Opposing GPCR signaling programs protein intake setpoint in Drosophila

Abstract

Animals defend a target level for their fundamental needs, including food, water, and sleep. Deviation from the target range, or "setpoint," triggers motivated behaviors to eliminate that difference. Whether and how the setpoint itself is encoded remains enigmatic for all motivated behaviors. Employing a high-throughput feeding assay in Drosophila, we demonstrate that the protein intake setpoint is set to different values in male, virgin female, and mated female flies to meet their varying protein demands. Leveraging this setpoint variability, we found, remarkably, that the information on the intake setpoint is stored within the protein hunger neurons as the resting membrane potential. Two RFamide G protein-coupled receptor (GPCR) pathways, by tuning the resting membrane potential in opposite directions, coordinately program and adjust the protein intake setpoint. Together, our studies map the protein intake setpoint to a single trackable physiological parameter and elucidate the cellular and molecular mechanisms underlying setpoint determination and modulation.

Keywords: GPCR signaling; RFamide; homeostasis; motivated behavior; protein intake setpoint; resting membrane potential.

Figure 1.. Tracking the sliding setpoint for protein feeding

(A) Schematic of lid feeding assay. Flies are fed from a 3D-printed 96-well plate lid with a divider in each well to hold two types of food labeled with blue or red dyes. After feeding, the lid is removed, and the excreted dye is extracted by adding water to the plate. Color intensity and composition are analyzed using a plate reader, measuring at 628 nm and 505 nm for blue and red, respectively. The absorbances are used to calculate the amount of intake for each food. (B-C) Daily intake per fly for protein (B) and sucrose (C) in age-matched wild-type iso31 male, virgin female, and mated female flies from day 1 to day 13 following 24-hour mating in mated female flies. n=24–32. Flies were flipped to new plates every 24hrs. (D-F) Daily protein intake per fly in male (D), virgin female (E), and mated female (F) flies with or without protein deprivation (PD) from day 1 to day 10 following protein deprivation. Protein deprivation durations were 9 days for males, 5 days for virgin females, and 3 days for mated female flies. n=24–32. A mixed-effects model with the Geisser-Greehouse correction was used to analyze the longitudinal daily food intake data across different groups in Figures 1B–1F, followed by Tukey’s multiple comparisons test for Figures 1B–1C and Šidák’s multiple comparisons test for Figures 1D–1F to determine significant daily differences between groups. In this and the subsequent figures, “*”, “**”, “***”, “****”, and “ns” denote P<0.05, P<0.01, P<0.001, P<0.0001, and not significant, respectively. Mean ± SEM is shown for longitudinal behavioral data. See also Figure S1.

Figure 2.. DA-WED neurons adjust the resting membrane potential with altered protein intake setpoint

(A-C) Representative traces (A), spontaneous action potential firing rate (B), and the resting membrane potential (C) for patch clamp recordings of DA-WED neurons from TH-D-Gal4, TH-C-FLP>UAS-frt-stop-frt-mCD8-GFP(3x) in age-matched male (n=11), virgin female (n=11), and mated female (n=10) flies. (D-F) Representative traces (D), spontaneous action potential firing rate (E), and the resting membrane potential (F) for patch clamp recordings of DA-WED neurons from TH-D-Gal4, TH-C-FLP>UAS-frt-stop-frt-mCD8-GFP(3x) virgin female flies with or without protein deprivation. n=6–12. (G) The fitted curve of evoked action potentials firing rate of DA-WED neurons in response to current injection at 10-pA increment ranging from −30 to 80 pA, in age-matched male (n=9), virgin female (n=9) and mated female (n=5) flies. (H) The estimated maximum firing rate of the DA-WED neurons using the fitted curve in (G) analyzed by the Boltzmann function. Kruskal-Wallis tests were performed with Dunn’s multiple comparisons test for Figure 2B. ANOVA followed by Tukey’s multiple comparisons test was used for Figures 2C and 2H. Unpaired t-tests were used for Figures 2E–2F. In this and the subsequent figures, for normally distributed datasets, scattered dot plots with mean ± SEM are shown. For non-normally distributed datasets, box plots are shown with scattered dots to display all raw data points. The line inside the box indicates the median. The top and bottom of the box represent the 75th and 25th percentiles, respectively. The whiskers represent the minimum and maximum values. See also Figure S2.

Figure 3.. Modifying resting membrane potential by manipulating FMRFaR-PKC pathway reprograms the protein intake setpoint.

(A) Schematic of the GPCR-PKC-ORK1 signaling pathway. (B) Resting membrane potential of DA-WED neurons in male flies of control (TH-D-Gal4>UAS-mCD8-GFP, UAS>KK control) or ORK1 knockdown (TH-D-Gal4>UAS-mCD8-GFP, UAS-ORK1-KK RNAi). n=4–7. (C-E) Daily protein intake per fly in male flies (C), virgin female flies (D), and mated female flies (E) of control (TH-C-Gal4>UAS-KK control) or ORK1 knockdown in DA-WED neurons (TH-C-Gal4>UAS-ORK1-KK RNAi). n=40–48. (F-G) Representative traces (F) and resting membrane potential (G) for patch-clamp recordings of DA-WED neurons from TH-D-Gal4, TH-C-FLP>UAS-frt-stop-frt-mCD8-GFP(3x) male flies with Saline or 4μΜ BIS applied inside the internal pipette solution. n=5–11. (H-I) Representative traces (H) and resting membrane potential (I) for patch-clamp recordings of DA-WED neurons from TH-D-Gal4, TH-C-FLP>UAS-frt-stop-frt-mCD8-GFP(3x) mated female flies with Saline or 100 nM PMA applied inside the internal pipette solution. n=6–9. (J) Resting membrane potential of DA-WED neurons in male flies of control (TH-F3-Gal4>UAS-mCD8-GFP, UAS-TRiP control) or PKC53E knockdown in DA-WED neurons (TH-F3-Gal4>UAS-mCD8-GFP, UAS-PKC53E TRiP RNAi). n=4–5. (K) Daily protein intake per fly in male flies of control (TH-C-Gal4>UAS-TRiP control) or PKC53E knockdown in DA-WED neurons (TH-C-Gal4>UAS-PKC53E-TRiP RNAi) from day 1 to day 10. n=42–80. (L) Resting membrane potential of DA-WED neurons in male flies of control (TH-F3-Gal4>UAS-mCD8-GFP, UAS-VSH control) or FMRFaR knockdown in DA-WED neurons (TH-F3-Gal4>UAS-mCD8-GFP, UAS-FMRFaR-VSH RNAi). n=5–6. (M) Daily protein intake per fly in male flies of control (TH-C-Gal4 >UAS-VSH control) or FMRFaR knockdown in DA-WED neurons (TH-C-Gal4 >UAS-FMRFaR-VSH RNAi) from day 1 to day 10. n=41–48. Unpaired t-tests were used for Figures 3B, 3I, 3J, 3L. Unpaired t-tests with Welch’s correction were used for Figure 3G. A mixed-effects model with the Geisser-Greehouse correction followed by Šidák’s multiple comparisons test was used to analyze daily food intake data in Figures 3C–3E, 3K, 3M. See also Figures S3 and S4.

Figure 4.. Upstream FMRFa neurons signal to lower the protein intake setpoint

(A) Whole-mount brain immunostaining of TH-F3-Gal4>UAS-mCD8-GFP with antibodies against Bruchpilot (nc82, magenta), GFP (green in the left panel), and FMRFa (green in the middle panel). High-magnification sections of wedge neuropil are shown on the right, with GFP in green and FMRFa in red. Scale bar, 50 μm. Dashed ellipses indicate the wedge neuropil. (B) Representative image from a GRASP experiment between DA-WED (labeled by TH-C-LexA) and FMRFa-WED (labeled by FMRFa-Gal4 BDSC_56837) neurons. Scale bar, 50 μm. Dashed ellipses indicate the wedge neuropil. Native fluorescence was shown. (C) Whole-mount brain images from MCFO analysis for FMRFa-WED neurons (R57C10-FLP2, FMRFa-Gal4 (56837)>HA_V5_FLAG) immunostained with anti-V5. Scale bar, 50 μm. Dashed ellipses indicate the wedge neuropil. Arrow indicates the cell body. (D) Location and EM image of a representative synapse between a presynaptic FMRFa-WED neuron (PLP.WED.2 in flywire database, annotated in green) and a postsynaptic DA-WED neuron (WED.162 in flywire database, annotated in magenta). In the EM image, green and magenta asterisks denote the pre- and post-synaptic sides of the synapse, respectively, and the orange arrow indicates the synaptic cleft. Scale bar, 500 nm. (E-G) Representative traces (E), spontaneous action potential firing rate (F), and the resting membrane potential (G) for patch-clamp recordings of FMRFa-WED neurons from FMRFa-Gal4 (56837)>UAS-mCD8-GFP in age-matched male (n=5), virgin female (n=8), and mated female (n=6) flies. (H) Membrane capacitance of FMRFa-WED neurons for males, virgin females, and mated females. n=6–9. (I-J) Representative images of the cell body (I) and quantification of the Feret’s parameters of the cell bodies (J) of FMRFa-WED neurons from FMRFa-Gal4 (56837)>UAS-mCD8-GFP in age-matched male, virgin female, and mated female flies. n=7–10. Scale bar, 10 μm. (K) Daily protein intake per fly in male flies of controls (WT>UAS-Kir2.1 or FMRFa-Gal4>WT) or flies with FMRFa neurons silenced (FMRFa-Gal4>UAS-Kir2.1) from day 1 to day 10. n=53–64. (L) Daily protein intake per fly in male flies of control (FMRFa-Gal4>UAS-KK control) or flies with FMRFa knockdown in FMRFa neurons (FMRFa-Gal4>UAS-FMRFa-KK RNAi) from day 1 to day 10. n=27–48. (M) Daily protein intake per fly in mated female flies of controls (WT>UAS-NachBac or FMRFa-Gal4>WT) or flies with FMRFa neurons activated (FMRFa-Gal4>UAS-NachBac) from day 1 to day 10. n=45–96. FMRFa-Gal4 driver line used in this figure is BDSC_56837. Kruskal-Wallis tests were performed with Dunn’s multiple comparisons test for Figure 4F. ANOVA followed by Tukey’s multiple comparisons test was used to analyze Figures 4G, 4H and 4J. A mixed-effects model with the Geisser-Greehouse correction was used to analyze daily food intake data in Figures 4K–4M, followed by Tukey’s multiple comparisons test for Figures 4K, 4M and Šidák’s multiple comparisons test for Figures 4L. See also Figure S5.

Figure 5.. MSR2-PKA signaling raises the protein intake setpoint.

(A) Resting membrane potential of DA-WED neurons in mated female flies of control (TH-F3-Gal4>UAS-mCD8-GFP, UAS>KK control) or MSR2 knockdown in DA-WED neurons (TH-F3-Gal4>UAS-mCD8-GFP, UAS-MSR2-KK RNAi). n=5–6. (B) Daily protein intake per fly in mated female flies of control (TH-C-Gal4>UAS-KK control) or MSR2 knockdown in DA-WED neurons (TH-C-Gal4 >UAS-MSR2-KK RNAi) from day 1 to day 10. n=44–48. (C) A schematic of the MS-MSR2-PKA signaling pathway. (D) Representative images of DA-WED neuron cell body from flies expressing PKA reporter SPARK (TH-F3-Gal4>UAS-PKA-SPARK) in age-matched male, virgin female, and mated female flies. Scale bar, 5 μm. (E) Normalized PKA SPARK, calculated as the ratio of GFP intensity in the droplets over total GFP in the cell body. n=10–16. (F) Daily protein intake per fly in mated female flies of controls (WT>UAS-PKA-mC* or TH-C-Gal4>WT) or flies overexpressing mC* in DA-WED neurons (TH-C-Gal4>UAS-PKA-mC*) from day 1 to day 5. The wing defects in these animals caused many of them to drown in the food, we therefore only had enough flies to survive for 5 days of recording. n=30–32. Unpaired t-tests were used for Figure 5A. A mixed-effects model with the Geisser-Greehouse correction was used to analyze daily food intake data in Figures 5B and 5F, followed by Šidák’s multiple comparisons test for Figure 5B and Tukey’s multiple comparisons test for Figure 5F. ANOVA followed by Tukey’s multiple comparisons test was used for Figure 5E. See also Figure S6.

Figure 6.. MS-WED neurons signal to raise the protein intake setpoint

(A) Whole-mount brain immunostaining of TH-F3-Gal4>UAS-mCD8-GFP with antibodies against GFP (green), and MS (red). High-magnification sections of wedge neuropil are shown on the right. Scale bar, 50 μm. Dashed ellipses indicate the wedge neuropil. (B) Representative images from a GRASP experiment between DA-WED (labeled by TH-F3-Gal4) and MS-WED (labeled by MS-2A-LexA) neurons. Scale bar, 50 μm. Dashed ellipses indicate the wedge neuropil. (C) Whole-mount brain images from MCFO analysis for MS-WED neurons (R57C10-FLP2, MS-2A-Gal4>HA-V5-FLAG) immunostained with anti-V5. Scale bar, 50 μm. Dashed ellipses indicate the wedge neuropil. Arrow indicates the cell body. (D) Location and EM image of a representative synapse between a pre-synaptic MS-WED neuron (SCL.WED.1 in flywire database, annotated in green) and a post-synaptic DA-WED neuron (IPS.290 in flywire database, annotated in magenta). In the EM image, green and magenta asterisks denote the pre- and post-synaptic sides of the synapse, respectively, and the orange arrow indicates the synaptic cleft. Scale bar, 500 nm. (E-H) Representative traces (E), spontaneous action potential firing rate (F), the resting membrane potential (G), and the burst firing frequency (H) for patch-clamp recordings of MS-WED neurons from MS-2A-Gal4>UAS-mCD8-GFP in age-matched male (n=8), virgin female (n=8), and mated female (n=8) flies. (I-J) Representative images (I) and Quantification (J)of BRP puncta in the wedge region of MS-WED neurons (MS-2A-Gal4>UAS-STaR-Brp) in male, virgin female, and mated female flies. Scale bar, 10 μm. n=14–18. (K) Daily protein intake per fly in mated female flies of controls (WT>UAS-Kir2.1 or MS-2A-Gal4>WT) or MS neurons silenced (MS-2A-Gal4>UAS-Kir2.1) from day 1 to day 10. n=38–64. (L) Daily protein intake per fly in virgin female flies of controls (WT>UAS-NachBac or MS-2A-Gal4>WT) or MS neurons activated (MS-2A-Gal4>UAS-NachBac). n=29–32. Kruskal-Wallis tests were performed with Dunn’s multiple comparisons test for Figures 6F, 6H and 6J. ANOVA followed by Tukey’s multiple comparisons test was used to analyze Figure 6G. A mixed-effects model with the Geisser-Greehouse correction followed by Tukey’s multiple comparisons test was used to analyze daily food intake data in Figures 6K–6L. See also Figure S7.

Figure 7.. Model

The resting membrane potential of DA-WED neurons is set to different values in male, virgin female, and mated female flies, similar to the gear shifts in manual cars. This defines different ranges for action potential firing rate and programs the protein intake setpoint. Within these individual ranges, protein starvation can increase the firing rate and the protein intake, without altering the protein intake setpoint. In males, stimulated FMRFa-FMRFaR and inhibited MS-MSR2 signaling increase the activities of both PKC and PKA. These kinases additively phosphorylate ORK1 proteins and lead to increased channel open probability and hyperpolarized resting membrane potential, which lowers the protein intake setpoint. In virgin females, MS-MSR2 signaling remains at low activity, while FMRFa-FMRFaR signaling is also inhibited. This causes PKA to be the only active kinase that phosphorylates part of the consensus sites, leading to decreased ORK1 channel open probability, which depolarizes the resting membrane potential and elevates the protein intake setpoint, compared to males. Following mating in female flies, stimulated MS-MSR2 signaling inhibits PKA. Inactivation of both PKC and PKA, by reducing the level of phosphorylation in ORK1, decreases the open probability of the channel furthermore. This leads to additional depolarization of the resting membrane potential and raising of the protein intake setpoint.