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



Ring neurons in the Drosophila central complex act as a rheostat for sensory modulation of aging [1]

['Christi M. Gendron', 'Department Of Molecular', 'Integrative Physiology', 'Geriatrics Center', 'University Of Michigan', 'Ann Arbor', 'Michigan', 'United States Of America', 'Tuhin S. Chakraborty', 'Cathryn Duran']

Date: 2023-06

Sensory perception modulates aging, yet we know little about how. An understanding of the neuronal mechanisms through which animals orchestrate biological responses to relevant sensory inputs would provide insight into the control systems that may be important for modulating lifespan. Here, we provide new awareness into how the perception of dead conspecifics, or death perception, which elicits behavioral and physiological effects in many different species, affects lifespan in the fruit fly, Drosophila melanogaster. Previous work demonstrated that cohousing Drosophila with dead conspecifics decreases fat stores, reduces starvation resistance, and accelerates aging in a manner that requires both sight and the serotonin receptor 5-HT2A. In this manuscript, we demonstrate that a discrete, 5-HT2A-expressing neural population in the ellipsoid body (EB) of the Drosophila central complex, identified as R2/R4 neurons, acts as a rheostat and plays an important role in transducing sensory information about the presence of dead individuals to modulate lifespan. Expression of the insulin-responsive transcription factor foxo in R2/R4 neurons and insulin-like peptides dilp3 and dilp5, but not dilp2, are required, with the latter likely altered in median neurosecretory cells (MNCs) after R2/R4 neuronal activation. These data generate new insights into the neural underpinnings of how perceptive events may impact aging and physiology across taxa.

Funding: o This work is supported by National Institutes of Health, National Institute on Aging RO1AG051649 (to SDP), National Institutes of Health, National Institute on Aging RO1AG030593 (to SDP), National Institutes of Health, National Institute on Aging RO1AG063371 (to SDP), and the Glenn Medical Foundation (to SDP). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Herein, we identify neural substrates and circuits that transduce sensory information from the perception of dead individuals to meaningful changes in lifespan. We identified a small subset of 5-HT2A-expressing neurons in the ellipsoid body (EB) of Drosophila (specifically R2/R4 neurons), a center of sensory information integration and motor coordination, that were both required and sufficient for the lifespan effects caused by death exposure. The transcription factor associated with insulin-signaling, foxo, was required in these neurons. We also discovered that the Drosophila insulin-like peptides (dilp) 3 and 5, but not dilp2, were required to mediate lifespan effects due to death perception. Dilp regulation appeared after changes in R2/R4 neuron activity, however, suggesting that these peptides do not directly impact Foxo activity in these neurons. By contributing to an understanding of the physiological effects of death exposure and the biological mechanism(s) that drive them, our results may provide insight for treating individuals who are routinely exposed to stressful situations surrounding death, including active combat soldiers and first responders.

With this in mind, we chose to investigate the neural circuits and central signaling processes that underlie the physiological effects associated with the perception of dead conspecifics, or death perception, in Drosophila. It has been shown that when live flies see, and to a lesser extent smell, an excess of dead flies in their environment, they become aversive to other flies, and they exhibit significant acute and chronic physiological changes, including rapid decreases in stored fat and starvation resistance as well as chronically increased mortality [ 2 ]. These effects are reminiscent of the behavioral and physiologic changes seen in other species exposed to similar circumstances, such as necrophoresis in eusocial insects, vocalization and corpse inspection in elephants, or increased glucocorticoid levels detected in nonhuman primates [ 3 ], suggesting similarity in the effector processes that are recruited in response to this perceptive event. Indeed, in Drosophila, the effects of death perception reportedly involve the highly conserved biogenic amine, serotonin, as well as neural signaling through one of its receptors, 5-HT2A. It remained unknown which 5-HT2A-expressing neurons, or the molecules expressed by these cells, are required to modulate the global physiologic and associated lifespan changes that are caused by death perception. Understanding the neural circuits through which death perception impacts these phenotypes may inform future work directed toward understanding the consequences associated with this, and perhaps other sensory experiences in individuals, including humans, and may provide insight into how specific neural states impact behavior and physiology.

A range of sensory processes, from the relatively simple (e.g., water, taste, or smell of food) to the more complex (e.g., the detection of the opposite sex), influences how long an organism lives as well as its health and vitality throughout life. Most of the research in this area has used simple model systems, such as Drosophila melanogaster or Caenorhabditis elegans, and has focused on identifying environmental cues with potent effects on aging or the sensory receptors and neurons that are involved in their initial detection [ 1 ]. Because peripheral sensory systems have been tuned by species-specific ecological conditions and evolutionary histories, these achievements are unlikely to provide direct insight into similar process in other systems, including humans. Nevertheless, they provide the necessary foundation for considering the more difficult task of understanding which neurons and associated neural states are influenced by sensory perception, how they relate to one another, and how they mechanistically relate to aging and physiology.

To test whether R4d neuron activation itself is sufficient to alter Dilp protein abundances in MNC, we used optogenetic techniques to activate R4d neurons for different lengths of time in the absence of dead conspecifics and quantified Dilp3 antibody staining. Flies expressing UAS-Chrimson (R4d-GAL4 x UAS-Chrimson) were stimulated with red light for either 2 days or 2 weeks, after which time Dilp3 protein abundance in MNCs was quantified. We observed increased anti-Dilp3 staining when R4d neurons were activated for 2 weeks but not for 2 days ( Fig 4G and 4H ). Changes in Dilp3 abundance were not due to red light exposure per se because there were no changes observed in control flies (R4d-GAL4 x w 1118 ) that were exposed to a similar intensity and duration ( Fig 4I ). Together, these data indicate death perception rapidly activates R4d neurons, which, after a period of chronic activity, stimulate Dilp3 expression in MNC to modulate aging.

Alternatively, we considered the possibility that Dilp3 and Dilp5 abundances may be altered by changes in ring neuron activity to enact changes in lifespan. To evaluate this notion, we took a closer look at when molecular and phenotypic changes occur. Our data (shown in Fig 4A ) demonstrated that dilp-expression was altered when the flies were exposed to dead for 2 weeks. We also knew that R2/R4 neurons appeared to be activated on a much shorter timescale, within 2 days (e.g., Fig 2A and 2B ). We therefore asked whether dilp3 and dilp5 mRNA or Dilp3 protein abundance were altered after only 2 days of exposure. We observed that dilp3 and dilp5 mRNA abundances, as well as Dilp3 protein abundance, were unchanged following 2 days of dead exposure compared to the unexposed group ( S5 Fig ). These data suggest that alterations in dilp3 and dilp5 expression occur after modulation of R4 activity and that they may result from changes in the activity of these cells.

It may be that Dilp3 and Dilp5 act on R4 neurons in response to dead exposure, thereby altering foxo activity in these neurons that subsequently influences lifespan. R4d neurons putatively express the only insulin receptor identified in flies (InR; [ 16 ]). We therefore examined whether the expression of a dominant negative form of InR specifically in R4d neurons (R4d-GAL4 x InR K1409A ) inhibited lifespan changes caused by death perception. We found that this manipulation had no effect ( S4 Fig ), suggesting that Dilps do not act directly on R4 neurons to modulate lifespan changes due to death perception.

(A) dilp2 mRNA was unchanged, while dilp3 and dilp5 mRNA abundances in Canton-S fly heads were increased upon exposure to dead conspecifics (n = 20 for each treatment). Plotted are the fold changes in dilp mRNA values when comparing the unexposed to the dead-exposed values. (B) Dilp3 protein abundance, as measured using an anti-Dilp3 antibody, was increased in Canton-S MNC upon exposure to dead conspecifics (n = 14 and 16 for unexposed and dead-exposed, respectively). (C) dilp2-3,5 triple mutant flies (n = 184 for unexposed and n = 175 for dead-exposed) did not show altered lifespan due to the presence of dead conspecifics compared to controls (w - Dahomey; n = 193 for unexposed and n = 183 for dead-exposed). (D) dilp2 mutants showed altered lifespan when co-housed with dead conspecifics (n = 60 for unexposed and n = 79 for dead-exposed), but not dilp3 mutants (n = 99 for unexposed and n = 90 for dead-exposed; panel E) or dilp5 mutants (n = 96 for unexposed and n = 98 for dead-exposed; panel F) compared to controls (w 1118 ; n = 93 for unexposed and n = 91 for dead-exposed). (G) Acute red light activation of R4d neurons (using R4d-GAL4 x UAS-Chrimson) for 2 days had no effect on Dilp3 abundance in MNC (n = 8 for each treatment). (H) Chronic red light activation of R4d neurons (using R4d-GAL4 x UAS-Chrimson) for 2 weeks significantly increased Dilp3 abundance in MNC (n = 19 for each treatment). (I) Chronic red light exposure of control flies (using R4d-GAL4 x w 1118 ) for 2 weeks had no effect on Dilp3 abundance in MNC (n = 8 for each treatment). Data for this figure can be found in the accompanying Supporting information ( S4 Datasheet ). MNC, median neurosecretory cell.

Considering that foxo is known to modulate aging through its role in the insulin-signaling pathway, we investigated whether the Drosophlia insulin-like peptides (dilps) were also involved in this process. Of the 8 dilps found in flies, we focused our studies on dilp2, dilp3, and dilp5 because they are synthesized and released by median neurosecretory cells (MNCs) in the brain, which have previously been implicated in modulating fly lifespan [ 15 ]. We found that dilp3 and dilp5 mRNA, but not dilp2 mRNA, were increased in the brains of flies that had been exposed to dead conspecifics for 2 weeks, compared to same-age, unexposed control animals ( Fig 4A ). Dilp3 protein abundance was also increased ( Fig 4B ). We were unable to examine Dilp5 protein abundance due to the lack of a Dilp5 antibody. These changes are likely important for modulating lifespan because we found that flies lacking all 3 dilps, or dilp3 and dilp5 individually, did not exhibit changes in lifespan when aged in the presence of dead conspecifics; dilp2 was dispensable for the lifespan effect ( Fig 4C–4F ).

To determine whether Foxo is required specifically in R2/R4 neurons for death perception to affect lifespan, we targeted expression of foxo-RNAi to these cells (R2/R4-GAL4 > UAS-foxo-RNAi). This manipulation eliminated the effect of death exposure on lifespan (Figs 3G and S3A ). RNAi-mediated knockdown of foxo expression did not alter the appearance or number of R2/R4 neurons ( S3B and S3C Fig ), suggesting that our results were not caused by pleiotropic effects of RNAi on cell survival and indicating a specific function of foxo in mediating the effects of death perception on lifespan.

To identify mechanisms that would illuminate how the EB modulates aging following death perception, we examined the requirement of signaling pathways that are known to be involved in metabolism, stress, and aging; including Dh44 (homologue of mammalian corticotropin-releasing hormone), NPF (homologue of mammalian NPY), and the transcription factor foxo, as well as molecules involved in general aspects of fly memory (e.g., dumb, rutabaga, and dunce). Most of the genes or neural manipulations tested were not required ( Table 2 ). However, we observed that flies homozygous for a null allele of foxo [ 14 ] did not exhibit altered lifespan when co-housed with dead ( Fig 3D ). A similar observation was made following RNAi-mediated knockdown of foxo expression in adult neurons using GS-elav-GAL4 ( Fig 3E ). Additionally, we observed a reduction in the expression of the Foxo target gene, Thor ( Fig 3F ), in fly head tissues of individuals exposed to dead conspecifics, suggesting that Foxo activity was altered in these animals.

Knock-down of 5-HT2A mRNA expression in R2 alone (A), in R4d alone (B), or in R2/R4 inhibited lifespan differences due to death perception. The following genotypes were used: R2-GAL4 x UAS-5-HT2A-RNAi (n = 194 for unexposed and n = 193 for dead-exposed), R2-GAL4 x w 1118 (control; n = 191 for unexposed and n = 189 for dead-exposed), R4d-GAL4 x UAS-5-HT2A-RNAi (n = 144 for unexposed and n = 134 for dead-exposed), R4d-GAL4 x w 1118 (control; n = 148 for unexposed and n = 143 for dead-exposed), R2/R4-GAL4 x UAS-5-HT2A-RNAi (n = 182 for unexposed and n = 180 for dead-exposed), and R2/R4-GAL4 x w 1118 (control; n = 178 for unexposed and n = 176 for dead-exposed). (D) foxoΔ94 mutant flies showed little lifespan differences when exposed to dead compared to unexposed flies (n = 184 and n = 180 for unexposed and dead-exposed, respectively), unlike control flies (w 1118 ; n = 196 for unexposed and n = 184 for dead-exposed, respectively). (E) Knock-down of foxo mRNA in neurons (GS-elav-GAL4 x UAS-foxo-RNAi given RU-486; n = 89 for unexposed and n = 95 for dead-exposed exposed) completely inhibited lifespan changes due to the presence of dead unlike controls (GS-elav-GAL4 x UAS-foxo-RNAi without RU-486; n = 69 for unexposed and n = 95 for dead-exposed). (F) The amount of Thor mRNA, a Foxo target gene, was decreased in the heads of flies exposed to dead compared to unexposed flies (n = 16 each treatment). Plotted is the fold change in Thor mRNA amount isolated from fly heads when comparing the unexposed values to the dead-exposed values. (G) Foxo was required in R2/R4 neurons to see lifespan effects caused by dead exposure. The genotypes of the flies used were: R2/R4-GAL4 x UAS-foxo-RNAi (n = 136 for unexposed and n = 130 for dead-exposed) and R2/R4-GAL4 x w 1118 (control; n = 136 for unexposed and n = 125 for dead-exposed). Data for this figure can be found in the accompanying Supporting information ( S3 Datasheet ). EB, ellipsoid body.

Considering that global loss of the 5-HT2A receptor abolished the effect of death exposure on lifespan and that the increased activity in EB neurons following exposure was first identified using a 5-HT2A-GAL4 promoter, we directly tested whether 5-HT2A expression in R2/R4 neurons is required for the effects of dead conspecifics on lifespan. We targeted RNAi-mediated suppression of 5-HT2A expression to R2/R4 neurons and found that knockdown of 5-HT2A in all R2/R4 neurons, as well as in R2 and R4d individually, either partially or fully reduced lifespan differences due to dead exposure ( Fig 3A–3C ).

If EB neurons are generally, or R2/R4 neurons more specifically, involved in mediating the effects of death perception on lifespan, we might expect that activating them in the absence of dead would be sufficient to affect lifespan. Expression of the transient receptor potential cation channel TRPA1 using EB-GAL4 resulted in a significant lifespan decrease when flies were aged at 29°C, a temperature sufficient for channel activation, but not when flies were kept below the threshold temperature of 18°C (Figs 2C and S2 ). Expression of TRPA1 in R2/R4 neurons or in R4d neurons alone also reduced the lifespan of adults aged at the activating temperature (29°C; Fig 2D and 2E ). To our surprise, TRPA1-mediated activation of R2 neurons alone significantly increased lifespan ( Fig 2F ). These data are in line with recent findings demonstrating that R2 and R4 neurons can differentially impact fly physiology and behavior, depending on the experimental context [ 13 ]. Altogether, we conclude that the EB, specifically R2/R4 neurons, are a dynamic component of the neural network that plays a prescriptive role in modulating aging following exposure to dead conspecifics.

Alterations in the abundance of the synaptic active zone protein Bruchpilot (Brp) are thought to reflect underlying activity-dependent changes in neural synaptic structure that are generally associated with changes in synaptic strength [ 12 ]. We therefore examined the abundance of Brp in the brains of unexposed and dead-exposed flies and observed increased Brp antibody staining in the EB neurons of dead-exposed flies compared to unexposed animals but not in other brain regions such as the antennal lobe ( Fig 2B ). These data are consistent with changes in the activity of R2/R4 neurons and suggest that exposure to dead may also influence neural plasticity.

(A) Representative images of brains that express UAS-CaMPARI in R2/R4 neurons and their quantification from flies that were unexposed (n = 14) or exposed (n = 12) to dead conspecifics. (B) Representative false color images and quantification of the antennal lobe or the EB that were stained using an anti-Bruchpilot antibody from flies that were unexposed (n = 15) or exposed (n = 14) to dead conspecifics. Activation of EB neurons (C), R2/R4d (D), or R4d alone (E) in the absence of dead decreased fly lifespan, whereas activation of R2 (F) was sufficient to increase lifespan. The following genotypes were used: EB-GAL4 x UAS-TRPA1 (n = 163), EB-GAL4 x w 1118 (control; n = 117), R2/R4-GAL4 x UAS-TRPA1 (n = 196), R2/R4-GAL4 x w 1118 (control; n = 179), R4d-GAL4 x UAS-TRPA1 (n = 197), R4d-GAL4 x w 1118 (control; n = 190), R2-GAL4 x UAS-TRPA1 (n = 140), and R2-GAL4 x w 1118 (control; n = 140). Data for this figure can be found in the accompanying Supporting information ( S2 Datasheet ). EB, ellipsoid body.

If R2 and R4 neuronal activation is required for death perception to modulate lifespan, we might expect that the activity of these neurons would be increased when flies are exposed to dead. To determine whether this was indeed the case, we used the calcium-modulated photoactivatable ratiometric integrator system, CaMPARI, which results in Ca 2+ -dependent conversion of GFP to red fluorescence protein (RFP) in activated neurons when exposed to 405 nm light [ 10 ]. For this experiment, we used CaMPARI rather than the CaLexA system described above because of its ability to temporally constrain calcium measures and control for tonic activity, both of which limit unrelated signals at the potential cost of reduced signal. When we targeted expression of the CaMPARI protein to both R2 and R4 neurons (using SS02769-GAL4, hereafter R2/R4-GAL4; [ 11 ]) and induced conversion during exposure, we observed a significant increase in the relative RFP signal in flies exposed to dead compared to unexposed controls, indicating that their activity was significantly increased ( Fig 2A ). CaLexA experiments using flies that were kept in the dark during the exposure period revealed no significant intensity differences between dead-exposed and control flies, consistent with the notion that sight is required for this treatment to induce changes in neural activity ( S1C Fig ).

The EB comprises a series of ring neurons, grouped into 11 subclasses based upon the staining of a global marker (DN-Cadherin) and GAL4 driver lineage analysis [ 8 ]. We obtained GAL4 driver lines that targeted 9 of these subclasses (R1, R2, R3a, R3d, R3p, R4m, R4d, R5, and R6; see S1 Table for the fly lines used) and tested whether silencing each group individually influenced the effect of death exposure on lifespan. We found that the survivorship of flies expressing Kir 2.1 in R3a, R3p, R3d, R5, or R6 neurons remained significantly affected by the presence of dead conspecifics, indicating that activation of these neurons was not required for the effects of dead exposure on lifespan ( Table 1 ). On the other hand, Kir 2.1 -mediated silencing of R4d neurons eliminated the significant effect on lifespan ( Fig 1D ). Expression of Kir 2.1 using GAL4 lines that targeted R1, R2, or R4m neurons resulted in developmental lethality, leading us to substitute optogenetic manipulations to induce their silencing after eclosion [ 9 ]. We observed that GtACR1-mediated, optogenetic silencing of R1 neurons did not influence the effect of death exposure on lifespan but that silencing of R2 and R4m led to partial and near-complete abrogation of lifespan effects, respectively ( Fig 1E and 1F , and Table 1 ). Of note, we also tested several neuronal populations outside the EB for potential roles in this phenotype but observed no effect of inhibiting these neurons on the lifespan effects caused by the presence of dead ( Table 2 ). We therefore conclude that the activity of R2 and R4 neurons are a specific and required component of the neural circuitry that modulates lifespan when flies are exposed to dead conspecifics.

We next tested whether EB neuronal activity was required to mediate the effects of death perception on lifespan. To accomplish this, we silenced these neurons by expressing an inward rectifier K+ channel (EB-GAL4 > UAS-Kir 2.1 , [ 6 ]) and examined how this would impact the lifespan differences we saw when animals were exposed to dead compared to their unexposed controls. We observed that when EB neurons were silenced, the survivorship of flies was unaffected by the presence of dead ( Fig 1C ). Notably, silencing tubercular-bulbar (TuBu) neurons (GMR88A06-GAL4 > UAS-Kir 2.1 ), which form synaptic connections with EB neurons linked to visual feature detection [ 7 ], also abrogated the effect of death perception on survivorship ( S1A and S1B Fig ), an observation that is consistent with the reported involvement of sight in this phenotype [ 2 ].

(A) The panel on the left are representative brain images from selective slices of 5-HT2A-GAL4 x NFAT unexposed and dead-exposed flies. The panel on the right is the quantification of the GFP-intensity seen in the ellipsoid cell bodies and axons from the regions indicated the picture of unexposed and dead-exposed brains. The pixel sum of all the stacks from the indicated region with background subtracted is shown. Each dot represents an individual brain (n = 10 brains for each treatment). (B) The panel on the left are representative brain images from selected slices of EB-GAL4 x NFAT unexposed and dead-exposed flies. The panel on the right is the quantification of the GFP-intensity seen in the ellipsoid cell bodies and axons from the brain regions indicated in the picture of unexposed and dead-exposed brains. The pixel sum of all the stacks from the indicated regions with background subtracted is shown. Each dot represents an individual brain (n = 11 brains for each treatment). (C) Inhibiting EB neurons using EB-GAL4 x UAS-Kir 2.1 flies inhibited lifespan differences caused by exposure to dead individuals (n = 59 or 52 for unexposed or dead-exposed, respectively). Control flies used were EB-GAL4 x w 1118 (n = 59 or 62 for unexposed or dead-exposed, respectively). Inhibiting R4d (D), R2 (E), or R4m (F) neurons specifically also inhibited lifespan differences caused by exposure to dead individuals. The following genotypes were used: R4d-GAL4 x UAS-Kir 2.1 (n = 189 for unexposed and n = 177 for dead-exposed), R4d-GAL4 x w 1118 (control; n = 187 for unexposed and n = 179 for dead-exposed), R2-GAL4 x UAS-GtACR1 (n = 139 for unexposed and n = 122 for dead-exposed), R2-GAL4 x w 1118 (control; n = 101 for unexposed and n = 103 for dead-exposed), R4m-GAL4 x UAS-GtACR1 (n = 190 for unexposed and n = 179 for dead-exposed), and R4m-GAL4 x w 1118 (control; n = 182 for unexposed and n = 193 for dead-exposed). Data for this figure can be found in the accompanying Supporting information ( S1 Datasheet ). EB, ellipsoid body; GFP, green fluorescence protein.

Having previously demonstrated that loss of the 5-HT2A receptor eliminated the ability of dead conspecifics to modulate lifespan [ 2 ], we sought to understand the underlying mechanisms involved. We focused our attention on the central nervous system (CNS) because we previously established that sight, and possibly olfaction, were involved in the response [ 2 ]. To identify individual neurons or neuronal populations that mediate the perceptive event or its consequences, we began by examining neural cell activation in 5-HT2A + neurons using 5-HT2A-GAL4 [ 4 ] and an NFAT-based tracing method (CaLexA) through which sustained neural activity drives expression of green fluorescence protein (GFP) [ 5 ]. We found that a short, 2-day exposure to freshly dead flies, which is known to affect the behavior of unexposed flies compared to dead-exposed individuals [ 2 ], led to a significant increase in fluorescent intensity relative to control animals in a population of neurons that was strongly indicative of the EB, suggesting that exposure to dead increased the activity of these neurons ( Fig 1A ). We observed a similar pattern when we expressed this same NFAT construct in EB neurons using the pan-EB-GAL4 driver, R73A06 (hereafter EB-GAL4), although the result was marginally less significant than that observed using the 5-HT2A driver, likely because of higher background signal. Nevertheless, these data provide evidence that the activity of 5-HT2A-expressing EB neurons are increased when flies are exposed to dead conspecifics ( Fig 1B ).

Discussion

Here, we have identified a discrete, 5-HT2A-expressing neural population in the EB of the fly brain that plays an important role in transducing sensory information about the presence of dead individuals in the environment to influence lifespan. The R2, R4m, and R4d subpopulations were activated upon exposure to dead conspecifics, and their function was also required to modulate lifespan. Components of the insulin-signaling pathway, specifically the transcription factor Foxo in R2/R4 neurons and the insulin-like peptides Dilp3 and Dilp5, were also required to mediate the effects of death perception on lifespan. Unexpectedly, several lines of evidence support a model in which Dilps act downstream of R2/R4 neurons and therefore downstream of Foxo: (i) knockdown of the insulin receptor in R2/R4 did not impact the effects of death perception on lifespan; (ii) R4 neurons show increased calcium influx soon after dead exposure (within 48 h), while Dilp3 and Dilp5 RNA/protein abundances change over a much longer time frame (2 weeks or more; Fig 4 compared to S5 Fig); and (iii) prolonged activation of R4d neurons is sufficient to increase Dilp3 abundance in the MNC (Fig 4H; S7 Fig summarizes our findings).

Our work stimulates new questions about how distinct neural circuits impinge on well-known mechanisms of aging. How does the transcription factor Foxo act independently of insulins in 5-HT2A+ R2/R4 neurons to transduce the effects of death perception? One possibility is that 5-HT2A signaling affects Foxo by activating downstream kinases that can phosphorylate Foxo, reducing Foxo accumulation in the nucleus and thereby altering Foxo activity [17]. 5-HT2A has previously been shown to activate phospholipase C, which can initiate the phosphoinositol second messenger cascade by producing inositol triphosphate (IP3) and diacylglycerol (DAG), stimulating the release of protein kinase C and opening L-type Ca2+ channels that lead to changes in neural plasticity and, hence, neural signaling [18]. Serotonin itself is known to impact Foxo activity; exogenous serotonin application down-regulated Foxo accumulation in the nucleus of wild-type C. elegans exposed to stress [19] and FoxO1 is required for serotonin to affect bone mass in mice [20]. Following death perception, it is likely that the increased intracellular Ca2+ levels detected in 5-HT2A-expressing EB ring neurons resulted in greater synaptic strength as reflected by the increased intensity of Brp antibody staining. We therefore put forth a model where the sight of dead conspecifics activates 5-HT2A serotonergic receptors in EB ring neurons thereby activating this neural population through changes in both Foxo activity and greater excitability.

It is unknown how R2/R4 neurons themselves influence insulin-like peptide production and/or release. The pattern of activity generated by R2/R4 ring neurons likely propagates to ellipsoid EPG neurons [21], as these neurons synapse with R4d [16]. From here, EPG neurons synapse with a dispersed network of other neurons found in the fan-shaped body and the protocerebral bridge that could impact dilp-expressing MNC. Alternatively, activation of R2/R4 neurons themselves may release, perhaps in a Foxo-dependent manner, chemical messengers that travel beyond the central complex to Dilp-producing neurons. Candidate molecules whose receptors are known to be expressed on Dilp-producing MNCs and have been shown to alter their expression include short neuropeptide F [22], serotonin [23], and GABA [24]. It is also plausible that activation of R2/R4 neurons directly impacts fly metabolism, which in turn elicit secondary changes in Dilp release.

We were surprised by the observation that activation of R2 neurons alone was sufficient to significantly extend fly lifespan. Given the similar requirements of R2 and R4 neurons in the effects of death perception, one might expect that, like activation of R4, activation of R2 would also decrease lifespan. This unexpected finding indicates complexity in signaling among EB neurons and suggests that this neuropil may act as a rheostat to increase or decrease the rate of aging in response to sensory input. Understanding the similarities and differences in their downstream functions may provide insight into how this may occur. Furthermore, we are just beginning to understand the different phenotypes controlled by one or both of these neuronal groups. R2 neurons promote sleep drive [25] and increase egg-laying preference [26], whereas R4 neurons are known to promote nutritive sugar feeding [27]. Both are essential for visual orientation memory for salient objects and simple pattern discriminations [28]. Neither group has been studied for its influence on longer-term phenotypes, such as lifespan.

While the EB is generally agreed to be a multifaceted sensory integration and motor coordination center, it is one of the neuronal centers that possibly drive motivation and may even be responsible for underlying “emotion-like” states in Drosophila, driven mechanistically by Foxo [17,29]. A low motivational state in male flies caused by losing a fight, for example, is rescued by the optogenetic activation of the serotonin receptor 5-HT1B in EB neurons, and this state is associated with R2/R4m neural activity [30]. Moreover, the EB plays a crucial role in the stress-induced arousal response of the fly [31], a key component of many emotional and affective behaviors [32]. Lastly, there is a strong correlation in neuroanatomical organization and function between the Drosophila central complex and the vertebrate basal ganglia [33] suggesting that it may contribute to functions similar to those attributed to the vertebrate structure, such as recognition of emotional stimuli, inferring the emotional state of others, and awareness of subjective well-being. Foxo likely has a role in these states, as it has been suggested to play a crucial role in the pathophysiology of depressive disorders [34]. The loss of FoxO3a, for example, caused mice to display antidepressant-like behaviors [35]. Given our findings, it seems plausible that the sight of dead conspecifics elucidates a “depressive-like” state that results in decreased longevity.

While clearly speculative, conceptual distinctions of neural states reflect causal influences on organism health, and they have real-world consequences that are clinically tangible. Death perception may be one of these, as it produces psychological changes in humans such as emotional dysregulation and depression, as well as important physiological changes that negatively impact overall health, including depression, headache, fatigue, and cardiovascular disease [36–38]. Could motivational therapy or pharmacologic intervention in reward systems, much like what is done for addiction, slow aging? Such ideas are testable today, in humans, using approved drugs once we have a clearer mechanistic insight into the neural circuits and signaling systems that are involved.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002149

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.