Conversion of dehydrocoelenterazine to coelenterazine in mammalian cells: Identification using cultured cells stably expressing coelenterazine-utilizing luciferase.

Dehydrocoelenterazine (dCTZ) is a dehydrogenated form of coelenterazine (CTZ), which is well-known as the luciferin responsible for the bioluminescence reaction in marine organisms. In this report, we demonstrate for the first time that dCTZ is readily reduced to CTZ in mammalian cells. Using an FDSS®/µCell functional drug screening system, the conversion of dCTZ to CTZ in cells was identified through the luciferin (CTZ)-luciferase reaction in Chinese hamster ovary K1 (CHO-K1) cell lines, which stably expressed CTZ-utilizing luciferases of Renilla luciferase (RLase) or QL-nanoKAZ (a mutant of the 19 kDa protein of Oplophorus luciferase). After loading dCTZ into CHO-K1 cells expressing RLase or QL-nanoKAZ, the luminescence from both cells was detected within 10 s and continued for over 30 min. Thus, dCTZ permeates mammalian cells and is immediately converted to CTZ. This suggests that dCTZ could potentially be used as a substitute for CTZ in in vivo assays of the CTZ-dependent luminescence systems.

The calcium-binding photoprotein clytin II: Expression of the preferred human codon-optimized clytin II gene in Chinese hamster ovary-K1 cells and its use in the G-protein-coupled receptor assays.

Clytin II (CLII) is a Ca2+-binding photoprotein and has been identified as an isotype of clytin I (CLI). CLII consists of apoCLII (an apoprotein) and 2-peroxide of coelenterazine (an adduct of molecular oxygen to coelenterazine), which is identical to the widely used Ca2+-binding photoprotein, aequorin (AQ). However, CLII triggered by Ca2+ exhibits a 4.5-fold higher maximum luminescence intensity (Imax) compared to both AQ and CLI, and it is approximately 5 times less sensitive to Ca2+ than AQ. To confirm the suitability of the preferred human codon-optimized CLII (pCLII) gene for cell-based G-protein-coupled receptor (GPCR) assays, a transformant stably expressing apoprotein of pCLII using the pCLII gene in the mitochondria of CHO-K1 cells was established and in situ regenerated pCLII in the cells were applied to the high-throughput screening system. An ATP-stimulated GPCR assay for endogenous P2Y purinergic receptors was confirmed using the established stable transformant.

Enzymatic sulfation of coelenterazine by human cytosolic aryl sulfotransferase SULT1A1: identification of coelenterazine C2-benzyl monosulfate by LC/ESI-TOF-MS.

Coelenterazine (CTZ) is known as a light-emitting source for the bioluminescence reaction in marine organisms. CTZ has two phenolic hydroxy groups at the C2-benzyl and C6-phenyl positions, and a keto-enol type hydroxy group at the C3-position in the core structure of imidazopyrazinone (= 3,7-dihydroimidazopyrazin-3-one). These hydroxy groups in CTZ could be sulfated by sulfotransferase(s), and the sulfates of Watasenia luciferin (CTZ disulfate at the C2- and C6-positions) and Renilla pre-luciferin (CTZ 3-enol sulfate) have been identified in marine organisms. To characterize the sulfation process of CTZ, human cytosolic aryl sulfotransferase SULT1A1 (SUTase) was used as a model enzyme. The sulfated products catalyzed by SUTase with 3'-phosphoadenosine 5'-phosphosulfate (PAPS) were analyzed by LC/ESI-TOF-MS. The product was the monosulfate of CTZ and identified as the C2-benzyl sulfate of CTZ (CTZ C2-benzyl monosulfate), but CTZ disulfate, CTZ 3-enol sulfate, and CTZ C6-phenyl monosulfate were not detected. The non-enzymatic oxidation products of dehydrocoelenterazine (dCTZ, dehydrogenated derivative of CTZ), coelenteramide (CTMD), and coelenteramine (CTM) from CTZ were also identified as their monosulfates.

Bioluminescent immunoassay for serine/threonine protein kinase activity using an aequorin-labeled monoclonal antibody and a synthetic peptide as a substrate.

A bioluminescent immunoassay system was developed to determine serine/threonine protein kinase activity using an aequorin-labeled monoclonal antibody and a synthetic peptide as the substrate. A monoclonal antibody against the synthetic phosphorylated serine peptide (K9P peptide) of histone H3 (19 amino acid residues), referred to as the H3S10P antibody, was chemically conjugated to maleimide-activated aequorin to prepare aequorin-labeled H3S10P (AQ-S-H3S10P). For the serine/threonine kinase assay, a non-phosphorylated serine peptide (K9C peptide) coated on a microplate was incubated with serine/threonine protein kinase in the presence of ATP and Mg2+. The resulting phosphorylated K9C peptides (K9P peptide) were identified using AQ-S-H3S10P. Thus, after the removal of unbound AQ-S-H3S10P though washing, the serine/threonine kinase activity was determined by the luminescence activity of aequorin from AQ-S-H3S10P bound to the K9P peptide. This assay system, in combination with the K9C peptide and AQ-S-H3S10P, could be used to screen inhibitors of various serine/threonine protein kinases in general.

Enzymatic conversion of dehydrocoelenterazine to coelenterazine using FMN-bound flavin reductase of NAD(P)H:FMN oxidoreductase.

Coelenterazine (CTZ) is known as luciferin (a substrate) for the luminescence reaction with luciferase (an enzyme) in marine organisms and is unstable in aqueous solutions. The dehydrogenated form of CTZ (dehydrocoelenterazine, dCTZ) is stable and thought to be a storage form of CTZ and a recycling intermediate from the condensation reaction of coelenteramine and 4-hydroxyphenylpyruvic acid to CTZ. In this study, the enzymatic conversion of dCTZ to CTZ was successfully achieved using NAD(P)H:FMN oxidoreductase from the bioluminescent bacterium Vibrio fischeri ATCC 7744 (FRase) in the presence of NADH (the FRase-NADH reaction). CTZ reduced from dCTZ in the FRase-NADH reaction was identified by HPLC and LC/ESI-TOF-MS analyses. Thus, dCTZ can be enzymatically converted to CTZ in vitro. Furthermore, the concentration of dCTZ could be determined by the luminescence activity using the CTZ-utilizing luciferases (Gaussia luciferase or Renilla luciferase) coupled with the FRase-NADH reaction.

Visualization of osteocalcin and bone morphogenetic protein 2 (BMP2) secretion from osteoblastic cells by bioluminescence imaging.

The secretions of osteocalcin and bone morphogenetic protein 2 (BMP2) from living osteoblastic cells were visualized for the first time using a method of video-rate bioluminescence imaging. The fusion proteins with Gaussia luciferase (GLase) for mouse osteocalcin and BMP2 (OC-GLase and BMP2-GLase, respectively) expressed in osteoblastic MC3T3-E1 cells were correctly processed and secreted. In the video images of exocytotic secretion, the luminescence spots of OC-GLase and BMP2-GLase disappeared rapidly and gradually, respectively, indicating different manners of these proteins in diffusion. Notably, a deletion mutant of BMP2 (Δ3BMP2-GLase) lacking three basic amino acid residues in the N-terminal region for binding to heparan sulfate showed rapidly disappearing luminescence spots. In our imaging conditions, the half-life of luminescence for the spots of Δ3BMP2-GLase (1.61 ± 0.20 s) was similar to that of OC-GLase (1.22 ± 0.14 s) but not to that of BMP2-GLase (4.31 ± 0.41 s). These results suggest that, in contrast to osteocalcin, the diffusion of BMP2 from cells occurred slowly after exocytosis. Thus, our bioluminescence imaging method is useful to study the diffusion properties of secreted proteins in exocytosis.

Efficient conversion to Cypridina luciferin from Cypridina luciferyl sulfate, coupled with enzymatic sulfation of acetic acid.

In Cypridina (Vargula) hilgendorfii, Cypridina luciferin is converted from Cypridina luciferyl sulfate by a sulfotransferase with adenosine 3', 5'-diphosphate (PAP), and is used for the luminescence reaction of Cypridina luciferase. We found that the luminescence activity of crude extracts of C. hilgendorfii was significantly stimulated by the addition of acetic acid. This stimulation may be explained by an efficient supply of PAP from 3'-phosphoadenosine 5'-phosphosulfate (PAPS) catalyzed by a sulfotransferase. Thus, acetic acid acts as a sulfate acceptor from PAPS, followed by forming acetyl sulfate and PAP. The structure of acetyl sulfate was identified using mass spectrometry and it spontaneously decomposed to acetic acid and free sulfate ion in aqueous solutions. This enzymatic conversion from Cypridina luciferyl sulfate to Cypridina luciferin could be coupled with acetic acid and PAPS by a sulfotransferase.

A novel yellow fluorescent protein of recombinant apoPholasin with dehydrocoelenterazine.

Pholasin is classified as a photoprotein and comprises apoPholasin (an apoprotein of pholasin) and an unknown prosthetic group as the light-emitting source. The luminescence reaction of pholasin is triggered by reactive oxygen species. Recombinant apoPholasin was recently expressed as a fusion protein of glutathione S-transferase (GST-apoPholasin) and purified from E. coli cells. By incubating non-fluorescent dehydrocoelenterazine (dCTZ, dehydrogenated form of CTZ) with GST-apoPholasin, the complex of GST-apoPholasin and dCTZ (GST-apoPholasin/dCTZ complex) was formed immediately and showed bright yellow fluorescence (λmax = 539 nm, excited at 430 nm). Unexpectedly, the fluorescent chromophore of the GST-apoPholasin/dCTZ complex was identified as non-fluorescent dCTZ. The luminescence intensity of the GST-apoPholasin/dCTZ complex was increased in a catalase-H2O2 system, but not in sodium hypochlorite.

Expression, purification, and characterization of recombinant apoPholasin.

Pholasin is a reactive oxygen-sensitive photoprotein that consists of an apoprotein (apoPholasin) and an unknown chromophore. The preferred human codon-optimized apoPholasin gene was transiently expressed in mammalian cells and apoPholasin was detected using an anti-recombinant apoPholasin antibody. For the first time, we found that apoPholasin secreted into the culture medium could catalyze the oxidation of coelenterazine (CTZ, a luciferin) to produce continuous luminescence. The fusion protein of apoPholasin and glutathione S-transferase (GST-apoPholasin) was successfully expressed as a soluble form in bacterial cells using the cold induction system. The purified GST-apoPholasin also had luminescence activity with CTZ, showing the bioluminescence emission peak at 461 nm, and the resultant product showed purple blue fluorescence under 365 nm light. Unexpectedly, the main oxidation product of CTZ was identified as coelenteramine (CTM), not coelenteramide (CTMD).

Identification of a novel oxidation product from yellow fluorophore in the complex of apoPholasin and dehydrocoelenterazine.

The complex of the recombinant fusion protein of apoPholasin and glutathione S-transferase (GST-apoPholasin) with non-fluorescent dehydrocoelenterazine (dCTZ) (GST-apoPholasin/dCTZ complex) shows yellow fluorescence at 539 nm by excitation at 430 nm. The GST-apoPholasin/dCTZ complex with a fluorophore (dCTZ*) has considerably weak luminescence activity, converting slowly to a blue fluorescence protein with the emission peak at 430 nm. The main oxidation products from dCTZ* for blue fluorescence were identified as coelenteramine (CTM) and an unreported pyrazine derivative, 3-benzyl-5-(4-hydroxyphenyl)pyrazin-2(1H)-one (CTO) that was confirmed by chemical synthesis.

Revalidation of recombinant aequorin as a light emission standard: Estimation of specific activity of Gaussia luciferase.

To validate the use of recombinant aequorin (reAequorin) as a light emission standard, the protein concentrations of highly purified reAequorin were determined by amino acid composition analysis, and the presence of active reAequorin was confirmed by the ratio of absorbance peak at 460 nm to that at 280 nm. The high correlation of the luminescence intensity with the protein concentration showed that reAequorin could be used for a light emission standard to study the luminescence properties of luciferases and to evaluate the detection sensitivity of luminometers. The specific activity of Gaussia luciferase with Imax was 7.5-fold higher than that of reAequorin and was calculated to be 3.8 × 1016 quanta/mg protein.

Single-step purification of recombinant Gaussia luciferase from serum-containing culture medium of mammalian cells.

A dihydrofolate reductase-deficient Chinese hamster ovary (CHO-K1/dhfr-) cell line stably expressing Gaussia luciferase with a histidine-tag sequence at the carboxyl terminus (GLase-His) was established. Recombinant GLase-His was purified from serum-containing culture medium by single-step Ni-chelate column chromatography in the presence of 2 M NaCl and 0.01% Tween 20. The protein yield of GLase-His with over 95% purity was 0.5 mg from 0.9 L of the cultured medium. The enzymatic properties of purified GLase-His were characterized. Interestingly, non-ionic detergent Tween 20 stabilized and stimulated GLase-His activity and its luminescence activity was stimulated 2-fold by the synergistic effect of 0.01% Tween 20 and 150 mM NaCl.

Quantitative visualization of synchronized insulin secretion from 3D-cultured cells.

Quantitative visualization of synchronized insulin secretion was performed in an isolated rat pancreatic islet and a spheroid of rat pancreatic beta cell line using a method of video-rate bioluminescence imaging. Video-rate images of insulin secretion from 3D-cultured cells were obtained by expressing the fusion protein of insulin and Gaussia luciferase (Insulin-GLase). A subclonal rat INS-1E cell line stably expressing Insulin-GLase, named iGL, was established and a cluster of iGL cells showed oscillatory insulin secretion that was completely synchronized in response to high glucose. Furthermore, we demonstrated the effect of an antidiabetic drug, glibenclamide, on synchronized insulin secretion from 2D- and 3D-cultured iGL cells. The amount of secreted Insulin-GLase from iGL cells was also determined by a luminometer. Thus, our bioluminescence imaging method could generally be used for investigating protein secretion from living 3D-cultured cells. In addition, iGL cell line would be valuable for evaluating antidiabetic drugs.

Visualization of glucagon secretion from pancreatic α cells by bioluminescence video microscopy: Identification of secretion sites in the intercellular contact regions.

We have firstly visualized glucagon secretion using a method of video-rate bioluminescence imaging. The fusion protein of proglucagon and Gaussia luciferase (PGCG-GLase) was used as a reporter to detect glucagon secretion and was efficiently expressed in mouse pancreatic α cells (αTC1.6) using a preferred human codon-optimized gene. In the culture medium of the cells expressing PGCG-GLase, luminescence activity determined with a luminometer was increased with low glucose stimulation and KCl-induced depolarization, as observed for glucagon secretion. From immunochemical analyses, PGCG-GLase stably expressed in clonal αTC1.6 cells was correctly processed and released by secretory granules. Luminescence signals of the secreted PGCG-GLase from the stable cells were visualized by video-rate bioluminescence microscopy. The video images showed an increase in glucagon secretion from clustered cells in response to stimulation by KCl. The secretory events were observed frequently at the intercellular contact regions. Thus, the localization and frequency of glucagon secretion might be regulated by cell-cell adhesion.

A fusion protein of the synthetic IgG-binding domain and aequorin: Expression and purification from E. coli cells and its application.

Aequorin is a Ca2+-binding photoprotein that is a complex of apoaequorin (apoAQ) and 2-peroxycoelenterazine. In this study, the fusion protein (ZZ-apoAQ) composed of the synthetic IgG-binding domain (ZZ domain) derived from Staphylococcus aureus protein A and apoAQ was expressed into the periplasmic space of Escherichia coli cells. ZZ-apoAQ was highly purified using Ni-chelate affinity chromatography followed by IgG affinity chromatography. ZZ-AQ was prepared from purified ZZ-apoAQ by incubation with coelenterazine and was characterized, including its luminescence properties. ZZ-AQ could be used as a reporter for detecting IgG and the measurable range of IgG coated on a 96-well plate was 1-1000 ng/mL.

Identification of amino acid residues responsible for high initial luminescence intensity in a calcium-binding photoprotein, clytin-II.

Clytin-II (CL-II) is an isotype of the calcium-binding photoprotein clytin-I (CL-I) from Clytia gregaria. CL-II shows approximately 4.5-fold higher initial luminescence intensity than CL-I with the same luminescence capacity, and is a potential candidate for a G-protein-coupled receptor assay among photoproteins. To investigate the high initial luminescence intensity of CL-II, the chimeric proteins between CL-I and CL-II were prepared and the responsible amino acid residues in CL-II were identified by site-specific mutagenesis of CL-I. The luminescence properties of CL-I were converted to those of CL-II by the replacement of only four amino acids in CL-I, and these amino acids did not interact with 2-peroxycolenterazine.

Crystal structure of nanoKAZ: The mutated 19 kDa component of Oplophorus luciferase catalyzing the bioluminescent reaction with coelenterazine.

The 19 kDa protein (KAZ) of Oplophorus luciferase is a catalytic component, that oxidizes coelenterazine (a luciferin) with molecular oxygen to emit light. The crystal structure of the mutated 19 kDa protein (nanoKAZ) was determined at 1.71 Å resolution. The structure consists of 11 antiparallel ß-strands forming a ß-barrel that is capped by 4 short α-helices. The structure of nanoKAZ is similar to those of fatty acid-binding proteins (FABPs), even though the amino acid sequence similarity was very low between them. The coelenterazine-binding site and the catalytic site for the luminescence reaction might be in a central cavity of the ß-barrel structure.

Protein expression of preferred human codon-optimized Gaussia luciferase genes with an artificial open-reading frame in mammalian and bacterial cells.

The protein expressions of three preferred human codon-optimized Gaussia luciferase genes (pGLuc, EpGLuc, and KpGLuc) were characterized in mammalian and bacterial cells by comparing them with those of wild-type Gaussia luciferase gene (wGLuc) and human codon-optimized Gaussia luciferase gene (hGLuc). Two synthetic genes of EpGLuc and KpGLuc containing the complete preferred human codons have an artificial open-reading frame; however, they had the similar protein expression levels to those of pGLuc and hGLuc in mammalian cells. In bacterial cells, the protein expressions of pGLuc, EpGLuc, and KpGLuc with approximately 65% GC content were the same and showed approximately 60% activities of wGLuc and hGLuc. The artificial open-reading frame in EpGLuc and KpGLuc did not affect the protein expression in mammalian and bacterial cells.

Codon optimization of genes for efficient protein expression in mammalian cells by selection of only preferred human codons.

A simple design method for codon optimization of genes to express a heterologous protein in mammalian cells is described. Codon optimization was performed by choosing only codons preferentially used in humans and with over 60% GC content, and the method was named the "preferred human codon-optimized method." To test our simple rule for codon optimization, the preferred human codon-optimized genes for six proteins containing photoproteins (aequorin and clytin II) and luciferases (Gaussia luciferase, Renilla luciferase, and firefly luciferases from Photinus pyralis and Luciola cruciata) were chemically synthesized and transiently expressed in Chinese hamster ovary-K1 cells. All preferred human codon-optimized genes showed higher luminescence activity than the corresponding wild-type genes. Our simple design method could be used to improve protein expression in mammalian cells efficiently.

Unconventional secretion of the mutated 19 kDa protein of Oplophorus luciferase (nanoKAZ) in mammalian cells.

The putative amino-terminal signal peptide of the catalytic 19 kDa protein (KAZ) of Oplophorus luciferase was found to be a functional secretory peptide in mammalian cells. A 16 amino acid substituted mutant of KAZ (nanoKAZ) could be secreted from mammalian cells using the amino-terminal signal peptide of KAZ, but KAZ could not be secreted at all. Notably, nanoKAZ lacking the amino-terminal signal peptide could be secreted from mammalian cells, and the distribution of nanoKAZ on the cell membrane was confirmed by video-rate bioluminescence imaging. Thus, nanoKAZ lacking the amino-terminal signal peptide was expressed in the cytoplasm, translocated to the cell membrane, and released into the culture medium through an endoplasmic reticulum-Golgi-independent pathway.