The proprotein convertase encoded by amontillado (amon) is required in Drosophila corpora cardiaca endocrine cells producing the glucose regulatory hormone AKH

Abstract

Peptide hormones are potent signaling molecules that coordinate animal physiology, behavior, and development. A key step in activation of these peptide signals is their proteolytic processing from propeptide precursors by a family of proteases, the subtilisin-like proprotein convertases (PCs). Here, we report the functional dissection of amontillado (amon), which encodes the Drosophila homolog of the mammalian PC2 protein, using cell-type specific inactivation and rescue experiments, and we show that amon is required in the islet-like adipokinetic hormone (AKH)-producing cells that regulate sugar homeostasis. In Drosophila, AKH acts analogously to vertebrate glucagon to increase circulating sugar levels from energy stores, while insulin-like peptides (DILPs) act to decrease sugar levels. amon mutant larvae have significantly reduced hemolymph sugar levels, and thus phenocopy larvae where the AKH-producing cells in the corpora cardiaca have been ablated. Reduction of amon expression in these cells via cell-specific RNA inactivation also results in larvae with reduced sugar levels while expression of amon in AKH cells in an amon mutant background rescues hypoglycemia. Hypoglycemia in larvae resulting from amon RNA inactivation in the AKH cells can be rescued by global expression of the akh gene. Finally, mass spectrometric profiling shows that the production of mature AKH is inhibited in amon mutants. Our data indicate that amon function in the AKH cells is necessary to maintain normal sugar homeostasis, that amon functions upstream of akh, and that loss of mature AKH is correlated with loss of amon activity. These observations indicate that the AKH propeptide is a proteolytic target of the amon proprotein convertase and provide evidence for a conserved role of PC2 in processing metabolic peptide hormones.

Figure 1. Larvae lacking functional amon have reduced hemolymph sugar levels.

(A) Bars indicate combined glucose and trehalose hemolymph levels in control siblings (black) and amon^Q178st mutants (white). Hemolymph carbohydrate levels were measured in control and amon mutant larvae collected 3 h (left bars) and 24 h (center bars) after the last in a series of three heatshocks (at 36, 60, and 84 h AEL) and in control and amon mutant larvae collected 3 h (right bars) after the last of a series of four heatshocks (at 36, 60, 84, and 108 h AEL). (B) amon-gal4 drives expression of uas-CD8-GFP in the corpora cardiaca (CC) cells of the ring gland (white arrows). (C) In situ hybridization of an amon probe to the ring gland. White arrows indicate signal in the ring gland CC cells. (D)AKH cells are visualized using an α-AKH antibody. Signal from amon-gal4 (F) co-localizes to the AKH cells (E). n = number of larvae assayed; larvae were pooled in groups of three. **p<0.0001, Students T-Test.

Figure 2. amon is required in the AKH producing cells for normal sugar regulation.

(A) The black bar indicates amon transcript levels in control larvae, while the white bar indicates amon transcript levels when amon-RNAi is ubiquitously expressed. Primers specific to amon exon 10 were used to assess amon transcript levels by quantitative real time PCR. (B) Dorsal and ventral views of a control pupa (top). Middle panels represent amon mutants that are unable to complete metamorphosis, and die with defects in head eversion and abdominal differentiation. amon RNAi knockdown animals also die with phenotypes similar to amon mutants (bottom). (C) Combined glucose and trehalose levels of control larvae are shown in the black bar. The center blue bar shows hemolymph sugar levels in AKH ablated larvae, while the gray bar represents animals in which amon expression has been reduced in the AKH producing cells by RNAi. n = number of larvae assayed; larvae were pooled in groups of three. **p<0.0001, one-way ANOVA.

Figure 3. Expression of amon in the AKH cells of an amon^C241Y mutant is sufficient to rescue hypoglycemia.

The gray bar represents larvae in which amon expression has been restored in the AKH producing cells (yw; uas-amon/hs-amon; Df(3R) Tl-X e/akh-gal4, amon^C241Y) as compared to amon mutants (yw; uas-amon/hs-amon; Df(3R) Tl-X e/amon^C241Y white bar) and control siblings (yw; uas-amon/hs-amon; Df(3R) Tl-X e or amon^C241Y/TM3 Sb Ser y+ e, black bar). n = number of larvae assayed; larvae were pooled in groups of three. p<0.0015, one way ANOVA.

Figure 4. Ubiquitous expression of AKH rescues the hypoglycemic defect seen in amon knockdown larvae.

The left black bar represents wild-type levels of combined glucose and trehalose (yw; hs-akh/+; akh-gal4/+). The left blue bar represents combined sugar levels of AKH ablated larvae (yw; hs-akh/uas-reaper; akh-gal4/+) while the left gray bar shows glucose and trehalose levels in which amon has been reduced in the AKH cells by RNAi (yw; hs-akh/uas-amon-RNAi^28b; akh-gal4). Bars denoted with a ‘+’ below the graph indicate combined glucose and trehalose levels following heatshock induced expression of akh via a hs-akh transgene. n = number of larvae assayed; larvae were pooled in groups of three. *p = 0.002, **p<0.0001, one-way ANOVA.

Figure 5. Direct peptide profiling of AKH and AKHGK in control and amon^C241Y flies.

(A) Model of the processing of the AKH prepropeptide (top) and profiling of the larval ring gland (left) and adult corpora cardiaca (right). AKH is processed by a concerted action of a signal peptidase (SP) and amon, likely followed by a two-step carboxypeptidase (CP) action that first removes the C-terminal R yielding the intermediate AKHGK. AKHG is than amidated to bioactive AKH (not shown). While AKH and AKHGK were detected in most preparations from control and rescued (continued heatshock once a day) flies, they were not detectable in amon larvae. (B) Original direct mass profiles from corpora cardiaca of adult control (above) and amon (below) flies. AKH only occurs as the characteristic [M+Na]⁺ and [M+K]⁺ adducts, whereas AKHGK also occurs as [M+H]⁺. In the control fly, both peptides show higher signal intensities as the stable isotope-labelled standard peptide (AKH*). In the amon fly, the signal intensity is clearly higher for AKH* than for the native peptides. As previously reported , , no other mass peaks occur in the range 990-1220 Da in direct mass spectrometric CC profiles. (C) Standard curve for adult corpora cardiaca obtained with a dilution series of AKH* added to the matrix salt, male OrR wild-type flies. The y axis shows the signal intensity ratio of native AKH/AKH*. Error bars are S.E.M. The relationship of AKH/AKH* is linear for AKH* concentrations of 50–500 nM. (D) Peptide quantification with the labeled AKH* standard at 400 nM. The concentrations of both AKH and AKHGK are significantly reduced in amon flies vs. controls five days after eclosion and final heatshock. *p<0.05, **p<0.01, Mann-Whitney.