm were recorded using the plate reader equipped with a red-sensitive PMT.Mutants that exhibited extreme K + -dependent bioluminescence changes were chosen for subsequent screening, which focused on their resistance to Na + .A similar procedure as described above was employed to test the bioluminescence responses of the mutants to 30 mM NaCl versus no NaCl.The mutant showing the highest response to K + and the lowest response to Na + was chosen as the template for the next screening round.To construct the focused library targeting residues 233, 260, and 261, oligos containing NNK degenerate codons (where N = A, T, G, or C and K = G or T) were utilized to amplify three short gene fragments.Subsequently, Gibson assembly 58 was employed to fuse these fragments with the predigested pBAD/His B plasmid.The remaining steps involved in library screening were identical to the procedures described above.
Protein Purification and In Vitro Assays.Recombinant proteins BREP and BRIPO were expressed and purified following a previous procedure, 28 and the purity was verified using SDS-PAGE (Figure S8).The purified proteins were diluted in MOPS buffer to a final concentration of 200 nM, with either 0 mM or 300 mM KCl.For the initial assay, 50 μL of each protein dilution was added to the wells of a 96-well plate.Then, 50 μL of potassiorin (50 μM) in MOPS buffer was dispensed into each well, resulting in final KCl concentrations of 0 mM or 150 mM.The bioluminescence spectra were recorded using the CLARIOstar Microplate Reader.For the K + or Na + concentration dependence assays, 50 μL of MOPS buffer containing 200 nM purified proteins and a specific concentration of NaCl or KCl was added to the wells of a 96-well plate.Then, 50 μL of potassiorin (50 μM) in MOPS buffer was injected into each well, establishing final ion concentrations ranging from 1800 to 0.5 mM.The bioluminescence spectra were recorded and the intensity values at 590 nm were plotted against ion concentrations.Data was fit using the one-site binding model in GraphPad Prism 9.For the ion selectivity assays, 50 μL of MOPS buffer containing 200 nM purified BRIPO and a specific metal ion was added to the wells of a 96-well plate.Then, 50 μL of potassiorin (50 μM) in MOPS buffer was injected into each well.Bioluminescence was recorded, with metal ions supplied as follows: NaCl (15 mM), ZnCl 2 (10 μM), CaCl 2 (2 mM), KCl (150 mM), MgCl 2 (2 mM), MnCl 2 (10 μM), FeCl 2 (10 μM), CuCl 2 (100 nM).Two additional CuCl 2 concentrations (1 and 10 μM) were tested in the presence of either potassiorin or DTZ.For the substrate concentration dependence assays, 50 μL of MOPS buffer containing 200 nM purified BRIPO and either no or 300 mM KCl was added to the wells of a 96-well plate.Various volumes of MOPS buffer and potassiorin solutions were injected into each well to achieve potassiorin concentrations ranging from 25 to 0.5 μM.After a 1 s shake, the bioluminescence of each well was recorded.The data was fit using the Michaelis−Menten nonlinear regression function in GraphPad Prism 9. To determine the bioluminescence quantum yields (QY) of BRIPO and potassiorin, an excess of enzyme was employed.The purified protein was diluted to a final concentration of 1.7 μM using a 0.1 M Tris buffer (pH 8.0) with or without 150 mM KCl.Subsequently, 50 μL of the diluted protein was dispensed into the wells of a 96-well plate.Then, 50 μL of potassiorin, diluted in the same buffer at a concentration of 10 nM, was added to each well to achieve a final enzyme and luciferin concentration of 0.85 μM and 5 nM, respectively.Bioluminescence from each well was promptly recorded at 1 s intervals for 1000 s.The bioluminescence intensity diminished to insignificance by the end of the measurement.The area under the intensity curve was computed to represent the relative total photon emission.Measurements were conducted concurrently with the FLuc enzyme (MilliporeSigma, Cat.#SRE0045) and D-luciferin (Glod Biotechnology, Cat.#LUCK100) under previously established conditions, 59 utilizing the reported QY of 0.41 as the reference to calculate the QYs for BRIPO and potassiorin in the presence and absence of 150 mM KCl.
Characterization in Mammalian Cell Lines.The BRIPO gene was amplified from the pBAD plasmid and inserted into a pcDNA3 vector, resulting in pcDNA3-BRIPO.HEK 293T cells (ATCC, Cat.#CRL-3216) were cultured in DMEM supplemented with 10% fetal bovine serum (FBS).A HEK 293T cell line stably expressing mTrek1 and the α1H subunit of Ca V 3.2, provided by Dr. Paula Barrett (University of Virginia), was cultured in DMEM supplemented with 10% FBS, 1% penicillin/streptomycin, 0.4 μg/mL puromycin, and 400 μg/mL G418.The generation and characterization of this cell line were previously described. 39Both types of cells were transfected using a previously described procedure. 28Imaging was conducted 2 to 3 days later.For the K + efflux experiments, cells were rinsed three times with a lab-made cell imaging buffer (15 mM D-glucose, 0.1 mM sodium pyruvate, 0.49 mM MgCl 2 , 2 mM CaCl 2 , 0.4 mM MgSO 4 , 0.44 mM KH 2 PO 4 , 5.3 mM KCl, 4.2 mM NaHCO 3 , 0.34 mM Na 2 HPO 4 , 138 mM NaCl, 10 mM HEPES, pH 7.2).Cells were then maintained in this buffer supplemented with 50 μM potassiorin or DTZ for bioluminescence.Time-lapse imaging was performed using an inverted Leica DMi8 microscope equipped with a Photometrics Prime 95B Scientific CMOS camera.The imaging settings included a 40× oil immersion objective lens (NA 1.2), no filter cube, 2 × 2 camera binning, 10 s exposure with no interval, camera sensor temperature of −20 °C, and 12-bit high-sensitivity mode.For HEK 293T cells, nigericin (10 mM), bumetanide (10 mM), and ouabain (10 mM) ethanol stocks were diluted in the imaging buffer mentioned above to achieve final concentrations of 20, 10, and 10 μM, respectively.For the stable HEK 293T cells, arachidonic acid (10 mM) ethanol stock was diluted in the same imaging buffer to a final concentration of 20 μM.For the K + influx experiments involving the HEK 293T cells, the procedures remained identical except for the utilization of a lab-made high-K + cell imaging buffer (15 mM D-glucose, 0.1 mM sodium pyruvate, 0.49 mM MgCl 2 , 2 mM CaCl 2 , 0.4 mM MgSO 4 , 0.44 mM KH 2 PO 4 , 200 mM KCl, 4.2 mM NaHCO 3 , 0.34 mM Na 2 HPO 4 , 10 mM HEPES, pH 7.2).Acquired images were processed using the Fiji version of ImageJ 1.53e as described. 28Data were plotted, and statistical analysis was performed using GraphPad Prism 9.The baselines caused by substrate decay were corrected according to the previously described procedure. 28iral Preparation and Characterization in Primary Mouse Neurons.The BRIPO gene was amplified from the corresponding pcDNA3 plasmid and subsequently inserted into a pAAV-hSyn vector, resulting in the creation of pAAV-hSyn-BRIPO.AAVs carrying the BRIPO gene were prepared using our previously reported procedure. 28he obtained AAV titers were about 1 × 10 13 GC/mL.Following preparation, the AAVs were aliquoted and stored at −80 °C for longterm preservation.Primary mouse neurons were prepared as described. 28Neurons were seeded on 35 mm glass-bottom dishes coated with poly-D-lysine, supplemented with 2 mL NbActiv4 medium (BrainBits).The culture was maintained at 37 °C with 5% CO 2 .On the fourth day postplating, half of the medium was changed to fresh NbActiv4.On the same day, 3 μL of the BRIPO virus and 1 μL of 1 M HEPES (pH 7.4) were added to each 35 mm culture dish.Neurons were imaged 4 days post-transduction.The growth medium was carefully replaced with 0.5 mL of the cell imaging buffer supplemented with 100 μM of potassiorin or DTZ before imaging.Time-lapse imaging was performed under the same settings described in the mammalian cell imaging section.During time-lapse imaging, glutamate dissolved in the above-mentioned imaging buffer was added to the dish at a final concentration of 1 mM.Image processing and data analysis were identical to the procedure described in the mammalian cell imaging section.
Imaging of K + Dynamics in Live Mice.For each BALB/cJ mouse, 500 nL of AAV was delivered to both sides of the hippocampus (AP − 1.7, ML ± 1.2, DV − 1.5) and cortex (AP − 1.7, ML ± 1.2, DV − 0.5) via intracranial stereotactic injection at a flow rate of 100 nL/min.The needle remained in the brain for an additional 5 min after the infusion was complete, and the wound was sealed with surgical adhesive.Two to 3 weeks after the virus injection, potassiorin (5 mM) or DTZ (15 mM) predissolved in a stock solution (ethanol/1,2-propanediol = 1:1 (v/v), supplemented with 0.88 mg/mL L-ascorbic acid) was diluted in saline to a concentration of 25 μM.The mice were anesthetized, and 500 nL of the diluted compound was injected into the virus infusion sites.Timelapse imaging was then performed using a UVP BioSpectrum dark box, a Computer Motorized ZOOM lens (M6Z1212MP3), and a Teledyne Photometrics Evolve 16 camera.The instrumental settings were as follows: camera sensor gain of 3, PMT gain of 600, 2 × 2 binning, camera sensor temperature of −20 °C, and 10 s exposure time with no interval.The ZOOM lens was set to be 100% open, 0% zoom, and 0% focus.The mice were positioned 20 cm away from the front of the lens without an emission filter.During the time-lapse imaging, the mice were briefly removed from the dark box and intracranially injected with 500 nL of glutamate (10 mM in saline) into the middle of the virus injection sites (AP − 0.7, ML 0, DV − 1.0) at a flow rate of 250 nL/min.The mice were immediately placed back in the dark box for subsequent imaging.Data analysis followed the same procedure described in the mammalian cell imaging section.
Synthesis and characterization data, BRIPO engineering, key mutations and sequence alignment, and additional in vitro characterization (PDF) BLI of BRIPO-expressing HEK 293T cells in low K + buffer in response to a combination of nigericin, ouabain, and bumetanide (Movie S1) (AVI) BLI of BRIPO-expressing HEK 293T cells stably expressing mTrek and several other ion channels in response to arachidonic acid (Movie S2) (AVI) BLI of a mouse with BRIPO expression in the brain in response to local glutamate stimulation (Movie S3) (AVI)

Figure 1 .
Figure 1.Mechanistic comparison of this work with other common bioluminescent indicators of in vivo imaging importance.(a) Reaction-based bioluminescent indicators: a caged luciferin is utilized, where a specific activity can remove the caging group and activate the luciferin, allowing for the specific detection of bioactivity.(b) Sensory luciferase-based bioluminescent indicators: a luciferase is engineered to be responsive to a specific analyte by strategically inserting and fusing a sensory domain to the luciferase.(c) Sensory luciferin-based bioluminescent indicators (this work): an analytebinding moiety (e.g., a K + -binding crown ether) is strategically introduced to the luciferin, leading to the analyte-responsive modulating of the bioluminescence reaction.CG, caging group; BL, bioluminescence.

Figure 2 .
Figure 2. Design of potassiorin and initial evaluation with BREP luciferase.(a) Illustration of the DTZ structure and the installation of a K + -binding crown ether ring to derive potassiorin.The C2, C6, and C8 derivatizations on DTZ are highlighted.(b) Schematic illustration of the domain arrangements of BREP, a fusion of mScarlet-I and teLuc through a three amino acid linker.(c) Illustration of a modeled structure of NanoLuc (cyan ribbon) in complex with CTZ (magenta sticks).The C2, C6, and C8 derivatizations on CTZ are highlighted.(d, e) Bioluminescence emission spectra of BREP in the presence of potassiorin (d) or DTZ (e) with or without 150 mM KCl. Presented are the averages from three technical replicates.(f) Bioluminescence intensities of BREP and potassiorin at 590 nm in the presence of the indicated concentrations of K + or Na + .n = 3 technical replicates.A one-site binding model was used to fit the data and derive the apparent dissociation constants (K d ).BL, bioluminescence.

Figure 3 .
Figure 3.In vitro characterization of BRIPO.(a) Schematic illustration of BRIPO with mutations from BREP highlighted.(b, c) Bioluminescence emission spectra of BRIPO in the presence of potassiorin (b) or DTZ (c) with or without 150 mM KCl. Presented are the averages from three technical replicates.(d) Bioluminescence spectra of BRIPO and potassiorin with the indicated concentrations of KCl.Presented are the averages from three technical replicates.(e) Bioluminescence intensities of BRIPO and potassiorin at 590 nm in the presence of the indicated concentrations of K + or Na + .n = 3 technical replicates.A one-site binding model was used to fit the data and derive the apparent dissociation constants (K d ).(f) Normalized bioluminescence intensity of BRIPO and potassiorin in the presence of different metal ions: Na + (15 mM), Zn 2+ (10 μM), Ca 2+ (2 mM), Fe 2+ (10 μM), Cu 2+ (100 nM), Mn 2+ (10 μM), K + (150 mM), Mg 2+ (2 mM).n = 3 technical replicates.BL, bioluminescence.

Figure 4 .
Figure 4. Imaging K + efflux in cultured cell lines.(a) Schematic illustration of nigericin-mediated K + efflux in HEK 293T cells in a low K + buffer.(b) Representative pseudocolored bioluminescence images of BRIPO-expressing HEK 293T cells in the presence of potassiorin (top) or DTZ (bottom) before (left) and after (right) treatment with a combination of nigericin, ouabain, and bumetanide.Scale bar: 50 μm.(c) Quantification of bioluminescence intensity changes of individual cells from experiments in (b).Data are presented as mean ± s.d.(n = 39 cells for the potassiorin group, n = 57 cells for the DTZ group).(d) Schematic illustration of a stable HEK 293T cell line treated with arachidonic acid to open the mTrek channel and induce K + efflux.(e) Representative pseudocolored bioluminescence images of BRIPO-expressing HEK 293T cells stably expressing mTrek and other ion channels in the presence of potassiorin (top) or DTZ (bottom) before (left) and after (right) treatment with arachidonic acid.Scale bar: 50 μm.(f) Quantification of bioluminescence intensity changes of individual cells from experiments in (e).Data are presented as mean ± s.d.(n = 33 cells for the potassiorin group, n = 42 cells for the DTZ group).In (c) and (f), the baselines were corrected using a monoexponential decay model, and the P value was derived from unpaired two-tailed t-tests.The GraphPad Prism software does not provide extract P values below 10 −15 .This figure is created with BioRender.com.BL, bioluminescence.Arb.units, arbitrary units.

Figure 5 .
Figure 5. Imaging K + efflux in primary mouse and the brains of live mice.(a) Representative fluorescence and pseudocolored bioluminescence images of BRIPO-expressing primary mouse neurons.Glutamate was used to induce potassium efflux.Scale bar, 50 μm.(b) Quantification of bioluminescence intensity changes of individual responsive neurons upon glutamate treatment.Data are presented as mean ± s.d.(n = 25 cells for the potassiorin group, n = 31 cells for the DTZ group).(c) Schematic illustration of stereotactic intracranial administration of AAVs containing the BRIPO gene and other general experiment procedures.(d) Quantification of bioluminescence intensity changes of individual animals.Data are presented as mean ± s.e.m. (n = 5 mice for the potassiorin group, n = 6 mice for the DTZ group).In (b) and (d), the baselines were corrected using a monoexponential decay model.and the P value was derived from unpaired two-tailed t-tests.The GraphPad Prism software does not provide extract P values below 10 −15 .This figure is created with BioRender.com.