(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



A sleep-active neuron can promote survival while sleep behavior is disturbed [1]

['Inka Busack', 'Biotec', 'Technical University Dresden', 'Dresden', 'Henrik Bringmann']

Date: 2023-03

Sleep is controlled by neurons that induce behavioral quiescence and physiological restoration. It is not known, however, how sleep neurons link sleep behavior and survival. In Caenorhabditis elegans, the sleep-active RIS neuron induces sleep behavior and is required for survival of starvation and wounding. Sleep-active neurons such as RIS might hypothetically promote survival primarily by causing sleep behavior and associated conservation of energy. Alternatively, RIS might provide a survival benefit that does not depend on behavioral sleep. To probe these hypotheses, we tested how activity of the sleep-active RIS neuron in Caenorhabditis elegans controls sleep behavior and survival during larval starvation. To manipulate the activity of RIS, we expressed constitutively active potassium channel (twk-18gf and egl-23gf) or sodium channel (unc-58gf) mutant alleles in this neuron. Low levels of unc-58gf expression in RIS increased RIS calcium transients and sleep. High levels of unc-58gf expression in RIS elevated baseline calcium activity and inhibited calcium activation transients, thus locking RIS activity at a high but constant level. This manipulation caused a nearly complete loss of sleep behavior but increased survival. Long-term optogenetic activation also caused constantly elevated RIS activity and a small trend towards increased survival. Disturbing sleep by lethal blue-light stimulation also overactivated RIS, which again increased survival. FLP-11 neuropeptides were important for both, induction of sleep behavior and starvation survival, suggesting that FLP-11 might have divergent roles downstream of RIS. These results indicate that promotion of sleep behavior and survival are separable functions of RIS. These two functions may normally be coupled but can be uncoupled during conditions of strong RIS activation or when sleep behavior is impaired. Through this uncoupling, RIS can provide survival benefits under conditions when behavioral sleep is disturbed. Promoting survival in the face of impaired sleep might be a general function of sleep neurons.

Funding: This study was funded by the European Research Council (ERC Starting Grant SLEEPCONTROL)(HB), the Max Planck Institute for Biophysical Chemistry (HB), the Philipps Universität Marburg (HB), and the TU Dresden (HB). The Article Processing Charge (APC) was funded by the joint publication fund of the TU Dresden, the Medical Faculty Carl Gustav Carus, and the SLUB Dresden (HB). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Copyright: © 2023 Busack, Bringmann. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

RIS hence appears to be able to promote survival independently of sleep behavior, indicating that promotion of sleep behavior and survival are separable functions of RIS that are normally coupled during sleep but that can be uncoupled. Uncoupling appears to occur during conditions of stimulation that disturb sleep and increase RIS activity. This suggests that an important function of RIS overactivation is to promote survival not only during normal sleep, but also and in particular when sleep behavior is disturbed. Sleep-active neurons might generally function to promote survival both during normal sleep and even more so when sleep behavior is disturbed.

In order to find out how RIS links sleep and survival, we manipulated RIS activity through transgenic expression of ion channel mutants in RIS. By strongly expressing unc-58gf in RIS, we achieved an overall increase of RIS activity that was caused by constantly elevated baseline calcium activity in RIS. RIS appeared to be locked at this higher activity level, as it was not able to activate further upon stimulation. Survival was increased in this strain, but sleep behavior was almost completely lacking. The sleep loss phenotype included a lack of dampening of the activity of interneurons, sensory neurons and muscles that is typical for normal sleep. Thus, strong expression of RIS led to an apparent uncoupling of sleep behavior and survival. Long-term optogenetic activation also generated constantly increased calcium activity and a small trend towards increased survival. Stimulation with blue light disturbed sleep and increased RIS activation, which in turn extended survival. Finally, we find that RIS promotes sleep behavior and promotes survival through FLP-11 neuropeptides.

Little is known about how sleep-active neurons such as RIS link sleep behavior and survival. Different hypotheses are conceivable. For example, sleep-active neurons might promote survival by reducing behavioral activities during sleep, thus conserving and allocating energy or optimizing behavior [ 27 , 28 ]. The close association of sleep and physiological benefits has suggested a causal relationship. For example, memory and information processing are thought to be improved in the human brain during sleep when information inputs are blocked [ 29 , 30 ], and energy is conserved by reduced behavioral activity [ 28 ]. A less explored additional hypothesis is that sleep-active neurons provide benefits during sleep, but independently of sleep behavior. According to this hypothesis, sleep-active neurons have two parallel functions. The first is to induce sleep behavior, and the second is to promote survival. This hypothesis implies that physiological benefits of sleep-active neuron functions can be exerted also when sleep behavior is impaired.

Sleep is found in most animals ranging from cnidarians to humans. Central to the induction of sleep in all species are so-called sleep-active neurons that activate to inhibit wakefulness behavioral circuits and thus cause sleep behavior [ 8 , 9 ]. C. elegans has a small and invariant nervous system and sleep is controlled by relatively few neurons. RIS is a GABAergic and peptidergic interneuron that activates during sleep, and its ablation virtually abolishes sleep behavior during molting of developing larvae, in arrested larvae, and in adults following cellular stress [ 10 – 21 ]. Optogenetic manipulations showed that acute activation of RIS inhibits wakefulness activity such as feeding and locomotion [ 10 , 15 ], whereas inactivation of RIS inhibits sleep [ 12 , 18 , 22 , 23 ]. RIS is important for survival during starvation and after wounding [ 12 , 18 , 24 ]. This neuron counteracts aging phenotype progression during starvation-induced arrest [ 12 ]. It supports the protective transcriptional response that underlies the arrest, and promotes the activity of the FOXO transcription factor DAF-16, which is required for survival [ 24 – 26 ]. Thus, RIS is a sleep-active neuron that is crucial for C. elegans sleep and stress survival.

Sleep is defined by behavioral inactivity and reduced responsiveness [ 1 ]. It is a physiological state that conserves and allocates energy, promotes restorative processes, and is important for brain and cognitive functions [ 2 – 4 ]. Sleep disorders and deviations from the normal sleep amount are widespread in modern societies and pose an unresolved medical and economical problem [ 5 – 7 ].

Results

Generation of a strain set for controlling RIS activity We set out to control RIS activity to study sleep and survival during L1 arrest. For this study, we generated strains that express constitutively active ion channels with different ion permeability and expression levels in RIS. To constitutively inactivate RIS, we expressed potassium channels that carry mutations that cause potassium leak currents using the flp-11 promotor that is strong and highly specific for RIS [11,17,31]. We used two different egl-23 channel mutations to achieve different strengths of inactivation of RIS (RIS::egl-23gf(weak) expresses egl-23(L229N) in RIS, RIS::egl-23gf(strong) expresses egl-23(A383V) in RIS) [32]. To create a strong tool for RIS inhibition, we expressed a twk-18 mutant allele that is known to cause a very strong potassium current (RIS::twk-18gf expresses twk-18(e1913) in RIS) [33]. In all three strains, the channel genes were knocked into the endogenous flp-11 locus behind the flp-11 gene to generate operons that express both flp-11 and the mutant ion channel. All fusion proteins localized to the cell membrane of RIS (Fig 1A–1D). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. Strains expressing ion channel mutants in RIS. A) Scheme of the genetic design of the different RIS potassium channel mutant transgenes. B-D) Localization and expression of mKate2, which is translationally fused to the ion channels, in the different inactivation strains. E) Schematic representation of the genetic design of the RIS sodium channel mutant transgenes. F) Quantification of the expression levels of UNC-58gf::mKate2 in the two RIS overactivation strains. ***p<0.001, Welch test. G) Localization and expression of the strong overactivation strain (RIS::unc-58gf(strong)). H) Localization and expression of the weak overactivation strain (RIS::unc-58gf(weak)). https://doi.org/10.1371/journal.pgen.1010665.g001 To constitutively over activate RIS, we used an unusual K2P channel mutant, unc-58(L428F), that causes a sodium leak current [32] (Thomas Boulin, personal communication). We expressed the ion channel mutant in RIS tagged with mKate2 at either a lower or a higher level in RIS (Fig 1E). The different expression levels were achieved by generating two different expression loci for the channel in the genome. To achieve a strong expression of unc-58(L428F) (RIS::unc-58gf(strong)), we turned the endogenous flp-11 locus into an operon as above. To achieve a weaker expression of the channel (RIS::unc-58gf(weak)), we expressed mKate2-tagged unc-58(L428F) from a single-copy knock-in locus on chromosome III [34], which contained only a minimal flp-11 promoter and 3’ UTR [11] (Fig 1E–1H). Fluorescent imaging after two days of L1 arrest showed that the stronger strain expressed at about double the level of the weaker strain and in both transgenes, the channel was expressed prominently in RIS and was targeted to the cell membrane (Fig 1E–1H and S1 Fig). Even after 21 days of arrest, mKate2-tagged unc-58(L428F) expression could be detected in the strong strain indicating that the sodium channel mutant is properly folded and targeted to the plasma membrane across the entire arrest (S1E Fig).

Ion channel mutant expression in RIS controls baseline RIS activity How are RIS activity, sleep behavior and survival linked? To shed light on this question, we imaged RIS calcium activity in these mutants. [10]. We used microfluidic chambers to keep and image the larvae. This method allowed for a quantification of RIS baseline activity, RIS calcium activation transients, and sleep behavior [35,37]. We were able to measure RIS activity for the first 12 days of starvation, and found that the effects of the channel transgenes were consistent across the entire time of starvation that we analyzed (Fig 3A for two-day starvation, 3B for all time points). We first analyzed the baseline activity of RIS in the absence of calcium transients. The mean RIS baseline calcium activity levels were reduced by the strongly inactivating transgenes twk-18gf and egl-23gf(strong). Baseline activity levels were modestly increased by unc-58gf(weak) and strongly increased by unc-58gf(strong). The magnitude of the baseline calcium signal alterations of the different transgenes were thus in accordance with the expression levels and the expected strengths of the leak currents (Fig 1F). These data demonstrate that transgenic unc-58gf can be used as a tool for increasing calcium activity in a neuron that does not normally express unc-58 [17,31,32]. We used these tools to create a set of strains that allows for probing associations of baseline activity levels and survival as well as sleep. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. Ion channel mutant expression in RIS controls baseline RIS activity. A) Ion channel expression in RIS results in a dose-response strain set with varying levels of RIS baseline activity. n.s. p>0.05, **p<0.001, ***p<0.001, Welch test with FDR correction for multiple testing. B) Long term RIS baseline activity for RIS activity strains from day 1 to day 12 of starvation. https://doi.org/10.1371/journal.pgen.1010665.g003

RIS::unc-58gf(strong) inhibits sleep behavior during L1 arrest We next quantified motion quiescence as a readout for sleep and tracked mobility behavior during L1 arrest inside microfluidic chambers. We used GCaMP fluorescence to quantify RIS activation transients across the entire measurement [10]. Consistent with previous reports [10,12,20], wild-type RIS showed activation transients over a constant baseline, and calcium transients were accompanied by phases of behavioral quiescence (Fig 4A–4G). Calcium transients were completely abolished in both the strongest deactivation strains (Fig 4A and 4B) as well as in the strong activation strain (4F). In the mild deactivation strain calcium transients were still visible and correlated with behavioral quiescence, but were decreased in magnitude (Fig 4C–4G and S3A Fig). In RIS::unc-58gf(weak), RIS calcium transient activity and bouts of behavioral activity were increased (Fig 4E). These results suggest that RIS activation transients are required for sleep induction and that strong inactivation prevents these transients. Mobility quiescence decreased with inactivation and was increased with modest activation as expected. We anticipated that the elevated RIS calcium baseline of RIS::unc-58gf(strong) might be associated with a further increase in sleep behavior compared with the weak activation. Mobility quiescence was, however, completely abolished by strong RIS activation in most of the animals (Fig 4H). These findings were additionally confirmed for a longer period of starvation up to day 12 (Fig 4I). The loss of quiescence of the strong RIS activation strain was similar to, but not as strong as in aptf-1(-) (S2B Fig). The sleep amounts for each strain are consistent for several days in L1 arrest (Fig 4I). In summary, baseline RIS activity correlates with behavioral quiescence over a wide range of conditions, except for the RIS::unc-58gf(strong) transgene, which appears to cause a lack of RIS calcium transients and sleep behavior. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. Both, strong inactivation of RIS as well as RIS::unc-58gf(strong) prevent sleep behavior during L1 arrest. A-F) Ion channel expression in RIS controls calcium activity and sleep behavior. Mobility quiescence bouts (grey shade) correlate with RIS transients. Movement speed was calculated by tracking the position of the head neuron RIS. A) RIS::twk-18gf and B) RIS::egl-23gf(strong) larvae do not have detectable mobility quiescence bouts or RIS transients. C) RIS::egl-23gf(weak) has reduced mobility quiescence bouts. D) Wild-type RIS activation transients and immobility bouts. E) RIS::unc-58gf(weak) spends more time in mobility quiescence bouts. F) RIS::unc-58gf(strong) has no detectable mobility quiescence bouts and no RIS transients but a high constant activity. G) RIS activation and immobility are correlated only for the strains that also show sleeping behavior. Alignment of RIS activity and speed to mobility quiescence bout onset. **p<0.01, ***p<0.001, Wilcoxon signed rank test. H) Mobility quiescence is a function of RIS activity levels n.s. p>0.05, **p<0.001, ***p<0.001, Welch test with FDR correction for multiple testing. I) Long-term sleep fraction for RIS activity strains from day 1 to day 12 of starvation. https://doi.org/10.1371/journal.pgen.1010665.g004

RIS::unc-58gf impairs sleep behavior during larval development and following cellular stress in the adult C. elegans sleeps during various stages and conditions. Previous studies showed that cellular stress increases adult sleep [17,38–41]. During development, sleep is coupled to the molting cycle and occurs during a period called lethargus, which coincides with formation of a new cuticle [10,42,43]. To test whether constant RIS baseline activation is also associated with a loss of behavioral quiescence during stress-induced sleep and during lethargus sleep, we tested whether RIS::unc-58gf affects sleep behavior during L1 development as well as following heat stress. For testing for stress-induced sleep, we kept young adult worms inside microfluidic devices and first measured baseline behavioral activity at 22°C, and then exposed the animals to heat stress at 37°C for 20 minutes. We then followed the behavior again after the end of the heat shock [17]. The wild type immobilized during the heat shock and displayed an increase in behavioral quiescence following the heat shock. The quiescence response was attenuated in all strains in which RIS was manipulated. Both unc-58gf expressing strains showed decreased behavioral quiescence. Thus, a reduction of behavioral sleep can also be caused by RIS::unc-58gf(weak) (Fig 5A–5G). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 5. RIS::unc-58gf impairs sleep following cellular stress and during the L1 molt in developing larvae. A-F) Sleep fraction of worms of different RIS activity strains during the heat shock experiment. Animals were recorded continuously throughout the experiment. First, a 1h baseline was recorded at 22°C, then a 20min heat shock period of 37°C (in red shade) was applied, after the heat shock, the temperature was lowered to at 22°C again to monitor the response to the heat shock. G) Quantification of time spent in sleep 20 min after the heat shock. Wild-type worms sleep on average 38.3%. All other RIS manipulations reduced quiescence. RIS::unc-58gf(strong) only sleeps on average 6.5% of the time. **p<0.001, ***p<0.001, Welch test with FDR correction for multiple testing. H) RIS::unc-58gf(strong) reduces lethargus sleep compared to wild type. ***p<0.001, Welch test. https://doi.org/10.1371/journal.pgen.1010665.g005 To test for sleep that occurs during development, we quantified behavioral quiescence of RIS::unc-58gf(strong) during lethargus in fed L1 larvae. Behavioral quiescence of RIS::unc-58gf(strong) was significantly reduced compared with the wild type. The loss of sleep behavior in lethargus was, however, not as strong as during L1 arrest or following stress-induced sleep (Fig 5H). Hence, in all the stages and conditions that we tested, sleep behavior was reduced by RIS::unc-58gf(strong). We also tested whether the lack of sleep behavior in RIS::unc-58gf(strong) during L1 arrest causes a compensatory increase in sleep when the animals resume development. For this experiment, we starved larvae for 12 days and then provided them with food, which allowed for a resumption of development. We then imaged sleep behavior during L1 lethargus in these recovered worms. There was no increase of sleep behavior in recovered RIS::unc-58gf(strong) larvae compared with wild type (S4 Fig). Thus there was no indication for rebound sleep occurring at the subsequent lethargus stage in RIS::unc-58gf(strong).

RIS::unc-58gf(strong) promotes muscle activity An absence of voluntary movement, reduced muscle contraction and a relaxed body posture are hallmarks of sleep across species including C. elegans [43,51–53]. If RIS::unc-58gf(strong) lacks deep sleep bouts then this should be reflected by the absence of extended periods of movement quiescence that coincide with reduced muscle activity. During wakefulness, C. elegans shows locomotion behavior and feeds by contractions of the pharynx muscle [54,55]. Locomotion and pharyngeal pumping stops during sleep and the animals assume a relaxed body posture [10,43,51,53]. Underlying these effects, RIS inhibits both locomotion and pharyngeal pumping [10,12,15,56,57]. To test the hypothesis that RIS::unc-58gf(strong) lacks long phases of muscle relaxation that are typical for sleep, we first imaged arrested larvae inside microfluidic chambers and quantified more precisely movement activity of the body and the pharynx. To quantify mobility behavior of the head and tail regions, we tracked the positions of neurons that are located in these body parts. To calculate movement speed of the head, we tracked the position of the RIM neurons. For tracking the tail we used the position of the PLM neurons [58]. The tracking data revealed that in RIS::twk-18gf, both head and tail movement was similar to the wild type during wakefulness. During the few quiescence phases of RIS::twk-18gf, movement speed was reduced less than in the wild type, indicating that the quiescence phases that remain in RIS::twk-18gf are less pronounces compared with the wild type. In RIS::unc-58gf(strong), head movement was decreased compared to wild-type worms, whereas there was no difference in tail movement (Fig 8A and 8B). RIS::unc-58gf(strong) worms were uncoordinated and the body was more curved (S7 Fig and S7 and S8 Movies), which might explain the difference of head and tail movement speeds. In the few locomotion quiescence phases of RIS::unc-58gf(strong), there appeared to be less of a reduction of movement speed. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 8. RIS::unc-58gf(strong) causes increased muscle activity. A) We calculated the movement speed of the RIM head neuron to measure head movement speed. RIM speed was reduced in RIS::unc-58gf(strong). *p<0.05, **p<0.01, ***p < 0.001, Welch test for comparison between strains, Wilcoxon signed rank test for comparison between the same strain quiescent and mobile bouts. B) We calculated the movement speed of the PLM tail neuron to measure tail movement speed. PLM movement speed was not significantly changed in mobile bouts in RIS::unc-58gf(strong) and increased in detected quiescent bouts. *p<0.05, **p<0.01, ***p < 0.001, Welch test for comparison between strains, Wilcoxon signed rank test for comparison between the same strain quiescent and mobile bouts. C-E) Sample traces of muscle activity for RIS::twk-18gf and RIS::unc-58gf(strong). F) RIS::unc-58gf(strong) causes increased muscle activity in quiescence as well as mobility bouts. *p<0.05, **p<0.01, ***p < 0.001, Welch test for comparison between strains, Wilcoxon signed rank test for comparison between the same strain quiescent and mobility bouts. G) A sleep bout alignment reveals that RIS::unc-58gf(strong) has increased muscle activity. **p<0.01, ***p < 0.001, Wilcoxon signed rank test. H) RIS::twk-18gf and RIS::unc-58gf(strong) reduced phases of muscle inactivity. *p<0.05, Welch test. https://doi.org/10.1371/journal.pgen.1010665.g008 We next imaged our transgenic strains in the arrested first larval stage for 1 min with a framerate of 10Hz and manually scored pumping rate. This analysis revealed no significant difference in pharyngeal contraction rates for all strains compared to wild type (S8 Fig). To test how RIS::unc-58gf(strong) and RIS::twk-18gf affect muscle activity, we measured and quantified muscle GCaMP [51]. In the wild type, muscle calcium activity was high during wakefulness, and strongly reduced during sleep. In RIS::twk-18gf, muscle calcium was almost constantly elevated and resembled the activity levels of the wild type. RIS::unc-58gf(strong) worms showed a general increase in muscle calcium activity (Fig 8C–8F). The increased muscle activity is consistent with and might contribute to the slightly uncoordinated movement of this mutant. Phases of behavioral and muscle activity dampening occurred less often, and if they occurred, then these dampened states appeared to be less deep. (Fig 8G and 8H).

In RIS::unc-58gf(strong), RIS calcium activity cannot be easily elevated further by optogenetic stimulation Calcium imaging suggested that the RIS baseline in RIS::unc-58gf(strong) is substantially elevated and that calcium transients are lacking. We aimed to better understand the nature of these physiological changes in RIS that are caused by expression of RIS::unc-58gf(strong). Based on the known characteristics of the unc-58gf mutant channel, the expression of unc-58gf might plausibly cause a constant depolarization of the membrane of RIS [32]. Depolarization in turn should lead to the activation of voltage-gated calcium channels, and RIS expresses the voltage-gated calcium channels egl-19 and unc-2 [31]. The conductivity of voltage-gated calcium channels is increased by depolarization with an optimum in the depolarized range. Further depolarization causes inhibition and long-term stimulation reduces conductivity by desensitization [59–64]. RIS depolarization in RIS::unc-58gf(strong) could thus perhaps be increased to allow for elevated RIS baseline calcium activity, but might also inhibit further activation of calcium channel activity. We tested the ability of RIS to activate by optogenetically stimulating RIS in RIS::unc-58gf(strong) inside microfluidic chambers and monitored the calcium activity of this neuron as well as the motion behavior of the larvae. For this experiment, we used the red-shifted channelrhodopsin variant ReaChR in combination with GCaMP3 [12,22,56]. In the wild type, optogenetic activation caused a strong increase in intracellular calcium and a strong induction of motion quiescence, which is consistent with previous results [12,15,22,56]. By contrast, optogenetic activation of RIS in RIS::unc-58gf(strong) did not increase RIS activity further and did not induce behavioral quiescence (Fig 9A and 9B). This result is consistent with the view that RIS::unc-58gf(strong) causes a heightened level of RIS activity that cannot be easily increased further with optogenetic activation. Hence, in RIS::unc-58gf(strong), RIS appears to be locked in a state of heightened activity while further activation is inhibited. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 9. RIS::unc-58gf(strong) is refractory to optogenetic activation of RIS. A) Optogenetic activation of RIS during the stimulation time (minute 1–2) leads to an increase of RIS activity and reduced mobility in wild-type worms but it does not affect the strong RIS activation strain. RIS speed was measured. A statistical comparison was made between baseline (0-1min) and stimulation (1-2min). n.s. p>0.05, **p<0.01, ***p<0.001, Wilcoxon signed rank test. B) Control for the optogenetic experiment. Without ATR treatment, neither the wild type nor the strong RIS activation strain, shows a response to green light. N.s. p>0.05, Wilcoxon signed rank test. https://doi.org/10.1371/journal.pgen.1010665.g009

Long-term optogenetic activation of RIS increases RIS baseline activity and inhibits activation transients RIS::unc-58gf(strong) appears to cause increased RIS activity that cannot easily be increased further. This could hypothetically be explained by a high level of depolarization in RIS::unc-58gf(strong) that causes both RIS baseline elevation as well as inhibition of RIS calcium transients. To test the idea that constant depolarization elevates baseline RIS activity while inhibiting activation transients, we used long-term optogenetic activation. We used again ReaChR to depolarize the membrane and followed the activity of RIS. Optogenetic tools like ReaChR typically display a brief photocurrent peak activity within the first second of light exposure and then plateau at a lower photocurrent level for an extended period of time [65]. We previously used ReaChR to establish long-term optogenetic manipulations in C. elegans [23]. To test for the activation of RIS during prolonged optogenetic activation, we activated ReaChR in RIS with orange light for 11h and followed the activity of RIS using GCaMP and measured mobility of the larvae that were kept inside microfluidic chambers. During the first hour of optogenetic stimulation of RIS, there was a strong increase of calcium activity as well as of behavioral quiescence (Fig 10A–10C). After this first peak period, the ongoing optogenetic stimulation led to lower but more constant increase of RIS baseline activation (which was, however, weaker than the activation observed in RIS::unc-58gf(strong) average calcium baseline activity increase in RIS::unc-58gf(strong) compared with wild type was 1.55, and average calcium activity baseline increase from 2-11h of activation in RIS::ReaChR compared with wild type was 1.25) (Fig 10D and 10E). Long-term optogenetic activation caused reduced calcium transients, and the residual calcium transients did not correlate with phases of immobility. This indicates that long-term optogenetic activation has an inhibitory effect on RIS calcium transients (Fig 10F). The optogenetic long-term RIS activation did not cause a measurable decrease in behavioral quiescence (Fig 10C). Thus, long-term optogenetic stimulation leads to a constantly increased RIS baseline activity and an inhibition of RIS calcium transients that is similar to, but weaker than, the activation observed in RIS::unc-58gf(strong). This difference in effect strength might be explained by the difference in ion conductivity of these channels [32,65]. These data indicate that long-term optogenetic stimulation can lead to a constant increase in baseline RIS calcium activity at a level that is above the baseline activity without optogenetic stimulation, yet lower than the maximum possible optogenetic RIS activation peak intensity. These optogenetic data are consistent with the idea that increased depolarization of RIS leads to a constant elevation of baseline calcium activity and impairs calcium transients. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 10. Long-term optogenetic activation of RIS increases overall RIS calcium activity and promotes survival. A) Sample trace of control larvae without retinal in which RIS cannot be activated optogenetically. RIS speed was measured. B) Long-term optogenetic RIS activation by orange light in the presence of retinal. C) Long-term activation of RIS by ReaChR leads to an initial strong increase in calcium activity in RIS and simultaneous mobility quiescence followed by a mild RIS activity increase yet no increase in movement speed or movement quiescence. D) Comparison of RIS activity level changes caused by RIS::unc-58gf(strong) and by optogenetic activation of RIS. The data was normalized to their respective controls. E) Long-term optogenetic activation of RIS significantly increases RIS baseline. *p<0.05 Welch test. F) Transients in RIS are significantly reduced in magnitude and don’t correlate with quiescence. **p<0.01 Welch test. G) Long-term optogenetic activation of RIS inside microfluidic chambers leads to a small increase in survival (p = 0.09 Logrank test, p = 0.02 Fisher’s Exact Test, the average of all individual worms of all three technical replicates was averaged, n = 92 for RIS::ReaChR(+ATR), n = 98 RIS::ReaChR(-ATR), n = 124 wild type (+ATR), n = 108 wild type (-ATR)). https://doi.org/10.1371/journal.pgen.1010665.g010

Long-term optogenetic activation of RIS causes a small increase in survival Our data using RIS::unc-58gf(strong) indicated that extended survival of L1 arrest is associated with increased overall RIS activity. To confirm this result with an independent approach, we used RIS::ReaChR for long-term activation of RIS using the OptoGenBox system [23]. We kept L1 larvae inside microchambers and constantly stimulated the entire chambers with orange light. The optogenetic stimulation was only paused briefly once every day to monitor the survival of the larvae. The stimulation protocol was applied until all larvae had died. The functionality of ReaChR in C. elegans requires the addition of all-trans retinal (ATR). Addition of ATR shortened the survival (Fig 10G). We hence compared the survival of N2 exposed to ATR to worms expressing ReaChR that were also exposed to ATR. By this comparison, optogenetic long-term RIS activation caused a small trend for increased survival that was statistically significant according to Fisher’s Exact Test but not significant according to the Logrank test (p = 0.02 Fisher’s Exact Test, p = 0.09 Logrank, Fig 10G). While this result borders statistical significance, it is consistent with the view that activation of RIS promotes survival. The small effect size is consistent with the small RIS activity increase caused by extended optogenetic stimulation. Thus, this experiment supports the view that RIS activation promotes survival of L1 arrest.

Lethal blue-light stimulation activates RIS, and RIS supports survival of lethal blue light Disturbing sleep by sensory stimulation leads to arousal that forces wakefulness and prevents the dampening of underlying neuronal circuits [12,20,22,43,66,67]. Sleep disturbance or deprivation can cause increased RIS activation [22,24,68]. The high level of RIS activity in turn could be protective under such conditions and promote survival [24]. Thus, RIS::unc-58gf(strong) might cause a state that is similar to a state caused by prolonged sleep deprivation or disturbed sleep. In both, disturbed sleep and RIS::unc-58gf(strong), the activity of wakefulness circuits is not dampened yet coincides with elevated RIS activity. To further illustrate the idea that sensory stimulation activates RIS while impairing sleep, we stimulated worms with a lethal intensity of blue light. We used arrested L1 larvae after 48h in the absence of food, a condition during which the animals normally show regular RIS activation transients and sleeping behavior [12]. We kept L1 larvae in microchambers and repeatedly applied pulses of blue light while monitoring RIS activity and motion behavior. For comparison, we used control animals that were exposed to shorter pulses of blue light of only one tenth the length, thus still allowing for calcium imaging. Control larvae displayed sleeping behavior detected by motion quiescence and RIS calcium activity during about a quarter of the time (Fig 11A and 11B). Strong blue-light stimulation caused an increase in RIS calcium activity compared with control animals (after 1h of strong blue-light stimulation, the activity of RIS was approximately 5.6-fold higher compared with worms that were stimulated with a lower dose). Many of the calcium transients were very short, however, and did not coincide with increased motion quiescence (Fig 11B and 11C). Hence, the stronger blue-light stimulus increased RIS activity but apparently did not increase sleep behavior. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 11. RIS activates upon lethal blue-light stimulation and extends survival. A) Sample trace for a control worm stimulated by a short blue light stimulus. RIS speed was measured. B) Sample traces of wild-type worms upon stimulation with a long blue-light stimulus. The grey shaded area presents detected time in motion quiescence. C) The strong stimulus increased normalized RIS activity over baseline compared to control worms. ***p<0.001, Welch test. D) Wild-type worms have a survival advantage over RIS-inhibited worms upon exposure to the lethal blue-light stimulus. *p<0.05, Welch test. https://doi.org/10.1371/journal.pgen.1010665.g011 To probe for the role of RIS in survival of lethal blue-light stimulation, we compared wild-type and RIS::twk-18gf animals, in which RIS activated less under blue-light stimulation (Fig 11C). We scored the time point at which the animals became terminally immobile, which presents an established proxy to score for death [69]. Control animals with functional RIS died around 7-8h after strong blue-light stimulus onset. RIS-impaired animals already died at around 6h after stimulus onset. The presence of functional RIS hence extended survival by 15% (Fig 11D). Thus, stimulation by a lethal blue-light stimulus can increase RIS calcium activity without increasing RIS-dependent sleep behavior. While all animals were eventually killed by the blue-light stimulus, RIS activity slightly extended survival. The blue-light experiment thus supports the idea that RIS is a protective neuron that activates upon stressful stimulation. As in RIS::ReaChR and during long-term optogenetic stimulation of RIS, increased RIS activity was not associated with increased sleep behavior.

Flp-11 is a key neuropeptide gene required for quiescence behavior in L1 arrest What are the key transmitters that RIS uses to control sleep during L1 arrest? GABA is expressed in RIS but has not been shown to play a substantial role in promoting sleep [10,15]. FLP-11 is a group of neuropeptides of RIS that present the major transmitters of RIS required for sleep induction during development [11]. Expression from the flp-11 promoter depends on neuronal activity levels [70], and is increased in RIS::unc-58gf(strong) (S3B Fig), suggesting that FLP-11 might also be a major transmitter for sleep induction during L1 arrest. To directly test the role of FLP-11 in quiescence during L1 arrest, we measured the effects of flp-11 deletion (flp-11(-)) using microfluidic chamber imaging [11]. flp-11(-) strongly reduced quiescence behavior (Fig 12A). RIS::unc-58gf(strong) still displayed a small fraction of quiescence behavior during L1 arrest. To test whether this residual quiescence was caused by flp-11, we deleted the entire flp-11 gene in the RIS::unc-58gf(strong) operon locus (S9A Fig) and quantified quiescence behavior. The sleep loss of RIS::unc-58gf(strong) was reduced further by flp-11(-), suggesting that the residual quiescence behavior in RIS::unc-58gf indeed depends on flp-11 (Fig 12A). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 12. FLP-11 is required for sleep during L1 arrest. A) Deletion of the neuropeptide gene flp-11 causes loss of sleep. Immobility in RIS::unc-58gf(strong) is increased when flp-11 is deleted. ***p<0.001, Welch test with FDR correction for multiple testing. B) Loss of the FLP-11 neuropeptides leads to strongly reduced survival (three-parameter logistic fit, see also S1 Table, Fisher’s Exact Test was conducted on day 18) for comparisons when flp-11(-) was the shortest lived condition, day 20) when RIS::unc-58gf(strong), flp-11(-) was the shortest lived condition or day 24) when wild type was the shortest lived condition. The p-values were FDR corrected with Benjamini-Hochberg procedure with a 5% false discovery rate. The plot includes data from three replicates. ***p<0.001. https://doi.org/10.1371/journal.pgen.1010665.g012

[END]
---
[1] Url: https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1010665

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.