Drosophila melanogaster NEP2 is a new soluble member of the neprilysin family of endopeptidases with implications for reproduction and renal function

Abstract

The mammalian neprilysin (NEP) family members are typically type II membrane endopeptidases responsible for the activation/inactivation of neuropeptides and peptide hormones. Differences in substrate specificity and subcellular localization of the seven mammalian NEPs contribute to their functional diversity. The sequencing of the Drosophila melanogaster genome has revealed a large expansion of this gene family, resulting in over 20 fly NEP-like genes, suggesting even greater diversity in structure and function than seen in mammals. We now report that one of these genes (Nep2) codes for a secreted endopeptidase with a highly restricted pattern of expression. D. melanogaster NEP2 is expressed in the specialized stellate cells of the renal tubules and in the cyst cells that surround the elongating spermatid bundles in adult testis, suggesting roles for the peptidase in renal function and in spermatogenesis. D. melanogaster NEP2 was found in vesicle-like structures in the syncytial cytoplasm of the spermatid bundles, suggesting that the protein was acquired by endocytosis of protein secreted from the cyst cells. Expression of NEP2 cDNA in D. melanogaster S2 cells confirmed that the peptidase is secreted and is only weakly inhibited by thiorphan, a potent inhibitor of human NEP. D. melanogaster NEP2 also differs from human NEP in the manner in which the peptidase cleaves the tachykinin, GPSGFYGVR-amide. Molecular modelling suggests that there are important structural differences between D. melanogaster NEP2 and human NEP in the S1' and S2' ligand-binding subsites, which might explain the observed differences in inhibitor and substrate specificities. A soluble isoform of a mouse NEP-like peptidase is strongly expressed in spermatids, suggesting an evolutionarily conserved role for a soluble endopeptidase in spermatogenesis.

Figure 1. ClustalW alignment of the amino acid sequences of D. melanogaster NEP2 (DmNEP2), human NEP (HuNEP), human MMEL2 (HuMMEL2) and human ECE1 (HuECE1)

—, signal peptide; |, predicted cleavage site for a signal peptidase; *, potential N-glycosylation sites conserved in all four proteins; =, conserved active site residues; ↑, conserved cysteines; □, cysteine involved in the formation of the disulphide-linked ECE-1 dimer; open up arrow, amino acids that contribute side-chains to the S₁′ subsite of human NEP.

Figure 2. Distribution of NEP2 mRNA and protein in adult Malpighian tubules

(A and B) In situ hybridization was carried out using a digoxygenin-labelled Nep2 anti-sense riboprobe. (C and D) Immunocytochemistry was performed using a Vectastain® ABC Kit and NEP2 antibodies. High levels of NEP2 mRNA and protein were detected in the stellate cells and the related bar-shaped cells, but not the principal cells of the Malpighian tubules. Control experiments performed with either a sense-strand riboprobe or non-immune IgG gave no specific staining of the Malpighian tubules (results not shown).

Figure 3. Distribution of NEP2 mRNA and protein in the adult testes

(A) In situ hybridization. Strong staining is present in cyst cells at the base of the testis (arrow). Staining is also apparent in tail cyst cells of elongating cysts (solid arrowhead) and in tail cyst cells surrounding the waste bags (open arrowhead). (B–D) Immunofluorescence studies. (B) Intact testis. Strong fluorescence is observed at the tail end of elongating cysts at approx. 75% of the length of the testis (arrow) and weaker fluorescence is observed at the tail end of fully elongated cysts (arrowhead). (C) Disrupted testis. Fluorescence is associated with the tail end of elongated cysts (e.g. closed arrowhead), but is weaker in more fully elongated cysts (open arrowhead). (D) Longitudinal optical section through the tail end of an elongating cyst. Punctate fluorescence is apparent within the spermatid bundle and is strongest at the tail end (arrowhead). Scale bars: (A–C) 50 μm; (D) 10 μm.

Figure 4. Demonstration of a soluble NEP2 in the Malpighian tubules and testes of adult flies

(A) Endopeptidase activity was measured in soluble extracts of Malpighian tubules (Mt) and testes by determining the rate of cleavage of the Gly–Val bond of the GPSGFYGVR-amide peptide using HPLC to quantify the reaction product GPSGFYG. Reactions were carried out in 100 mM Tris/HCl, pH 7.5, in either the absence or presence of 100 μM phosphoramidon (P), as described in the Experimental section. The results are expressed as the means±S.E.M. (n=3). (B, C) Western blot analysis of the distribution of NEP2 in membrane (P) and soluble (S) fractions prepared from Malpighian tubules from 10 adult wild-type flies and testes (5 pairs per lane) from wild-type and tudor flies. Equivalent amount of membrane and soluble fractions were loaded on to the SDS/PAGE gel. Blotted proteins were probed with purified NEP2 antibody as the primary antibody to detect a protein band of approx. 90 kDa.

Figure 5. Immunoprecipitation of the soluble endopeptidase activity of Malpighian tubules and testes from adult flies

(A) Endopeptidase activity was measured in soluble fractions from Malpighian tubules and testes by determining the rate of hydrolysis of the Gly–Val peptide bond of LomTK-1 before and after immunoprecipitation with purified D. melanogaster NEP2 antibodies/Protein A–Sepharose. In control experiments, the D. melanogaster NEP2 antibody was replaced with non-immune rabbit IgG at the same protein concentration. The results are expressed as the mean of two independent experiments and were within a range ±5%. (B) The Protein A–Sepharose beads were analysed by Western blotting using the D. melanogaster NEP2 antibody (NEP2Ab) to probe for immunoprecipitated NEP2 in the soluble fraction prepared from Malpighian tubules (Mt) and testes (T). Specificity was controlled using IgG from non-immunized rabbits (Sigma–Aldrich).

Figure 6. Characterization of soluble recombinant D. melanogaster NEP2 expressed in S2 cells

(A) Time-course for the hydrolysis of LomTK-I by the conditioned medium from a culture of S2 cells over-expressing NEP2 (◆) and control S2 cells transformed with an empty vector (■). In this experiment aliquots (5 μl) of the unconcentrated media were incubated with LomTK-1 (100 μM, final concentration) in 100 mM Tris/HCl, pH 7.5, for 3 h before termination of the reaction and HPLC analysis, as described in the Experimental section. The extent of hydrolysis was determined by quantifying the decline in the parent peptide with time. No hydrolysis was observed when the peptide was incubated with medium from cells transfected with empty vector. (B) Western blot of recombinant NEP2 expressed in S2 cells. S2 cells transfected with either empty vector or Nep2 cDNA were grown to confluency before being maintained for 24 h in serum-free medium. Medium was harvested, dialysed and concentrated before analysis as described in the Experimental section. Collected cells were lysed with 1% (w/v) Triton X-100 in PBS and an equivalent amount of the media and cell extract were loaded on to SDS/PAGE gels for direct comparison of the distribution of NEP2. (C and D) HPLC chromatograms showing the cleavage of LomTK-1 by the medium (C) and detergent-lysed cells (D) from a culture of S2 cells expressing NEP2 under the control of the inducible metallothionein promoter. The only hydrolysis product detected by HPLC was identified as GPSGFYG by co-chromatography with authentic material. The predicted second cleavage product, VR-amide, was not detectable with our chromatography system. Equivalent amounts of medium and cells were incubated with the peptide substrate in 100 mM Tris/HCl, pH 7.5, for 1h.

Figure 7. Inhibition of recombinant D. melanogaster NEP2 activity by inhibitors of human NEP/ECE

Culture medium (2 μl) from S2 cells transiently expressing Nep2 was incubated with LomTK-1 (100 μM) in 100 mM Tris/HCl, pH 7.5, in the absence and presence of different concentrations of phosphoramidon (■) and thiorphan (□). The rate of cleavage of the Gly–Val peptide bond of LomTK-1 was determined by HPLC quantification of the reaction product (GPSGFYG) at 214 nm. The results are expressed as a percentage of the uninhibited activity.

Figure 8. Model of D. melanogaster NEP2 with LomTK-1 docked at the active site

A ribbon diagram showing the active-site space with substrate and metal ion co-ordinating residues. The peptide substrate LomTK-1 is shown in yellow and the ligand-binding subsites S₁′ and S₂′ are shown with their side-chains in grey and light blue respectively. Associated helices and zinc co-ordinating ligands are shown in magenta.