The BRET2/arrestin assay in stable recombinant cells: a platform to screen for compounds that interact with G protein-coupled receptors (GPCRS)

In BRET2 (Bioluminescence Resonance Energy Transfer), a Renilla luciferase (RLuc) is used as the donor protein, while a Green Fluorescent Protein (GFP2) is used as the acceptor protein. In the presence of the cell permeable substrate DeepBlueC, RLuc emits blue light at 395 nm. If the GFP2 is brought into close proximity to RLuc via a specific biomolecular interaction, the GFP2 will absorb the blue light energy and reemit green light at 510nm. BRET2 signals are therefore easily determined by measuring the ratio of green over blue light (510/395nm) using appropriate dual channel luminometry instruments (e.g., Fusion Universal Microplate Analyzer, Packard BioScience). Since no light source is required for BRET2 assays, the technology does not suffer from high fluorescent background or photobleaching, the common problems associated with standard FRET-based assays. Using BRET2, we developed a generic G Protein-Coupled Receptor (GPCR) assay based on the observation that activation of the majority of GPCRs by agonists leads to the interaction of beta-arrestin (a protein that is involved in receptor desensitization and sequestration) with the receptor. We established a cell line stably expressing the GFP2:beta-arrestin 2 fusion protein, and showed that it can be used to monitor the activation of various transiently expressed GPCRs, in BRET2/arrestin assays. In addition, using the HEK 293/GFP2:beta-arrestin 2 cell line as a recipient, we generated a double-stable line co-expressing the vasopressin 2 receptor (V2R) fused to RLuc (V2R:RLuc) and used it for the pharmacological characterization of compounds in BRET2/arrestin assays. This approach yields genuine pharmacology and supports the BRET2/arrestin assay as a tool that can be used with recombinant cell lines to characterize ligand-GPCR interactions which can be applied to ligand identification for orphan receptors.     

Effect of enhanced Renilla luciferase and fluorescent protein variants on the Förster distance of Bioluminescence resonance energy transfer (BRET)

Bioluminescence resonance energy transfer (BRET) is an important tool for monitoring macromolecular interactions and is useful as a transduction technique for biosensor development. Förster distance (R₀), the intermolecular separation characterized by 50% of the maximum possible energy transfer, is a critical BRET parameter. R₀ provides a means of linking measured changes in BRET ratio to a physical dimension scale and allows estimation of the range of distances that can be measured by any donor–acceptor pair. The sensitivity of BRET assays has recently been improved by introduction of new BRET components, RLuc2, RLuc8 and Venus with improved quantum yields, stability and brightness. We determined R₀ for BRET¹ systems incorporating novel RLuc variants RLuc2 or RLuc8, in combination with Venus, as 5.68 or 5.55 nm respectively. These values were approximately 25% higher than the R₀ of the original BRET¹ system. R₀ for BRET² systems combining green fluorescent proteins (GFP²) with RLuc2 or RLuc8 variants was 7.67 or 8.15 nm, i.e. only 2–9% greater than the original BRET² system despite being ～30-fold brighter.     

Comparison of enhanced bioluminescence energy transfer donors for protease biosensors

Bioluminescence energy transfer (BRET) is a powerful tool for the study of protein-protein interactions and conformational changes within proteins. We directly compared two recently developed variants of Renilla luciferase (RLuc), RLuc2 and RLuc8, as BRET donors using an in vitro thrombin assay. The comparison was carried out by placing a thrombin-specific cleavage sequence between the donor luciferase and a green fluorescent protein (GFP(2)) acceptor. Substitution of native RLuc with the RLuc mutants, RLuc2 and 8, in a BRET(2) fusion protein increased the light output by a factor of ~10. Substitution of native RLuc with either of the RLuc mutants resulted in a decrease in BRET(2) ratio by a factor of ~2 when BRET(2) components were separated by the thrombin cleavage sequence. BRET(2) ratios changed by factors of 18.8±1.2 and 18.2±0.4 for GFP(2)-RG-RLuc2 and GFP(2)-RG-RLuc8 fusion proteins, respectively, on thrombin cleavage compared to 28.8±0.20 for GFP(2)-RG-RLuc. The detection limits for thrombin were 0.23 and 0.26 nM for RLuc2 and RLuc8 BRET(2) systems, respectively, and 15 pM for GFP(2)-RG-RLuc. However, overall, the mutant BRET systems remain more sensitive than FRET and brighter than standard BRET(2).     

THE BRET2/ARRESTIN ASSAY IN STABLE RECOMBINANT CELLS: A PLATFORM TO SCREEN FOR COMPOUNDS THAT INTERACT WITH G PROTEIN-COUPLED RECEPTORS (GPCRS)*

ABSTRACT

In BRET² (Bioluminescence Resonance Energy Transfer), a Renilla luciferase (RLuc) is used as the donor protein, while a Green Fluorescent Protein (GFP²) is used as the acceptor protein. In the presence of the cell permeable substrate DeepBlueC?, RLuc emits blue light at 395?nm. If the GFP² is brought into close proximity to RLuc via a specific biomolecular interaction, the GFP² will absorb the blue light energy and reemit green light at 510?nm. BRET² signals are therefore easily determined by measuring the ratio of green over blue light (510/395?nm) using appropriate dual channel luminometry instruments (e.g., Fusion? Universal Microplate Analyzer, Packard BioScience). Since no light source is required for BRET² assays, the technology does not suffer from high fluorescent background or photobleaching, the common problems associated with standard FRET-based assays. Using BRET², we developed a generic G Protein–Coupled Receptor (GPCR) assay based on the observation that activation of the majority of GPCRs by agonists leads to the interaction of β-arrestin (a protein that is involved in receptor desensitization and sequestration) with the receptor. We established a cell line stably expressing the GFP² : β-arrestin 2 fusion protein, and showed that it can be used to monitor the activation of various transiently expressed GPCRs, in BRET²/arrestin assays. In addition, using the HEK 293/GFP² : β-arrestin 2 cell line as a recipient, we generated a double-stable line co-expressing the vasopressin 2 receptor (V₂R) fused to RLuc (V₂R : RLuc) and used it for the pharmacological characterization of compounds in BRET²/arrestin assays. This approach yields genuine pharmacology and supports the BRET²/arrestin assay as a tool that can be used with recombinant cell lines to characterize ligand–GPCR interactions which can be applied to ligand identification for orphan receptors.     

Design of near-infrared single-domain fluorescent protein GAF-FP based on bacterial phytochrome

Fluorescent proteins (FPs) are widely used as genetically encoded markers for quantitative and noninvasive study of biological processes. Development of biomarkers that are fluorescent in the near-infrared spectral range allows the tissues of animals to be studied at a deeper level because they are more permeable to the light of this wavelength range than that of visible range. Such properties as low molecular weight and monomeric state are important for widespread use of FPs. In this paper, we managed to obtain FP based on the chromophore-binding domain of bacterial phytochrome (BphP) from Rhodopseudomonas palustris (RpB-phP1), named GAF-FP, with a molecular weight of ~19 kDa, which is half that of other FP based on BphP and 1.4 times lower than that of commonly used GFP-like proteins, which are fluorescent in the near-infrared range. In contrast to most other near-infrared FPs, GAF-FP is a monomer, which has a high photostability, and its structure is stable to the incorporation of small peptide inserts. Moreover, GAF-FP is capable of covalent attachment of two different tetrapyrrole chromophores: phycocyanobilin (PCB) and biliverdin (BV), which is contained in mammalian tissues. GAF-FP with attached BV as a chromophore (GAF-FP–BV) has the main absorption band with a maximum at 635 nm. The fluorescence maximum falls at 670 nm, whereby GAF-FP has a high ratio of the fluorescence signal to the background signal even if FP is localized at a depth of several mm below the tissue surface. Together with the near-infrared absorption band, GAF-FP–BV also has an absorption band in the violet region of the spectrum with a maximum at 378 nm. We used this property to design a chimeric protein consisting of modified luciferase from Renilla reniformis (RLuc8) and GAF-FP. We showed resonance energy transfer from the substrate, the excited state of which occurs when oxidized by luciferase, to the chromophore GAF-FP–BV in the designed fusion protein. In the absence of an energy acceptor, RLuc8 catalyzes the cleavage of the substrate with the emission of the light with a maximum at 400 nm. At the same time, the energy from the substrate is transferred to the FP chromophore and then emitted in the near-infrared range corresponding to the spectrum of GAF-FP fluorescence in the GAF-FP–RLuc8 chimeric protein. These results open the way for the development of new small near-infrared FPs based on various natural BphPs with a view to their widespread use in cell and molecular biology.     

The cAMP-Dependent Protein Kinase Inhibitor H-89 Attenuates the Bioluminescence Signal Produced by Renilla Luciferase

Background

Investigations into the regulation and functional roles of kinases such as cAMP-dependent protein kinase (PKA) increasingly rely on cellular assays. Currently, there are a number of bioluminescence-based assays, for example reporter gene assays, that allow the study of the regulation, activity, and functional effects of PKA in the cellular context. Additionally there are continuing efforts to engineer improved biosensors that are capable of detecting real-time PKA signaling dynamics in cells. These cell-based assays are often utilized to test the involvement of PKA-dependent processes by using H-89, a reversible competitive inhibitor of PKA.

Principal Findings

We present here data to show that H-89, in addition to being a competitive PKA inhibitor, attenuates the bioluminescence signal produced by Renilla luciferase (RLuc) variants in a population of cells and also in single cells. Using 10 µM of luciferase substrate and 10 µM H-89, we observed that the signal from RLuc and RLuc8, an eight-point mutation variant of RLuc, in cells was reduced to 50% (±15%) and 54% (±14%) of controls exposed to the vehicle alone, respectively. In vitro, we showed that H-89 decreased the RLuc8 bioluminescence signal but did not compete with coelenterazine-h for the RLuc8 active site, and also did not affect the activity of Firefly luciferase. By contrast, another competitive inhibitor of PKA, KT5720, did not affect the activity of RLuc8.

Significance

The identification and characterization of the adverse effect of H-89 on RLuc signal will help deconvolute data previously generated from RLuc-based assays looking at the functional effects of PKA signaling. In addition, for the current application and future development of bioluminscence assays, KT5720 is identified as a more suitable PKA inhibitor to be used in conjunction with RLuc-based assays. These principal findings also provide an important lesson to fully consider all of the potential effects of experimental conditions on a cell-based assay readout before drawing conclusions from the data.     

Firefly luciferase and RLuc8 exhibit differential sensitivity to oxidative stress in apoptotic cells

Over the past decade, firefly Luciferase (fLuc) has been used in a wide range of biological assays, providing insight into gene regulation, protein-protein interactions, cell proliferation, and cell migration. However, it has also been well established that fLuc activity can be highly sensitive to its surrounding environment. In this study, we found that when various cancer cell lines (HeLa, MCF-7, and 293T) stably expressing fLuc were treated with staurosporine (STS), there was a rapid loss in bioluminescence. In contrast, a stable variant of Renilla luciferase (RLuc), RLuc8, exhibited significantly prolonged functionality under the same conditions. To identify the specific underlying mechanism(s) responsible for the disparate sensitivity of RLuc8 and fLuc to cellular stress, we conducted a series of inhibition studies that targeted known intracellular protein degradation/modification pathways associated with cell death. Interestingly, these studies suggested that reactive oxygen species, particularly hydrogen peroxide (H(2)O(2)), was responsible for the diminution of fLuc activity. Consistent with these findings, the direct application of H(2)O(2) to HeLa cells also led to a reduction in fLuc bioluminescence, while H(2)O(2) scavengers stabilized fLuc activity. Comparatively, RLuc8 was far less sensitive to ROS. These observations suggest that fLuc activity can be substantially altered in studies where ROS levels become elevated and can potentially lead to ambiguous or misleading findings.     

Luciferase-YFP fusion tag with enhanced emission for single-cell luminescence imaging

Taking advantage of the phenomenon of bioluminescence resonance energy transfer (BRET), we developed a bioluminescent probe composed of EYFP and Renilla reniformis luciferase (RLuc)--BRET-based autoilluminated fluorescent protein on EYFP (BAF-Y)--for near-real-time single-cell imaging. We show that BAF-Y exhibits enhanced RLuc luminescence intensity and appropriate subcellular distribution when it was fused to targeting-signal peptides or histone H2AX, thus allowing high spatial and temporal resolution microscopy of living cells.     

Development and application of hepatitis C reporter viruses with genotype 1 to 7 core-nonstructural protein 2 (NS2) expressing fluorescent proteins or luciferase in modified JFH1 NS5A

To facilitate genotype-specific high-throughput studies of hepatitis C virus (HCV), we have developed reporter viruses using JFH1-based recombinants expressing core-nonstructural protein 2 (NS2) of genotype 1 to 7 prototype isolates. We introduced enhanced green fluorescent protein (EGFP) into NS5A domain III of the genotype 2a virus J6/JFH1 [2a(J6)]. During Huh7.5 cell culture adaptation, 2a(J6)-EGFP acquired a 40-amino-acid (aa) (Δ40) or 25-aa (Δ25) deletion in NS5A domain II, rescuing the impairment of viral assembly caused by the EGFP insertion. Δ40 conferred efficient growth characteristics to 2a(J6) tagged with EGFP, DsRed-Express2, mCherry, or Renilla luciferase (RLuc), yielding peak supernatant infectivity titers of 4 to 5 log(10) focus-forming units (FFU)/ml. 2a(J6) with Δ40 or Δ25 was fully viable in Huh7.5 cells. In human liver chimeric mice, 2a(J6)-EGFPΔ40 acquired various deletions in EGFP, while 2a(J6)Δ40 did not show an impaired viability. We further developed panels of JFH1-based genotype 1 to 7 core-NS2 recombinants expressing EGFP- or RLuc-NS5AΔ40 fusion proteins. In cell culture, the different EGFP recombinants showed growth characteristics comparable to those of the nontagged recombinants, with peak infectivity titers of 4 to 5 log(10) FFU/ml. RLuc recombinants showed slightly less efficient growth characteristics, with peak infectivity titers up to 10-fold lower. Overall, the EGFP and RLuc recombinants were genetically stable after one viral passage. The usefulness of these reporter viruses for high-throughput fluorescence- and luminescence-based studies of HCV-receptor interactions and serum-neutralizing antibodies was demonstrated. Finally, using RLuc viruses, we showed that the genotype-specific core-NS2 sequence did not influence the response to alfa-2b interferon (IFN-alfa-2b) and that genotype 1 to 7 viruses all responded to treatment with p7 ion channel inhibitors.     

Crystal structures of the luciferase and green fluorescent protein from Renilla reniformis

Due to its ability to emit light, the luciferase from Renilla reniformis (RLuc) is widely employed in molecular biology as a reporter gene in cell culture experiments and small animal imaging. To accomplish this bioluminescence, the 37-kDa enzyme catalyzes the degradation of its substrate coelenterazine in the presence of molecular oxygen, resulting in the product coelenteramide, carbon dioxide, and the desired photon of light. We successfully crystallized a stabilized variant of this important protein (RLuc8) and herein present the first structures for any coelenterazine-using luciferase. These structures are based on high-resolution data measured to 1.4 Å and demonstrate a classic α/β-hydrolase fold. We also present data of a coelenteramide-bound luciferase and reason that this structure represents a secondary conformational form following shift of the product out of the primary active site. During the course of this work, the structure of the luciferase's accessory green fluorescent protein (RrGFP) was also determined and shown to be highly similar to that of Aequorea victoria GFP.     

Profile of the GSK Published Protein Kinase Inhibitor Set Across ATP-Dependent and-Independent Luciferases: Implications for Reporter-Gene Assays

A library of 367 protein kinase inhibitors, the GSK Published Kinase Inhibitor Set (PKIS), which has been annotated for protein kinase family activity and is available for public screening efforts, was assayed against the commonly used luciferase reporter enzymes from the firefly, Photinus pyralis (FLuc) and marine sea pansy, Renilla reniformis (RLuc). A total of 22 compounds (∼6% of the library) were found to inhibit FLuc with 10 compounds showing potencies ≤1 µM. Only two compounds were found to inhibit RLuc, and these showed relatively weak potency values (∼10 µM). An inhibitor series of the VEGFR2/TIE2 protein kinase family containing either an aryl oxazole or benzimidazole-urea core illustrate the different structure activity relationship profiles FLuc inhibitors can display for kinase inhibitor chemotypes. Several FLuc inhibitors were broadly active toward the tyrosine kinase and CDK families. These data should aid in interpreting the results derived from screens employing the GSK PKIS in cell-based assays using the FLuc reporter. The study also underscores the general need for strategies such as the use of orthogonal reporters to identify kinase or non-kinase mediated cellular responses.     

Anomalous fluorescence of yeast 3-phosphoglucerate kinase.

The 3-phosphoglycerate kinase (EC 2.7.2.3) of yeast which contains two tryptophyl and eight tyrosyl residues per molecule, displayed an unusualy fluorescence emission spectrum with a maximum at 308 nm when excited at 280 nm. The emission peak shifted to 329 nm when excited at 295 nm. We could confirm that it was due to the efficient quenching of tryptophyl fluorescence as well as to the incomplete energy transfer from tyrosyl to tryptophyl residues. The average fluorescence quantum yield of this protein was 0.076 (excitation at 280 nm) and that of tryptophyl residues was 0.046 (excitation at 295 nm). As the pH of the solution was lowered, the fluorescence intensity of phosphoglycerate kinase at 329 nm dramatically increased between pH 5 and 4, while the position of the peak remained unchanged. When denatured in 4 M guanidine hydrochloride, the protein showed two emission peaks, one at 343 nm and the other at 303 nm.     

A comparison of the intrinsic fluorescence of red kangaroo, horse and sperm whale metmyoglobins

Several metmyoglobins (red kangaroo, horse and sperm whale), containing different numbers of tyrosines, but with invariant tryptophan residues (Trp-7, Trp-14), exhibit intrinsic fluorescence when studied by steady-state front-face fluorometry. The increasing tyrosine content of these myoglobins correlates with a shift in emission maximum to shorter wavelengths with excitation at 280 nm: red kangaroo (Tyr-146) emission maximum 335 nm; horse (Tyr-103, -146) emission maximum 333 nm; sperm whale (Tyr-103, -146, -151) emission maximum 331 nm. Since 280 nm excites both tyrosine and tryptophan, this strongly suggests that tyrosine emission is not completely quenched but also contributes to this fluorescence emission. Upon titration to pH 12.5, there is a reversible shift of the emission maximum to longer wavelengths with an increase greater than 2-fold in fluorescence intensity. With excitation at 305 nm, a tyrosinate-like emission is detected at a pH greater than 12. These studies show that: (1) metmyoglobins, Class B proteins containing both tyrosine and tryptophan residues, exhibit intrinsic fluorescence; (2) tyrosine residues also contribute to the observed steady-state fluorescence emission when excited by light at 280 nm; (3) the ionization of Tyr-146 is likely coupled to protein unfolding.     

Intrinsic fluorescence of B and Z forms of poly d(G-m5C).poly d(G-m5C), a synthetic double-stranded DNA: spectra and lifetimes by the maximum entropy method.

A study has been made of the fluorescence of poly d(G-m5C).poly d(G-m5C), a synthetic double-stranded DNA, in buffered neutral aqueous solution at room temperature, excited by synchrotron radiation at 280 nm and 250 nm and by a frequency-doubled pulse dye laser at 290 nm. Exciting at 280 nm, the B form shows a uni-modal UV spectrum with lambdaf(max) approximately 340 nm. The Z form has in addition a visible emission lambdaf(max) at 450 nm. The spectral positions remain unchanged on exciting at 250 nm but the relative intensities change considerably. Decay profiles have been obtained at 360 nm and 450 nm for both the B and Z forms and have been analyzed by fitting to a pseudo-continuous distribution of 100 (and occasionally 200) exponentials, ranging from 10 ps to 20 ns, by optimizing the 'entropy' of the signal (the method of maximum entropy). We find the mean lifetimes for both wavelengths of emission and for both structural forms fall into three well-separated regions in the ranges indicated tau1 approximately 0.04-0.21 ns, tau2 approximately 0.9-1.26 ns, and tau3 approximately 5.1-6.5 ns. The UV emission, from its spectral position and half-width, correlates with monomeric emission from m5C (and from C for poly d(G-C)). However the lifetime tau1 is approximately 2 orders of magnitude longer than the monomers and points to an involvement of protonated guanosine (GH+, tauf approximately 200 ps) in the overall absorption/emission sequence. In the UV the tau3 emission is predominant, with fractional time-integrated emission approximately 86% for B DNA and approximately 64% for Z. We suggest it results from exciton (stacked) absorption followed by dissociative emission. For Z DNA the visible (450 nm) emission is dominated by a tau3 species (approximately 91%) with a lifetime of 6.5 ns and we suggest it represents a hetero-excimer emission consequent upon absorption by the strongly overlapped base-stacking, which differs from that in B DNA. The weak emission corresponding to tau2 is made more apparent by scanned gated detection of the emission from laser excitation (290 nm) of single-crystal d(m5C-G)3. A central role is attributed to the tight stacking of the bases in the Z form which correlates with enhanced hypochromism at 250 nm vs. 280 nm and with the reversal of the fluorescence intensity ratios UV-visible between these wavelengths.     

Substituted 4-[4-(dimethylamino)styryl]pyridinium salt as a fluorescent probe for cell microviscosity

In aqueous solution, 4-[4-(dimethylamino)styryl]pyridine (DMASP) derivatives displayed dual fluorescence, in which excitation at either 469 or 360 nm produced an emission band near 600 nm. Increasing the viscosity of the environment intensified the fluorescence emission obtained at the longer wavelength of excitation, whereas the emission at the lower wavelength of excitation showed little change in intensity. Thus, using the ratio of the 600 nm emission obtained by exciting at 469 nm to that obtained with 360 nm excitation, it is possible to obtain a value related to the local viscosity that does not depend on the system parameters. The fluorescence emission of the dye in aqueous solution, as well as in living cells, is well suited for use with visible fluorescence spectroscopy. The N-carboxymethyl butyl ester DMASP derivative (1) was found to be irreversibly loaded into living smooth muscle cells, presumably because it is hydrolyzed by cellular esterases, transforming it into a membrane-impermeable fluorescent carboxylate DMASP derivative. (2) After calibrating 2 against glycerol/water and sucrose/water mixtures of known viscosity, the fluorescence ratio generated from cultured smooth muscle cells in dual-excitation mode gave an average intracellular viscosity of 4.5 cP. This value corresponds to those reported in the literature.     

Spatial energy transfer with observation of bimolecular singlet oxygen emission using quantum dots as donors and zinc‐phthalocyanine as acceptors

This study reports the influence of CdSe–ZnS core–shell quantum dots (QDs) for formation of singlet oxygen using zinc‐phthalocyanine (ZnPc) dyes in colloidal solutions. Using a microluminescence surface scan technique it was possible to measure accurately the photon diffusion length, or photon mean free path, inside the medium. Analyses were performed for a range of QD concentrations. Photon diffusion length was assigned to the bimolecular singlet oxygen emission at 707 nm. Related singlet oxygen emission was predicted by observing quenching of the photon diffusion length measured at the specific oxygen emission as a function of QD concentration, being a nontrivial phenomenon related to the QD donors. Diffusion length measured at 707 nm increased with QD concentration; in the absence of QDs, as in pure ZnPc samples, the emission peak at 707 nm was not observed.     

X-ray photoelectron spectroscopy and tunable photoluminescence study of gold nanoparticles embedded in PVA films

This article reports the systematic photoluminescence study of the various contents of gold nanocomposites in polyvinyl alcohol (PVA) films. The variations in the gold content in PVA film were 0.2, 0.5, 1.0, and 1.5 wt%. All the samples were excited at two selected wavelengths; those are at 400 nm and 532 nm. On exciting the gold-PVA nanocomposite films at 400 nm the photoluminescence was observed in the region of 430–500 nm in comparison to pure PVA films that show an emission at 400 nm. However, on exciting the gold-PVA nanocomposites at 532 nm, the emission was observed at 560–650 nm with a long tail till 700 nm that is unlike the pure PVA films that do not show any emission peak in this region. This suggests that emission between 430 and 500 nm regions is due to the coordination of PVA with gold nanoparticles because PVA has an emission at 400 nm. However, the emission peak between 560 and 650 nm is entirely due to the gold nanocomposite particle. The peak also shows a smaller red-shift that is usually with the increasing nanoparticles size with the increasing content in the PVA films. The formation of gold nanoparticles was justified by X-ray diffraction (XRD) analysis which is further supported by X-ray photoelectron spectroscopy (XPS) analysis.     

Investigations of the Blue-green Fluorescence Emission of Plant Leaves

The UV light (337 nm) induced blue-green fluorescence emission of green leaves is characterized at room temperature (298 K) by a maximum near 450 nm (blue region) and a shoulder near 525 nm (green region) and was here also studied at 77 K. At liquid nitrogen temperature (77 K) the blue (F450) and green fluorescence (F525) are much enhanced as is the red chlorophyll fluorescence near 735 nm. During development of green tobacco leaves the blue fluorescence F450 (77 K) is shifted towards longer wavelengths from about 410 nm to 450 nm. The isolated leaf epidermis of tobacco showed only slight fluorescence emission with a maximum near 410 nm. The green fluorescence F525 was found to mainly originate from the mesophyll of the leaf, its intensity increased when the epidermis was removed. The red chlorophyll fluorescence emission was also enhanced when the epidermis was stripped off; this considerably changed the blue/red fluorescence ratios F450/F690 and F450/F735. The epidermis, with its cell wall and UV-light-absorbing substances in its vacuole, plays the role of a barrier for the exciting UV-light. In contrast to intact and homogenized leaves, isolated intact chloroplasts and thylakoid membranes did not exhibit a blue-green fluorescence emission.