1. Cell Biology
  2. Neuroscience

Large-scale cell-type-specific imaging of protein synthesis in a vertebrate brain

  1. Or David Shahar
  2. Erin Margaret Schuman  Is a corresponding author
  1. Max Planck Institute for Brain Research, Germany
https://doi.org/10.7554/eLife.50564
Share this article
Cite this article
  1. Or David Shahar
  2. Erin Margaret Schuman
(2020)
Large-scale cell-type-specific imaging of protein synthesis in a vertebrate brain
eLife 9:e50564.
https://doi.org/10.7554/eLife.50564
  2. Download BibTeX
  3. Download .RIS

Abstract

Despite advances in methods to detect protein synthesis, it has not been possible to measure endogenous protein synthesis levels in vivo in an entire vertebrate brain. We developed a transgenic zebrafish line that allows for cell-type-specific labeling and imaging of nascent proteins in the entire animal. By replacing leucine with glycine in the zebrafish MetRS-binding pocket (MetRS-L270G), we enabled the cell-type-specific incorporation of the azide-bearing non-canonical-amino-acid azidonorleucine (ANL) during protein synthesis. Newly synthesized proteins were then labeled via 'click chemistry'. Using a Gal4-UAS-ELAV3 line to express MetRS-L270G in neurons, we measured protein synthesis intensities across the entire nervous system. We visualized endogenous protein synthesis and demonstrated that seizure-induced neural activity results in enhanced translation levels in neurons. This method allows for robust analysis of endogenous protein synthesis in a cell-type-specific manner, in vivo at single-cell resolution.

Introduction

Protein synthesis is critical for remodeling synaptic proteomes, especially when this process is associated with information storage (Sutton and Schuman, 2006). Chemical stimuli and changes in behavioral states alter protein expression in the nervous system. It has been shown in different model organisms that protein synthesis, during or shortly after learning, is essential for the formation of long-term memory (Davis and Squire, 1984; Agranoff et al., 1966; Agranoff and Klinger, 1964; Costa-Mattioli et al., 2009). Despite the importance of neuronal protein synthesis for many biological processes such as learning (Flexner et al., 1962; Hinz et al., 2013; Roberts et al., 2013), stress responses (Langebeck‐Jensen et al., 2019), and epilepsy (Brooks-Kayal et al., 1998; Hinz et al., 2012; Baraban et al., 2005; Del Bel et al., 1998), little is known about the endogenous neuronal protein synthesis levels and their changes in vivo.

Zebrafish are vertebrates that exhibit a variety of complex behaviors (Orger and de Polavieja, 2017) including swimming and rheotaxis (e.g. Olszewski et al., 2012; Oteiza et al., 2017; Marques et al., 2018), hunting (e.g. Bianco et al., 2011; Semmelhack et al., 2014), learning (eg. Hinz et al., 2013; Roberts et al., 2013; Aizenberg and Schuman, 2011; Kenney et al., 2017; Ahrens et al., 2012; Valente et al., 2012) and social behaviors (e.g. Hinz et al., 2013; Peichel, 2004; Oliveira, 2013; Gerlai, 2014; Teles et al., 2016; Stednitz et al., 2018; Dreosti et al., 2015). Moreover, a variety of neurological syndromes including epilepsy have been investigated (Kundap et al., 2017). The zebrafish larval brain is small and translucent, enabling high-resolution imaging of cells. The complexity of brain tissue, however, is still an issue. Neurons, in particular, have long processes, which are tightly entangled in their respective tissues. As such, monitoring protein synthesis levels with cell-type and temporal resolution has so far been impossible in zebrafish neurons, highlighting the need for the methodology developed here (Figure 1A).

Figure 1 with 3 supplements see all
Download asset Open asset
Cell-type-specific labeling of newly synthesized proteins.

(A) Schematic demonstrating the method to visualize newly synthesized proteins in the brain in a cell-type-specific manner. (B) The protein sequence of the catalytic core domain of the MetRS in a number of species including Danio Rerio shows strong conservation (green). Leucine 270 (bold green) was mutated to Glycine to develop cell-type-specific metabolic labeling in zebrafish. (C) Schematic of the binding pocket of the MetRS and the ribosome during translation. The wt MetRS allows the charging of Met (black) that can be incorporated during translation initiation and elongation (left). The non-canonical amino acid ANL (blue), which contains an azide group, does not fit into the binding pocket of the wt MetRS, and thus is notincorporated into nascent protein (center). The mutant MetRSL270G can charge ANL, which is then incorporated into newly synthesized proteins (in cells expressing the MetRSL270G). (D) Schematic of the UAS-CFP-MetRSL270G line transgene. Crossing the line with any Gal4-expressing line allows for the metabolic labeling of newly synthesized proteins in any accessible cell type. (E) A scheme demonstrating the use of the ELAVL3-Gal4:UAS-CFP-MetRSL270G line. Left: a zebrafish larva expressing the transgene in neurons (cyan). Following addition of ANL to the water bath, newly synthesized proteins in neurons incorporate ANL (blue). Right: a whole mount click reaction with a fluorescent alkyne reveals the newly synthesized proteins (red). (F) The effect of different ANL concentrations on swim speed after 24 hr of ANL exposure (measurement was done in the presence of ANL). 10 mM ANL, which had no significant effect on larvae swimming, was used in further experiments. N = 5 to 6 larvae for each concentration. (G) Projections of confocal images of zebrafish larval brains after click reactions demonstrating the specificity of fluorescently labeled nascent protein in the MetRSL270G larva treated with ANL, but not in controls. Scale bar = 50 μm.

Bio-orthogonal approaches based on metabolic precursors enable the labeling of nascent proteins or protein modifications (Hinz et al., 2012; Beatty et al., 2006; Laughlin et al., 2008), and have been combined with immunostaining to measure protein synthesis in specific cell types (Liu and Cline, 2016). These platforms have recently been coupled with genetic control, allowing access to particular cell types. For example, the wild-type Methionyl-tRNA synthetase (MetRS) can be modified to enable the charging of a different azide-bearing non-canonical amino acid, the methionine analog azidonorleucine (ANL), which cannot be charged by the wild-type MetRS. By using cell-type-specific promoters, the mutant MetRS can be expressed in cell types of interest and the nascent proteins can be labeled via the administration of ANL, as has been demonstrated in Caenorhabditis elegans (Yuet et al., 2015), Drosophila melanogaster (Erdmann et al., 2015) and Mus musculus (Alvarez-Castelao et al., 2017).

To date, overall levels of protein synthesis within neurons across the entire intact brain have not yet been measured and imaged in any vertebrate. The existing protein synthesis reporters used for most single cell analyses rely on fluorescent tagging of individual protein species and therefore do not measure endogenous nascent protein levels. Here we demonstrate for the first time the ability to label and image in situ newly synthesized proteins in vivo in a cell-type-specific manner. We visualized nascent neuronal proteins across the entire animal. Combining a specific Gal4 reporter line with the UAS-MetRSL270G and adding ANL for various durations enabled cell-type-specific labeling with temporal control. We also demonstrate the sensitivity of the protein synthesis signal to alterations in neuronal activity.

Results

Cell-type-specific nascent protein tagging

To enable cell-type-specific non-canonical amino acid tagging in zebrafish, we cloned and mutated the zebrafish MetRS to introduce a point mutation (L270G) at a conserved position in the methionine binding pocket (Figure 1B–C). We developed a transgenic line in which the conditional expression of CFP and the MetRSL270G (separated by T2A) is under the control of the UAS enhancer. When these fish are crossed to a Gal4 line with a cell-type-specific promoter, cell-type-specific incorporation of azidonorleucine (ANL) into nascent proteins can be achieved (Figure 1C–D). In order to realize cell-type-specific labeling of newly synthesized proteins in neurons, we crossed the UAS-CFP- MetRSL270G line with a well-established pan neuronal driver Gal4 line, ELAVL3-Gal4 (Figure 1D–E). Following the addition of ANL, the charging of ANL by MetRSL270G and its incorporation into protein, a click reaction was performed and newly synthesized proteins were visualized (Figure 1E). In a previous study, we showed that zebrafish nascent proteins can be globally labeled with the non-canonical amino acid azidohomoalanine (AHA) via its addition to the swim water (Hinz et al., 2012). We added ANL to the swim water and determined whether there were any apparent behavioral effects. Freely swimming larvae were incubated with different concentrations of ANL for 24 hr (hrs) and the average swim speed was measured. We found that ANL concentrations of 10 mM or less did not affect larval swimming behavior (Figure 1F). Zebrafish larvae exhibit a preference for illuminated regions of their habitat (Hinz et al., 2013). We also observed that the ELAVL3-MetRSL270G larvae maintained their light preference in the presence of ANL (Figure 1—figure supplement 1). We next incubated larvae with 10 mM ANL for 24 hr, fixed the larvae and performed a whole-mount click reaction (see Materials and methods). Confocal images acquired across the brain revealed newly synthesized proteins in neurons in ELAVL3-MetRSL270G larvae that were incubated with ANL. Only weak fluorescence was detected in WT larvae that were incubated with ANL, or ELAVL3-MetRSL270G larvae that were not incubated with ANL (Figure 1G, for images of single z-sections see Figure 1—video 1 and Figure 1—video 2).

Detection of endogenous nascent proteins in different neuronal populations

We performed bio-orthogonal non-canonical amino-acid tagging (BONCAT) (Dieterich et al., 2007; Dieterich et al., 2006) on proteins extracted from WT or MetRSL270G larvae (4 dpf) after metabolic labeling (24 hr, 10 mM ANL) or MetRSL270G fish without metabolic labeling. Western blot analysis revealed an abundance of biotinylated nascent proteins, spanning various molecular weights, in head tissue from ANL-treated MetRSL270G larvae (Figure 2A, Figure 2—figure supplement 1) and only low background levels of biotinylated proteins in the controls (WT ANL+, MetRSL270G ANL- in Figure 2A). To examine the robustness and specificity of the labeling, we visualized newly synthesized proteins in different brain regions. We incubated 3 dpf larvae with 10 mM ANL for 24 hr, fixed the larvae (at 4 dpf) and performed a whole-mount click reaction using a fluorescent alkyne tag. We performed immunostaining for CFP to visualize neurons expressing CFP (and therefore MetRSL270G) in the same larvae. We then imaged the entire nervous system visualizing both nascent protein and CFP (Figure 2, Figure 2—figure supplements 2–5). We detected nascent protein signal across the nervous system including in the subpallium, habenula, anterior pretectum, optic tectum, hindbrain, medulla and the spinal cord (Figure 2, Figure 2—figure supplement 2). In order to quantify the nascent protein signal, we segmented the cell somata volumes using the 3D CFP signal (Figure 2—figure supplements 6–7 and Figure 2—video 1). We then measured the average voxel intensity in the fluorescent-alkyne channel to determine the level of nascent protein in each cell. We segmented hundreds to over two thousand cells in each larva. We calculated the average signal intensity in a specific brain region, the habenula (Figure 2G), and in the entire nervous system (Figure 2H). The same click reaction performed on larvae that were not incubated with ANL revealed only low levels of background fluorescence (Figure 2C–D,G–H). In some neurons, we could detect a nascent protein signal in neurites indicating the sensitivity of the method to visualize newly synthesized proteins in dendrites or axons (Figure 2—figure supplement 2C–D, Figure 2—figure supplement 8). These newly synthesized proteins could have either been synthesized in somata and moved to the processes, or could have been synthesized locally in the processes.

Figure 2 with 9 supplements see all
Download asset Open asset
High resolution imaging newly synthesized proteins in different neuronal populations.

3 dpf larvae were incubated with 10 mM ANL for 24 hr in vivo, fixed (at 4 dpf) and clicked to a biotin tag for BONCAT or a fluorescent tag to visualize nascent proteins in situ. (A) Immunoblot detecting newly synthesized proteins in WT larvae treated with ANL (WT ANL+, control), MetRSL270G (MerRS*) larvae treated with ANL (ANL +) or not (ANL-, control). (B) Dorsal view collage projection of confocal images showing fluorescently labeled newly synthesized neuronal proteins (red). (C-F) High magnification view of different brain regions. 4–6 confocal planes are shown (~10 microns in depth). Note the overlap between the CFP channel (Ab staining, green) and the nascent protein channel (click labeling, red), indicating that the signal is specific to cells expressing the MetRSL270G. (C-D) Optic tectum (Ot). (C) Shown are 4 planes of the region indicated by the square in B. (D) The same region in a larva not incubated with ANL demonstrating the CFP but not nascent protein labeling. (E) Maximal projection of labelled newly synthesized proteins in an entire brain (dorsal view and lateral view) (see Figure 2—figure supplement 2A–B for lower brightness). White frames indicate the subpallium (Sp), habenula (Hb) and anterior pretectum (Apt), regions zoomed in (in F). (F) CFP Ab staining and nascent protein labeling in 4–6 confocal images indicated in the white frames in E. See Figure 2—figure supplement 2 for more brain regions. (G-H) Quantification of the average nascent protein levels in the habenula (G) and the entire nervous system (H). Neurons were segmented in 3D using the CFP channel (see supplementary material and Figure 2—figure supplements 6–7) and the average voxel fluorescence intensity for the CFP and fluorescently labeled nascent protein was measured in each cell. Plotted are the average fluorescence intensities in single cells. 30 to 60 neurons were segmented for each habenula of 4 ANL-treated and 3 control larvae (G). (H) Quantification of the average CFP and nascent protein fluorescence intensity in neurons across the entire nervous system. More than 1000 neurons were segmented in 3D using the CFP channel (similar to G). Plotted are the mean fluorescence intensities in single cells from 3 larvae treated (ANL+, squares) or not treated (ANL-, circles) with ANL. See Figure 2—figure supplement 5 for statistical differences between the groups. Squares: ANL- (control), circles: ANL+, blue: CFP, red: Nascent protein, black line – mean fluorescence intensity within a single larva. One cell had a nascent protein intensity value below ten and is shown on the x-axis. Scale bars = 20 μm.

Figure 2—source data 1

Source data for Figure 2G–H.

https://cdn.elifesciences.org/articles/50564/elife-50564-fig2-data1-v1.xlsx

Detecting nascent proteins with different periods of labeling

We found that 24 hr of ANL exposure was sufficient to allow the detection of nascent protein signal (Figure 2). In order to determine if an even shorter incubation period would result in labeling, we incubated 3 dpf zebrafish larvae for 12 hr with the same ANL concentration (10 mM). Quantification of the average nascent protein intensity in neurons revealed significant labeling compared to the control, but these levels were significantly lower than levels observed following 24 hr of ANL incubation (Figure 3). This relatively short labeling window could thus be used for comparing protein synthesis intensities between experimental and control fish in various behavioral paradigms.

Figure 3
Download asset Open asset
Newly synthesized proteins following different durations of metabolic labeling.

Larvae were incubated with 10 mM ANL in their water bath for the indicated durations before fixation and click reaction. (A) Neurons in the habenula of MetRSL270G larvae were segmented in 3D using the CFP channel and the fnascent proteinintensity was measured in each neuron. Plotted are the average intensities of 3 larvae for each treatment (N = 3 for each treatment). Error bars indicate the SEM (N = 3). *p<0.05 (0.005 for 12 hr and 0.009 for 24 hr). (B) Representative images. Shown are maximal projections of 10 planes (~4 μm) of confocal images in the nascent proteins channel of the forebrain focusing on the olfactory bulb. Calibration bar – top left, scale bar = 10 μm.

Figure 3—source data 1

Source data for Figure 3A.

https://cdn.elifesciences.org/articles/50564/elife-50564-fig3-data1-v1.xlsx

Detection of altered neuronal protein synthesis levels following seizures

We next addressed whether the nascent protein signal was sensitive to global alterations in neural activity. The GABAergic receptor antagonist Pentylenetetrazole (PTZ) induces epileptic-like neuronal discharges and seizure-like behaviors in rodents and zebrafish (Baraban et al., 2005; Baraban et al., 2007; Naumann et al., 2010). PTZ has been shown to induce expression of immediate early genes in larval zebrafish (Baraban et al., 2005). To determine the effect of elevated neural activity and behavioral seizures on protein synthesis levels in neurons in vivo, we incubated 3 dpf MetRSL270G larvae with ANL for 12 hr and induced seizures by adding PTZ for the last 2.5 hr (Figure 4A). We then measured neuronal protein synthesis levels within single cells in two different brain regions, the spinal cord and the habenula (Figure 4B–E). We segmented tens to hundreds of neurons in the spinal cord or right habenula and calculated the average protein synthesis signal in each cell (Figure 4D). Following PTZ exposure, we detected a significant (~60%) increase in protein synthesis levels in the habenula and a modest, though not significant, increase in the spinal cord (Figure 4D–E) when calculating the average intensity between several larvae.

Figure 4 with 1 supplement see all
Download asset Open asset
Seizure-induced neuron-specific protein synthesis.

(A) Schematic of the experiment. Freely swimming larvae were incubated with ANL. After 10 hr incubation, PTZ was added for 2.5 hr, inducing seizures. Following fixation, whole-mount click with a fluorescent alkyne and confocal imaging were performed. (B-C) Representative images of the ANL signal in the spinal cord (B) or habenula (C). Shown are maximum projections of 4 confocal planes of the specific regions of larvae treated or non-treated with the protein synthesis inhibitor puromycin (PSI) and treated or non-treated with PTZ. (D) Quantification of the images shown in (B-C). Cells were segmented in 3D using the CFP channel, and the mean nascent protein labeling was measured in each cell using the fluorescent-alkyne channel. The dots represent the mean intensity in cells in the corresponding image (of B-C respectively). Red line – mean, whiskers - STDEV. (E) Bar plot showing the average nascent protein intensity in 3 to 5 larvae PTZ- and PTZ+, respectively. CFP positive cells in the spinal cord or the habenula were segmented in 3D using the CFP antibody staining. The levels of newly synthesized proteins were measured using the fluorescent alkyne (similar to D). More than 100 neurons in the spinal cord and 30 neurons in the habenula were analyzed. The average intensity was calculated for each larva. Plotted are the averaged intensities for each treatment (N = 3 to 5 larvae for each treatment). Error bars indicate SEM. *p<0.05, **p<0.01***p<0.001. Scale bars = 5 μm (B, C).

Figure 4—source data 1

Source data for Figure 4E.

https://cdn.elifesciences.org/articles/50564/elife-50564-fig4-data1-v1.xlsx

Discussion

We have developed a system that enables the labeling of nascent proteins in living zebrafish larvae in a cell-type-specific manner. We describe a UAS line that allows one to tag newly synthesized endogenous proteins in a cell type-of-interest simply by crossing it with any specific Gal4 driver line.

Given the importance of protein synthesis for many biological processes, labeling nascent proteins for imaging within the intact organism will be useful for future studies. Non-canonical amino acids have been used to elucidate different biological processes including protein turnover (Cohen et al., 2013), protein dynamics (Zhang et al., 2010; Schanzenbächer et al., 2016) and local protein synthesis (Dieterich et al., 2006; Tcherkezian et al., 2010; Yoon et al., 2012). We have previously used AHA to label nascent proteins in zebrafish larvae (Hinz et al., 2012) without cell type specificity. To achieve cell-type-specific labeling in mice and other species, we have mutated the MetRS in the evolutionarily well-conserved methionine-binding pocket, resulting in cell type specificity of nascent protein labeling in vivo (Erdmann et al., 2015; Mahdavi et al., 2016; Alvarez-Castelao et al., 2017).

We detected CFP expression levels above background in all cells with ANL labeling (Figure 2). Generally, we observed a positive correlation between the intensity levels of CFP (a proxy for the MetRSL270G mutant expression) and nascent protein, but there were a few cells detected with a very low expression of CFP and high nascent protein intensity or vice versa (Figure 2). The general positive correlation between CFP and nascent protein is expected because CFP is synthesized within the cells of interest and its level of protein synthesis will likely reflect the general translational activity of the cell. Cases in which the expression of CFP is not proportional to the nascent protein labeling may be explained by the following: the CFP labeling intensity is a result of both expression and degradation. CFP may thus bedegraded to a greater or lesser extent in some neurons. Zebrafish larvae exhibit high levels of neural activity during swimming and other natural behaviors (Chen et al., 2018). Many neuronal activity-dependent regulatory processes use both protein synthesis and proteasome-dependent protein degradation (Sutton and Schuman, 2006; Hinz et al., 2013; Langebeck‐Jensen et al., 2019; Djakovic et al., 2009; Bingol and Schuman, 2006) to remodel the proteome. Recently, evidence for the coordination of protein synthesis and proteasome dependent degradation has been observed, including the degradation of nascent proteins by a neuron-specific 20S membrane proteasome complex (for example Ramachandran et al., 2018). It is possible that dynamic translation and degradation processes result in diverging levels of CFP and other proteins.

The fact that we detect neurons in which there is strong intensity of nascent proteins with only a faint CFP signal may suggest that even a low amount of the MetRSL270G mutant is sufficient for ANL incorporation into nascent proteins. We observed variation in the nascent protein intensities between larvae and between cells within the same larva and even within the same region and between neighboring cells (Figure 2). This is consistent with previous findings visualizing fluorescently labeled newly synthesized proteins using the non-canonical amino acid AHA (Liu and Cline, 2016).

ANL incorporation into nascent proteins was detectable in situ following 12 hr of exposure in the swim water, circumventing the need for ANL administration through food or drink (Erdmann et al., 2015; Alvarez-Castelao et al., 2017), and allowing better control over the concentration and duration of ANL exposure. This relatively short time window opens opportunities to measure protein synthesis levels in many biological paradigms. We have tried shorter incubation durations but the signal was sparse, the signal-to noise was too low and the variability between cells and between different larvae was high, indicating that shorter incubation time is not sufficient to measure nascent protein levels under these conditions. As far as we know, the 12 hr duration demonstrated here is the shortest available time frame so far for cell-type-specific metabolic labeling.

We demonstrate that this method can be used to address biological questions by labeling nascent proteins in neurons during an induced seizure-like behavior. PTZ-induced seizure-like behavior in zebrafish larvae results in higher expression levels of immediately early genes (Hinz et al., 2012; Baraban et al., 2005). We have previously shown a general increase in protein synthesis, using the non-canonical amino acid AHA which can incorporate into nascent proteins in all cell types (Hinz et al., 2012), following exposure of zebrafish larvae to PTZ. However, it was not possible in that study to determine whether the increased protein synthesis signal arose from neurons or other cell types. Here, we used PTZ to induce seizures and revealed an increase in neuronal protein synthesis. Furthermore, the specific labeling of newly synthesized proteins in neurons reduces background signals and noise from other cell types, thereby allowing us to compare different brain regions and neuron groups. For example, the habenula is a highly conserved structure that has been implicated in decision-making (Hikosaka, 2010). We found a significant increase in protein synthesis in the habenula following PTZ-induced seizures. Our findings are consistent with an increase in immediately early gene expression following PTZ-induced seizures in the habenula as has been reported in rats (Del Bel et al., 1998).

The method described here is robust: ANL, which is incorporated into nascent proteins, and the fluorescent alkyne used for imaging form a strong covalent bond allowing for stringent washes and hence low background signal. While the click chemistry used here precludes live imaging, the ANL incorporation takes place in vivo while the larvae are freely swimming and behaving. Therefore, the duration of the labeling can be chosen according to the studied cell-type, the protein synthesis levels (and degradation), and the biological question. The ability to label newly synthesized proteins in vivo for a pre-chosen duration and to then ‘freeze’ the result before imaging has advantages. Because of the strong stability of the fluorescently labeled newly synthesized proteins following the click reaction one can image the entire larvae, as demonstrated here. This is especially useful for labeling neural networks during dynamic physiological processes such as cumulative calcium influx activity (Fosque et al., 2015) or protein synthesis following neural activity (demonstrated here). Beyond the contribution to neuroscience, this platform can be adopted by crossing the UAS-MetRSL270G zebrafish line described here to any existing Gal4 driver line of interest.

Zebrafish larvae have the advantage of being both translucent and smaller than mammals, therefore allowing one to label and image nascent proteins in an entire intact brain or other tissues without the need to slice the tissue. Future experiments using cell-type-specific newly synthesized proteins in zebrafish could include investigating the loci of learning and memory formation. Protein synthesis is essential for the formation of many types of long-term memory (Davis and Squire, 1984; Agranoff et al., 1966; Flexner et al., 1962; Hinz et al., 2013; Roberts et al., 2013). The method described here, combined with zebrafish larva learning paradigms (reviewed in Roberts et al., 2013), will enable future studies to reveal the neural networks in which protein synthesis occurs during learning and memory formation. Therefore, methods combining zebrafish larva with cell-type-specific protein synthesis labeling will be widely applicable.

Materials and methods

Key resources table
Reagent type
(species) or
resource		DesignationSource or
reference		IdentifiersAdditional
information
AntibodyRabbit polyclonal anti GFPInvitrogenCat# A11122IF(1:600)
AntibodyChicken polyclonal anti GFPAvesCat# GFP 1010WB(1:1000)
AntibodyDonkey polyclonal anti-chicken IR680LicorCat# 926–68075WB(1:10000)
AntibodyGoat anti-rabbit IR800LicorCat# 925–32211WB(1:10000)
Zebrafish lineHuC-Gal4Stevenson, T. J. et alSchuman lab,
See Materials and methods section
Zebrafish lineUAS-MetRSL270Tefor-Amagen/
This paper		Schuman lab,
See Materials and methods section
Zebrafish lineELAVL3-MetRSL270GTefor-Amagen/
This paper		Schuman lab,
See Materials and methods section

Zebrafish husbandry

Request a detailed protocol

Adult fish strains AB, UAS-MetRSL270, ELAVL3-MetRSL270G and HuC-Gal4 were kept at 28°C on a 14 hr light/10 hr dark cycle, in a Techinplast Zebtec system. Embryos were obtained from natural spawning using a Techinplast breeding system, and were maintained in E3 embryo medium (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, 0.33 mM MgSO4) at 28°C on a 14 hr light/10 hr dark cycle.

Constructs and transgenic zebrafish

Request a detailed protocol

The following plasmid was injected to wt (AB strain) eggs at one cell stage: pBT2_4xnr UAS-CFP (Tol2-4xnr UAS:cerulean-2A-MetRS-6x His –Tol2). MetRS (Methionyl-tRNA synthetase (AAH57463.1) Danio rerio with mutation to L270G. The 4x non-repetitive (nr) UAS sequence was designed by the Halpern lab (Akitake et al., 2011). The following primers were used for genotyping: forward- gcaagggcgaggagctg, reverse: gctcaggtagtggttgtcg. The PCR product size was 602 bp. Huc:Gal4 was generated by the Piotrowski lab (Stevenson et al., 2012).

ANL administration

Request a detailed protocol

Azidonorleucine (ANL) was synthesized as previously described Mahdavi et al. (2016). ANL was kept as powder and was freshly dissolved in E3 solution prior to experiments at the indicated concentrations (0, 5, 10 and 20 mM). For the dosage determination experiment, Konstanz wt or ELAVL3-MetRSL270G freely swimming zebrafish larvae were supplemented with ANL or mock E3 exchange for 24 hr. During the last 20 min, larvae were moved to a chamber with swimming lanes that allowed single larvae to freely swim. The positions of the larvae were recorded with a camera at 1 Hz, and tracked automatically with a Matlab script to measure swimming distance and speed. In all the fluorescent labeling experiments, the ANL concentration was 10 mM. 0.1 mM 1-phenyl 2-thiourea (PTU) was added to the E3 water at 1 dpf for all larvae in Figures 2–4.

Click chemistry for fluorescent tagging

Request a detailed protocol

3–4 dpf old larvae were anaesthetized on ice for 45 min. Larvae were transferred to 1.5 ml tubes and washed once with clean E3. E3 was removed and replaced with 1 ml fixation solution (4% PFA, 4% Sucrose in PBS (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4, 1.4 mM KH2PO4)), and incubated overnight at 4°C with gentle shaking. Next, larvae were dehydrated in Methanol at −20°C overnight. Samples were gradually rehydrated through successive 5 min washes with 75% methanol in PBST (PBS+0.1% Tween-20), 50% methanol in PBST, 25% methanol in PBST and PBST. Samples were incubated 3 times for 5 min in PBDTT (PBST + 1% DMSO and 0.5% Triton X-100) and digested with 1 mg/ml collagenase (Sigma-Aldrich) in PBST for 45 min for permeabilization. Following one wash with PBST, larvae were post-fixed for 20 min in 4% PFA and 4% Sucrose in PBS. Samples were washed 3 times for 5 min in PBDTT and incubated for 3 hr at 4°C in blocking buffer (5% BSA, 10% goat serum in PBDTT). Samples were then washed 3 times for at least 10 min in PBST (pH 7.8). ANL labeled proteins were tagged using a click reaction. To 1 ml PBS (pH 7.8) TBTA (Sigma-Aldrich) 1:500 (stock 200 mM in DMSO, final 0.2 mM) was added followed by a 10 s vortex, TCEP (Thermo Scientific) 1:400 (stock 40 mM in H2O, final 0.5 mM), 10 s vortex, 1:500 AlexaFluor-647-alkyne (Invitrogen, stock 2 mM in DMSO, final 2 μM), 10 s vortex, and 1:500 CuSO4 (stock 200 mM in H2O, final 0.2 mM) followed by 30 s vortex. The above click reaction buffer was immediately added to tubes containing the larvae and kept overnight in the dark, at 4°C with gentle agitation. Samples were washed 4 times for 30 min in PBDTT with 0.5 mM EDTA and twice for one hour in PBDTT and kept in PBS (pH 7.4) until mounting. For immunofluorescence, samples were incubated in blocking buffer (10% serum in PBS) for 1 hr and incubated with 1:600 rabbit anti GFP antibody (Invitrogen A11122; to detect CFP) in blocking buffer overnight with gentle agitation at 4°C. Samples were washed in PBST twice and PBS 3 times (~10 min each wash), incubated with blocking buffer for ten minutes and incubated with an Alexa-488 fluorescent secondary antibody (goat anti rabbit, ThermoFisher A1008) overnight with gentle agitation at 4°C. Samples were washed twice with PBST and PBS (pH 7.4). After click or immunofluorescence, the samples were gradually moved to glycerol through successive 5 min washes (25% glycerol in PBS, 50% glycerol in PBS, 75% glycerol in PBS) and finally 100% glycerol and kept at 4°C in dark.

BONCAT

Request a detailed protocol

Larvae were incubated with 10 mM ANL for 24 hr. After incubation, the media was replaced with fresh E3. Larvae were sacrificed on ice-cold water (30 min). Heads were dissected using a scalpel and immediately snap frozen in tubes (1.5 ml, Eppendorf) that were pre-cooled using dry ice and stored at −80C until lysis. Tissue was homogenized and lysed using a pestle in lysis buffer (1% in Triton X100, 0.4% (w/v) SDS in PBS pH 7.8, 1:1000 EDTA free protease inhibitors (Calbiochem) and benzonase (Sigma, 1:1,000), and denatured at 75°C, 13,000 rpm for 10 min. Lysates were then cleared by centrifugation. BONCAT was performed as previously described (Dieterich et al., 2007). In brief, 60 μg proteins were dissolved in 120 μL PBS pH 7.8 supplemented with 0.01% SDS, 0.1% Triton, 300 μM Triazol (Sigma, 678937), 50 μM biotin-alkyne tag (Thermo, B10185) and 83 μg/mL CuBr (prepared by dilution of fresh 10 mg/mL solution in DMSO) at 4°C overnight in the dark. Biotinylated proteins were then separated by gel electrophoresis and immunoblotted with 1:1000 chicken anti-GFP (Aves), 1:1000 rabbit anti-biotin (Cell signaling) and Donkey anti-chicken IR680, goat anti-rabbit IR800 (IB, 1:10,000, Licor) antibodies.

Microscopy

Request a detailed protocol

Larvae were directly, or after dissection to remove the eyes and yolk sack (Figures 2–4), mounted on Mattek dishes with the dorsal side facing the glass bottom of the dish prior to imaging. Larvae were imaged using a Zeiss LSM880 confocal microscope and a 25X glycerol objective (NA 0.8, PSF: X 0.389, Y 0.336, Z 1.62). Spacing between z planes = 1.89 micron. 488 and 633 lasers were used for Alexa488 and Alexa647, respectively (filters: 504–563 and 652–702, respectively). To image the entire brain, tiling was done with 10% overlap using the ZEN program with a scaling of 0.332 micron per pixel and a 12-bit mode; tiles were stitched together using the ZEN program after imaging. In Figure 2, in order to allow for the clear visualization of 3-dimensional data of an entire larva in 2 dimensions, a collage was made to exclude non-relevant planes (e.g: in each region of the tail, only 3–5 planes out of ~150 contain neurons, therefore a maximal projection of the entire stack would impede a clear visualization of the neurons).

3D fluorescence intensity analysis

Request a detailed protocol

Stitched images were imported to Imaris 9.2. Cell segmentation was conducted using CFP channel intensities. Surface segmentation of cells in the entire image was based on the Marching cube algorithm (Lorensen and Cline, 1987) using the Imaris program. In some regions where we noticed that the segmentation did not perform well, a regional segmentation was performed again using the same algorithm with a manual threshold in smaller specific ROIs. False positive objects that were not real cells, such as salt crystals, were deleted based on morphology. Splitting of touching objects was allowed using the watershed algorithm. In some cases where splitting of touching objects failed due to a high density of neurons, a few cells were inevitably considered as one. The threshold of the minimal volume was set to 10 μm3, and of the minimal ‘sphericity’ was set to 0.75. An example of segmentation in a low density ROI can be found in Figure 2—figure supplement 6, and in Figure 2—video 1. Segmentation of a densely labeled ROI in the habenula is demonstrated in Figure 2—figure supplement 7. After segmentation in the CFP channel, the intensity values were measured in both the CFP and the ANL channels (for Figure 2) or for the ANL channel (Figures 3–4). For each cell, the average voxel intensity was calculated and documented. Next, based on the average cell intensities, the average intensity of cell in a region or an entire larva was calculated. Finally the mean intensity between 3 to 4 larvae was calculated for each region or entire larvae and plotted (n = 3 or 4). The error bars in all the graphs represent Standard Error of the Mean (SEM), unless otherwise noted. In Figure 4D, the bars represent SEM between cells. In all the other graphs, error bars represent SEM between the average intensities of several larvae. One-tailed, two-sample unequal variance (heteroscedastic) TTESTs were used for calculating all the p values (p).

Data availability

All data generated or analyzed during this study are included in the manuscript and supporting files. Image data is available on the Max Planck Digital Library (https://doi.org/10.17617/1.8L).

The following data sets were generated

References

Decision letter

  1. Vatsala Thirumalai
    Reviewing Editor; National Centre for Biological Sciences, India
  2. Didier YR Stainier
    Senior Editor; Max Planck Institute for Heart and Lung Research, Germany
  3. Hollis T Cline
    Reviewer; The Scripps Research Institute, United States

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

This article describes a tool for marking newly synthesized proteins in a cell-type specific manner in larval zebrafish. Such a tool would likely be of use to mark nascent proteins in a cell-type specific or tissue-specific manner in order to understand how novel protein synthesis is involved in development or disease for example.

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting your work entitled "Large scale cell-type-specific imaging of protein synthesis in a vertebrate brain" for consideration by eLife. Your article has been reviewed by three peer reviewers, and the evaluation has been overseen by a Reviewing Editor and a Senior Editor. The reviewers have opted to remain anonymous.

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we regret to inform you that your work will not be considered further for publication in eLife.

While all three reviewers agreed that this tool was useful for the zebrafish field and could be easily adopted by investigators, the fact that this method has already been established in other model organisms and is an extension of previous work diminished the overall novelty and significance of this tool. In addition, reviewers also had suggestions for improving demonstration of specificity of this tool (see full comments below).

Reviewer #1:

Shahar and Schuman apply a bio-orthogonal approach to study the dynamics and extent of protein synthesis in a cell type specific manner in zebrafish. By driving MetRSL270G expression under the ELAVL3 neuron-specific promoter, and adding the unnatural amino acid, azidonorleucine (ANL) to the swim water, the authors observe ANL incorporation indicative of new protein synthesis in neurons in the heterogenous environment of the brain and spinal cord. Using this approach. They then show that protein expression is increased upon seizure induction using the GAG receptor antagonist, PTZ. The authors suggest that the approach can, in principle, be extended to other tissues and cell types and questions, with the UAS line generated serving as a tool for the community.

Previously, the Schuman lab showed that it is possible to examine protein synthesis using a similar approach in whole larvae (Hinz et al., 2011). They now extend their previous findings to a tissue-specific manner (i.e larval brain and spinal cord neurons) and which (as they report in the Introduction) has been shown previously in other organisms. The Gal4 UAS system for cell type specific expression has been used in zebrafish for a long time. The ability to image sites of protein synthesis in the brain of intact animals post behaviour is likely to be of use to identify regions of the brain important for behaviour (- which is currently visualised in vivo by large through calcium imaging). Overall, the manuscript is straightforward and easy to understand. However, it does not represent a significant advance.

– The authors should examine ANL incorporation upon inhibition of protein synthesis using cycloheximide or other protein synthesis inhibitor. This will determine if the labelling is specific.

– The 12 hour incorporation window for the UNAA appears rather long in the context of a fast developing zebrafish larva. What is the half -life of ANL? And what is the shortest ANL exposure window within which incorporation can be detected. Though the 2 hour PTZ treatment shows clear differences over control larvae, a shorter ANL exposure might decrease noise from protein synthesis.

– It would be helpful to discuss the advantages/disadvantages of this approach as compared to other approaches to estimate newly synthesized proteins.

– Figure 4 graph D is specific for the animal shown in images B and C, but does not appear to be representative for all the animals analysed, as shown in graph E. It would be useful to do this experiment without ANL and with ANL in wild type controls to examine background autofluorescence or non-specific click reactions post PTZ incubation.

Reviewer #2:

Or and Schuman describe a method for cell type specific evaluation of new protein synthesis in intact Zebrafish using transgene expression of a mutant MetRS, that is modified to only charge the non-canonical amino acid, azidonorleucine (ANL). This is a valuable method that will likely be widely used in the Zebrafish community. The method and analysis are described in sufficient detail to allow most investigators to apply it successfully.

Addressing the following points would strengthen the paper.

1) Previous studies in Drosophila suggested that expression of mutant MetRS during development was detrimental. Does use of HuC-gal4 address this problem? In addition to swimming speed, the authors should test behaviors indicative of brain function, such as sensory information processing, to address the concern of developmental toxicity.

2) The data (for instance Figure 2G) indicate that NSP labeling is not consistently proportional to CFP levels, as a proxy for mutant MetRS expression. Although there are many potential reasons for this, the authors should comment on this potentially interesting observation, and in particular mention variation in rates of protein synthesis and rapid turnover of NSPs within the 24h period examined.

3) It is not correct to refer to 'the intensity of protein synthesis' since NSP lifetimes can be <24h.

4) Similarly 'averaged translation intensity' is not an accurate term.

5) Can the authors comment on the efficiency of ANL labeling NSPs compared to other available methods?

6) Is application of this method limited to Zebrafish larvae or can is be used in older animals? Does ANL penetrance to the brain change over larval development?

7) The text and figure legends differ in statements about use 3 or 4 dpf larvae.

8) The subsection “Induced seizures result in higher protein synthesis levels in neurons” refers to 2.5 h in PTZ. Figure 4A legend and schematic say 2 h.

9) Subsection “Detection of endogenous nascent proteins in neurons across the brain”. The FUNCAT label in neuronal processes could be NSPs that were synthesized in somata and transported into processes. It is not accurate to claim demonstration of sensitivity to detect dendritic protein synthesis in vivo.

10) The demonstration of the cell body segmentation in sparsely labeled brain is very nice, but segmentation using Imaris is challenging in densely labeled regions like the habenula. Can you include an example of the segmented images from habenula?

11) Discussion, third paragraph. The observation that PTZ induces NSPs in habenula in Zebrafish is consistent with the previous study by Del Bel et al. in rats, not the other way around.

12) Subsection “Zebrafish husbandry”. For non-Zebrafish aficionados, it would be helpful to provide more information on the breeding strategy. For instance, how were the HuC-Gal4 and nacre animals used?

13) Subsection “Constructs and transgenic zebrafish”. What are 'AB' eggs?

14) Subsection “BONCAT”. 'pistil' should be 'pestle'?

15) Figure 1G. The black images suggest extreme contrast settings.

16) Figure 2. Figure panel labels don't all match text in figure legend.

17) Figure 4D. Explain the box and whiskers and red line. Are the data plotted from a single animal, only from images shown in B and C?

18) The Liu and Cline paper showing cell type analysis of AHA labeling should be cited.

19) The Schanzenbacher paper is not listed numerically in the references.

Reviewer #3:

In this paper, Shahar and Schuman report a cell-type specific method for labeling newly synthesized proteins in zebrafish larvae. The method relies on expressing a mutated Methionyl tRNA synthetase in neurons using the well-established elav promoter. The mutated enzyme loads the unnatural amino acid azidonorleucine (ANL) instead of methionine into polypeptides. ANL can then be visualized using click-chemistry with a fluorescent alkyne. Using this protocol, authors show brain wide labeling in neurons and increase in intensity following chemically-induced seizures. As mentioned by the authors in the Introduction, cell-type specific labeling of new proteins has been shown before in C. elegans, Drosophila and mice and its establishment in zebrafish larvae is not a major advancement. All of the images in the manuscript are of fixed larvae. Given the obvious advantages of zebrafish for in vivo imaging, the paper would have been exciting if the authors were able to track novel protein synthesis in live animals. But the permeabilization steps required for click-labeling preclude such a possibility. There are other possibilities too such as cell-type specific proteomics characterization of the ANL-labeled proteins but this manuscript does not go that far.

In addition, the authors could address the following to make the manuscript stronger:

1) The authors will do well to establish better the specificity of this tool by showing more control data. Figure 2A shows that there is some labeling even in wild type. Authors need to show the corresponding fluorescence image (WT+ANL+) in Figure 1G. WT+ANL+ control data must also be added to Figure 3. Additionally, even without ANL addition, there is some labeling (Figure 2A, middle lane), probably corresponding to non-specific binding of the fluorescent tag. Thus, not all the signal observed corresponds to novel protein synthesis. A quantification of this non-specific background is required.

2) Related to the point above, were the ANL- samples in Figures 1 and 2 processed for click-dye labeling? Why are the images completely dark?

3) In Figure 2G, authors show colocalization of CFP and click-label in confocal stacks made from 4-6 planes spanning several cell layer thickness. Single optical slices need to be shown.

4) Authors report average of averages in Figures 2H and 2I. As such, the variability in signal amplitude seen in different neurons will likely be much higher than the spread shown. It is not clear how much variability is seen within one region, say habenula, in a single larva. It will also be better to show the data scatter for the bar plots shown in 2H and 2I. This is now standard practice.

5) Why were the spinal cord and habenula alone chosen for the PTZ experiment?

6) Authors call data in Figure 4D and E "Protein Synthesis Level". This is too strong – the signal likely reflects NET incorporation of ANL deriving from both synthesis and degradation of proteins.

7) Supplementary Figure 3 does not seem to be referred anywhere in the text.

8) There are several places where the figure panels and their description in legends or the Results section are mismatched.

https://doi.org/10.7554/eLife.50564.sa1

Author response

[Editors’ note: The authors appealed the original decision. What follows is the authors’ response to the first round of review.]

Reviewer #1:

[…] Previously, the Schuman lab showed that it is possible to examine protein synthesis using a similar approach in whole larvae (Hinz et al., 2011). They now extend their previous findings to a tissue-specific manner (i.e. larval brain and spinal cord neurons) and which (as they report in the Introduction) has been shown previously in other organisms. The Gal4 UAS system for cell type specific expression has been used in zebrafish for a long time. The ability to image sites of protein synthesis in the brain of intact animals post behaviour is likely to be of use to identify regions of the brain important for behaviour (- which is currently visualised in vivo by large through calcium imaging). Overall, the manuscript is straightforward and easy to understand. However, it does not represent a significant advance.

– The authors should examine ANL incorporation upon inhibition of protein synthesis using cycloheximide or other protein synthesis inhibitor. This will determine if the labelling is specific.

We added data showing the significant inhibition of the nascent protein signal when the protein synthesis inhibitor Puromycin was added. We note that inhibition was also observed in the PTZ (seizure-induction) experiment where very high levels of protein synthesis were detected in the PTZ-treated condition. See revised Figure 4.

– The 12 hour incorporation window for the UNAA appears rather long in the context of a fast developing zebrafish larva. What is the half -life of ANL? And what is the shortest ANL exposure window within which incorporation can be detected. Though the 2 hour PTZ treatment shows clear differences over control larvae, a shorter ANL exposure might decrease noise from protein synthesis.

There is no data regarding the half-life of ANL. We have tried shorter incubation times but the signal was sparse, the signal-to-noise was weak and the variability between cells within a larva and between different larvae was high, indicating that under current experimental conditions a short incubation time is not sufficient to measure nascent protein levels. The 12-hour duration demonstrated in this manuscript is the shortest labelling period reported so far for in vivo cell-type-specific metabolic labeling.

– It would be helpful to discuss the advantages/disadvantages of this approach as compared to other approaches to estimate newly synthesized proteins.

We have now added paragraphs in the Introduction (last paragraph) and in the Discussion (last two paragraphs) describing the advantages and disadvantages of the method as compared to other approaches.

– Figure 4 graph D is specific for the animal shown in images B and C, but does not appear to be representative for all the animals analysed, as shown in graph E. It would be useful to do this experiment without ANL and with ANL in wild type controls to examine background autofluorescence or non-specific click reactions post PTZ incubation.

We have done these control experiments and added them to the revised manuscript: Figure 4D, E now contains controls with protein synthesis inhibitor (PSI). We believe that the ANL+/PSI+ control clarifies the background signal in the mutant line, including the remaining non-specific signal (after the washes) and the signal of non-specific click reactions. Additionally, we added a supplementary figure for Figure 4 with WT larvae incubated with ANL and PTZ and subjected to the FUNCAT protocol together with the samples shown in Figure 4. We did not detect a significant signal above background. We note that the WT + ANL control experiments are shown in Figure 4—figure supplement 1.

Reviewer #2:

[…] Addressing the following points would strengthen the paper.

1) Previous studies in Drosophila suggested that expression of mutant MetRS during development was detrimental. Does use of HuC-gal4 address this problem? In addition to swimming speed, the authors should test behaviors indicative of brain function, such as sensory information processing, to address the concern of developmental toxicity.

We did not see any detrimental developmental effects in larvae or fish expressing the MetRS mutant. We have already more than 3 generations of both the UAS-MetRSL270G and then ELAVL3- MetRSL270G lines and we even have them raised in other laboratories for future collaborations. In addition, to the swim speed data in the original submission, we have now added additional behavioral data regarding the intact light preference of larvae. Furthermore, we detect active swimming of freely swimming larvae according to their natural preference in a chamber upon frequent light/dark changes in defined regions within the chamber. Please see the new Figure 1—figure supplement 1. We use this assay routinely and we don’t see any difference between the transgenes to wt larvae (for example in Hinz FI, et al., 2013).

2) The data (for instance Figure 2G) indicate that NSP labeling is not consistently proportional to CFP levels, as a proxy for mutant MetRS expression. Although there are many potential reasons for this, the authors should comment on this potentially interesting observation, and in particular mention variation in rates of protein synthesis and rapid turnover of NSPs within the 24h period examined.

We agree with the comment and now add the explanations for both the general correlation between the CFP\MetRS* levels to the nascent protein intensities as well as for the lack of consistency in some of the cells. Please see the revised Discussion and specifically the third paragraph.

3) It is not correct to refer to 'the intensity of protein synthesis' since NSP lifetimes can be <24h.

Corrected.

4) Similarly 'averaged translation intensity' is not an accurate term.

Corrected.

5) Can the authors comment on the efficiency of ANL labeling NSPs compared to other available methods?

We did not directly measure efficiency levels. Nevertheless, we thank the reviewer for this comment and we add a paragraph in the Discussion comparing the method to fluorescently labeled reporters. We also highlight that this method is the shortest labeling for cell-type-specific non canonical amino acid tagging in vivo.

6) Is application of this method limited to Zebrafish larvae or can is be used in older animals? Does ANL penetrance to the brain change over larval development?

We used the ELAVL3 promoter, which is an early pan-neuronal marker. We did so in order to be able to conduct imaging when the larvae are still translucent. We therefore did not test the method in older fish. This would require a different promoter and therefore a new fish line. As mentioned, the strength of the system is that it can be used for any promoter of interest and indeed it would be interesting to examine in future studies the incorporation of ANL at later ages and whether dissolving the ANL in the water is sufficient.

7) The text and figure legends differ in statements about use 3 or 4 dpf larvae.

8) The subsection “Induced seizures result in higher protein synthesis levels in neurons” refers to 2.5 h in PTZ. Figure 4A legend and schematic say 2 h.

9) Subsection “Detection of endogenous nascent proteins in neurons across the brain”. The FUNCAT label in neuronal processes could be NSPs that were synthesized in somata and transported into processes. It is not accurate to claim demonstration of sensitivity to detect dendritic protein synthesis in vivo.

We clarified all the text issues and typos mentioned by the reviewer.

10) The demonstration of the cell body segmentation in sparsely labeled brain is very nice, but segmentation using Imaris is challenging in densely labeled regions like the habenula. Can you include an example of the segmented images from habenula?

We agree with the reviewer, that the segmentation is challenging and not always perfect but it was good for our goal. As requested, we added in the revised manuscript an example of the segmentation in the densely labeled habenula in Figure 2—figure supplement 7.

11) Discussion, third paragraph. The observation that PTZ induces NSPs in habenula in Zebrafish is consistent with the previous study by Del Bel et al. in rats, not the other way around.

12) Subsection “Zebrafish husbandry”. For non-Zebrafish aficionados, it would be helpful to provide more information on the breeding strategy. For instance, how were the HuC-Gal4 and nacre animals used?

13) Subsection “Constructs and transgenic zebrafish”. What are 'AB' eggs?

14) Subsection “BONCAT”. 'pistil' should be 'pestle'?

All the text mistakes (in points 11 to 14) were corrected.

15) Figure 1G. The black images suggest extreme contrast settings.

All the images in Figure 1G were acquired with the same settings. These are maximal intensity projections of stacks in order to show a large portion of the brain. Therefore, we used settings that allows for the visualization of both controls and the positive sample. We understand the concern of the reviewer and we add in the revised manuscript a video with the raw data of this image including the controls, plane by plane along the z axis (please see Figure 1—videos 1 and 2).

16) Figure 2. Figure panel labels don't all match text in figure legend.

Corrected

17) Figure 4D. Explain the box and whiskers and red line. Are the data plotted from a single animal, only from images shown in B and C?

In Figure 4D, the data plotted is the analysis of the images shown in B, C. Each dot is one cell. The red line is the average of the cells seen in the images. The whiskers are the STDEV. In Figure 4E, each bar represents the average of several larvae. In each larva, the average intensity was calculated and then an average of the average is plotted. Bars are standard error of the mean (n=# larvae). This is explained in the revised text and figure legend.

18) The Liu and Cline paper showing cell type analysis of AHA labeling should be cited.

19) The Schanzenbacher paper is not listed numerically in the references.

Technical formatting issues were corrected.

Reviewer #3:

In this paper, Shahar and Schuman report a cell-type specific method for labeling newly synthesized proteins in zebrafish larvae. The method relies on expressing a mutated Methionyl tRNA synthetase in neurons using the well-established elav promoter. The mutated enzyme loads the unnatural amino acid azidonorleucine (ANL) instead of methionine into polypeptides. ANL can then be visualized using click-chemistry with a fluorescent alkyne. Using this protocol, authors show brain wide labeling in neurons and increase in intensity following chemically-induced seizures. As mentioned by the authors in the Introduction, cell-type specific labeling of new proteins has been shown before in C. elegans, Drosophila and mice and its establishment in zebrafish larvae is not a major advancement. All of the images in the manuscript are of fixed larvae. Given the obvious advantages of zebrafish for in vivo imaging, the paper would have been exciting if the authors were able to track novel protein synthesis in live animals. But the permeabilization steps required for click-labeling preclude such a possibility. There are other possibilities too such as cell-type specific proteomics characterization of the ANL-labeled proteins but this manuscript does not go that far.

Whereas the click chemistry precludes live imaging, the ANL incorporation is done in vivo while the larvae are freely swimming and behaving. The experimentalist can choose the duration of the labeling according to the studied cell-type, the protein synthesis levels (and degradation) and the biological question. The ability to label newly synthesized proteins in vivo for a prechosen duration and then “freeze” the result before imaging has advantages. Because of the strong stability of the fluorescently labeled newly synthesized proteins following the click reaction (which forms covalent bonds), one can image the entire larvae as we demonstrated. In the revised manuscript, we added paragraphs in the Discussion regarding the advantages and state of the art.

In addition, the authors could address the following to make the manuscript stronger:

1) The authors will do well to establish better the specificity of this tool by showing more control data. Figure 2A shows that there is some labeling even in wild type. Authors need to show the corresponding fluorescence image (WT+ANL+) in Figure 1G. WT+ANL+ control data must also be added to Figure 3. Additionally, even without ANL addition, there is some labeling (Figure 2A, middle lane), probably corresponding to non-specific binding of the fluorescent tag. Thus, not all the signal observed corresponds to novel protein synthesis. A quantification of this non-specific background is required.

We performed the WT ANL+ control for all the experiments and now add the add images of WT ANL+. The WT ANL+ control for Figure 2F (previously 2G) is in the new Figure 2—figure supplement 3. We also added the relevant image to the modified Figure 3, now including the WT ANL+ control. We did not add the quantification because the CFP channel (not existing in WT) was used for the segmentation. We used “lookup table fire”, because otherwise the image would look too dark. Quantification of the ANL- fluorescence can be found in Figure 2H. We added quantification of Figure 2A in the new Figure 2—figure supplement 1.

2) Related to the point above, were the ANL- samples in Figures 1 and 2 processed for click-dye labeling?

Yes.

Why are the images completely dark?

They are not completely dark. It looks dark because the background is low in the absence of ANL.

3) In Figure 2G, authors show colocalization of CFP and click-label in confocal stacks made from 4-6 planes spanning several cell layer thickness. Single optical slices need to be shown.

We added as a new figure. Please see Figure 2—figure supplement 4.

4) Authors report average of averages in Figures 2H and 2I. As such, the variability in signal amplitude seen in different neurons will likely be much higher than the spread shown. It is not clear how much variability is seen within one region, say habenula, in a single larva. It will also be better to show the data scatter for the bar plots shown in 2H and 2I. This is now standard practice.

We modified the figure to include the average intensities of all the individual neurons. Please see the modified Figure 2. We moved the previous graphs to the supplementary figure: Figure 2—figure supplement 5.

5) Why were the spinal cord and habenula alone chosen for the PTZ experiment?

They were chosen as two exemplar regions with clear morphology and different density of neurons as well as different roles.

6) Authors call data in Figure 4D and E "Protein Synthesis Level". This is too strong – the signal likely reflects NET incorporation of ANL deriving from both synthesis and degradation of proteins.

We thank the reviewer for the comment. This is corrected.

7) Supplementary Figure 3 does not seem to be referred anywhere in the text.

It is referred to in the original manuscript: “Supplementary Figures 1C-D, 3”. In order to make it clear we now changed to “(Figure 2—figure supplement 2C-D, Figure 2—figure supplement 8).” (The previous Supplementary Figure 3 is now called “Figure 2—figure supplement 8”).

8) There are several places where the figure panels and their description in legends or the Results section are mismatched.

We apologize for this. It was corrected.

https://doi.org/10.7554/eLife.50564.sa2

Article and author information

Author details

  1. Or David Shahar

    Max Planck Institute for Brain Research, Frankfurt, Germany
    Contribution
    Conceptualization, Formal analysis, Investigation, Visualization, Methodology
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-5039-8307

  2. Erin Margaret Schuman

    Max Planck Institute for Brain Research, Frankfurt, Germany
    Contribution
    Conceptualization, Supervision, Funding acquisition, Project administration
    For correspondence
    erin.schuman@brain.mpg.de
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-7053-1005

Funding

European Molecular Biology Organization (ALTF-643-2014)

  • Or David Shahar

Seventh Framework Programme (Marie Curie 628003)

  • Or David Shahar

Max-Planck-Gesellschaft (Research grant)

  • Erin Margaret Schuman

Horizon 2020 Framework Programme (743216)

  • Erin Margaret Schuman

Deutsche Forschungsgemeinschaft (CRC 902)

  • Erin Margaret Schuman

Deutsche Forschungsgemeinschaft (CRC 1080)

  • Erin Margaret Schuman

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank Jan Gluesing and Anja Staab for technical assistance, Anett-Yvon Loos for fish husbandry and managment, Doug Campbell for plasmid design, Alejandro Pinzon-Olejua for help with line screening, Susanne tom Dieck for ANL synthesis and Florian Vollrath for image processing assistance. Work in the laboratory of EMS is supported by the Max Planck Society and The European Research Council under the European Union’s Horizon 2020 research and innovation program (grant agreement No 743216). We also acknowledge the support of DFG CRC 902 and 1080. ODS was supported by an EMBO Long-Term Fellowship and by the European Union’s Horizon 2020 research and innovation programme under the Marie Sklowdoska-Curie grant agreement 628003.

Senior Editor

  1. Didier YR Stainier, Max Planck Institute for Heart and Lung Research, Germany

Reviewing Editor

  1. Vatsala Thirumalai, National Centre for Biological Sciences, India

Reviewer

  1. Hollis T Cline, The Scripps Research Institute, United States

Publication history

  1. Received: July 26, 2019
  2. Accepted: January 20, 2020
  3. Version of Record published: February 24, 2020 (version 1)

Copyright

© 2020, Shahar and Schuman

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 3,150
    Page views
  • 386
    Downloads
  • 5
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Or David Shahar
  2. Erin Margaret Schuman
(2020)
Large-scale cell-type-specific imaging of protein synthesis in a vertebrate brain
eLife 9:e50564.
https://doi.org/10.7554/eLife.50564

Categories and tags

Research organism

Further reading

    1. Cell Biology

    Insights into cargo sorting by SNX32 and its role in neurite outgrowth

    Jini Sugatha, Amulya Priya ... Sunando Datta
    Research Article Updated

    Sorting nexins (SNX) are a family of proteins containing the Phox homology domain, which shows a preferential endo-membrane association and regulates cargo sorting processes. Here, we established that SNX32, an SNX-BAR (Bin/Amphiphysin/Rvs) sub-family member associates with SNX4 via its BAR domain and the residues A226, Q259, E256, R366 of SNX32, and Y258, S448 of SNX4 that lie at the interface of these two SNX proteins mediate this association. SNX32, via its PX domain, interacts with the transferrin receptor (TfR) and Cation-Independent Mannose-6-Phosphate Receptor (CIMPR), and the conserved F131 in its PX domain is important in stabilizing these interactions. Silencing of SNX32 leads to a defect in intracellular trafficking of TfR and CIMPR. Further, using SILAC-based differential proteomics of the wild-type and the mutant SNX32, impaired in cargo binding, we identified Basigin (BSG), an immunoglobulin superfamily member, as a potential interactor of SNX32 in SHSY5Y cells. We then demonstrated that SNX32 binds to BSG through its PX domain and facilitates its trafficking to the cell surface. In neuroglial cell lines, silencing of SNX32 leads to defects in neuronal differentiation. Moreover, abrogation in lactate transport in the SNX32-depleted cells led us to propose that SNX32 may contribute to maintaining the neuroglial coordination via its role in BSG trafficking and the associated monocarboxylate transporter activity. Taken together, our study showed that SNX32 mediates the trafficking of specific cargo molecules along distinct pathways.

    1. Cell Biology
    2. Stem Cells and Regenerative Medicine

    Atf3 defines a population of pulmonary endothelial cells essential for lung regeneration

    Terren K Niethamer, Lillian I Levin ... Edward E Morrisey
    Research Article

    Following acute injury, the capillary vascular bed in the lung must be repaired to reestablish gas exchange with the external environment. Little is known about the transcriptional and signaling factors that drive pulmonary endothelial cell (EC) proliferation and subsequent regeneration of pulmonary capillaries, as well as their response to stress. Here, we show that the transcription factor Atf3 is essential for the regenerative response of the mouse pulmonary endothelium after influenza infection. Atf3 expression defines a subpopulation of capillary ECs enriched in genes involved in endothelial development, differentiation, and migration. During lung alveolar regeneration, this EC population expands and increases the expression of genes involved in angiogenesis, blood vessel development, and cellular response to stress. Importantly, endothelial cell-specific loss of Atf3 results in defective alveolar regeneration, in part through increased apoptosis and decreased proliferation in the endothelium. This leads to the general loss of alveolar endothelium and persistent morphological changes to the alveolar niche, including an emphysema-like phenotype with enlarged alveolar airspaces lined with regions that lack vascular investment. Taken together, these data implicate Atf3 as an essential component of the vascular response to acute lung injury that is required for successful lung alveolar regeneration.

    1. Cell Biology
    2. Microbiology and Infectious Disease

    Profiling the bloodstream form and procyclic form Trypanosoma brucei cell cycle using single-cell transcriptomics

    Emma M Briggs, Catarina A Marques ... Keith R Matthews
    Research Article Updated

    African trypanosomes proliferate as bloodstream forms (BSFs) and procyclic forms in the mammal and tsetse fly midgut, respectively. This allows them to colonise the host environment upon infection and ensure life cycle progression. Yet, understanding of the mechanisms that regulate and drive the cell replication cycle of these forms is limited. Using single-cell transcriptomics on unsynchronised cell populations, we have obtained high resolution cell cycle regulated (CCR) transcriptomes of both procyclic and slender BSF Trypanosoma brucei without prior cell sorting or synchronisation. Additionally, we describe an efficient freeze–thawing protocol that allows single-cell transcriptomic analysis of cryopreserved T. brucei. Computational reconstruction of the cell cycle using periodic pseudotime inference allowed the dynamic expression patterns of cycling genes to be profiled for both life cycle forms. Comparative analyses identify a core cycling transcriptome highly conserved between forms, as well as several genes where transcript levels dynamics are form specific. Comparing transcript expression patterns with protein abundance revealed that the majority of genes with periodic cycling transcript and protein levels exhibit a relative delay between peak transcript and protein expression. This work reveals novel detail of the CCR transcriptomes of both forms, which are available for further interrogation via an interactive webtool.