Further information and requests for resources and reagents should be directed to and will be fulfilled by the corresponding author, SP (spletch@umich.edu).

The w-; CS stock was established by mixing a population of w ¹¹¹⁸ flies with standard Canton-S (CS) flies for more than 10 generations and re-isolating white-eyed flies. The 5-HT2A ^−/− null mutants containing a GAL4 element in place of the first coding exon (Qian et al., 2017) were generously donated by Yi Rao, Peking University. The broad-expressing 5-HT2A-GAL4 was created by replacing a MiMIC insertion in the 5-HT2A locus with a GAL4 element (Gnerer et al., 2015) and was kindly shared by Herman Dierick, Baylor College of Medicine. The restricted 5-HT2A-GAL4 expressing in the superior medial protocerebrum (BDSC #38744) contains a GAL4 element in the promoter region of 5-HT2A. Npf-GAL4 contains a GAL4 element fused to the regulatory sequence region for npf (BDSC# 25682). Neuronal activation experiments were performed using; UAS-CsChrimson (BDSC #55135), which was back-crossed to the w-; CS control stock for 10 generations. Neuronal inhibition experiments were performed using the anion channelrhodopsin (GtACR1) fused to a UAS element (Mauss et al., 2017), which was generously shared by Monica Dus, University of Michigan and back-crossed to the w-; CS control stock for 10 generations.

All fly stocks were maintained on a standard cornmeal-based larval growth medium (produced by LabScientific Inc. and purchased from Fisher Scientific) in a constant environment (25°C, 60% humidity) with a 12:12 h light:dark cycle. We controlled the developmental larval density by aliquoting 32 µl of collected eggs into individual bottles containing 25 ml of food. Following eclosion, mixed-sex flies were kept on SY10 (10% [w/v] sucrose and 10% [w/v] yeast) medium for 2–3 days until they were used for experiments. Pioneer table sugar (purchased from Gordon Food Service, MI) and MP Biomedicals Brewer’s Yeast (purchased from Fisher Scientific) were used in our study.

See table below for information on each diet used in lifespan and/or ConEx experiments (values are based on 1 L total volume):

All flies for qPCR analysis were flash frozen in liquid nitrogen for storage at −80^°C. Total mRNA was extracted in FastPrep Lysing Matrix D tubes (MP Biomedicals) using cold TRIzol (ThermoFisher Scienctific, Inc.) according to the manufacturer’s directions. The RNA-containing fraction was then removed by pipet to a fresh, autoclaved microcentrifuge tube for RNA precipitation using 100% ethanol. After 30 min at −80^°C, the RNA pellet was collected by spinning at 13,000 rpm in a benchtop centrifuge. The pellet was washed 2X using 70% ethanol in DEPC water and then dissolved in 20 μl DEPC water for spectrophotometer analysis at 260 nm. Equal amounts of RNA from each sample were then subjected to reverse transcription using SuperScript III RT enzyme and protocol (Invitrogen) according to the manufacturer’s directions. The subsequent cDNA was then subjected to qPCR analysis in a StepOne Plus 96-well Thermocycler (Applied Biosystems) using SYBR Green PCR Master Mix (Applied Biosystems) according to the manufacturer’s instructions. The primer sequences used for amplification of the 5-HT2A cDNA are as follows:

Forward primer: 5′ TCG​CCA​GCC​GTT​TAT​TGA​CT 3′

Reverse primer: 5′ CTC​CGC​TTC​GTC​GAT​AGC​TT 3′

After the flies were aged on SY10 for 10–14 days (food was changed every 2–3 days), experimental flies were quickly frozen, collected into groups of five, and then homogenized in 200 μl of ice-cold phosphate-buffered saline containing 0.1% Triton X-100 for 30 s at 30 Hz using a QIAGEN TissueLyser. For triacylglycerol (TAG) quantification, the homogenate (20 μl) was added into 200 μl of Infinity Triglyceride Reagent (Thermo Electron Corp.) and incubated at 37°C for 10 min with constant agitation. TAG concentrations were determined by measuring the absorbance at 520 nm and estimated by a known triglyceride standard. For protein measurement, 5 μl of fly homogenate was incubated with 200 μl of (1:50) 4% (w/v) cupric sulfate/bicinchoninic acid solution (Novagen) at room temperature for 30 min. For protein excretion measurements, flies were fed SY10 food for 72 h (20 flies per vial, 7-8 replicates for each group). The food was then removed from the vials and 1 ml of Milli-Q water was used to wash the excrement from the vials. This solution was then passed through a dechorionation sieve to remove any eggs and 10 μl of the wash volume was used to measure protein concentration. Protein concentrations were calculated by measuring the absorbance at 562 nm through the comparison with bovine serum albumin standards for all protein assays. For mass measurements, flies were aged on SY10 for 10–14 days, then separated into Eppendorf tubes for weight measurements. The tubes were kept at −20°C for 30 min to ensure the flies were not moving prior to wet mass measurements. The tubes were then placed in a 37°C oven for 3 days before measuring dry mass. Average weight, TAG, and protein values were based on at least 10 independent biological replicates from multiple vials.

Virgin flies of each genotype were collected and mated to one week-old Canton-S males in a 1:1 male to female ratio (five females and five males were co-housed in SY10 vials with ten vials per treatment). Exposure to males began 48 h prior to the start of the experiment. Vials were flipped daily, and the total number of eggs were counted each day then summed for each vial over 7 days.

We used the protocol described previously (Shell et al., 2018). Briefly, experimental flies were co-housed in vials (10 flies per vial, 8–10 replicates for each group) for 2–3 days following eclosion and then sorted into individual sex cohorts on SY10 food. After 10–14 days, the flies were transferred onto diets with 1% (w/v) FD and C Blue No. 1 in varying percentages from 5% to 20% of Yeast or Sucrose (w/v) based on the experiment (see table below under Diet Compositions for details). Vials were discarded if one or more dead flies were observed after the 24 h feeding period. For ConEx experiments coupled to optogenetics, 3 min of 40 Hz red light was supplied every 15 min to minimize depolarization block (Shao et al., 2017). Excreted dye (ExVial) was collected by adding 3 ml of Milli-Q water to each vial followed by vortexing for 10 s. The concentration of the ExVial dye in water extracts was determined by reading the absorbance at 630 nm, which was used to infer macronutrient consumption.

Flies were tested on the Fly Liquid-Food Interaction Counter (FLIC) system as previously described to monitor feeding behaviors (Ro et al., 2014) coupled to optogenetic activation technology. Specifically, custom lids were produced containing one LED for each well in the FLIC chamber, allowing precise control over the dynamics of the light timing, duration, and frequency. Easycargo Aluminium Radiator Cooler Heat Sinks were used to dissipate any heat generated by the LEDs [DigiKey manufacturer numbers: LXM2-PD01-0050 (red), LXML-PM01-0100 (green)]. Each liquid-food reservoir contained either 2% (w/v) yeast extract +1% (w/v) sucrose or 1% (w/v) sucrose in 1% Tegosept (v/v) and 4 mg/l MgCl₂. For closed-loop experiments, 100 µM denatonium in 1% Tegosept (v/v) and 4 mg/l MgCl₂ was used. Flies were anesthetized briefly on ice and manually aspirated into the Drosophila feeding monitors (DFMs). Each DFM was loaded with flies from at least two treatment groups to reduce technical bias from the DFM signals. LED lights were always pulsed at 40 Hz with a 32% duty cycle. In the open-loop activation experiments, red light was provided 12 s every six minutes for experiments using the broad 5-HT2A-GAL4 driver and for three minutes every fifteen minutes for the more restricted 5-HT2A-GAL4 (BDSC #38744). For open-loop inhibition experiments, green light was supplied for three minutes every fifteen minutes. Red light was pulsed when flies interacted with the liquid food in closed-loop experiments. All LEDs were pulsed at 40 Hz to minimize depolarization block (Shao et al., 2017). For closed-loop experiments, red light was pulsed at 40 Hz upon interaction with the liquid food well and included a one second decay after food interaction ceased. Food interactions were analyzed using custom R code, which is available on GitHub at https://github.com/PletcherLab/FLIC_R_Code. Default thresholds were used for analysis. Flies that had zero feeding events over the testing interval were removed from the analysis.

Flies emerged on CT food and were collected after 48 h to allow for mating. Female flies were then separated from males and placed in individual vials during nutrient exposure to minimize potential neuronal signals from social cues. Vials were prepared containing 2% agar coated with 200 mM sucrose ±2.5x complete amino acid solution, prepared as previously described (Piper et al., 2014) Nutrient exposure time was 48 h, after which the female brains were dissected and imaged as described below using a wavelength of 488 nm to visualize the NFAT GFP signal.

Four days post-eclosion brains from adult females exposed to 200 mM sucrose ±2.5x complete amino acid solution conditions were dissected in ice-cold phosphate buffer saline (PBS) using sharpened tweezers and fixed in PBS containing 4% paraformaldehyde for 60 min at room temperature. The brains were washed thoroughly with 1 ml of PBS with 0.1% Triton-X (PBS-T), then moved using a wide-bore pipet tip to a glass slide with Vectashield (Vector Laboratories) and sealed using a coverslip and clear nail polish for immediate imaging.

Imaging was carried out using an Olympus FLUOVIEW FV3000 confocal microscope. The brains were brought into focus with 10x (0.40 NA) objective lens before switching to 20x (0.75 NA). Extra care was taken in order not to saturate the image. Images were acquired at 1,024 × 1,024 pixels with a step size of 3.0 micron. The laser power and the parameters for image acquisition was kept the same between control and treatment groups.

For data analysis, the imaging files were analyzed in the publicly available imaging software, Fiji. Selective slices were combined and collapsed into a single image using SUM slices. Brightness and contrast were adjusted manually when required for better visualization of the image. Background was calculated from the brain region adjacent to the ROI and subtracted. Representative images are displayed in mpl-inferno pseudo color.

Lifespans were measured using established protocols (Linford et al., 2013). Unless otherwise noted, 10 replicate vials (∼200 experimental flies) were established for each treatment. Flies were transferred to fresh media every 2–3 days, at which time dead flies were removed and recorded using the DLife system developed in the Pletcher Laboratory (Linford et al., 2013). Flies were kept in constant temperature (25°C) and humidity (60%) conditions with a 12:12 h light:dark cycle. For optogenetic lifespans, flies were kept in specially designed rigs containing red or green LEDs, to activate or inhibit neurons, respectively. Controls were kept in a dark box in the same incubator and flipped under dim red light. Red LEDs were pulsed at 2 Hz for activation experiments and green LEDs were pulsed at 40 Hz for inhibition experiments.

Unless otherwise indicated, pairwise comparisons between different treatment survivorship curves were carried out using the statistical package R within DLife (Linford et al., 2013). Each p‐value was obtained using a Log‐Rank analysis. For testing the interaction between genotypes and diets, we used Cox-regression analysis to report p-value for the interaction term. To test the effects of diet and genotype involved in food consumption, we performed Two-way ANOVA followed by a Bonferroni post-hoc significance test or ANCOVA. For OptoFLIC data, total licks for a given time period were fourth-root transformed, which was empirically found to normalize the majority of FLIC data. Significance was determined using a one-sided t-test. Significance for metabolic assays was measured using a two-sided t-test. For imaging experiments, a one-sided t-test was used to determine significance.

To better understand the role of serotonin receptor 5-HT2A in aging, we studied flies homozygous for a putative null mutation (5-HT2A ^−/− ), which contained a GAL4 element in place of the first coding exon (Qian et al., 2017). The 5-HT2A ^−/− line used in these studies is a verified mutant used in other published studies and we confirmed that it lacked measurable 5-HT2A mRNA by quantitative PCR (Gasque et al., 2013; Howard et al., 2019). We first asked whether 5-HT2A ^−/− mutant flies were long-lived when aged on a standard laboratory diet consisting of 10% sucrose and 10% yeast (w/v, SY10) relative to a laboratory control strain (Canton-S) to which it had been extensively backcrossed. We found that 5-HT2A ^−/− mutant females lived significantly longer than control flies (Figure 1A). Loss of 5-HT2A did not have a significant effect on male lifespan (Figure 1B), leading us to focus on females for the majority of the study. Previous experiments had shown that 5-HT2A is required for modulating fly lifespan when the major dietary components of sucrose and yeast are presented separately, compared to when they are mixed and presented as a single, homogeneous mixture (Ro et al., 2016). We therefore wondered whether altering the levels of sucrose or yeast in our fixed diet would influence the magnitude of the mutant lifespan phenotype. To examine this, we measured the lifespans of 5-HT2A ^−/− mutant and control females on both yeast- and sucrose-rich diets, consisting of 15% yeast/5% sucrose or 5% yeast/15% sucrose, respectively. We observed that 5-HT2A ^−/− mutants were only long-lived on the yeast-rich diet and showed no difference in lifespan on the 5% yeast/15% sucrose diet (Figures 1C,D), suggesting that either a low sugar or high yeast environment promotes extended lifespan in 5-HT2A ^−/− mutant flies.

Dietary protein as yeast is particularly impactful in fly aging, as it largely drives the dietary restriction (DR) lifespan extension phenotype (Chippindale et al., 2004; Min and Tatar, 2006). Decreasing the levels of protein typically increases lifespan in a dose-dependent manner, up to a point where protein levels are presumably too low to support essential metabolism (Min and Tatar, 2006; McCracken et al., 2020), and we hypothesized that loss of 5-HT2A might alter the magnitude of this response. Holding sucrose levels constant, we therefore titrated the amount of protein in the diet and measured the lifespan of 5-HT2A ^−/− mutant and control flies. Not unexpectedly, the mean lifespan of control animals increased significantly as dietary protein levels decreased (Figure 1E). The mean lifespan of 5-HT2A ^−/− mutant flies, however, was less affected by dietary protein and remained high even in protein-rich conditions (Figure 1E). There was also a statistically significant interaction between genotype and diet (p = 0.03), supporting the conclusion that the mean lifespans of the two genotypes respond differently to diet. This suggests that loss of 5-HT2A alters the response to protein, potentially making flies more resistant to reductions in lifespan that normally accompany increases in dietary protein.

To further examine the interaction between the effects of 5-HT2A loss and dietary yeast, we investigated its role in feeding behavior. We titrated levels of yeast in fly diets and estimated food consumption in 5-HT2A ^−/− mutant and background control flies by spiking diets with a non-metabolizable blue dye (FD&C Blue 1) and used the amount of excreted dye as an estimate of consumption [i.e., the ConEx assay (Shell et al., 2018)]. Using the standard 10% sugar/yeast laboratory medium used in lifespan measurements as a reference diet, we measured consumption on diets containing 10% sucrose and ranging from 5% to 20% yeast (w/v). In flies from both genotypes, we observed that the volume of food consumed decreased as the amount of yeast in the diet increased (Figure 2A). When presenting these data in terms of the mass of yeast consumed, we observed a significant effect of genotype and no significant effect of dietary yeast concentration, supporting the notion that flies have a protein consumption target, ie., the amount of protein voluntarily ingested that we speculate satisfies their needs, driving their feeding behavior in fixed diets (Figure 2B) (Lee Kwang et al., 2008). The genotype effect was indicative of a consistent reduction of protein consumption in 5-HT2A ^−/− mutant flies, suggesting the possibility that they exhibit a lower protein consumption target. Differences in the mass of sucrose consumed matched the pattern of total consumption because sucrose was held constant in these diets (Supplementary Figure S1A). Similar to lifespan, loss of 5-HT2A ^−/− did not significantly affect feeding in males. Both mutant and control genotypes responded similarly to increased dietary protein by consuming less, and there was no significant effect of diet or genotype on the mass of yeast consumed (Suppulementary Figures S1B,C). Similar to females, sucrose mass consumption decreased as dietary protein increased, but there was no effect of genotype in males (Supplementary Figure S1D).

A similar experiment in which feeding was measured on diets ranging from 5% to 20% sucrose (w/v) only (to avoid confounding effects from yeast) revealed that female flies of both genotypes decreased total volume consumed as the concentration of dietary sucrose increased (Figure 2C). However, when scaled to present the mass of sucrose consumed, female flies of both genotypes consumed a higher mass of sucrose as concentration increased (Figure 2D). Unlike the experiments in which we manipulated dietary yeast, however, we observed no significant difference in pattern of consumption between 5-HT2A ^−/− mutant and control flies on sucrose diets (Figures 2C,D).

A reduced protein consumption target in 5-HT2A ^−/− mutants might be reflected in lower protein content in the body. Indeed, we found that 5-HT2A ^−/− mutant females exhibited decreased whole-body protein content (Figure 3A). To evaluate if nutrient absorption was involved in female body composition differences, we measured protein content in their excretion and observed no differences between 5-HT2A ^−/− mutant and control females (Figure 3B). We also asked whether changes in body composition in 5-HT2A ^−/− mutant females was specific to protein or whether it extended to other forms of nutrient storage. In Drosophila, excess carbohydrates are converted and stored primarily as triglycerides (TAG), and we found no differences in female TAG levels (Figure 3C). To determine whether the reduction in total protein influenced overall body mass, we measured both wet and dry mass and observed a significant reduction in both measures in 5-HT2A ^−/− mutant females (Figures 3D,E). The reduction in body mass is likely due to lower protein levels in 5-HT2A ^−/− mutants, with the lack of differences in TAG suggesting fat storage levels and carbohydrate usage are similar.

Our behavioral and physiological data led us to speculate that the absence of 5-HT2A receptor results in a chronic state of perceived protein limitation, independent of its availability in the diet. This situation might be expected to stimulate adaptive processes that reduce protein utilization, which would manifest as increased lifespan, particularly when dietary protein is replete. In this view, 5-HT2A ^−/− mutant flies would exhibit lifespan phenotypes that are associated with physiological states of low protein utilization and would be resistant to manipulations that subsequently increase it. Female reproductive status is associated with such a state; virgin female flies lack developing embryos, exhibit a reduced drive for protein consumption, and are long-lived (Ro et al., 2014). Mating increases reproductive output and drives protein feeding. We found that as little as three days of mating following eclosion (following our typical lifespan measurement protocol) significantly reduced the lifespan of control flies fed our standard 10% sugar-yeast diet (Figure 3F). 5-HT2A ^−/− mutant flies, however, were long-lived regardless of mating status; average mutant lifespan was not influenced by mating and was not significantly different from that of the control genotype virgin flies (Figure 3F). As postulated by Ro et al. (2016), several lines of evidence support the notion that extended lifespan in mutant females is independent of reduced nutrient intake and likely reflects a defect in normal physiological responses to perceived protein availability. First, feeding differences did not depend on mating status (Supplementary Figure S2A). Second, mutant females showed no differences in egg-laying relative to control flies (Supplementary Figure S2B). Third, the lifespan extension of 5-HT2A ^−/− mutants persisted in the absence of protein in the diet, eliminating differences in protein consumption as a required component of the lifespan extension (Supplementary Figure S2C).

We next sought to investigate whether 5-HT2A signaling influences motivational or reward circuitry related to protein consumption, which would affect responses to perceived availability. Published data suggest that 5-HT2A signaling is not involved in specifying protein demand but instead influences protein feeding by perhaps 1) stimulating its consumption or 2) by acting subsequent to consumption to reinforce feeding (Ro et al., 2016). The first model predicts that activation of 5-HT2A neurons would precede protein feeding and that neuronal activation would promote protein, but not carbohydrate, feeding and would not be generally rewarding. On the other hand, if 5-HT2A were acting to reinforce protein feeding, we might expect that activation of 5-HT2A neurons would follow protein feeding and that their activation would reinforce behavior when closely paired with feeding events.

To evaluate when and where 5-HT2A ⁺ neuronal activity is responsive to protein availability in food, we used 5-HT2A-GAL4 (Gnerer et al., 2015) to express a transcriptional reporter system involving an NFAT-based tracing method (CaLexA) through which sustained neural activity drives expression of GFP (Masuyama et al., 2012). We then quantified fluorescent intensity to identify populations of neurons that showed increased activity upon 48h protein manipulation. Specifically, 5-HT2A > NFAT flies were provided 2% agar coated with 200 mM sucrose +/- 2.5x complete amino acid mix (Piper et al., 2014). Flies fed amino acids exhibited decreased GFP intensity, relative to flies fed sucrose alone, in a population of 5-HT2A ⁺ neurons in the supramedial protocerebrum (SMP; Figure 4A), indicating these neurons were responsive to decreases in dietary amino acids.

We next determined whether acute activation or inhibition of 5-HT2A ⁺ neurons influence how the animals interact with different nutrients. For these studies, we used the Fly Liquid-Food Interaction Counter (FLIC), which measures the precise timing and duration of fly feeding behaviors through the closure of a circuit when a fly inserts its proboscis into a liquid food source (Ro et al., 2014). For neuronal activation, we drove expression of a red light-sensitive channelrhodopsin protein (CsChrimson) in all 5-HT2A ⁺ neurons and coupled the FLIC technology to optogenetic stimulation equipment. To accomplish this, we installed red LED lights on top of individual FLIC feeding chambers and executed an open-loop design, whereby light was illuminated for 12 s (40 Hz and 32% duty cycle) every five minutes, activating neurons regardless of the fly’s behavior. We compared feeding interactions between 5-HT2A>CsChrimson flies and their genotypic controls (see Section 2 for strain details) exposed to either sucrose or yeast food and in the presence or absence of red-light stimulation. When putatively all 5-HT2A ⁺ neurons were activated in this open-loop configuration, we observed a rapid, significant increase in interactions with a protein-containing solution in 5-HT2A>CsChrimson flies relative to controls, but not with a sucrose-only solution (Figures 4B,C). When these same experiments were conducted in the absence of light, no differences were observed between 5-HT2A-neuronal activated flies and controls in the presence of either protein or sucrose (Supplementary Figures S3A,B). For inhibition, we expressed an inhibitory anion channelrhodopsin (GtACR1) in all 5-HT2A ⁺ neurons and substituted green LEDs in an open-loop configuration where green light was pulsed for 3 min every 15 min. This manipulation did not significantly impact interactions with a protein-containing solution (Figure 4D), showing that the activity of 5-HT2A ⁺ neurons does not bi-directionally control protein feeding and suggesting that 5-HT2A signaling is specifically recruited to promote protein feeding behaviors.

To determine whether activation of 5-HT2A ⁺ neurons is sufficient to reinforce behavior regardless of the presence of protein, perhaps as a result of it conveying a more general sense of reward, we exposed flies to a denatonium solution (100 µM), which is bitter and aversive, and activated 5-HT2A ⁺ neurons in response to individual flies’ interaction with it (i.e., a closed-loop configuration). Unlike the robust increase in interactions observed upon activation of putative reward neurons that express the neuropeptide NPF (Shao et al., 2017), we observed only a modest increase in interactions with the bitter solution when 5-HT2A ⁺ neurons were activated (Figure 4E), suggesting 5-HT2A ⁺ neuronal activation alone does not provide a potent reward.

Finally, we also asked whether activation or inhibition of 5-HT2A ⁺ neurons impacted lifespan. We observed that both manipulations significantly shortened lifespan (Supplementary Figures S3C,D). We hesitate to interpret these results in the context of our model because 5-HT2A ⁺ neurons express many receptors and signaling peptides, all of which would be affected by optogenetic manipulations. These effects on lifespan may, therefore, be non-specific and due to the pleiotropic effects of activating or inhibiting a large group of neurons.

In summary, increased neuronal activity was observed when flies were amino acid deprived (Figure 4A), not when then had fed on protein, and 5-HT2A ⁺ neuronal activation was sufficient to drive protein consumption (Figure 4B) but was not generally sufficient to reinforce feeding behavior (Figure 4E). Taken together, therefore, these data support the notion that 5-HT2A signaling is recruited to promote protein consumption, perhaps by establishing a heightened protein consumption target, and to enact a physiological state of higher protein utilization that subsequently accelerates aging.

We next aimed to narrow the population of 5-HT2A ⁺ neurons that influence protein feeding behaviors with the hope of obtaining a better understanding of the relationship between the consumption set point and lifespan. Driving GFP expression with the broad 5-HT2A-GAL4 marks diverse populations of nearly 80 neurons in the Drosophila brain (Davie et al., 2018) (Figure 5A). Our neural activity data (Figure 4A) indicated that 5-HT2A ⁺ SMP neurons might influence feeding behavior because the increase in activity upon amino acid deprivation suggested they are responsive to protein demand. We therefore sought to determine whether manipulation of their activity alone might phenocopy the results that we observed following broader 5-HT2A neuronal activation. The FlyLight collection contains thousands of GAL4 lines that contain fragments of gene-specific promotors (enhancer-trap, or et-GAL4), thus, labeling smaller populations of cells that express that gene. One such GAL4 line for 5-HT2A (R50D04, BDSC #38744) specifically labels SMP neurons and recapitulates the expression of 5-HT2A ⁺ neurons with altered activity in the NFAT data (Jenett et al., 2012). We used this GAL4 line to express CsChrimson and to ask whether optogenetic stimulation was sufficient to increase behavioral interactions with dietary protein. Using a similar open-loop configuration with slightly longer activation periods to ensure against lesser activation of deeper neurons, we observed that activation of R50D04⁺ neurons in the FLIC resulted in a significant increase in interactions with a protein-containing solution relative to controls (Figure 5B), recapitulating the broad 5-HT2A-neuronal activation phenotype. The increase in feeding behaviors was not as rapid as the broad 5-HT2A-neuronal activation phenotype, suggesting the R50D04 line may not provide full coverage over 5-HT2A ⁺ neurons that promote protein feeding behaviors. This difference between genotypes was not observed in the absence of light activation (Figure 5C).

We also asked whether activation of these R50D04⁺ neurons would impact feeding in the same nutritional context as the standard lifespan conditions: a solid food diet consisting of 10% sucrose and 10% yeast (w/v). To test whether activation of 5-HT2A ⁺ SMP neurons also led to increased consumption of solid food, we coupled optogenetic approaches with the ConEx assay, allowing us to activate specific neurons during solid food consumption. In this assay, flies expressing CsChrimson in R50D04⁺ neurons were provided with a blue dye spiked diet of 10% sucrose and yeast and housed in vials that were exposed to high-intensity red LEDs. When R50D04⁺ neurons were activated (at 40 Hz and 32% duty cycle) periodically for three minutes every fifteen minutes over 24 h, we saw a significant increase in total consumption relative to the genetic controls, which restored to baseline in a subsequent 24-h dark period (Figure 5D). Both inhibition and activation of these neurons had no effect on lifespan (Figures 5E,F), suggesting protein consumption and lifespan are controlled through, at least partially, distinct mechanisms. Together, these data indicate that this limited subset of 5-HT2A ⁺ neurons in the SMP are critical to promote protein feeding behaviors, and thus, may help encode a protein drive that affects protein feeding behaviors; however, additional 5-HT2A ⁺ neurons may determine how these protein-dependent states impact lifespan.