The green fluorescent protein (GFP) from the hydromedusa Aequorea victoria and its derivatives have become indispensable imaging devices in cell biology. In previous years, a wide variety of GFP-like proteins were discovered in non-bioluminescent anthozoa. Some of them displayed exciting new properties, including photoactivated changes of the fluorescence emission intensity and wavelength. Photoactivatable proteins offer a high potential as tools for regional optical marking in live cells and tissues. This review aims to give an overview of photoactivatable marker proteins, focusing on the molecular basis of light-induced green to red photoconversion in EosFP.
We have investigated the autocatalytic mechanism of green fluorescent protein (GFP) maturation. To this end, we have used techniques such as site-directed mutagenesis, X-ray crystallography and in vitro kinetics, and have monitored the reaction by fluorescence, HPLC and MALDI (matrix-assisted laser desorption ionization) mass spectrometry. In summary, we find that chromophore formation, which generally occurs within 40 to 60 min, can be accelerated dramatically under some conditions. In the E222Q variant, the rate-limiting process appears to be a function of slow proton transfer steps. Other mutagenesis data indicate that chromophore biogenesis is not driven by the aromatic character of residue 66. The GFP self-modification process involves a rate-limiting oxidation reaction that results in the production of H2O2. The data are most consistent with a reaction mechanism that proceeds via cyclization-oxidation-dehydration during in vitro maturation under aerobic conditions. The ejection of water from the heterocycle that is formed from main chain protein atoms appears to depend on the degree of π-overlap of the five-membered ring with the side chain adduct.
We report on the photophysical properties of a far-red intrinsic fluorescent protein by means of single molecule and ensemble spectroscopic methods. The green fluorescent protein (GFP) from Aequorea victoria is a popular fluorescent marker with genetically encoded fluorescence and which can be fused to any biological structure without affecting its function. GFP and its variants provide emission colors from blue to yellowish green. Red intrinsic fluorescent proteins from Anthozoa species represent a recent addition to the emission color palette provided by GFPs. Red intrinsic fluorescent markers are on high demand in protein-protein interaction studies based on fluorescence-resonance energy transfer or in multicolor tracking studies or in cellular investigations where autofluorescence possesses a problem. Here we address the photophysical properties of a far-red fluorescent protein (HcRed), a mutant engineered from a chromoprotein cloned from the sea anemone Heteractis crispa, by using a combination of ensemble and single molecule spectroscopic methods. We show evidence for the presence of HcRed protein as an oligomer and for incomplete maturation of its chromophore. Incomplete maturation results in the presence of an immature (yellow) species absorbing/fluorescing at 490/530-nm. This yellow chromophore is involved in a fast resonance-energy transfer with the mature (purple) chromophore. The mature chromophore of HcRed is found to adopt two conformations, a Transoriented form absorbing and 565-nm and non-fluorescent in solution and a Cis-oriented form absorbing at 590-nm and emitting at 645-nm. These two forms co-exist in solution in thermal equilibrium. Excitation-power dependence fluorescence correlation spectroscopy of HcRed shows evidence for singlet-triplet transitions in the microseconds time scale and for cis-trans isomerization occurring in a time scale of tens of microseconds. Single molecule fluorescence data recorded from immobilized HcRed proteins, all point to the presence of two classes of molecules: proteins with Cis and Trans-oriented chromophores. Immobilization of HcRed in water-filled pores of polyvinyl alcohol leads to a polymer matrix - protein barrel interaction which results in a 'freezing' of the chromophore in a stable conformation for which non-radiative deactivation pathways are either suppressed or reduced. As a result, proteins with both Cis- and Trans-oriented chromophores can be detected at the single molecule level. Polymer chain motion is suggested as a mediator for an eventual cis-trans isomerization of the chromophore in the case of single immobilized proteins.
Green fluorescent protein (GFP) is highly fluorescent with blue excitation despite the creation of a buried charge resulting from photoionization of the chromophore and neutralization of Glu 222. These major electrostatic rearrangements do not lead to rapid internal conversion processes. A competing phototransformation reaction, which ionizes the chromophore and decarboxylates Glu 222, causes electrostatic and structural changes which are very similar to those in the fluorescence photocycle. The X-ray crystallography and IR spectroscopy of phototransformed GFP provides evidence for relaxations involving protein, chromophore, solvent, and CO2. Time-resolved and static infrared spectroscopy and X-ray crystallography to 1.85 Å resolution identify structural mechanisms common to phototransformation and to the fluorescence photocycle. A detailed study of the ps time-resolved IR absorption during the fluorescence photocycle has been reported. Through global fitting the species associated difference spectra were determined showing the vibrational response to excited state proton transfer in addition to the transient population of a late ground state photocycle intermediate 'I2'. We additionally used pump-dump-probe spectroscopy to dump the deprotonated radiative state 'I*' and directly provide the I2-I* difference spectrum, which strongly resembled its photocycle counterpart. We discuss spectral markers that specifically report on structural relaxation during the photocycle which may be compared with structural relaxations in the competing phototransformation reaction.
The fluorescence decay of the biologically important enhanced green fluorescent protein (GFP) is a function of the refractive index of its environment (Suhling et al, Biophys J 83, 3589-3595, 2002). To address the question whether this effect can be exploited to image the local environment of specific proteins in cell biology, we need to determine the distance over which the GFP fluorescence decay is sensitive to the refractive index. To this end, we employ Fluorescence Lifetime Imaging (FLIM) of GFP in buffer solution at an air and at an oil interface. This approach allows us to map the fluorescence lifetime as a function of distance from the interface. Preliminary data show that the average fluorescence lifetime of GFP increases near a buffer/air interface and decreases near a buffer/oil interface. Similar results showing the same trend are obtained using fluorescein in buffer at an oil and at an air interface. The range over which this fluorescence lifetime change occurs is found to be of the order several micrometers which is consistent with theoretical models. In addition, GFP-tagged MHC proteins in fixed cells were imaged in different refractive index media using FLIM. It appears that the average GFP fluorescence lifetime in cells is also sensitive to different refractive index environments, and is inversely proportional to the square of the refractive index.
Despite the explosive growth of the Internet (in terms of the World Wide Web) as an informational resource for the original scientific literature pertaining to fluorescent protein investigations, there remains an obvious void in educational Websites targeted at beginning students and novices in the field. To address this issue, educational sites dedicated to optical microscopy and digital imaging being constructed and hosted at The Florida State University are turning their attention to the increasing application of fluorescent proteins for live-cell imaging studies. The primary focus of this effort is to create new sections of the sites that address the structure and properties of fluorescent proteins as well as optimizing their utility in imaging experiments.
A variety of fluorescent and chromophoric proteins homologous to the green fluorescent protein (GFP) has been recently discovered and cloned from non-bioluminescent marine animals, such as corals, and now provide a multitude of colors for use in fluorescence imaging applications. Recently, a novel fluorescence imaging methodology has emerged that utilizes the unique photoactivatory property of several GFP-like proteins, which respond to irradiation by altering their optical properties, thereby providing a new spatio-temporal capability to the GFP-based imaging applications. During our studies of GFP-like proteins from the Great Barrier Reef corals, several novel photoactivatable (PA) GFP-like proteins have been discovered. These include fluorescence photo-amplifiers and reversible photoswitchers, similar to PA jelly-fish derived PA-GFP and Dronpa, that greatly increase their emissions following ultraviolet-A (UVA) irradiation; the red-to-green (R-to-G) converters, similar to DsRed, that rapidly change to green color following single- or 2-photon irradiation; the green-to-red (G-to-R) converters, that acquire bright red fluorescence following UV-violet irradiation, similar to Kaede-like proteins; and the kindling GFP-like proteins, that are non fluorescent, but rapidly acquire bright fluorescence after green light irradiation. We report on the various optical characteristics of these coral PA proteins that may be used to expand the scope of the available fluorescence bio-imaging technologies.
The main applications of fluorescent proteins (FPs) are monitoring tumor growth, angiogenesis, metastases formation and effects of new classes of drugs. Different types of tomography allow fluorescence imaging of tumors located deep in human or animal tissue. These techniques were used for investigation of the distribution of near-infrared fluorescent probes, but only a few works are devoted to fluorescence tomography in visible light. In this work, preliminary results of the frequency domain fluorescent diffuse tomography (FD FDT) method in application to DsRed2 protein as a fluorescent agent are presented. For the first step of our experiments we utilized second harmonic generation of Nd:YAG laser (532 nm) modulated by low frequency (1 kHz) in the experimental setup. The transilluminative planar configuration was used in the setup. A series of model experiments has been conducted and show good agreement between theoretical and experimental fluorescence intensity. Post mortem experiments with capsules containing DsRed2 and scattering solution introduced into esophagus of rats to simulate tumor formation have been conducted. The results of these experiments show that sensitivity of the setup is sufficient to detect DsRed2 in concentrations similar to those in FP-expressed tumor, but the contrast is not enough high to separate fluorescence of DsRed2 and surrounding tissues. The setup can be significantly improved by utilizing high-frequency modulation (110 MHz using acousto-optical modulator) of the excitation light and precise phase measurements due to difference in fluorescence life-time of FPs and surrounding tissues. An algorithm of processing a fluorescent image based on calculating zero of maximum curvature was employed for detection of fluorescent inclusions boundaries in the image.
We have genetically engineered dual-color fluorescent cells with one color in the nucleus and the other in the cytoplasm that enables real-time nuclear-cytoplasmic dynamics to be visualized in living cells in the cytoplasm in vivo as well as in vitro. To obtain the dual-color cells, red fluorescent protein (RFP) was expressed of the cancer cells, and green fluorescent protein (GFP) linked to histone H2B was expressed in the nucleus. Mitotic cells were visualized by whole-body imaging after injection in the mouse ear. Common carotid artery or heart injection of dual-color cells and a reversible skin flap enabled the external visualization of the dual-color cells in microvessels in the mouse where extreme elongation of the cell body as well as the nucleus occurred. The migration velocities of the dual-color cancer cells in the capillaries were measured by capturing individual images of the dual-color fluorescent cells over time. Human HCT-116-GFP-RFP colon cancer and mouse mammary tumor (MMT)-GFP-RFP cells were injected in the portal vein of nude mice. Extensive clasmocytosis (destruction of the cytoplasm) of the HCT-116-GFP-RFP cells occurred within 6 hours. The data suggest rapid death of HCT-116-GFP-RFP cells in the portal vein. In contrast, MMT-GFP-RFP cells injected into the portal vein mostly survived and formed colonies in the liver. However, when the host mice were pretreated with cyclophosphamide, the HCT-116-GFP-RFP cells also survived and formed colonies in the liver after portal vein injection. These results suggest that a cyclophosphamide-sensitive host cellular system attacked the HCT-116-GFP-RFP cells but could not effectively kill the MMT-GFP-RFP cells. With the ability to continuously image cancer cells at the subcellular level in the live animal, our understanding of the complex steps of metastasis will significantly increase. In addition, new drugs can be developed to target these newly visible steps of metastasis.
We developed the Alb-DsRed2 transgenic (Tg) rat designed with liver-specific expression of the red fluorescent protein, DsRed2. Herein, we report high expression of DsRed2 in neonate liver of both sexes, although they were sexually dimorphic and exhibited a male-specific pattern in adult rats. In an effort to examine the expression in each animal under development, we employed an in vivo Bio-imaging system to quantitatively estimate hepatic DsRed2 expression levels. The temporal profiles pertaining to DsRed expression were similar in male and female Tg rats until 28 days old. The levels in both sexes decreased gradually following birth, and were not detectable at 21 days. Subsequently, expression in males increased again at 35 days and was maintained at a persistently high level thereafter. On the other hand, expression in females disappeared steadily. Although hepatic DsRed expression levels in gonadectomized Tg rats was not significantly different, DsRed expression in hypophysectomized female Tg rats appeared dramatically 72 hr following operation. Hepatocytes were collected from adult Tg rats and cultured in conditioning medium. DsRed expression in female hepatocytes could be detected 72 hr following culturing. These results suggest that hepatic DsRed expression in female rats is regulated in vivo by the pituitary. This report is shows use of Alb-DsRed2 Tg rats in conjunction with a novel bio-imaging system represents a powerful experimental system.
GFP is a fluorescent product of the jellyfish Aequorea victoria and has been used for a variety of biological experiments as a reporter molecule. While GFP possesses advantages for the non-invasive imaging of viable cells, GFP-positive cells are still considered potential xeno-antigens. It is difficult to observe the precise fate of transplanted cells/organs in recipients without immunological control. The aim of this study was to determine whether intrathymic injection of GFP to recipients and the depletion of peripheral lymphocytes could lead to donor-specific unresponsiveness to GFP-expressed cell. LEW rats were administered intraperitoneally with 0.2 ml of anti-rat lymphocyte serum (ALS) 1 day prior to intrathymic injection of donor splenocytes or adeno-GFP vector. Donor cells and vector were non-invasively inoculated into the thymus under high frequency ultrasound imaging using an echo-guide. All animals subsequently received a 7 days GFP-expressed skin graft from the same genetic background GFP LEW transgenic rat. Skin graft survival was greater in rats injected with donor splenocytes (23.6+/-9.1) compared with adeno-GFP (13.0+/-3.7) or untreated control rats (9.5+/-1.0). Intrathymic injection of donor antigen into adult rats can induce donor-specific unresponsiveness. Donor cells can be observed for a long-term in recipients with normal immunity using this strategy.
In vivo imaging strategies provide cellular and molecular events in real time that helps us to understand biological processes in living animals. The development of molecular tags such as green fluorescent proteins and luciferase from the firefly Photinus pyralis has lead to a revolution in the visualization of complex biochemical processes. We developed a novel inbred transgenic rat strain containing firefly luciferase based on the transgenic (Tg) technique in rats. This Tg rat expressed the luciferase gene ubiquitously under control of the ROSA26 promoter. Cellular immune responsiveness against the luciferase protein was evaluated using conventional skin grafting and resulted in the long-term acceptance of Tg rat skin on wild-type rats. Strikingly, organ transplant with heart and small bowel demonstrated organ viability and graft survival, suggesting that cells from luciferase-Tg are transplantable to track their fate. Taking advantage of the less immunogenic luciferase, we also tested the role of hepatocyte-infusion in a liver injury model, and bone marrow-derived cells in a skin defect model. Employed in conjunction with modern advances in optical imaging, this luciferase-Tg rat system provides an innovative animal tool and a new means of facilitating biomedical research such as in the case of regeneration medicine.
Luminescent proteins originally isolated from marine or terrestrial organisms have played a key role in the development of several biosensing systems. These proteins have been used in a variety of applications including, immunoassays, binding assays, cell-based sensing, high throughput screening, optical imaging, etc. Among the luminescent proteins isolated, the bioluminescent protein aequorin has been one of the proteins at the forefront in terms of its use in a vast number of biosensing systems. In our laboratory, we have employed aequorin as a label in the development of highly sensitive assays through chemical and genetic modifications from single step analysis of physiologically important molecules in biological fluids. An important aspect of optimizing these assays for clinical use involves understanding the stability of the various aequorin variants that are available. To this end we have designed several stability studies involving three important aequorin mutants, Mutant S, Mutant 5, and Mutant 53. The cysteine free aequorin, Mutant S, has been the most ubiquitously used aequorin variant in our laboratory because of its increased stability and activity as compared to native aequorin. Mutant 5 and Mutant 53 contain a single cyteine residue at position 5 and 53 in the protein, respectively. Because of the presence of a single cysteine residue, Mutant 5 and Mutant 53 both can be site-specifically conjugated. This site specific conjugation capability gives Mutant 5 and Mutant 53 an advantage over native aequorin when developing assays. Additional studies optimizing the expression, purification, and charging of aequorin Mutant S were also performed. A thorough understanding of the efficient expression, purification, and storage of these aequorin mutants will allow for the more practical utilization of these mutants in the development of future biosensing systems.
Analytical biophotonics techniques such as steady-state fluorescence spectroscopy and fluorescence microscopy have been proven to be important tools in the study of genetically engineered bacterial sensors for plant antigenotoxicity. The assay involves the use of E. coli RS4U tagged with the red fluorescent protein (RFP). The cells emit red fluoresce in direct proportion to the genotoxicant present. Antigenotoxicity is seen as an act of preventing the DNA-damage induced expression of RFP. Thus, co-treatment of the cells with the genotoxicant and antigenotoxicant plant samples resulted to the reduction of the RFP fluorescence.
A new antioxidant activity assay based on the reactive oxygen species (ROS)-inducible bacterial strain (E. coli DPD2511) is described. The strain harbors the plasmid pKatG::luxCDABE and responds to hydrogen peroxide treatment by increasing light emission at 490 nm. Antioxidant capacity is evaluated through the ability of an agent to inhibit the hydrogen peroxide-induced bioluminescence of E. coli DPD2511. Applicability of the developed assay in detecting levels of antioxidants in various aqueous plant extracts is demonstrated. The assay was validated against 2,2-diphenylpicrylhydrazyl (DPPH) assay, a known antioxidant assay.
Nisin is a lantibiotic, an antibacterial peptide produced by certain Lactococcus lactis strains that kills or inhibits the growth of other bacteria. Nisin is widely used as a food preservative, and its long-time use suggests that it can be generally regarded as safe. We have developed a method for determining the amount of nisin in food samples that is based on luminescent biosensor bacteria. Bacterial luciferase operon luxABCDE was inserted into plasmid pNZ8048, and the construct was transformed by electroporation into Lc. lactis strain NZ9800, whose ability to produce nisin has been erased by deletion of the gene nisA. The operon luxABCDE has been modified to be functional in gram-positive bacteria to confer a bioluminescent phenotype without the requirement of adding an exogenous substrate. In the plasmid pNZ8048, the operon was placed under control of the nisin-inducible nisA promoter. The chromosomal nisRK genes of Lc. lactis NZ9800 allow it to sense nisin in the environment and relay this signal via signal transduction proteins NisK and NisR to initiate transcription from nisA promoter. In the case of our sensor bacteria, this leads to production of luciferase and, thus, luminescence that can be directly measured from living bacteria. Luminescence can be detected as early as within minutes of induction. The nisin assay described here provides a detection limit in the sub-picogram level per ml, and a linear area between 1 - 1000 pg/ml. The sensitivity of this assay exceeds the performance of all previously published methods.
The glucose-mediated conformational changes in the glucose binding protein (GBP) have been exploited in the development of fluorescence based glucose sensors. The fluorescence response is generated by a polarity sensitive dye attached to a specific site. Such fluorescent sensors respond to submicromolar glucose at diffusion-controlled rates mimicking the wild type. However, such sensors have been limited to in vitro glucose sensing because of the preliminary dye-labeling step. In the study described here, the dye-labeling step is omitted by genetically encoding the GBP with two green fluorescent mutants namely, the green fluorescent protein (GFP) and the yellow fluorescent protein (YFP) in the N- and C-terminal ends, respectively. These two GFP mutants comprise a fluorescence resonance energy transfer (FRET) donor and acceptor pair. Thus, when glucose binds with GBP, the conformational changes affect the FRET efficiency yielding a dose-dependent response. A potential application for this FRET-based glucose biosensor is online glucose sensing in bioprocessing and cell culture. This was demonstrated by the measurement of glucose consumption in yeast fermentation. Further development of this system should yield in vivo measurement of glucose in bioprocesses.
Practically all fluorescent proteins from corals are oligomerized or aggregated in solutions. For zFP538 it was demonstrated that intensity of fluorescence and pH-profile change upon dilution that was explained by the dissociation of aggregates. Particle size distribution of zFP538 aggregates adsorbed on the surface of graphite from solutions in the concentration range 10-7 - 10-9 M was studied by AFM. Multi-population size distribution was observed for all three protein concentrations. Within one concentration, size distribution is characterized by three populations: small, medium, and large. For 10-7M concentration, the largest size is 10 times less than that in the 10-5M solution or solid film. For 10-8M concentration, the maximum population is of medium size, but smaller in aggregation number than the minimal population for 10-7M. Aggregation numbers for small, medium, and large populations for 10-9M concentration are the same as for 10-8M, but the population is shifted to smaller aggregation number. Thus, upon dilution, an average aggregation number is gradually shifted to a smaller value. Difference in small, medium, and large particles is observed within 10-7 - 10-8M concentration range.