The thirsty fly: Ion transport peptide (ITP) is a novel endocrine regulator of water homeostasis in Drosophila

Abstract

Animals need to continuously adjust their water metabolism to the internal and external conditions. Homeostasis of body fluids thus requires tight regulation of water intake and excretion, and a balance between ingestion of water and solid food. Here, we investigated how these processes are coordinated in Drosophila melanogaster. We identified the first thirst-promoting and anti-diuretic hormone of Drosophila, encoded by the gene Ion transport peptide (ITP). This endocrine regulator belongs to the CHH (crustacean hyperglycemic hormone) family of peptide hormones. Using genetic gain- and loss-of-function experiments, we show that ITP signaling acts analogous to the human vasopressin and renin-angiotensin systems; expression of ITP is elevated by dehydration of the fly, and the peptide increases thirst while repressing excretion, promoting thus conservation of water resources. ITP responds to both osmotic and desiccation stress, and dysregulation of ITP signaling compromises the fly's ability to cope with these stressors. In addition to the regulation of thirst and excretion, ITP also suppresses food intake. Altogether, our work identifies ITP as an important endocrine regulator of thirst and excretion, which integrates water homeostasis with feeding of Drosophila.

Fig 1. ITP regulates water homeostasis.

(A) Short term (6 h) desiccation stress efficiently reduces body fluids. Two-tailed Student’s t–test: P < 0.001. (B) Desiccation stress increases expression of the ITP gene. A primer pair that detects all 5 isoforms (transcripts RC, RD, RE, RF and RG, FlyBase FB2017_06) was used. Two-tailed Student’s t–test: P < 0.05. (C) Desiccation stress increases the abundance of the RE transcript. Primer pair that detects solely this isoforms was used. Two-tailed Student’s t–test: P < 0.05. (D) Daughterless-GAL4 driven ITP RNAi (da>ITPi) results in developmental lethality. Fischer’s exact test: P < 0.001 for comparison with each control. Animals were analyzed in three replicates, Fischer’s exact test was done on pooled data. Sample size: da-GAL4 n = 595; ITPi n = 439; da>ITPi n = 566. (E) Daughterless-GeneSwitch-driven over-expression of ITP (daGS>ITP) increases the proportion of body fluids. Two-tailed Student’s t–test: P < 0.001. (F) daGS driven ITPi (daGS>ITPi) decreases the proportion of body fluid. Two-tailed Student’s t–test: P < 0.001. (G) Over-expression of ITP decreases desiccation resistance. Log-rank test: P < 0.001. Sample size: daGS>ITP off n = 83; daGS>ITP on = 70. (H) ITP RNAi decreases desiccation resistance. Log-rank test: P < 0.001. Sample size: daGS>ITPi off n = 88; daGS>ITPi on n = 89.

Fig 2. ITP regulates desiccation survival independently of its action on the water storage under ad libitum feeding.

(A) Impl2-GAL4 is a suitable driver to target several of the ITP-expressing neurons. Impl2-GAL4-driven expression of GFP partially overlaps with the ITP immunoreactivity (ITPir); scale bars 50 μm. Upper panel: brain neurons. Note the ipc-1 and ipc-2a neurosecretory cells are covered by the Impl2 expression pattern, whereas the rest of the ipc-2 (marked in middle figure) and ipc-3 interneurons (asterisks) are not. Median panel: neurons in the terminal abdominal ganglia that express ITP do not express of Impl2>GFP. iag = ITPir abdominal ganglia neurons. Lower panel: peripheral lateral bipolar dendrite (LBD) neurons in the seventh and eighth abdominal segment express both Impl2>GFP and ITP (ant anterior, lat lateral). (B) Summary table listing partially overlapping expression pattern of Impl2-GAL4 and ITP. (C) ITPi driven by both Impl2-based TARGET and da-GeneSwitch effectively decreases ITP mRNA. Two-tailed Student’s t–test: P < 0.01 for the TARGET manipulations and P < 0.001 for the GeneSwitch manipulation. F1 generation of the cross between the ITPi line and w¹¹¹⁸ was used as a control (off conditions) for the TARGET-based experiment. (D) ITPi driven by the Impl2- based TARGET does not affect the proportion of body fluids. Two-tailed Student’s t–test: P > 0.05 for both comparisons with controls. (E) ITPi driven by the Impl2-based TARGET reduces survival under desiccation. Log-rank test: P < 0.001 for comparisons with both controls. Sample size: tub-GAL80^ts; Impl2-GAL4 n = 87; ITPi n = 83; tub-GAL80^ts; Impl2>ITPi n = 49. (F) Mild ITPi driven by the daGS induced by 50 μM RU-486 does not affect body fluids. Two-tailed Student’s t–test: P > 0.05 for comparisons with the non-induced conditions. Strong ITPi driven by the daGS induced by 200 μM RU-486 reduces body water. Two-tailed Student’s t–test: P < 0.05. (G) Mild ITPi driven by the daGS induced by 50 μM RU-486 reduces survival under desiccation. Log-rank test: P < 0.05. The detrimental effect of ITPi on desiccation resistance is dose-dependent; induction by 50 μM RU-486 has a milder effect than induction by the 200 μM RU-486. Log-rank test: P < 0.01. Sample size: 0 μM RU-486 n = 61; 50 μM RU-486 n = 45; 200 μM RU-486 n = 40.

Fig 3. ITP regulates survival under osmotic stress.

(A) Osmotic stress appears to increase expression of the ITP gene. However, when a primer pair that detects all 5 isoforms (transcripts RC, RD, RE, RF and RG, FlyBase FB2017_06) is used, a clear tendency is observed but the P value does not reach statistical significance. Two-tailed Student’s t–test: P > 0.05. (B) Osmotic stress increases abundance of the ITP RE transcript. A primer pair that detects solely this isoform was used. Two-tailed Student’s t–test: P < 0.05. (C) Osmotic stress decreases body fluid. Two-tailed Student’s t–test: P < 0.01. (D) Over-expression of ITP decreases survival during osmotic stress. Log-rank test: P < 0.001. Sample size: daGS>ITP off n = 66; daGS>ITP on = 66. (E) Over-expression of ITP increases body water, but does not affect the osmotic stress-induced loss of body water. Two-way ANOVA, ITP over-expression and osmotic treatment as fixed factors. Effect of ITP: P < 0.001, effect of osmotic stress: P < 0.001, effect of ITP × osmotic stress interaction: P > 0.05. See S1 Table for further details. (F) daGS-driven ITP RNAi (ITPi) decreases resistance to osmotic stress. Log-rank test: P < 0.05. Sample size: daGS>ITPi off n = 66; daGS>ITPi on n = 66. (G) daGS-driven ITPi decreases water content under standard conditions (two-tailed Student’s t–test: P < 0.001), and interacts with the effect of osmotic stress. Two-way ANOVA, ITPi and osmotic treatment as fixed factors; effect of ITPi: P < 0.01, effect of osmotic stress: P < 0.05, effect of the interaction: P < 0.001. See S2 Table for further details. (H) ITPi driven by the Impl2-based TARGET decreases osmotic stress resistance. Log-rank test: P < 0.01 for each comparison with controls. Sample size: tub-GAL80^ts; Impl2-GAL4 n = 69; ITPi n = 66; tub-GAL80^ts; Impl2>ITPi n = 75. (I) ITPi driven by the Impl2-based TARGET does not affect water content under standard conditions (two-tailed Student’s t–test: P > 0.05 for each comparison with controls), nor interacts with the effect of osmotic stress (two-way ANOVA, ITPi and osmotic treatment as fixed factors; effect of the interaction: P > 0.05. See S3 Table for further details. Thus, the Impl2-based TARGET enables disentangling the requirement for ITP under osmotic stress from its requirement for water preservation.

Fig 4. ITP regulates feeding.

(A) Schematic drawing of the method to measure hunger as propensity to initiate feeding. (B) Over-expression of ITP does not affect the propensity to start feeding. Fischer’s exact test: P > 0.05 at all tested time points. Animals were analyzed in four replicates, Fischer’s exact test was conducted using pooled data. Sample size: daGS>ITP off: n = 100 (1 h), n = 96 (1.5 h), n = 78 (3 h); daGS>ITP on: n = 100 (1 h), n = 86 (1.5 h), n = 76 (3 h). (C) ITPi does not affect the propensity to start feeding. Fischer’s exact test: P > 0.05 at all tested time points. Animals were analyzed in four replicates, Fischer’s exact test was done using pooled data. Sample size: daGS>ITPi off: n = 100 (1 h), n = 100 (1.5 h), n = 87 (3 h); daGS>ITPi on: n = 100 (1 h), n = 100 (1.5 h), n = 85 (3 h). (D) Schematic drawing of the capillary feeding system, which measures the total volume of food eaten during a given period of time. (E) Over-expression of ITP decreases food intake. Two-tailed Student’s t–test: P < 0.01. (F) ITPi increases food intake. Two-tailed Student’s t–test: P < 0.05.

Fig 5. ITP regulates the pace of transit through the digestive tract and the number of defecation events.

(A) Schematic drawing of the assay to measure defecation rate. All flies started to feed on the blue-dyed food at the same time, and the numbers of colored feces were counted every hour after the transfer on the blue dye food. (B) Over-expression of ITP decreases defecation rate. Two-way ANOVA, ITP and time as fixed factors; effect of ITP P < 0.01, effect of time: P < 0.001, effect of the interaction: P > 0.05. See S4 Table for further details. (C) ITPi increases defecation rate. Two-way ANOVA, ITPi and time as fixed factors; effect of ITPi P < 0.001, effect of time: P < 0.001, effect of the interaction: P > 0.05. See S5 Table for further details. (D) Over-expression of ITP decreases defecation events per fly. Two-tailed Student’s t–test: P < 0.01. (E) ITPi increases defecation events per fly. Two-tailed Student’s t–test: P < 0.05. (F) Over-expression of ITP decreases the size of feces. Two-tailed Student’s t–test: P < 0.05. (G) ITPi decreases the size of feces. Two-tailed Student’s t–test: P < 0.001.

Fig 6. ITP regulates thirst and water intake.

(A) Schematic drawing of a system to measure thirst as propensity to initiate water intake. (B) Over-expression of ITP increases thirst, i.e. reduces the time to start spontaneous drinking. Significant differences are indicated by asterisk symbols. Fischer’s exact test: * P < 0.05; *** P < 0.001. Animals were analyzed in three replicates. Fischer’s exact test was done on pooled data. Sample size: daGS>ITP off: n = 47 (1 h), n = 45 (1.5 h), n = 45 (3 h); daGS>ITP on: n = 46 (1 h), n = 44 (1.5 h), n = 43 (3 h). (C) ITPi reduces thirst, i.e. extends the time until flies initiate drinking. Fischer’s exact test: * P < 0.05. Animals were analyzed in three replicates. Fischer’s exact test was done using pooled data. Sample size: daGS>ITPi off: n = 48 (1 h), n = 42 (1.5 h), n = 43 (3 h); daGS>ITPi on: n = 48 (1 h), n = 44 (1.5 h), n = 43 (3 h). (D) Schematic drawing of the capillary drinking system, which measures the total volume of water ingested during a given period of time. Panels (E) and (F) show the mean water intake per day of feeding on a water-deprived food. (E) Over-expression of ITP increases the amount of ingested water. Two-tailed Student’s t–test: P < 0.05. (F) ITPi decreases the amount of ingested water. Two-tailed Student’s t–test: P < 0.001.

Fig 7. Scheme summarizing the roles of ITP in Drosophila, as revealed by this study.

ITP is a central regulator of water balance. Conditions leading to reduced volume of body fluids result in increased expression of ITP gene. ITP subsequently promotes water intake, while inhibiting feeding and water loss by excretion, promoting thus the increase in body fluids and restoration of water balance.