Characterization of a set of abdominal neuroendocrine cells that regulate stress physiology using colocalized diuretic peptides in Drosophila

Multiple neuropeptides are known to regulate water and ion balance in Drosophila melanogaster. Several of these peptides also have other functions in physiology and behavior. Examples are corticotropin-releasing factor-like diuretic hormone (diuretic hormone 44; DH44) and leucokinin (LK), both of which induce fluid secretion by Malpighian tubules (MTs), but also regulate stress responses, feeding, circadian activity and other behaviors. Here, we investigated the functional relations between the LK and DH44 signaling systems. DH44 and LK peptides are only colocalized in a set of abdominal neurosecretory cells (ABLKs). Targeted knockdown of each of these peptides in ABLKs leads to increased resistance to desiccation, starvation and ionic stress. Food ingestion is diminished by knockdown of DH44, but not LK, and water retention is increased by LK knockdown only. Thus, the two colocalized peptides display similar systemic actions, but differ with respect to regulation of feeding and body water retention. We also demonstrated that DH44 and LK have additive effects on fluid secretion by MTs. It is likely that the colocalized peptides are coreleased from ABLKs into the circulation and act on the tubules where they target different cell types and signaling systems to regulate diuresis and stress tolerance. Additional targets seem to be specific for each of the two peptides and subserve regulation of feeding and water retention. Our data suggest that the ABLKs and hormonal actions are sufficient for many of the known DH44 and LK functions, and that the remaining neurons in the CNS play other functional roles. Electronic supplementary material The online version of this article (doi:10.1007/s00018-017-2682-y) contains supplementary material, which is available to authorized users.


Figure S1 :
Figure S1: LK and DH44 expression in the larval Drosophila CNS.(A) Lk-GAL4 drives expression in five pairs of neurons in the brain; however, four of these pairs do not display any LK-immunoreactivity [20].(B) These four pairs of neurons display ITP-immunoreactivity. (C) Three pairs of neurons in the subesophageal ganglion express Lk in larval Drosophila.(D) Seven pairs of neurons in the larval ventral nerve cord (VNC) express Lk.(E) A schematic of LK-expressing neurons in the larval brain and VNC of Drosophila.Neurons displaying LK-immunoreactivity are labelled in red and neurons displaying ITP-immunoreactivity are labelled in black.(F) DH44-GAL4 driven GFP and DH44-immunoreactivity is present in three pairs of median neurosecretory cells in the larval brain.(G) DH44 is expressed in several neurons, with strong expression seen in seven pairs of neurosecretory cells in the larval VNC.In both F and G, there are some neurons that contain GFP but do not contain DH44immunoreactivity.

Figure S2 :
Figure S2: LK expression in adult Drosophila brain.LK-GAL4 drives weak GFP expression in four pairs on neurons in the adult brain.The location of these cells is indicated by white boxes.Scale bar: 50 µm.

Figure S3 :
Figure S3: LK and DH44 are co-expressed in neurons of the ventral nerve cord but not in the brain of larval Drosophila.(A) DH44-GAL4 driven GFP is not co-localized with LK-immunoreactivity in the larval brain.(B) DH44-GAL4 driven GFP is co-localized with LK-immunoreactivity in all seven pairs of abdominal LK neurons (ABLKs) in the ventral nerve cord (VNC) (C) Lk-GAL4 driven GFP is not co-localized with DH44-immunoreactivity in the larval brain.(D) Lk-GAL4 driven GFP is co-localized with DH44immunoreactivity in ABLKs in the larval VNC.

Figure S5 :
Figure S5: Number of LK-immunoreactive neurons in the VNC following knockdown of DH44 and Lk using DH44-GAL4.Lk knockdown but not DH44 knockdown causes a significant decrease in the number of LKimmunoreactive neurons that could be detected in the adult VNC.(**** p < 0.0001, as assessed by One-way ANOVA).

Figure S6 :
Figure S6: Knockdown of DH44 using DH44-GAL4 impacts stress resistance and feeding in Drosophila.DH44-GAL4 driven DH44 knock down results in a significant increase in survival compared to control flies under (A) desiccation (compared to the GAL4 control), (B) starvation and (C) ionic stress (artificial food supplemented with 4% NaCl).Data are presented in survival curves and the error bars represent standard error (**** p < 0.0001, as assessed by Log-rank (Mantel-Cox) test) (D) Lk knock down results in a delayed recovery from chill coma.(* p < 0.05, as assessed by Log-rank (Mantel-Cox)test) (E)There is no significant difference (One-way ANOVA) in feeding as measured by capillary feeding (CAFE) assay between Lk knock down and control flies.Results are presented as cumulative food intake over four days.(F) There is no significant difference in wet weight, dry weight and water content of DH44-knockdown and control flies.Legend for B-F is the same as the one in A.

Figure S7 :
Figure S7: Knockdown of Lk in ABLKs with DH44-GAL4 does not influence LK-stimulated Malpighian tubule secretion ex vivo.(A) Secretion rates of 10 -9 M LK stimulated MTs isolated from DH44 > w 1118 (n = 14) or DH44 > Lk RNAi flies (n = 25).(B) Secretion rates of 10 -10 M LK stimulated MTs isolated from DH44 > w 1118 (n = 10) or DH44 > Lk RNAi flies (n = 12).For both A and B, secretion rates were measured at 10 min intervals for 30 min before and after the addition of peptide (indicated with an arrow).(C, D) Change (%) in secretion determined by comparing the secretion rate over the first 30 min to the maximum secretion rate following peptide application.The legend and sample size for C and D are the same as the one in A and B, respectively.Asterisk indicates significantly different secretion rate compared to basal secretion rate (secretion rate prior to the addition of peptide).(NS = not significant, * p < 0.05, ** p < 0.01, *** p < 0.001; Mann-Whitney U test).