$$\rightleftharpoonup{xx}$$
$$\longleftharp{xx}$$,
$$\longrightharp{xx}$$,
Reactive oxygen species (ROS) signaling regulates development and functioning of the nervous system1. An important cellular ROS source are NADPH oxidases (NOX), which are transmembrane proteins generating superoxide and hydrogen peroxide (H2O2)2. NOX enzymes are found throughout the central nervous system (CNS), and NOX-derived ROS contribute to neuronal development3,4,5,6. Maintenance and differentiation of neural stem cells, establishing neuronal polarity, neurite outgrowth, and synaptic plasticity have been shown to require adequate levels of ROS7,8,9,10,11. On the other hand, uncontrolled production of ROS by NOXes contribute to neurodegenerative disorders including Alzheimer's Disease, multiple sclerosis, and traumatic brain injury12,13,14. Hence, production of physiologically relevant ROS is critical to maintaining healthy conditions.
Development of genetically encoded biosensors facilitated the detection of cellular ROS greatly. One important advantage of genetically encoded biosensors is the increased temporal and spatial resolution of the ROS signal, as these sensors can be specifically targeted to distinct locations. Redox-sensitive GFP (roGFP) is one type of such ROS biosensors. The roGFP2-Orp1 variant specifically detects H2O2 through its Orp1 domain, which is a glutathione peroxiredoxin family protein from yeast15,16. The oxidation of the Orp1 protein is transferred to roGFP2 to alter its conformation (Figure 1A). The probe exhibits two excitation peaks near 405 nm and 480 nm, and a single emission peak at 515 nm. Upon oxidation, the fluorescence intensity around excitation peaks changes: while 405 nm excitation increases, 480 nm excitation decreases. Thus, roGFP2-Orp1 is a ratiometric biosensor, and H2O2-levels are detected by the ratio of fluorescence intensities excited at two different wavelengths (Figure 1B). Overall, roGFP2-Orp1 is a versatile tool for ROS imaging that can be utilized efficiently in vivo.

Figure 1: Schematic representation and excitation spectra of roGFP2-Orp1. (A) Oxidant transfer occurs between Orp1 and roGFP2 in response to H2O2, leading to conformational changes in roGFP2. (B) The excitation spectra of the roGFP2-Orp1 exhibits two excitation peaks at 405 nm and 480 nm and single emission peak at 515 nm. Upon oxidation by H2O2, the 405 nm excitation increases while 480 nm excitation decreases. This results in a ratiometric readout for H2O2 presence. The figure has been modified from Bilan and Belousov (2017)16 and Morgan et al. (2011)25. Please click here to view a larger version of this figure.
The Danio rerio (zebrafish) model system has several advantages for applying genetically encoded biosensors. The optical transparency of the embryos and larvae enables non-invasive in vivo imaging. New imaging tools are being developed to achieve higher resolution and deeper penetration17. Furthermore, there are established tools for genetic manipulation (ectopic mRNA expression, Tol2 transgenesis, etc.) and genome editing (TALENs, CRISPR/Cas9, etc.), which promotes the generation of transgenic animals18. As the zebrafish embryos develop outside of the mother, this system further allows easier access and manipulation of the embryos. For instance, one-cell stage injections and drug treatments can easily be done.
Here, we used zebrafish to transiently express the H2O2-specific biosensor roGFP2-Orp1 by injecting in vitro-transcribed mRNA. These embryos can be used for both in vitro imaging of cultured neurons and in vivo imaging (Figure 2). We describe a protocol for dissecting and plating retinal ganglion cells (RGCs) from zebrafish embryos followed by assessing H2O2-levels in cultured neurons. Then, we present a method for in vivo imaging of roGFP2-Orp1-expressing embryos and larvae using confocal microscopy. This approach not only allows to determine physiological H2O2-levels but also potential changes occurring in different developmental stages or conditions. Overall, this system provides a reliable method for detecting H2O2 in live cells and animals to study the role of H2O2 in development, health, and disease.

Figure 2. Outline of the experimental approach. Briefly, after embryo collection, roGFP2-Orp1 mRNA is injected into the yolk of one-cell stage zebrafish embryos. Developing embryos can be used for both (A) in vitro and (B) in vivo imaging. (A) GFP-positive embryos are used to dissect retinas for RGC collection at 34 hpf. Dissociated RGCs are plated on PDL/laminin-coated coverslips in ZFCM (+) media. Growth cone imaging can be conducted as RGCs extend their axons after 6-24 h of plating. Cells can be subjected to different treatments to measure the potential changes in H2O2-levels. Here, we measured H2O2-levels in the growth cones of RGCs (red). (B) GFP positive embryos are used for in vivo imaging. At the desired age, embryos can be anesthetized and mounted on 35 mm glass bottom dishes for confocal imaging. Here, embryos are mounted ventrally for retinal imaging. Schematic shows retinal development in zebrafish. RGCs form ganglion cell layer (GCL), which is the innermost layer in retina. RGC axons develop into optic nerve to cross midline, forming optic chiasm. Then, RGC axons grow dorsally to make synapses in the optic tectum in the midbrain. Please click here to view a larger version of this figure.