Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Mar 1.
Published in final edited form as: J Neurogenet. 2018 Nov 29;33(1):1–9. doi: 10.1080/01677063.2018.1531857

Genetic Analysis of KillerRed in C. elegans Identifies a Shared Role of Calcium Genes in ROS-mediated Neurodegeneration

Lyndsay E A Young 1, Chelsea Shoben 1, Kyra Ricci 1, Daniel C Williams 1,*
PMCID: PMC6486406  NIHMSID: NIHMS1520197  PMID: 30489172

Abstract

In C. elegans, neurodegeneration induced by excitotoxicity or aggregation of misfolded proteins is dependent on genes involved in calcium release from the endoplasmic reticulum. Reactive oxygen species (ROS) can also induce neurodegeneration, but the relationship between ROS-mediated neurodegeneration and calcium has not been established. We activated KillerRed in the GABA neurons of C. elegans to produce ROS that leads to functional loss and structural degeneration of these neurons and demonstrated that the severity of neurodegeneration was dependent on extent of KillerRed activation. To genetically examine the role of calcium in ROS-mediated neurodegeneration, we measured functional neurodegeneration in itr-1 (inositol trisphosphate receptor), crt-1 (caltreticulin), and unc-68 (ryanodine receptor) mutants. Similar to other neurotoxic conditions, neurodegeneration triggered by KillerRed was reduced in itr-1 and crt-1 mutants. Somewhat unexpectedly, genetic or pharmacological disruption of unc-68 had a minimal effect on neurodegeneration. Our results indicate ROS-mediated neurodegeneration occurs through a conserved calcium regulated mechanism and suggest that components of the degeneration process have different sensitivities to ROS.

Keywords: C. elegans, neurodegeneration, reactive oxygen species, calcium

Introduction

Degeneration of neurons is a hallmark of neurodegenerative diseases and can be caused by factors such as stroke, physical trauma, or aging. Reactive oxygen species (ROS) and oxidative stress are associated with age and degenerating neurons. Because ROS are normal by-products of metabolism and neurons are highly metabolically active, these cells are particularly susceptible to oxidative damage caused by ROS. Excessive cellular ROS can result in protein oxidation, increased DNA mutations, and lipid peroxidation, all of which have a detrimental effect on neuronal function (Emerit, Edeas, & Bricaire, 2004; Pollari, Goldsteins, Bart, Koistinaho, & Giniatullin, 2014). Although ROS are associated with neurodegeneration, the cellular pathways involved in initiation of degeneration due to ROS are not well understood.

C. elegans has been used to identify genes and molecular mechanisms of degeneration triggered by various genetic and environmental insults. Initial studies focused on degeneration resulting from gain-of-function mutations that result in hyperactive ion channels. These channels, such as mec-4, mec-10 and deg-1, belong to the DEG/ENaC family of conserved ion channels that cause necrotic-like degeneration of touch receptor neurons when mutated to a constitutive open state (Driscoll & Chalfie, 1991; Syntichaki & Tavernarakis, 2004; Waldmann, Champigny, Voilley, Lauritzen, & Lazdunski, 1996). Similarly, dominant alleles that disrupt desensitization of the nicotinic acetylcholine receptor deg-3 also cause neurodegeneration (Treinin & Chalfie, 1995). In addition to hyperactive ion channel toxicity, motor neurons of the ventral cord degenerate in response to constitutive activity of a heterotrimeric G protein (Gαs), encoded by gsa-1 (Berger, Hart, & Kaplan, 1998; Korswagen, Park, Ohshima, & Plasterk, 1997). Other experiments have initiated degeneration using pharmacological treatments that model human neurodegenerative diseases or heterologous expression of human disease genes in C. elegans. Worms treated with the neurotoxins 6-hydroxydopamine or 1-methyl-4-phenlypyridinium display similar degeneration of dopaminergic neurons as observed in mammalian models of Parkinson’s disease (Nass, Hall, Miller, & Blakely, 2002; Pu & Le, 2008). Transactive response DNA-binding protein 43 (TDP-43) has been associated with amyotrophic lateral sclerosis and expression of mutant TDP43A315T in GABA neurons results in functional and structural degeneration of these neurons (Vaccaro et al., 2012). Because of the genetic and molecular tools available in C. elegans, the worm is an attractive model organism for genetic dissection of neurodegeneration and identification of cellular mechanisms of degeneration.

Genetic and pharmacological analysis of neurodegeneration in C. elegans indicates that calcium (Ca2+) signaling has an important functional role in degeneration triggered by various toxic conditions. Forward genetic screens isolated crt-1 loss-of-function alleles that suppress neurodegeneration induced by hyperactive degenerin channels or constitutively active Gαs (Barbagallo, Prescott, Boyle, Climer, & Francis, 2010; Xu, Tavernarakis, & Driscoll, 2001). crt-1 encodes calreticulin, a multi-functional endoplasmic reticulum (ER) resident protein that maintains ER Ca2+ levels via a high-capacity Ca2+ binding domain (Lee, Singaravelu, Park, & Ahnn, 2007; Nakamura et al., 2001). Suppression of neurodegeneration by disruption of crt-1 is likely due to altered Ca2+ homeostasis in the ER as knock-down of cnx-1 (calnexin) or pharmacological manipulation of Ca2+ can also suppress neurodegeneration (Xu et al., 2001). In addition, disruption of the ryanodine receptor (unc-68) or the inositol trisphosphate receptor (itr-1) that are involved in Ca2+ release from the ER can reduce neurodegeneration (Barbagallo et al., 2010; Xu et al., 2001). Genetic and pharmacological perturbation of Ca2+-regulated calpain and aspartyl proteases also reduce neurodegeneration induced by hyperactive ion channels or protein misfolding (Aggad, Vérièpe, Tauffenberger, & Parker, 2014; Syntichaki, Xu, Driscoll, & Tavernarakis, 2002). Collectively, these results have led to a model in which diverse neurotoxic conditions trigger Ca2+ release from the ER that leads to proteolytic cell destruction.

KillerRed is a genetically encoded photosensitizer that produces toxic levels of ROS in response to photoactivation (Bulina et al., 2006). Illumination of transgenic worms expressing KillerRed in specific neurons results in cell-autonomous neuronal cell death (Williams et al., 2013). The neuronal cell death and degeneration due to KillerRed-mediated ROS toxicity resembles other means of inducing neurodegeneration in C. elegans. First, the cell bodies of affected neurons exhibit cell swelling and alterations in nuclear structure when visualized by differential interference contrast microscopy. This abnormal neuronal morphology is similar to the effects caused by mec-4(gf) and deg-1(gf) mutations (Hall et al., 1997). Second, GFP expressing neurons that have been damaged by KillerRed exhibit fragmented neuronal processes and blebbed fluorescent puncta that is also observed in dopaminergic neurons that degenerated in response to manganese treatment (Ijomone, Miah, Peres, Nwoha, & Aschner, 2016). Finally, neurodegeneration triggered by KillerRed and other neurotoxic conditions, such as TDP43A315T or ion channel gain-of-function alleles, is caspase independent and proceeds via a necrosis-like process (Aggad et al., 2014; Barbagallo et al., 2010). Because the degeneration induced by KillerRed is morphologically and genetically similar to other neurotoxic insults, this optogenetic method provides a novel means to investigate the molecular mechanisms and cellular pathways of ROS-mediated neurodegeneration.

The experiments described here utilize worm strains that express soluble KillerRed in the GABA neurons of C. elegans. The cell bodies of 19 GABA neurons are located longitudinally within the ventral nerve cord and send circumferential commissures to the dorsal nerve cord. These neurons innervate ventral and dorsal body-wall muscles and are inhibitory motor neurons at the neuromuscular junction (Richmond & Jorgensen, 1999). Worms lacking GABA neuron function cannot relax their body-wall muscles and exhibit a characteristic “shrinker” phenotype when touched on the head (McIntire, Jorgensen, Kaplan, & Horvitz, 1993). Illumination of worms expressing KillerRed specifically in GABA neurons causes neurotoxic production of ROS and the affected animals shrink (Williams et al., 2013; Young & Williams, 2015). Using an optimized method of KillerRed activation and measuring functional degeneration using the shrinker assay, we genetically test the role of genes involved in calcium storage and release from the ER in ROS-mediated neurodegeneration.

Methods

C. elegans strains

Worms were cultured on Nematode Growth Media (NGM) at 20°C using established methods (Brenner, 1974; Stiernagle, 2006). Strains expressing KillerRed and GFP in GABA neurons (wpIs15 [unc-47p::KillerRed], juIs76[unc-25p::GFP] have been previously described (Jin, Jorgensen, Hartwieg, & Horvitz, 1999; Williams et al., 2013). The alleles crt-1(bz30), itr-1(sa73), and unc-68(r1161) were obtained from the Caenorhabditis elegans Genome Center. Relevant strains were generated by crossing and verified by direct visualization of KillerRed and GFP, phenotypic characterization of crt-1 and itr-1, or PCR of unc-68(r1161) as previously described (Young & Williams, 2015). The r1161 allele deletes ~7.2kb of the 5’ end of unc-68 and represents the null allele based on allelic interaction analysis (Maryon, Coronado, & Anderson, 1996).

Illumination of KillerRed animals

Ten to fifty L4 to young-adult aged worms were placed in 20 μl of M9 buffer that was layered on 30 μl of 1% agarose in the lid of a 200 μl PCR tube. Animals were exposed to green light using a 10x objective lens of an Olympus BX51 fluorescent microscope via TRITC filters and a 100 W mercury bulb. The power of each illumination was measured using an optical power meter (ThorLabs, Newton, NJ) at 585 nm and was constant for each experiment. After illumination, worms were recovered overnight on a NGM plate with OP50 bacteria before scoring for the shrinker phenotype. For dantrolene experiments, after illumination, animals were recovered on plates containing 50 μM dantrolene or vehicle (50% DMSO) and scored 24 hours later.

Scoring of shrinker

Recovered worms were analyzed for loss of GABAergic neuron function by gently tapping them on the head and scoring the response. Animals that fail to move backward, but instead exhibit body-wall muscle contraction were scored as shrinker (Jorgensen, 2005). Data are presented as percentage of animals that display the shrinker phenotype +/− 95% confidence intervals and significance was determined using two-tailed Fisher’s exact test (GraphPad Quick Calcs). Videos were captured using a SZX10 microscope and DP27 digital camera (Olympus) with cellSens software.

Microscopy and fluorescence intensity measurements

Worms were mounted on 8% agarose pads and immobilized using 0.1 μm polystyrene beads (Polysciences Inc.). Individual worms expressing GFP in GABA neurons were visualized using an Olympus BX51 microscope with GFP filters. Commissures that were markedly thin or had multiple blebbing or beaded areas were scored as damaged. Images of KillerRed were obtained using TRITC filter set and DCF images were obtained with GFP filters and captured with a Q-Color 3 CCD (Olympus). Mean pixel intensity of individual neuronal cell bodies was measured using ImageJ.

Single worm RT-qPCR

Established methods for single worm RT-qPCR were used (Ly, Reid, & Snell, 2015). In brief, individual worms were lysed in 5μl of lysis buffer containing 5 mM Tris - pH 8.0, 0.5% Triton X-100, 0.5% Tween-20, 25 mM EDTA, and 1 mg/ml proteinase-K. Equal volume lysates of individual worms were separated and 10 μl RT-qPCR reactions were run using the iTaq Universal SYBR Green One-Step Kit (Bio-Rad). Amplification of KillerRed (92bp) used (5’to 3’) FW (GATGGAGGACTTATGATGGGAC), REV (TTGTAACGAAATGTGGTCCTGG) and the reference gene act-1 (111bp) (Ly et al., 2015) used FW (ACGCCAACACTGTTCTTTCC) and REV (CGATGATCTTGATCTTCATGGTTG). Samples were run on the Bio-Rad CFX96 Real-time System and melting curve analysis was performed over a range of 45-95°C to verify amplification of a single product.

Measurement of ROS production

KillerRed expressing animals were treated in the dark with 10μM H2DCFDA (Invitrogen) suspended in M9 buffer for 3 hours. To observe DCF fluorescence during an illumination, worms were mounted on a microscope slide and illuminated and visualized with a 60x objective on an Olympus BX51 microscope. At timepoints within the illumination, KillerRed images (TRITC filter) and DCF fluorescence (GFP filter) were obtained. To compare DCF fluorescence between strains, H2DCFDA treated animals were illuminated under a 10x objective for 15 minutes and then recovered on an NGM plate before mounting and imaging of KillerRed and DCF fluorescence. Mean pixel intensity of DCF present in cell bodies was determined with ImageJ using regions of interest defined by KillerRed fluorescence.

Results

Optimization of KillerRed induced neurodegeneration for genetic analysis

Illumination of KillerRed results in ROS production and the phototoxic effects of KillerRed are dependent on the duration of illumination (Bulina et al., 2006). We sought to optimize illumination conditions that reduce KillerRed activity to moderate levels and allow identification of changes in ROS-mediated neurodegeneration based on genetic background. We illuminated wild-type worms expressing KillerRed in GABA neurons with green light and scored these animals for loss of GABA neuron function after a 24-hour recovery period. After being gently touched on the head with a thin platinum wire, wild-type animals respond by moving backwards away from the stimulus, while animals that lack GABA neurons contract their body-wall muscles and exhibit the shrinker phenotype (Figure 1 and Supplemental Video 1 & 2). We determined the percentage of shrinker animals from populations that were illuminated at constant power for varying amounts of time (Figure 2A). There was an illumination dependent increase in shrinker frequency. Illumination times of 5 minutes at 0.19 W resulted in ~70% animals having the shrinker phenotype, while increasing illumination times led to saturating effects with almost all of the 20 minute illuminated animals displaying the shrinker phenotype. This effect was dependent on KillerRed because illumination of wild-type animals that lack KillerRed failed to produce shrinker animals (data not shown and Williams et al. 2013). These results indicate the functional effects of KillerRed can be experimentally regulated by altering the levels of illumination.

Figure 1.

Figure 1.

Open in a new tab

Illumination of GABA::KillerRed animals results in shrinkers. Sequential images of control and illuminated GABA::KillerRed worms in response to gentle touch on the head. Illuminated animals were exposed to green light at 0.16 W for 30 minutes to induce a robust shrinker phenotype. Animals were touched in Frame Number 0 and each frame represents one second after stimulation. Dashed lines indicate position of the animal’s head (thick) and tail (thin) in Frame Number 0.

Figure 2.

Figure 2.

Open in a new tab

KillerRed-mediated neurodegeneration varies with illumination time. (A) Functional neurodegeneration. Populations of GABA::KillerRed animals were exposed to green light at 0.19 W for the indicated illumination times and then scored for the shrinker phenotype after a 24-hour recovery period. At least 15 animals were scored for each illumination condition and data represents the percentage of shrinker animals (+/− 95% confidence intervals). P < 0.0001 for each illumination condition compared to no illumination control (Fisher’s exact test, two-tailed). (B) Structural neurodegeneration. Representative images of GABA::KillerRed worms illuminated at 0.14 W for the indicated times. After a 24-hour recovery period, animals were visualized for GFP fluorescence to assess structural integrity of GABA neuronal processes (arrows). Asterisks indicate neuron cell bodies and arrowheads mark the ventral nerve cords that are out of focus. Scale bar = 5μm. (C) Quantification of neuronal damage. Worms were illuminated at 0.14 W for the indicated time and after a 24-hour recovery period, neuronal commissures processes were scored for damage. Commissures that were significantly thin, fragmented, or displayed blebbing were scored as damaged. Data represents the total percentage of damaged commissures (+/− 95% confidence intervals) from at least 10 animals for each illumination condition. P < 0.005 for each illumination time compared to no illumination control (Fisher’s exact test, two-tailed).

We assessed the effects of KillerRed on neuron structure by imaging GABA neurons that express GFP. Animals that express KillerRed and GFP in the GABA neurons were illuminated for varying amounts of time and then visualized by fluorescent microscopy 24 hours later (Figure 2B). In the absence of illumination, the neuron cell bodies had a normal oval morphology and the nerve cords and commissures were intact. We observed signs of neuronal damage that increased with longer illumination times. Short illumination times produced GABA neuron commissures that appeared thinner when compared to unilluminated animals. These affected commissures often had large accumulations of fluorescence at various locations along the process. Longer illumination times resulted in more extensive neuronal damage. The commissures of affected neurons were often much thinner than illuminated controls and were broken at multiple locations with numerous blebs of fluorescence. In addition, neuronal cell bodies of these animals were often rounded relative to unilluminated controls. Severely damaged neurons had commissures that were barely visible and consisted of small fluorescence puncta. We quantified the structural effects of KillerRed, by scoring commissures of worms that were illuminated for varying amounts of time (Figure 2C). Individual commissures that were thin, fragmented, and had multiple blebs were scored as damaged. Illumination conditions that resulted in functional degeneration, as measured by shrinker frequency, caused increased structural damage to the GABA neurons. Because the effects of KillerRed are dependent on extent of illumination, these results indicate neurodegeneration can be modulated to a baseline level that will allow testing the requirement of specific genes for ROS-mediated neurodegeneration.

Disruption of calcium signalling suppresses KillerRed neurodegeneration

We previously demonstrated similar effects of KillerRed in wild-type and unc-68 mutants, which suggested distinct pathways of neurodegeneration based on cellular injury (Young & Williams, 2015). After establishing an illumination protocol associated with intermediate neuronal damage, we expanded our genetic analysis of ROS-mediated neurodegeneration by testing the requirement of other genes associated with calcium signalling. We generated KillerRed expressing animals in an itr-1, crt-1, or unc-68 mutant background and compared the effects of KillerRed to wild-type animals. In the absence of illumination, these genetic backgrounds do not confer a shrinker phenotype (Supplemental Figure 1) and thus, neurodegeneration can be measured using the shrinker assay. Animals were illuminated, then scored for the shrinker phenotype after 24 hours. Relative to wild-type animals, there was a statistically significant reduction in the shrinker frequency of both itr-1 and crt-1 mutants after illumination (Figure 3A). Consistent with previous results, we found no significant difference in the shrinker frequency between wild-type and unc-68 mutant animals (Figure 3A). In addition to genetic disruption, we tested pharmacological perturbation of unc-68 by treating animals with the ryanodine receptor agonist dantrolene (Fruen, Mickelson, & Louis, 1997; Song, Karl, Ackerman, & Hotchkiss, 1993). Dantrolene treatment failed to have a marked effect on the percentage of illuminated animals that were shrinker (Figure 3B).

Figure 3.

Figure 3.

Open in a new tab

Genetic and pharmacological analysis of KillerRed-mediated neurodegeneration. (A) GABA::KillerRed worms of the indicated genotype were illuminated for 5 minutes at 0.17 W and scored for the shrinker phenotype. Data represents the percentage of shrinkers +/− 95% confidence intervals of the indicated number of animals pooled from at least 3 independent illuminations. Significance was determined using Fisher’s exact test (two-tailed) for each genotype vs. wild type, ** = P<0.01, * = P<0.05. (B) Shrinker frequency of animals treated with dantrolene or vehicle (DMSO). Data represents the percentage shrinkers +/− 95% confidence intervals. At least 15 animals were scored for each condition. P < 0.001 for illuminated vs. dark (Fisher’s exact test, two-tailed)

itr-1 and crt-1 mutations do not affect KillerRed activity

One possibility for the reduction of ROS-mediated neurodegeneration seen in itr-1 and crt-1 backgrounds is that these genes could influence KillerRed activity. To test this, we first measured KillerRed mRNA and fluorescence in these different genetic backgrounds. We performed single-worm RT-qPCR to assess KillerRed mRNA levels (Figure 4A and Supplemental Figure 2). Using act-1 as a reference gene (Ly et al., 2015),we found no difference in KillerRed mRNA levels between wild-type and itr-1 or crt-1 mutant animals. This result suggests the protein products of itr-1 and crt-1 do not have a functional role in transcription or stabilization of KillerRed mRNA. To more directly assess KillerRed expression, we also measured KillerRed fluorescence in living animals (Figure 4B and C). Images of KillerRed expressing animals were collected and fluorescence intensity was determined within individual neurons. Consistent with the RT-qPCR results, we found similar average fluorescence intensity in itr-1 and crt-1 animals compared to wild type.

Figure 4.

Figure 4.

Open in a new tab

KillerRed expression is normal in itr-1 and crt-1 mutants. (A) RT-qPCR analysis of KillerRed mRNA. Average ΔCq values between KillerRed and the reference gene act-1 were determined from individual worms. Bars represent mean ΔCq values +/− SEM. Circles represent the Cq value determined from individual worms. (B) Representative images of KillerRed expressing GABA neuron cell bodies captured with fluorescence microscopy and (C) average neuronal fluorescence intensity. Images were collected from at least 9 worms and mean pixel density determined from at least 3 neuron cell bodies per image. Data represents mean pixel density from over 50 neurons, +/− SEM. Scale bar = 10μm.

To test whether ROS production by KillerRed is affected by genetic background, we measured fluorescence of a ROS sensing molecular probe. Animals expressing KillerRed in the GABA neurons were treated with 2,7-dichlorodihydrofluorescein diacetate (H2DCFDA), which is a membrane-permeant reporter that is converted to the fluorescent dichlorofluorescein (DCF) when oxidized by ROS (Labuschagne & Brenkman, 2013). Illumination of treated animals resulted in green fluorescence that overlapped with KillerRed fluorescence (Figure 5A). The intensity of fluorescence increased with longer illumination times. Notably, tissues that surround the GABA neurons, including other neuronal cell bodies along the ventral cord, did not have any detectable DCF fluorescence. This demonstrates that ROS production is dependent on both KillerRed and illumination and is consistent with the absence of neurodegeneration in illuminated animals that do not express KillerRed. We measured DCF fluorescence intensity in neuronal cell bodies of wild-type and mutant animals after illumination. The amount of DCF fluorescence was similar in all genetic backgrounds tested (Figure 5 bottom). Collectively, these data demonstrate that itr-1 or crt-1 do not affect KillerRed activity and therefore indicate the suppression of neurodegeneration in these genetic backgrounds is not due to reduced ROS production.

Figure 5.

Figure 5.

Open in a new tab

ROS production. (A) Neurons of a GABA::KillerRed animal treated with 10μM H2DCFDA and illuminated with green light under a 60× objective. At the indicated time, images of KillerRed or DCF fluorescence were obtained and merged (bottom row). Arrows indicate GABA neuron cell bodies and arrow head indicates ventral nerve cord. Asterisk in DCF images is non-specific gut autofluorescence. Scale bar = 20μm. (B) Quantification of DCF fluorescence in different genetic backgrounds. Populations of worms were treated with H2DCFDA and then illuminated at 0.13 W for 15 minutes and then recovered and mounted for microscopy. Images were collected from at least 6 worms in each genotype and mean pixel density determined from at least 3 neuron cell bodies per image. Data represents mean pixel density from at least 30 neurons, +/− SEM.

Discussion

In this paper, we used the photosensitizer KillerRed to produce ROS in the GABA neurons of C. elegans. Illumination of living animals resulted in functional and structural degeneration of these neurons that was dependent on the extent of KillerRed activation. Using illumination times that caused a moderate level of neurodegeneration, we tested the requirement of genes involved in calcium storage and release from the ER in ROS-mediated neurodegeneration. We observed significantly lower levels of functional degeneration, as measured by the shrinker assay, in calreticulin and inositol trisphostate receptor mutants. Importantly, we demonstrated similar levels of KillerRed expression in these mutant backgrounds. Genetic or pharmacological perturbation of the ryanodine receptor had a negligible effect on levels of neurodegeneration. Collectively, our results add KillerRed to models of toxicity that triggers neurodegeneration in C. elegans and suggest that degeneration induced by various cellular injuries occurs through a calcium regulated process.

Calcium and neurodegeneration

Calcium signalling is associated with neurodegeneration that is triggered by various intrinsic and exogenous neurotoxic insults. In mammalian systems, degeneration via glutamate induced excitotoxicity is caused by excessive Ca2+ influx though hyperactive ionotropic glutamate receptors (Arundine & Tymianski, 2004; Brassai, Suvanjeiev, Bán, & Lakatos, 2015). Accumulation of misfolded proteins is associated with many neurodegenerative diseases and can induce the ER unfolded protein response and result in aberrant Ca2+ release from the ER (Hetz & Mollereau, 2014; Soto, 2003). Genetic analysis in C. elegans has identified Ca2+ related genes as being required for neurodegeneration induced by diverse neurotoxic conditions. Initially a genetic selection for mutations that suppress the paralyzed phenotype associated with ectopic expression of mec-4(d) ion channels resulted in the isolation of crt-1 (calreticulin) alleles (Xu et al., 2001). Similarly, deletion of crt-1 is able to partially suppress motor neuron loss that is associated with dominant mutations in nicotinic acetylcholine receptors that display increased activation and slow inactivation (Barbagallo et al., 2010). Ectopic expression of TDP-43A315T results in age dependent neurodegeneration of GABA neurons that is delayed in crt-1 mutants (Aggad et al., 2014; Vaccaro et al., 2012). The suppression of neurodegeneration by crt-1 is likely due to altered release of calcium from the ER as degeneration induced by mutations in the plasma membrane acetylcholine receptor deg-3 that leads to increased Ca2+ influx is not affected by crt-1 mutations (Treinin & Chalfie, 1995; Xu et al., 2001). Subsequent studies have demonstrated that neurodegeneration is also reduced by mutations in genes that encode calcium regulated proteases (Aggad et al., 2014; Syntichaki et al., 2002). Collectively, these results have led to a model in which neurotoxic injuries cause inappropriate release of Ca2+ from the ER that causes activation of cytoplasmic proteases leading to necrotic cell death.

KillerRed causes ROS-mediated neurodegeneration

We used the ROS producing fluorescent protein KillerRed to cause neurodegeneration of GABA neurons. Illumination of living animals expressing KillerRed results in cell-autonomous ROS production that induces functional and structural degeneration of these neurons. Loss of GABA function is measured by scoring illuminated animals for the shrinker phenotype (Figure 1). Fluorescence microscopy of affected neurons that express GFP reveals abnormal morphology; the nerve cords and commissures become thin fragments and contain numerous fluorescent accumulations. The most damaged neuronal processes are comprised of numerous tiny blebs that are separated by thin wisps of fluorescence. Because the degenerative effects of KillerRed are similar to other neurotoxic conditions, KillerRed expands the panel of insults that trigger neurodegeneration in C. elegans and allows for further dissection of the molecular pathways and cellular mechanisms of neurodegeneration that is due to ROS. Furthermore, the effects of KillerRed can be controlled by varying the amount of time animals are illuminated, allowing for identification of genetic conditions that influence degeneration.

Calcium and ROS-mediated neurodegeneration

In this study, we optimized using KillerRed in C. elegans to test the role of genes involved in Ca2+ homeostasis and release from the ER in ROS-mediated neurodegeneration. We found that mutations in itr-1 or crt-1 reduced the toxic effects of KillerRed on GABA neuron function. This reduction in neurodegeneration is not due to altered KillerRed activity as we observed similar levels of KillerRed mRNA and fluorescence in itr-1 and crt-1 mutants when compared to wild-type animals. In addition, the amount of ROS produced was not affected in these genetic backgrounds as measured by oxidation of a fluorescent ROS probe. The effects of KillerRed-mediated neurodegeneration by itr-1 and crt-1 mutants is similar to suppression of degeneration induced by other neurotoxic conditions, such as mec-4(d), ACR-2(L/S) and TDP-43A315T. Genetic analysis in C. elegans leads to a model by which various neurotoxic insults induce Ca2+ release from the ER and subsequent activation of Ca2+ dependent proteases (Aggad et al., 2014; Syntichaki et al., 2002). Although our results are consistent with this model, future studies examining structural neurodegeneration induced by KillerRed in crt-1 and itr-1 mutant backgrounds are necessary to rule out suppression of KillerRed by an alternative mechanism.

The extent of neurodegeneration suppression by crt-1, itr-1, or unc-68 mutations is variable with different neurotoxic insults. Nonsense mutations in crt-1 (bz29 or bz30) almost completely abolish loss of touch receptor neurons observed in mec-4(u231) animals while loss of itr-1 or unc-68 results in a ~50% reduction in of mec-4-induced neurodegeneration (Xu et al., 2001). The bz30 allele likely represents the null phenotype because missense mutations of crt-1 show less suppression of mec-4(u231) than bz30 (Xu et al., 2001) and bz30 suppresses TDP-43A315T degeneration to the same extent as the deletion allele jh101 (Aggad et al., 2014). By contrast, degeneration of cholinergic motor neurons caused by ACR-2(L/S) channels is partially suppressed by the deletion mutation crt-1(ok948) (Barbagallo et al., 2010). Our results with the crt-1-dependent suppression of KillerRed-mediated neurodegeneration are similar to the partial suppression of ACR-2(L/S) toxicity (Barbagallo et al., 2010). However, genetic or pharmacological disruption of unc-68 does not affect KillerRed-mediated neurodegeneration (Figure 3 and Young and Williams, 2015), but does affect ACR-2(L/S) neurodegeneration (Barbagallo et al., 2010). Together, these data could suggest different mechanisms of degeneration based on neuronal type (touch receptor neurons vs. motor neurons), or alternatively distinct degenerative effects based on neurotoxic conditions. We prefer the latter hypothesis because of similarities in the genetic requirements for mec-4(d)-induced degeneration of touch receptor neurons (Xu et al., 2001) and TDP-43A315T-induced degeneration of GABAergic neurons (Aggad et al., 2014). Degeneration of both these neuronal types is more sensitive to crt-1-dependent suppression than to suppression by itr-1. By contrast, we find that GABAergic neurons subject to KillerRed-induced degeneration are equally affected by crt-1 and itr-1.

In contrast to the suppression by crt-1 and itr-1 mutations, genetic or pharmacological disruption of the C. elegans ryanodine receptor (unc-68) had minimal effect on KillerRed-mediated neurodegeneration. Our results suggest that ROS induces Ca2+ release from the ER to initiate neurodegeneration via the inositol trisphosphate receptor and that the ryanodine receptor does not have a role in ROS-mediated neurodegeneration. Although Ca2+ release from the ER could be a generic response to neurotoxic conditions, such as oxidative stress, the inositol trisphosphate receptor is a potential target of ROS signalling. In vertebrate cells, addition of ROS generating compounds can induce Ca2+ release from ER stores in a manner is that dependent on expression of specific inositol trisphosphate isoforms (Bánsághi et al., 2014). Similar levels of KillerRed-mediated neurodegeneration in wild-type and unc-68 mutants is surprising as disruption of unc-68 is able to suppress neurodegeneration caused by mec-4(d), ACR-2(L/S), or TDP-43A315T (Aggad et al., 2014; Barbagallo et al., 2010; Xu et al., 2001). Lack of KillerRed suppression by unc-68 could reflect the cell-autonomous activity of KillerRed and absence of UNC-68 in GABA neurons. Expression of unc-68 is primarily in body-wall, pharyngeal, and vulval muscles based on antibody staining, however neuronal expression of a lacZ reporter is observed with truncated unc-68 promoter transgenes (Maryon, Saari, & Anderson, 1998; Sakube, Ando, & Kagawa, 1997). By contrast staining with anti-ITR-1 antibodies or visualization of GFP under the itr-1 promoter revealed broad tissue expression of itr-1 with signal present in unidentified neurons in the ventral nerve cord, where the GABA cell bodies are located (Baylis, Furuichi, Yoshikawa, Mikoshiba, & Sattelle, 1999). Alternatively, the differential requirement of unc-68 for neurodegeneration induced by other neurotoxic conditions compared to KillerRed could reflect the inducible nature of KillerRed-mediated neurodegeneration. Our experiments activated KillerRed in L4 to young adult animal and we scored neurodegeneration 24 hours after induction. By contrast other neurotoxic conditions are due to expression of mutant channels or mis-foleded proteins under promoters that are active in embryos and early larval stages (Vaccaro et al., 2012). Notably, ACR-2(L/S)-mediated neurodegeneration is observed in L1 larva soon after hatching (Barbagallo et al., 2010). Future experiments examining neurodegeneration induced by KillerRed activation during embryonic development could test this hypothesis.

Conclusions

KillerRed can be used to induce neurodegeneration within targeted cells and adds ROS to the list of neurotoxic insults available in C. elegans. Moreover, KillerRed activity can be experimentally controlled allowing for investigations into genes or drugs that affect neurodegeneration due to excessive ROS. Like other neurotoxic conditions, KillerRed-mediated neurodegeneration is reduced by disruption of crt-1 or itr-1, indicating a shared Ca2+ dependent mechanism of neurodegeneration.

Supplementary Material

Supp1

Supplemental Figure 1. itr-1, crt-1, and unc-68 mutants are not GABA defective. Animals of the indicated genotype were analysed for GABA function by the shrinker assay. unc-47 mutants are GABA defective because they lack the vesicular GABA transporter. Data represents the percentage shrinkers +/− 95% confidence intervals.

Supp2

Supplemental Figure 2. Single worm RT-qPCR analysis of KillerRed. (A) Representative fluorescence measured for each cycle of the indicated samples. Color represents target amplicon and symbols represent genotypes: KillerRed (red) and act-1 (green) wild-type (square), itr-1 (triangle), and crt-1 (circles). (B) Melting curve analysis demonstrating amplification of single amplification products.

Supp3

Supplemental Video 1. Normal worm response to gentle touch on the head.

Download video file (7.3MB, mp4)
Supp4

Supplemental Video 2. Shrinker phenotype. Animals that lack GABA neuron function display a characteristic shrinker phenotype when gently touched on the head. Stimulus results in uncoordinated body-wall muscle contraction without relaxation that is manifest as the shrinker phenotype. Video is of a KillerRed animal that has been illuminated for 12 minutes.

Download video file (10.4MB, mp4)

Acknowledgements

This work utilized strains provided by the CGC, which is funded by NIH Office of Research Infrastructure Programs (P40 OD010440). This publication was supported in part by grant P20GM103499 (SC-INBRE) from the National Institutes of General Medical Sciences, National Institutes of Health. The authors are grateful to Dr. Monica Driscoll for guidance and comments on this work.

References

  1. Aggad D, Vérièpe J, Tauffenberger A, & Parker JA (2014). TDP-43 toxicity proceeds via calcium dysregulation and necrosis in aging Caenorhabditis elegans motor neurons. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 34(36), 12093–12103. 10.1523/JNEUROSCI.2495-13.2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arundine M, & Tymianski M (2004). Molecular mechanisms of glutamate-dependent neurodegeneration in ischemia and traumatic brain injury. Cellular and Molecular Life Sciences: CMLS, 61(6), 657–668. 10.1007/s00018-003-3319-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bánsághi S, Golenár T, Madesh M, Csordás G, RamachandraRao S, Sharma K, Hajnóczky G (2014). Isoform- and species-specific control of inositol 1,4,5-trisphosphate (IP3) receptors by reactive oxygen species. The Journal of Biological Chemistry, 289(12), 8170–8181. 10.1074/jbc.M113.504159 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Barbagallo B, Prescott HA, Boyle P, Climer J, & Francis MM (2010). A dominant mutation in a neuronal acetylcholine receptor subunit leads to motor neuron degeneration in Caenorhabditis elegans. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 30(42), 13932–13942. 10.1523/JNEUROSCI.1515-10.2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Baylis HA, Furuichi T, Yoshikawa F, Mikoshiba K, & Sattelle DB (1999). Inositol 1,4,5-trisphosphate receptors are strongly expressed in the nervous system, pharynx, intestine, gonad and excretory cell of Caenorhabditis elegans and are encoded by a single gene (itr-1). Journal of Molecular Biology, 294(2), 467–476. 10.1006/jmbi.1999.3229 [DOI] [PubMed] [Google Scholar]
  6. Berger AJ, Hart AC, & Kaplan JM (1998). G alphas-induced neurodegeneration in Caenorhabditis elegans. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 18(8), 2871–2880. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Brassai A, Suvanjeiev R-G, Bán E-G, & Lakatos M (2015). Role of synaptic and nonsynaptic glutamate receptors in ischaemia induced neurotoxicity. Brain Research Bulletin, 112, 1–6. 10.1016/j.brainresbull.2014.12.007 [DOI] [PubMed] [Google Scholar]
  8. Brenner S (1974). The genetics of Caenorhabditis elegans. Genetics, 77(1), 71–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bulina ME, Chudakov DM, Britanova OV, Yanushevich YG, Staroverov DB, Chepurnykh TV, Lukyanov KA (2006). A genetically encoded photosensitizer. Nature Biotechnology, 24(1), 95–99. 10.1038/nbt1175 [DOI] [PubMed] [Google Scholar]
  10. Driscoll M, & Chalfie M (1991). The mec-4 gene is a member of a family of Caenorhabditis elegans genes that can mutate to induce neuronal degeneration. Nature, 349(6310), 588–593. 10.1038/349588a0 [DOI] [PubMed] [Google Scholar]
  11. Emerit J, Edeas M, & Bricaire F (2004). Neurodegenerative diseases and oxidative stress. Biomedicine & Pharmacotherapy = Biomedecine & Pharmacotherapie, 58(1), 39–46. [DOI] [PubMed] [Google Scholar]
  12. Fruen BR, Mickelson JR, & Louis CF (1997). Dantrolene inhibition of sarcoplasmic reticulum Ca2+ release by direct and specific action at skeletal muscle ryanodine receptors. The Journal of Biological Chemistry, 272(43), 26965–26971. [DOI] [PubMed] [Google Scholar]
  13. Hall DH, Gu G, García-Añoveros J, Gong L, Chalfie M, & Driscoll M (1997). Neuropathology of degenerative cell death in Caenorhabditis elegans. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 17(3), 1033–1045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hetz C, & Mollereau B (2014). Disturbance of endoplasmic reticulum proteostasis in neurodegenerative diseases. Nature Reviews. Neuroscience, 15(4), 233–249. 10.1038/nrn3689 [DOI] [PubMed] [Google Scholar]
  15. Ijomone OM, Miah MR, Peres TV, Nwoha PU, & Aschner M (2016). Null allele mutants of trt-1, the catalytic subunit of telomerase in Caenorhabditis elegans, are less sensitive to Mn-induced toxicity and DAergic degeneration. Neurotoxicology, 57, 54–60. 10.1016/j.neuro.2016.08.016 [DOI] [PubMed] [Google Scholar]
  16. Jin Y, Jorgensen E, Hartwieg E, & Horvitz HR (1999). The Caenorhabditis elegans gene unc-25 encodes glutamic acid decarboxylase and is required for synaptic transmission but not synaptic development. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 19(2), 539–548. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Jorgensen EM (2005). GABA. WormBook: The Online Review of C. Elegans Biology, 1–13. 10.1895/wormbook.1.14.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Korswagen HC, Park JH, Ohshima Y, & Plasterk RH (1997). An activating mutation in a Caenorhabditis elegans Gs protein induces neural degeneration. Genes & Development, 11(12), 1493–1503. [DOI] [PubMed] [Google Scholar]
  19. Labuschagne CF, & Brenkman AB (2013). Current methods in quantifying ROS and oxidative damage in Caenorhabditis elegans and other model organism of aging. Ageing Research Reviews, 12(4), 918–930. 10.1016/j.arr.2013.09.003 [DOI] [PubMed] [Google Scholar]
  20. Lee D, Singaravelu G, Park B-J, & Ahnn J (2007). Differential requirement of unfolded protein response pathway for calreticulin expression in Caenorhabditis elegans. Journal of Molecular Biology, 372(2), 331–340. 10.1016/j.jmb.2007.06.071 [DOI] [PubMed] [Google Scholar]
  21. Ly K, Reid SJ, & Snell RG (2015). Rapid RNA analysis of individual Caenorhabditis elegans. MethodsX, 2, 59–63. 10.1016/j.mex.2015.02.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Maryon EB, Coronado R, & Anderson P (1996). unc-68 encodes a ryanodine receptor involved in regulating C. elegans body-wall muscle contraction. The Journal of Cell Biology, 134(4), 885–893. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Maryon EB, Saari B, & Anderson P (1998). Muscle-specific functions of ryanodine receptor channels in Caenorhabditis elegans. Journal of Cell Science, 111 (Pt 19), 2885–2895. [DOI] [PubMed] [Google Scholar]
  24. McIntire SL, Jorgensen E, Kaplan J, & Horvitz HR (1993). The GABAergic nervous system of Caenorhabditis elegans. Nature, 364(6435), 337–341. 10.1038/364337a0 [DOI] [PubMed] [Google Scholar]
  25. Nakamura K, Zuppini A, Arnaudeau S, Lynch J, Ahsan I, Krause R, Michalak M (2001). Functional specialization of calreticulin domains. The Journal of Cell Biology, 154(5), 961–972. 10.1083/jcb.200102073 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Nass R, Hall DH, Miller DM, & Blakely RD (2002). Neurotoxin-induced degeneration of dopamine neurons in Caenorhabditis elegans. Proceedings of the National Academy of Sciences of the United States of America, 99(5), 3264–3269. 10.1073/pnas.042497999 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Pollari E, Goldsteins G, Bart G, Koistinaho J, & Giniatullin R (2014). The role of oxidative stress in degeneration of the neuromuscular junction in amyotrophic lateral sclerosis. Frontiers in Cellular Neuroscience, 8, 131 10.3389/fncel.2014.00131 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Pu P, & Le W (2008). Dopamine neuron degeneration induced by MPP+ is independent of CED-4 pathway in Caenorhabditis elegans. Cell Research, 18(9), 978–981. 10.1038/cr.2008.279 [DOI] [PubMed] [Google Scholar]
  29. Richmond JE, & Jorgensen EM (1999). One GABA and two acetylcholine receptors function at the C. elegans neuromuscular junction. Nature Neuroscience, 2(9), 791–797. 10.1038/12160 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Sakube Y, Ando H, & Kagawa H (1997). An abnormal ketamine response in mutants defective in the ryanodine receptor gene ryr-1(unc-68) of Caenorhabditis elegans. Journal of Molecular Biology, 267(4), 849–864. 10.1006/jmbi.1997.0910 [DOI] [PubMed] [Google Scholar]
  31. Song SK, Karl IE, Ackerman JJ, & Hotchkiss RS (1993). Increased intracellular Ca2+: a critical link in the pathophysiology of sepsis? Proceedings of the National Academy of Sciences of the United States of America, 90(9), 3933–3937. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Soto C (2003). Unfolding the role of protein misfolding in neurodegenerative diseases. Nature Reviews. Neuroscience, 4(1), 49–60. 10.1038/nrn1007 [DOI] [PubMed] [Google Scholar]
  33. Stiernagle T (2006). Maintenance of C. elegans. WormBook: The Online Review of C. Elegans Biology, 1–11. 10.1895/wormbook.1.101.1 [DOI] [PMC free article] [PubMed]
  34. Syntichaki P, & Tavernarakis N (2004). Genetic models of mechanotransduction: the nematode Caenorhabditis elegans. Physiological Reviews, 84(4), 1097–1153. 10.1152/physrev.00043.2003 [DOI] [PubMed] [Google Scholar]
  35. Syntichaki P, Xu K, Driscoll M, & Tavernarakis N (2002). Specific aspartyl and calpain proteases are required for neurodegeneration in C. elegans. Nature, 419(6910), 939–944. 10.1038/nature01108 [DOI] [PubMed] [Google Scholar]
  36. Treinin M, & Chalfie M (1995). A mutated acetylcholine receptor subunit causes neuronal degeneration in C. elegans. Neuron, 14(4), 871–877. [DOI] [PubMed] [Google Scholar]
  37. Vaccaro A, Tauffenberger A, Aggad D, Rouleau G, Drapeau P, & Parker JA (2012). Mutant TDP-43 and FUS cause age-dependent paralysis and neurodegeneration in C. elegans. PloS One, 7(2), e31321 10.1371/journal.pone.0031321 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Waldmann R, Champigny G, Voilley N, Lauritzen I, & Lazdunski M (1996). The mammalian degenerin MDEG, an amiloride-sensitive cation channel activated by mutations causing neurodegeneration in Caenorhabditis elegans. The Journal of Biological Chemistry, 271(18), 10433–10436. [DOI] [PubMed] [Google Scholar]
  39. Williams DC, Bejjani RE, Ramirez PM, Coakley S, Kim SA, Lee H, Hammarlund M (2013). Rapid and permanent neuronal inactivation in vivo via subcellular generation of reactive oxygen with the use of KillerRed. Cell Reports, 5(2), 553–563. 10.1016/j.celrep.2013.09.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Xu K, Tavernarakis N, & Driscoll M (2001). Necrotic cell death in C. elegans requires the function of calreticulin and regulators of Ca(2+) release from the endoplasmic reticulum. Neuron, 31(6), 957–971. [DOI] [PubMed] [Google Scholar]
  41. Young LE, & Williams DC (2015). Reactive oxygen species-mediated neurodegeneration is independent of the ryanodine receptor in Caernorhabditis elegans. Journal of the South Carolina Academy of Science, 13(2). [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp1

Supplemental Figure 1. itr-1, crt-1, and unc-68 mutants are not GABA defective. Animals of the indicated genotype were analysed for GABA function by the shrinker assay. unc-47 mutants are GABA defective because they lack the vesicular GABA transporter. Data represents the percentage shrinkers +/− 95% confidence intervals.

NIHMS1520197-supplement-Supp1.jpg (776.7KB, jpg)
Supp2

Supplemental Figure 2. Single worm RT-qPCR analysis of KillerRed. (A) Representative fluorescence measured for each cycle of the indicated samples. Color represents target amplicon and symbols represent genotypes: KillerRed (red) and act-1 (green) wild-type (square), itr-1 (triangle), and crt-1 (circles). (B) Melting curve analysis demonstrating amplification of single amplification products.

NIHMS1520197-supplement-Supp2.jpg (2.1MB, jpg)
Supp3

Supplemental Video 1. Normal worm response to gentle touch on the head.

Download video file (7.3MB, mp4)
Supp4

Supplemental Video 2. Shrinker phenotype. Animals that lack GABA neuron function display a characteristic shrinker phenotype when gently touched on the head. Stimulus results in uncoordinated body-wall muscle contraction without relaxation that is manifest as the shrinker phenotype. Video is of a KillerRed animal that has been illuminated for 12 minutes.

Download video file (10.4MB, mp4)

RESOURCES