In Vivo Imaging with Genetically Encoded Redox Biosensors

Abstract

Redox reactions are of high fundamental and practical interest since they are involved in both normal physiology and the pathogenesis of various diseases. However, this area of research has always been a relatively problematic field in the context of analytical approaches, mostly because of the unstable nature of the compounds that are measured. Genetically encoded sensors allow for the registration of highly reactive molecules in real-time mode and, therefore, they began a new era in redox biology. Their strongest points manifest most brightly in in vivo experiments and pave the way for the non-invasive investigation of biochemical pathways that proceed in organisms from different systematic groups. In the first part of the review, we briefly describe the redox sensors that were used in vivo as well as summarize the model systems to which they were applied. Next, we thoroughly discuss the biological results obtained in these studies in regard to animals, plants, as well as unicellular eukaryotes and prokaryotes. We hope that this work reflects the amazing power of this technology and can serve as a useful guide for biologists and chemists who work in the field of redox processes.

Keywords: NADPH; fluorescent proteins; genetically encoded sensors; glutathione (GSH), hydrogen peroxide (H2O2), in vivo imaging; mycothiol (MSH), NADH; reactive oxygen species (ROS); redox biology.

Figure 1

(A,B) Ratio images of whole intact larvae expressing cyto-Grx1-roGFP2 (A), mito-roGFP2-Grx1 (B). Probe responses in selected tissues were quantified in ten individual third-instar larvae for each biosensor (A’,B’). Boxes, lower/upper quartile; whiskers, 5th/95th percentile. * p < 0.05, ** p < 0.01, *** p < 0.001. M, muscle; H, hemocytes; G, gut; T, (Malpighian) tubules; F, fat. Scale bars, 600 µm. (C) A representative detailed image to highlight tissue-specific differences evident in mito-roGFP2-Grx1-expressing larvae. Arrows indicate hemocytes (H) and muscle tissue (M). Scale bar, 60 µm. (D) Feeding (top) and wandering (bottom) third-instar larvae expressing the cytosolic roGFP2-Orp1 probe. Scale bars, 600 µm. Reprinted by permission from Elsevier: Cell Metabolism [18], copyright 2011.

Figure 2

(A) Survival (scatter plot with Gompertz fit) and H₂O₂ levels (stacked bars) during development and aging. Diet-restricted (DR) worms live 76% longer than fully fed (FF) HyPer transgenic worms (p < 0.0001; log rank test). H₂O₂ levels did not change significantly during postembryonic development (L1–day 1; light green bars; slope = 0.0004 ± 0.006; p = 0.94; linear mixed model (LMM)). H₂O₂ levels increase significantly with age in FF (dark green bars; slope = 0.083 ± 0.028; p = 0.0002; LMM) and DR HyPer worms (red bars; slope = 0.030 ± 0.009; p = 0.005). Although FF and DR worms have similar H₂O₂ levels at the first day of adulthood (p = 0.78), DR significantly attenuated the age-related increase in H₂O₂ (p = 0.016). The portion of autofluorescence in the total ratio value is indicated with black bars. Error bars represent standard error of the mean (SEM). (B) Confocal analysis of young and old HyPer worms. Dot plot of mean individual H₂O₂ levels of young (day 0) and old (day 6) adults scanned with the same laser settings and quantified over the whole worm body; the population mean is indicated by a horizontal line. H₂O₂ levels, averaged over the population, significantly increase with age. Day 0, 0.42 ± 0.008; day 6, 0.57 ± 0.016; p < 0.01 (paired t test). (C) Spatial patterns of GSSG/2GSH ratios in young (first day of adulthood) Grx1-roGFP2 transgenic adults. Intensity-normalized ratio false-colored image of one Z-level of 3D-stitched worms. The anterior and posterior spermathecae (sp) show low GSSG/2GSH ratios. Calibration bar indicates the ratio of 525-nm emission after excitation at 405 nm vs. 488 nm. Reprinted by permission from Elsevier: Free Radical Biology & Medicine [120], copyright 2011.

Figure 3

Inflammatory response in zebrafish larvae. Tail fin amputation leads to H₂O₂ production and release of arachidonate metabolite in the injured tissue. These compounds are sensed by leukocytes and play an essential role in their recruitment to the wound. H₂O₂ signal is dampened by arriving neutrophils as H₂O₂ is consumed in a reaction catalyzed by myeloperoxidase (MPO) [57,68,141,271].

Figure 4

Schematic illustration of redox events during light-inducible reaction in photosynthetic leaves of A. thaliana. Plants sense light with the pigments of the photosynthetic electron transport chain (ETC) in the chloroplast or with photoreceptors. Chloroplasts generate stromules in response to the changes in the internal chloroplast redox status in a pathway regulated by the chloroplast NADPH-dependent thioredoxin reductase, NTRC. Stromules deliver H₂O₂ directly to the nucleus bypassing the cytosol. Chloroplast-sourced H₂O₂ in the nucleus may act as a signal to induce gene expression, which facilitates the acclimation of cells to high light intensities [163,187,188]. Abbreviations: 2CP—2-Cys peroxiredoxin; Myo—myosin; FTR—(Fd)-dependent Trx reductase; DBMIB—2,5-dibromo-6-isopropyl-3-methyl-1,4-benzoquinone; FD—ferredoxin; Trx—thioredoxins; DCMU—3-(3,4-dichlorophenyl)-1,1-dimethylurea; FNR—ferredoxin-NADP⁺ reductase; SHAM—salicylhydroxamic acid.

Figure 5

Compartment-specific features of selected redox processes in yeast cells. The picture summarizes the mechanisms of glutathione redox potential (E_GSH) maintenance and its dependence on the temperature, as well as the dynamics of H₂O₂ concentration shifts in response to changes in oxygenation degree in cytosol, mitochondria and peroxisomes [12,14,32,46,205,212]. Abbreviations: ETC—electron transport chain; GR—glutathione reductase; GSH—reduced glutathione; GSSG—oxidized glutathione; Mt DNA—mitochondrial DNA.

Figure 6

Inheritance of mitochondria in yeast cells. During division daughter cells on average inherit mitochondria with lower glutathione redox potential (E_GHS) than mother cells, these mitochondria are anchored in the bud tip by Mmr1p. Mitochondria are transported to the daughter cell via actin cytoskeleton in the direction opposite to retrograde actin cable flow (RACF), which serves as a “filter” that prevents inheritance of less fit mitochondria to the bud. However, the mother cell retains a pool of high-functioning mitochondria anchored by Mfb1p in the mother tip [25,26,27,202,470].

Figure 7

Many membrane transport systems are regulated in a redox-dependent manner. (A) In Salmonella, an oxidative shift in the E_GSH value leads to a rapid decline in membrane permeability due to redox-dependent closure and opening of OmpC and OmpA proteins, respectively [258]. (B) Blue background: transport of glutathione into the endoplasmic reticulum (ER) in S. cerevisae. Glutathione is transported into the ER via Sec61 membrane protein complex. Glutathione transport is regulated by a negative feedback loop: (1) reduced glutathione (GSH) enters the ER via concentration gradient; (2) in the ER GSH induces activation of Ero1 which produces H₂O₂; (3) Kar2 protein is oxidized by H₂O₂ and in oxidized form blocks GSH transport into the ER [43]. Green background: Regulation of calcium channels in S. cerevisae. Oxidation of the cellular glutathione pool leads to activation of Cch1p and Yvc1p calcium channels by glutathionylation. Deglutathionylaton is required for inactivation of these channels [209,211]. (C) So-AqpA of S. oligofermentans is involved in transport of H₂O₂ and expression of this gene is up-regulated in the presence of H₂O₂. So-AqpA promotes detoxification of endogenously produced H₂O₂ and enhances competitiveness of S. oligofermentans [263]. Abbreviations: E_GSH—glutathione redox potential; GSSG—oxidized glutathione.

Figure 8

The picture represents selected redox processes during bacteria and immune cells interactions. (A) When residing in macrophages, Salmonella Typhimurium is capable of avoiding redox stress due to injection of Salmonella pathogenicity island 2 (SPI2) effectors via type 3 secretion system (T3SS) into the host cytosol. These proteins disrupt the correct localization of reactive oxygen/nitrogen species (ROS/RNS) generating machinery [259]. (B) Any implementation of roGFP-based probes in neutrophilic phagosome faces significant difficulties since high HOCl concentration in this compartment leads to a nonspecific oxidation of the sensors. Interestingly, it seems that myeloperoxidase (MPO)-generated oxidants do not alter cytosolic glutathione redox potential (E_GSH) as revealed by roGFP2. When experiments with neutrophil activation are planned, it is important to take into account that phorbol 12-myristate 13-acetate (PMA) treatment and bacteria exposure stimulate this process via different signaling pathways [51,239]. Abbreviations: DPI—diphenyleneiodonium; GSH—reduced glutathione; GSSG—oxidized glutathione; iNOS—inducible NO synthase; MPO—myeloperoxidase; NOX—NADPH oxidase; PKC—protein kinase C; PI3K—phosphoinositide 3-kinase; SOD—superoxide dismutase.

Figure 9

Several anti-trypanosomal compounds were tested for their ability to affect redox homeostasis in Trypanosoma brucei brucei. Interestingly, in many cases, small variations in the molecular structure of a drug (highlighted in color) lead to a dramatic change in the mechanism of its functioning. (A) Anti-trypanosomal tri-thiazoles linked by amide [227,231]. (B) Anti-trypanosomal organometallic compounds [229,230]. (C) Anti-trypanosomal sugar diselenides [228]. Abbreviations: dppf—1,1′-bis(diphenylphosphino) ferrocene; E_T(SH)2—trypanothione redox potential; GSH—reduced glutathione; MT DNA—mitochondrial DNA; ROS—reactive oxygen species.

Figure 10

Internalization of Mycobacterium tuberculosis (Mtb) cells into macrophages leads to the emergence of subpopulations with different mycothiol redox potential (E_MSH). Subsequent phagosome maturation is followed by an increase of the “oxidized” fraction size. Phagosomal acidification and antibiotic treatment also affect the average E_MSH value. It is important to note, that the latter correlates with antibiotic resistance which is partially attributed to the differences in drug-efflux pumps expression [49,52]. Abbreviations: CQ—chloroquine; EMB—ethambutol; IFN—interferon; INH—isoniazid; iNOS—inducible NO synthase; LPS—lipopolysaccharides; NOX—NADPH oxidase; RIF—rifampicin; RNS—reactive nitrogen species; ROS—reactive oxygen species.

Figure 11

Redox processes that lead to the emergence of antibiotic resisters in mycobacteria. (A) Prolonged exposure of mycobacterial cells to antibiotic is characterized by three phases (“killing”, “persistence” and “regrowth”). The second phase is accompanied by elevated reactive oxygen species (ROS) production which stimulates genome-wide mutagenesis and consequent emergence of resisters. However, it should be mentioned that since DNA modification by ROS is random, the surviving cells experience a fitness cost due to undesired mutations [249,253]. (B) In mycobacterial populations, two fractions of cells can be found in regard to the cellular length, which result from asymmetric divisions. Short cells are characterized by elevated ROS production, which on the one hand lowers their resistance to environmental stresses, but on the other allows them to act as ROS-producing “factories” that stimulate the emergence of antibiotic resisters among normal cells [247,248]. Abbreviations: CFU—colony forming unit; E_MSH—mycothiol redox potential; INH—isoniazid; NOX— NADPH oxidase; RIF—rifampicin; SOD—superoxide dismutase.