Proteins belonging to the family of neprilysins are typically membrane bound M13 endopeptidases responsible for the inactivation and/or activation of peptide signaling events on cell surfaces. Mammalian neprilysins are known to be involved in the metabolism of various regulatory peptides especially in the nervous, immune, cardiovascular and inflammatory systems. Although there is still much to learn about their participation in various diseases, they are potential therapeutic targets. Here we report on the identification and first characterization of neprilysin 4 (NEP4) from Drosophila melanogaster. Reporter lines as well as in situ hybridization combined with immunolocalization demonstrated NEP4 expression during embryogenesis in pericardial cells, muscle founder cells, glia cells and male gonads. Western blot analysis confirmed the prediction of one membrane bound and one soluble isoform, a finding quite unusual among neprilysins with presumably strong physiological relevance. At least one NEP4 isoform was found in every developmental stage indicating protein activities required throughout the whole life cycle of Drosophila. Heterologously expressed NEP4 exhibited substrate preferences comparable to human neprilysin 2 with distinct cleavage of substance P and angiotensin I.

Since the identification and characterization of the first neprilysin from rabbit renal brushborder membranes (Kerr and Kenny, 1974), the family of M13 zinc metallopeptidases has become increasingly important. Members of this family in general and neprilysin in particular are subjects of various medical investigations regarding Alzheimer's disease (Iwata et al., 2000), hypertension(Molinaro et al., 2002),analgesia (Whitworth, 2003)and the progression of cancers (Turner et al., 2001). These investigations focus on the enzyme's ability to hydrolyze signaling peptides like enkephalins, bradykinins, tachykinins, the atrial natriuretic factor, substance P or even the neurotoxic amyloidβ-peptide (Roques et al.,1993). A disregulation of the extracellular peptide homeostasis may influence the corresponding signaling pathways and account for the above mentioned diseases.

In mammals, seven members of the M13 family are known including neprilysin,endothelin-converting enzymes (ECE-1, ECE-2), the KELL blood group protein and PHEX (Turner and Tanzawa,1997). However, currently only limited data are available for most of these proteins and a crystal structure is known only in individual cases(Oefner et al., 2000). Among the members of the M13 family, human neprilysin is the best characterized. The protein specifically cleaves N-terminal peptide bonds at aromatic and bulky hydrophobic amino acids (Hersh and Morihara, 1986), and is potently inhibited by phosphoramidon from Streptomyces (Oefner et al.,2000). Human neprilysin is a type II integral membrane protein of 750 amino acids, structured into a short cytoplasmic domain, a membrane spanning region and a large extracellular domain containing the active site with its characteristic HExxH motif, which typically can be found in various other zinc peptidases. In contrast to zinc proteases, however, two distinct protein domains prevent neprilysin from cleaving larger substrates simply by restricting active site access to oligopeptides(Oefner et al., 2000). With the exception of neprilysin 2, which is expressed in a membrane bound state but becomes soluble through proteolytic cleavage(Ghaddar et al., 2000; Ikeda et al., 1999), almost every member of the mammalian M13 family analyzed so far shares this membrane bound topology. In the vast majority of cases, however, the identification of in vivo functions and substrates is the main task for future investigations.

Here we report on the characterization of neprilysin 4 (NEP4) from Drosophila melanogaster Meigen. In contrast to almost every other neprilysin analyzed so far, NEP4 from Drosophila is expressed not only as a membrane bound but also as a soluble protein. The two solubility states are due to two different splice variants with variable N-termini: in contrast to the larger isoform A, isoform B lacks the N-terminal cytosolic and transmembrane regions and is therefore presumably expressed in a soluble state. RT-PCR as well as northern and western blot analysis validated protein expression in every developmental stage. Reporter lines and in situhybridization combined with immunostaining identified NEP4 in a subset of pericardial cells, in three dorsal muscle founder cells and in numerous types of glia cells during embryonic development. In larval and adult flies, NEP4 is present in the nervous system and in the testes. Activity assays with heterologously expressed enzyme demonstrate that the peptides substance P and angiotensin I are cleaved with high efficiency while other peptides like bradykinin, tachykinin or pigment dispersing factor (PDF) have to be considered as poor substrates.

Sequence analysis, RT-PCR, in situ hybridization and northern blot

nep4 transcript predictions (FBgn0038818, http://flybase.org)were verified by RT-PCR. Total-RNA (RNeasy Mini Kit, Qiagen, Hilden, Germany)from different developmental stages was treated with DNase I (Invitrogen,Carlsbad, CA, USA) according to the manufacturer's instructions and used as a template for cDNA synthesis (AMV First Strand cDNA Synthesis Kit for RT-PCR,Roche, Mannheim, Germany). The two transcript variants were amplified with the following primer pairs: transcript A (3123 bp): atgagtcgccacagccaactg(forward, FW), ctaccaaacgctgcacttttt (reverse, RV); transcript B (2937 bp):tgaagtgtggtgcaaccataa (FW), ctaccaaacgctgcacttttt (RV). To ensure template specific priming, transcript B was amplified with a forward primer annealing in the 5′-UTR, immediately upstream of the predicted translational start of this splice variant. Amplification products were cloned into the pGEM-T vector (Promega, Madison, WI, USA) and sequenced.

Templates for riboprobe synthesis were generated with primer pairs specific to either nep4 transcript A (FW: atgagtcgccacagccaactg; RV:ctggaagaagtagaagcattt, 187 bp) or both transcripts (FW: tggtaatgctgccactgaccc;RV: cgcggcgctcccgtattctga, 489 bp). For reasons of limited specific sequence length, adequate riboprobes against transcript B, suitable for whole mount in situ hybridization or northern analysis, could not be generated. Sense and antisense RNA probes were synthesized with a DIG RNA labeling kit(Roche). Hybridizations on whole mount embryos, brains of 3rd instar larvae and adult testes were performed as described(Duan et al., 2001). For fluorescence whole mount in situ hybridization an antisense riboprobe was generated using the nep4 DGRC clone LD25753. Triple staining was done as described previously (Jagla et al., 1997). Northern blots were conducted with total RNA (15 μg per lane) according to standard protocols at a hybridization temperature of 67°C.

Generation, purification and specificity of antibodies

For the generation of polyclonal antibodies a 630 bp fragment of nep4 (FW: tactcagaattcatgagggatctgcggaac; RV:tactcaaagcttgaagccggccttatcctt) was cloned into the pET29b vector and transformed into Escherichia coli Rosetta (DE3) cells (Novagen,Darmstadt, Germany). Following protein expression (3 h, 37°C), cells were harvested by centrifugation (10 min, 5000 g) and sonicated(Branson sonifier 250; Branson Ultrasonics, Danbury, CT, USA). The insoluble protein fraction was isolated by centrifugation (10,000 g,4°C, 20 min), solubilized in CAPS buffer (50 mmol l–1CAPS, 0.3% NLS, 1 mmol l–1 DTT, pH 11.0) and refolded in dialysis buffer (300 mmol l–1 NaCl, 50 mmol l–1 NaH2PO4, 0.1 mmol l–1 DTT, 2.5 mmol l–1 imidazole, pH 8.0). Subsequent to Ni-NTA purification the antigen was used for immunization of two rabbits (Pineda Antibody-Service, Berlin, Germany). The resulting sera were affinity purified and their monospecificity was confirmed by western blotting. Immunohistochemical control stainings were done with preimmune sera, secondary antibody only and antigen blocked purified sera (50 μl of sera were incubated with 40 μg of purified antigen in 1 ml PBS for 1 h at room temperature).

Protein preparation and immunoblotting

Protein extracts from different developmental stages were isolated by three repetitive cycles of homogenization with each cycle including freezing of tissues in liquid nitrogen, thawing on ice and homogenization in a glass–teflon homogenizer. Proteins from larva and pupa were heated at 70°C for 10 min in Laemmli buffer (method B), while embryonic and adult proteins were boiled at 99°C for 3 min in the same buffer (method A). Membrane fractions from homogenized tissues or cells were obtained by differential centrifugation: subsequent to centrifugation at 5000 g (10 min) the supernatant was subjected to ultracentrifugation (100,000 g, 1 h). The resulting membrane and soluble fractions were further processed according to method A. Following SDS-PAGE, separated proteins (10 μg per lane) were transferred to nitrocellulose membranes and analyzed by immunodetection. Purified NEP4 antiserum was applied at a dilution of 1:2000 and visualized by anti-rabbit alkaline phosphatase conjugated antibody (1:10,000; Sigma, St Louis, MO,USA).

Immunohistochemistry

Whole mount stainings on embryos were performed as described(Sellin et al., 2009). Primary antibodies used were mouse anti-Repo (1:5, Developmental Studies Hybridoma Bank, DSHB; University of Iowa, IA, USA), mouse anti-Prospero (1:5, DSHB),rabbit anti-Eve [1:2000, from D. Kosman(Kosman et al., 1998)],guinea-pig anti-Runt [1:2000, from J. Reinitz(Kosman et al., 1998)],guinea-pig anti-Krüppel [1:500, from J. Reinitz(Kosman et al., 1998)] and mouse anti-GFP (1:500, JL-8; Clontech, Mountain View, CA, USA). Prior to application of NEP4 antiserum in PBS (1:500), fixed tissues were incubated in PBS containing 0.05–0.1% SDS (20 min) and blocked with 1% BSA. Brains prepared from 3rd instar larvae were permeabilized as described above before fixation in 1% formaldehyde in PBS for 10 min. Secondary antibodies were conjugated with: Alexa Fluor 488 (1:500), Cy2 (1:200), Cy3 (1:100) or Cy5(1:200; Jackson ImmunoResearch, Newmarket, Suffolk, UK) and diluted in PBS.

Bacterial expression and purification of NEP4

To express NEP4 isoform B (978 amino acids) fused to glutathione S-transferase (GST), the corresponding sequence was cloned into the pGEX-5X-1 vector (GE healthcare, Little Chalfont, Bucks, UK). Escherichia coliRosetta (DE3) were transformed either with the empty vector as a control or with the vector inducing the expression of GST-NEP4B. Cells were grown at 37°C to OD600≈0.5. To induce expression, 0.2 mmol l–1 IPTG was added and cells were incubated at 28°C (5 h). Proteins were purified with glutathione agarose according to the manufacturer's instructions (Machery Nagel, Düren, Germany). For subsequent activity assays, the eluted protein fractions were dialyzed against 50 mmol l–1 Tris, 100 mmol l–1 NaCl(variable pH values).

Insect cell culture and protein expression

Heterologous expression was done in SF21 cells using the Bac-to-Bac baculovirus expression system (Invitrogen). nep4 transcript A (3123 bp) was cloned into the pFastBacDual vector downstream of the polyhedrin promoter. To track transfection efficiency, an enhanced GFP (eGFP) reporter gene was inserted into the same vector (p10 promoter). Infected and non-infected SF21 cells were grown in 175 cm2 flasks to 95%confluency and harvested by centrifugation (100 g, 10 min). Cell disruption was done in PBS with a glass–teflon homogenizer. Membrane fractions were collected by differential centrifugation as explained above and used for activity assays.

Peptidase activity assay

Peptides tested for digestion were substance P(RPKPQQFFGLM-NH2), angiotensin I (DRVYIHPFHL-NH2), Locusta migratoria tachykinin 1 (GPSGFYGVR-NH2),bradykinin (RPPGFSPFR-NH2) and pigment dispersing factor (PDF,NSELINSLLSLPKNMNDA-NH2). For hydrolysis assays, 10 μg of SF21 membrane fractions or 1 μg of GST or GST-NEP4B was incubated with 750 ng of peptide in 50 mmol l–1 Tris, 100 mmol l–1NaCl (variable pH values and incubation times, 35°C). To check for peptidase specific inhibition, phosphoramidon and thiorphan (variable concentrations) were preincubated with the enzyme for 30 min at 35°C prior to addition of peptide. Protein activity was stopped by boiling the samples for 10 min. Cleavage was assayed by HPLC as follows: reaction products were loaded onto a C4 column (Vydac 214TP54; MZ-Analysentechnik, Mainz, Germany) in 5% acetonitrile, 0.03% TFA in H2O and eluted via a linear acetonitrile gradient (final concentration: 80%) over 25 min. UV spectra (211 nm) were recorded during elution and used for quantifications. Mass spectrometry (MS) and tandem mass spectrometry (MS/MS) data (collision induced dissociation) were acquired by an ESI-ion trap (Esquire-HCT, Bruker Daltonics,Bremen, Germany) with sequence information deduced from the obtained MS/MS data by the Mascot search algorithm (Matrix Science; www.matrixscience.com)and the MSDB database(ftp://ftp.ncbi.nih.gov/repository/MSDB/msdb.nam).

Laser scanning microscopy

Confocal images were captured with either a Zeiss LSM 5 Pascal confocal microscope (Zeiss, Jena, Germany) or a Zeiss LSM 510 Meta. Z-stacks are depicted as maximum projections if not denoted otherwise. High magnification images were calculated using Velocity software (Improvision; www.improvision.com).

Drosophila stocks and transgenic fly lines

W1118 was used as wild-type. P{eve-GAL4.eme} was provided by Rolf Bodmer (Burnham Institute, La Jolla, CA, USA), UAS-eGFP was obtained from the Bloomington Stock Keeping Center (BL6874; http://flystocks.bio.indiana.edu/). The nep4-eGFP reporter line was generated by cloning an intronic region of the nep4 gene into the pH Stinger vector(Barolo et al., 2004). Primers used for amplification of the regulatory element (2941 bp) were tactcatctagatccactgaaacgaatccagcg (FW) and tactcagctagccccacgagccgtaattga(RV), giving rise to nep4-up01-eGFP(Fig. 1A). The construct was subjected to P-element based transformation using the MYORES Network of Excellence, technical platform for fly injection at Clermont-Ferrand University (France). Three independent transgenic lines were tested for reporter activity.

NEP4 occurs in two splice variants

Sequence analysis using standard software(http://flybase.org, http://www.bioinformatics.org/sms/index.html)predicts the existence of two NEP4 isoforms. One of them, isoform A, consists of 1040 amino acids (3123 base pairs) and shows a prevalent structure among neprilysin-like proteins: a short N-terminal intracellular domain (56 amino acids) followed by a transmembrane segment (19 amino acids) and a large extracellular part (963 amino acids), which contains the active site with its typical HExxH zinc binding motif. Isoform B, however, consists of only 978 amino acids (2937 base pairs). This reduction in size is due to the absence of the intracellular domain as well as most of the transmembrane domain and leads to the prediction of a soluble protein (TMHMM, v2.0, http://www.cbs.dtu.dk/services/TMHMM-2.0),a state quite unusual among neprilysins in general, which are typically membrane bound enzymes. Apart from these N-terminal differences, the two isoforms are identical (Fig. 1). Initial evidence for the existence of at least two different splice variants can be inferred from expressed sequence tag (EST) data: while several EST clones indicate the existence of transcript A (e.g. BI483296 or BI580864), the clone AI389953 contains sequence data unique to transcript B(http://flybase.org). Using transcript specific primers, we could confirm the presence of both predicted splice variants in all developmental stages by RT-PCR(Fig. 2A). By cloning and sequencing the amplicons, we verified that the isolated cDNAs correspond to the predicted transcripts A and B (FBgn0038818, http://flybase.org). The absence of introns in the respective sequences excluded the possibility of genomic DNA priming.

NEP4 is expressed in every developmental stage

The expression of nep4 during development was further analyzed with specific RNA probes for transcript detection(Fig. 1A) as well as antisera that recognize both NEP4 isoforms (Fig. 1B). In accordance with the results obtained by RT-PCR, nep4 transcripts were detected in embryonic, larval, pupal and adult RNA preparations, suggesting NEP4 plays a role during the whole life cycle of Drosophila. In the course of development, the transcript amounts gradually increase with least abundance in embryonic and greatest in adult stages (Fig. 2B). Since the resolution of the northern gel was not sufficient to clearly separate the predicted transcripts from each other, we did northern blots with a transcript A specific RNA probe. In contrast to transcript B, transcript A harbors a unique 5′ sequence of 186 nucleotides, which allowed the generation of a probe with specificity for this splice variant(Fig. 1A). The blots performed with this probe showed distinct signals in all stages (supplementary materialFig. S2), but with a considerably reduced intensity compared with blots performed with probes detecting both transcripts(Fig. 2B). This result demonstrates that transcript A is expressed in all developmental stages. The reduced intensity, however, indicates that the stronger signal generated by probes without transcript specificity is based on the combined expression of transcripts A and B. This simultaneous expression of the two transcripts was also shown by RT-PCR (Fig. 2A).

To ascertain protein abundance, western blots were done. Monospecificity of the antisera was validated by preimmune sera and antigen blocked controls(supplementary material Fig. S1A). Consistent with the distribution of mRNA,NEP4 protein was present in every developmental stage with expression starting in embryos about 8 h after oviposition. Whilst only the smaller isoform B was detectable in embryos, during larval stages isoform A became traceable as well, yet with a greatly reduced abundance compared with isoform B. At the pupal stage, the two isoforms reached similar expression levels whereas in the adult, NEP4A became the dominant isoform(Fig. 2C). Because of variable protein preparation techniques, necessary to visualize NEP4 in every developmental stage, a significant comparison of protein amounts is only valid for relative isoform abundance within the respective stages. Stage spanning protein amounts could be distorted by variations in preparation procedures. Nevertheless, the results from western blot analysis clearly support the data obtained by RT-PCR that already indicate a dynamic and simultaneous expression pattern of the two splice variants. The observation that, despite the RT-PCR and northern blot based detection of both transcripts, no NEP4A could be detected in embryonic protein extracts(Fig. 2C) is presumably based on weak embryonic expression levels of this isoform and the higher sensitivity of the alternative methods. Thus, in the course of development, in addition to an increase in NEP4 expression levels in general, the predominant isoform changes from NEP4B in early developmental stages to NEP4A in later ones.

NEP4 is expressed in a membrane bound but also in a soluble state

Based on the sequence data mentioned above, one membrane bound and one soluble NEP4 isoform were anticipated. Using a purified antiserum, we were able to confirm this prediction. As expected, in crude lysates of adult flies both isoforms were detected (Fig. 2C). In a membrane fraction, however, only the larger isoform A was present and it migrated almost exactly at its predicted molecular mass of 120 kDa while the smaller isoform B occurred exclusively in the soluble fraction, again close to its theoretical size of 113 kDa(Fig. 3A). This result,together with the presence of nep4B mRNA(Fig. 2A), demonstrates that NEP4B is expressed in a soluble state, which stands in clear contrast to the current opinion that neprilysins are expressed almost exclusively as membrane bound proteins. The presence of a double signal in the soluble fraction(Fig. 3A, arrowhead) remains to be elucidated by additional data. However, posttranslational modifications like glycosylation, which is known to occur in the case of Drosophilaneprilysin 2 (NEP2) (Bland et al.,2006), or protein degradation could account for this observation. Thus, in addition to NEP2, NEP4B may be one of the first members of a new subfamily of soluble neprilysins. In contrast to Drosophila NEP2,however, which exists solely in a soluble state(Thomas et al., 2005), or mammalian NEP2, which becomes a soluble peptidase only after cleavage of a membrane bound precursor protein (Ikeda et al., 1999), the isoform specific solubility of NEP4 shown in this study is apparently based on alternative splicing instead of posttranslational modification, a finding that has never been reported for any neprilysin before.

NEP4 expression is restricted to pericardial cells, dorsal muscle founder cells, the central nervous system and the male germline

Expression of NEP4 in embryonic, larval and adult tissues was analyzed by three different approaches. Firstly, we visualized nep4 mRNA. Antisense riboprobes, corresponding to coding regions shared by transcript A and B or being unique to transcript A (Fig. 1A), were generated and subsequently used for in situhybridization. Because of the fact that the respective probes showed basically identical in situ hybridization patterns, yet with reduced signal intensity in the case of transcript A specific probes, only stainings on the basis of probes that do not discriminate between transcripts are shown. Secondly, immunofluorescence microscopy was performed with an anti-NEP4 antiserum. The specificity of the antiserum was verified by antigen blocking and by control stainings with preimmune sera and without primary antibodies(supplementary material Fig. S1B).

The first appearance of NEP4 is obvious at stage 12 in the dorsal mesoderm in a segmental pattern (Fig. 4A,G). From stage 13 to stage 17, expression of NEP4 is maintained solely in cells of the cardiac mesoderm, which are located in two one-cell wide rows along the anterior–posterior axis of the embryo. In addition,we found NEP4 expression from embryonic stage 14 onwards in cells of the central nervous system (Fig. 4E,K). To elucidate the identity of NEP4 expressing cells precisely, we performed double and triple immunostainings with antibodies that specifically mark individual muscle founder cells in the dorsal mesoderm. Even-skipped (Eve) is expressed in founder cells that give rise to muscle founder DA1, DO2 and a subset of pericardial cells. Initially, Eve is expressed in two cell clusters, in one of which (cluster 2) Eve initiates the expression of Krüppel (Kr). This progenitor, transiently coexpressing Eve and Kr, divides to yield two founder cells that express Runt. One of these founders gives rise to muscle DO2 and shows continuous Runt expression. The other one gives rise to the so-called even-skipped positive pericardial cells(EPCs). Shortly after, DO2 loses Eve expression, whereas the EPCs remain Eve positive. One cell from the second Eve cluster (cluster 15) divides to yield the DA1 muscle founder (that maintains Eve expression) and a second cell that is assumed to die (Alvarez et al.,2003; Carmena et al.,1998; Fujioka et al.,2005; Han and Bodmer,2003). Besides DA1, Krüppel additionally labels a second muscle founder within the dorsal mesoderm, which is the DO1 founder(Ruiz-Gómez et al.,1997). During stage 12–14, NEP4 transiently colocalizes with Eve, Krüppel and Runt in some, but not all dorsal muscle founder cells. Thus we conclude that NEP4 is expressed in the progenitor of the Eve pericardial cells, muscle founder DA1, muscle founder DO2 and a third muscle founder of as yet unknown identity (Fig. 5A–L). The expression of NEP4 in dorsal muscle founder cells is transient; in contrast, NEP4 expression in the Eve positive pericardial cells is maintained until the end of embryogenesis(Fig. 5M).

From stage 14 to stage 17, expression of NEP4 is seen in cells of the central nervous system (Fig. 4E,K). Double stainings for NEP4 and Repo (reversed polarity),which is specifically expressed in all glia cells of the nervous system(Halter et al., 1995),demonstrated NEP4 expression in glia cells in the embryonic and larval CNS but not in neuronal cells (Fig. 5O). Further colocalization studies with antibodies labeling individual subsets of glia cells, including anti-Prospero antibody(Doe et al., 1991)(Fig. 5N), revealed that NEP4 is present in three major types of glia cell. These are the cell body glia,the lateral glia cells, which are partially marked by Prospero, and the medial intersegmental nerve root glia (Fig. 5N,O) (Beckervordersandforth et al., 2008). A reporter line expressing GFP under the control of the native nep4 promoter element confirmed the expression of NEP4 in glia cells (Fig. 4N–P)and furthermore revealed persistent expression in the larval, pupal and adult central nervous system (Fig. 4P, pupal and adult stages not shown). Control stainings with anti-NEP4 antibodies showed an overlap between the immunosignal and GFP expression (Fig. 4N). It should be noted that the identified regulatory region of the nep4 gene recapitulates the neuronal expression of endogenous NEP4 but lacks the regulatory elements for muscle and pericardial gene expression. Anti-NEP4 staining and in situ hybridization on the CNS of 3rd instar larva substantiated the observed reporter expression(Fig. 4F,L). Western blots on protein extracts isolated from adult heads confirmed the expression of both isoforms in the adult brain (Fig. 3B).

In addition to the described tissue specific expression of NEP4 in pericardial, muscle founder and glia cells, NEP4 is also obvious in gonads of late embryonic stages (Fig. 4I,arrow). The NEP4 signal in embryonic gonads is apparent in about 50% of embryos of that stage, indicating sex specificity. Previous data from genome wide microarray approaches have already provided evidence for sex specific NEP4 expression (McIntyre et al.,2006). Additionally, nep4 in situ hybridization experiments conducted on adult testes as part of the FlyTED-project(www.fly-ted.org)corroborate a potential role for NEP4 in reproduction. Therefore, we performed NEP4 immunostainings and in situ hybridizations on adult ovaries and testes. While we did not detect NEP4 in ovaries (not shown), we found strong expression in adult testes, where NEP4 is located in early spermatocytes to late stage spermatids. At the apical tip of the testis NEP4 was detected in mitotically amplifying gonial cells and was still present postmeiotically in early spermatids, as well as in mature sperm, as indicated by a strong NEP4 immunosignal in seminal vesicles (Fig. 4M). To ascertain the isoform distribution in this tissue, we did western blots with abdominal protein preparations isolated from males and females, respectively. Consistent with the absence of immunostainings in ovaries, female abdominal preparations did not harbor any NEP4 protein in detectable amounts. In male preparations, however, a strong expression of isoform A was apparent, while the smaller isoform B was not detectable(Fig. 3B). This result demonstrates a negligible function for soluble NEP4 in testes.

Heterologously expressed NEP4 exhibits distinct substrate specificities

In order to assay the enzymatic activities of NEP4, we expressed isoform A in SF21 cells, isolated the membranes and tested them for protein expression. As expected, western blots confirmed the presence of a considerable amount of NEP4A protein in the membrane fraction. However, in untransfected control cells a protein of about the same size but of considerably less abundance was detected as well (not shown), which might be an indication of the presence of endogenous NEP4 in this cell type. This observation makes it difficult to distinguish between endogenous and heterologous neprilysin activity and clearly renders SF21 cells problematic for heterologous NEP4 expression. For this reason we used E. coli as an expression system and performed activity assays with purified NEP4 protein and different peptides as putative substrates (Fig. 6). These peptides included tachykinin, substance P, bradykinin, angiotensin I and PDF. The first two, both belonging to well characterized groups of neuropeptides,were chosen because of the strong NEP4 expression in the CNS. This expression pattern points to neuropeptides as potential substrates. Bradykinin and angiotensin I were chosen because of their known activity as tissue hormones in several vertebrate species. PDF was previously discussed to be a putative substrate of Drosophila neprilysins(Isaac et al., 2007) and was therefore included in the activity assay.

It appeared that the peptides substance P (85.8% degradation) and angiotensin I (51.6%) in particular were cleaved quite efficiently in a time dependent manner (Fig. 6C)while the remaining peptides were either cleaved to a much lesser extent (PDF,27.8%) or not cleaved at all (bradykinin, tachykinin, Fig. 6A). With respect to the main substrate, substance P (RPKPQQFFGLM), the decreasing amount of undegraded peptide exhibited a linear correlation with an increasing abundance of two cleavage products: a major one, lacking three amino acids (RPKPQQFF) and a minor one lacking two amino acids (RPKPQQFFG) at the C-terminus. These data demonstrate that hydrolysis occurs predominantly at the Phe8–Gly9 and,to a lesser extent, at the Gly9–Leu10 bond and therefore adjacent to bulky hydrophobic residues, a preference that is shared by many mammalian peptidases (Turner et al.,2001). The latter site of hydrolysis is also reported for human NEP and NEP2 (Rose et al.,2002), while hydrolysis between Phe8–Gly9 represents a cleavage site not reported previously. These cleavage characteristics were substantiated by results from hydrolysis assays performed with NEP4A expressing SF21 cells. Identified peptide substrates (substance P and angiotensin I) added to SF21 cell membrane preparations were degraded rapidly in samples from infected cells but also from non-infected control cells. In both cases we detected the cleavage fragments mentioned above. However,relative cleavage activity was enhanced by 30% in membranes from infected cells compared with control cells and the amount of resulting peptide degradation products increased correspondingly (not shown). Thus, the data obtained by SF21 expression confirmed the activities and specificities already measured with the purified protein from E. coli.

With respect to the typical inhibitors of neprilysins, phosphoramidon and thiorphan, we found that these have only limited capacity to inhibit NEP4. Apparently, inhibitor concentrations have to be in high micromolar ranges (100μmol l–1) to reduce enzyme activity significantly. However, even at this concentration, inhibition rates did not exceed 20–30% (Fig. 6D). Again in accordance with purified NEP4 protein expressed in E. coli, the addition of the respective inhibitors to SF21 membranes containing heterologously expressed NEP4 hampered protein activity at comparable rates(not shown). Of note, this reduced susceptibility to phosphoramidon and thiorphan was also shown for human NEP2.

In terms of pH dependence, NEP4 characteristics are similar to those of other neprilysins. As shown in Fig. 6B, the highest catalytic efficiency was measured in the neutral range (pH 7) with a dramatic reduction in activity above pH 8 and below pH 6.

NEP4 exhibits enzyme characteristics similar to mammalian NEP2

A comparison of the enzymatic properties of NEP4 from Drosophilawith those of other neprilysins from different species reveals a clear analogy to mammalian NEP2. Intriguingly, murine NEP2 was recently reported to be involved in sperm formation and embryonic development(Carpentier et al., 2004), two physiological functions that are likely to be shared by DrosophilaNEP4, as the described expression pattern suggests (Figs 4 and 5). In particular, the expression in embryonic dorsal muscle founder and pericardial cells renders an involvement in developmental processes rather likely. However, further investigation is necessary to ascertain this issue properly. A remarkable similarity between Drosophila NEP4 and human NEP2 is the limited susceptibility to phosphoramidon and thiorphan, inhibitors that have been shown to be effective against many neprilysins. Thiorphan in particular has frequently been used to inhibit neprilysin-like activity and is reported to be neprilysin specific at nanomolar concentrations(Turner et al., 2001). The observation that in contrast to human NEP, human NEP2(Whyteside and Turner, 2008)but also Drosophila NEP4 (Fig. 6D) are relatively resistant against these two inhibitors could be an indication of similarities in the structure of their active sites. Despite distinct differences in substrate specificity and inhibitor sensitivity(Whyteside and Turner, 2008),a comparison of the amino acid residues that line the hydrophobic pockets of human NEP (Oefner et al.,2000; Oefner et al.,2004) and human NEP2(Whyteside and Turner, 2008)showed identical ligand binding S1′ and S2′subsites (Table 1). NEP4 from Drosophila and human NEP2 on the other hand revealed considerably less homology (S1′ subsite: five conserved residues,S2′ subsite: no conserved residues; Table 1). Nevertheless, in addition to comparable inhibitor susceptibilities the two enzymes exhibit quite similar substrate specificities. In line with Drosophila NEP4(this work), the main substrates cleaved by human NEP2 are substance P and angiotensin I (Whyteside and Turner,2008). The fact that human NEP and NEP2, despite completely conserved subsite residues, exhibit major differences in their specificities clearly indicates that residues other than those mentioned above are responsible for regulating access to the catalytic center. From modeling of the active site of rat NEP2, Voisin and colleagues recently proposed two additional critical residues (Ser133 and Leu739) that are present in rat NEP2(Voisin et al., 2004) whereas human NEP harbors glycines at the respective positions. Sequence alignments(http://bioinfo.genotoul.fr/multalin/multalin.html)indicate that Drosophila NEP4 shares one glycine (Gly1005) with human NEP at the respective position, while the second glycine is replaced by Glu416 in Drosophila NEP4 and Ser133 in the case of rat NEP2. As glutamate as well as serine are, unlike glycine, well known to be involved in hydrogen bond formation with the protein backbone and thereby stabilize the protein structure, this position could be of particular importance to substrate access. The identification of specific inhibitors, that act equally potently on the two proteins human NEP2 and Drosophila NEP4, would be strong support for the hypothesis of a structural relationship between these two peptidases. Unfortunately, in neither case is such an inhibitor currently known.

With respect to other neprilysins from Drosophila, only NEP2 has been characterized so far (Bland et al.,2006; Thomas et al.,2005). Unlike Drosophila NEP2, Drosophila NEP4 does not cleave the peptide tachykinin from Locusta migratoria,demonstrating considerably different substrate specificities and thereby physiological functions within the Drosophila neprilysin family. Especially interesting in this context is the fact that DrosophilaNEP2 and NEP4 are partially expressed in the same tissues. As mentioned, Drosophila NEP2 is expressed in the testes(Thomas et al., 2005), which is also true for Drosophila NEP4(Fig. 4M). The obvious difference in substrate specificity together with their expression in the same tissue indicates a physiological requirement for highly specialized peptidases with individual substrate specificities. Although further investigation is necessary to ascertain this issue in more detail, the data presented in this study for the first time allow a direct comparison between two neprilysins from Drosophila melanogaster.

It is noteworthy that former work with Drosophila head membranes has already introduced an enzyme with neprilysin-like activities but low susceptibility to phosphoramidon (Isaac et al., 2002). Although the identity of this enzyme was not elucidated, it represents the first evidence for the existence of neprilysin-like peptidases in Drosophila that are relatively resistant to common neprilysin inhibitors.

Isoform specific solubility is quite unusual among neprilysins

Neprilysins are generally considered to be membrane bound proteins(Turner et al., 2001). However, in individual cases, the existence of soluble neprilysins has been reported (Ikeda et al., 1999; Thomas et al., 2005), which apparently become soluble by proteolytic cleavage of a membrane bound precursor protein.

Our data demonstrate that NEP4 from D. melanogaster exists as both a membrane bound and a soluble protein. In contrast to other neprilysins,however, the solubility apparently depends on alternative splicing and corresponding protein biosynthesis instead of posttranslational proteolytic cleavage. In addition to the verification of transcript specific mRNAs(Fig. 2A), the fact that the NEP4A sequence does not contain a prohormone-convertase recognition site(Lys–Arg) close to the transmembrane domain(Fig. 7) further contradicts the possibility of proteolytic release of isoform B from membrane bound isoform A. This recognition site is considered to be essential for proteolytic processing of NEP2 in mammals (Ikeda et al., 1999). Despite this difference, the mere existence of both membrane bound and soluble NEP4 is quite unusual among neprilysins and bears potentially strong physiological relevance. Based on the current data, we expect a scavenger function for the soluble isoform that regulates peptide homeostasis in the hemolymph and in the nervous system independently of the more stationary membrane bound isoform. However, further studies on this issue are hindered by the fact that the two isoforms share an almost identical structure (Fig. 1), which makes it difficult to distinguish between them. Efforts to utilize a nep4Aspecific RNA probe for in situ detection of the corresponding mRNA in embryos were successful, however; while these data represent good evidence for the presence of transcript A in the embryonic nervous system, somatic muscle founders and the EPCs, the additional expression of transcript B can be measured only indirectly by the stronger in situ signal generated by a probe against both transcripts (Fig. 4E). Nevertheless, from this together with a western blot that shows expression of isoform B in embryonic tissues(Fig. 2C) we come to the conclusion that both splice variants are present in the central nervous system during embryogenesis indicating distinct physiological relevance, which is presumably unique to the respective isoforms. This interpretation is corroborated by the expression of both isoforms in the adult nervous system as shown by western blot (Fig. 3B).

With respect to physiological functions, a potential in vivorelevance can be attributed in particular to protein presence in the central nervous system. NEP4 expression in this tissue begins during embryogenesis and is maintained until adult stages, demonstrating a permanent role for neprilysin-like endopeptidases. As shown previously, members of the neprilysin family are responsible for terminating the actions of neuropeptides like enkephalins (Malfroy et al.,1978; Schwartz et al.,1980) or tachykinins (Barnes et al., 1993; Matsas et al.,1983) on neuronal surfaces and especially in the perisynaptic region. A similar activity can also be assumed for Drosophila NEP4 as its potential to degrade neuropeptides(Fig. 6), together with a highly specific expression in glia cells(Fig. 5N,O) strongly indicate such a physiological relevance. This hypothesis is further corroborated by the general functions of the glia cell subtypes NEP4 is expressed in: while cell body associated glia are structurally similar to mammalian astrocytes(Freeman and Doherty, 2006),lateral glia cells are similar to oligodendrocytes(Stork et al., 2008), with both mammalian cell types reported to be responsible for ion and neurotransmitter homeostasis (Mentlein and Dahms, 1994; Vilijn et al.,1989; Stacey et al.,2007). As no system responsible for the reuptake of neuropeptides at the nerve terminal is known in Drosophila, the biological activity of these transmitters is presumably controlled by extracellular degradation. In this context, glia cells and corresponding peptidases might play a decisive role in modulating secreted peptides present in the extracellular spaces of the CNS. While the majority of this processing is presumably accomplished by membrane bound peptidases like NEP4A, the identification of a soluble NEP4 isoform expressed at least in the adult nervous system(Fig. 3B) is a strong indication of the additional requirement for soluble endopeptidases. Because of an efficient insulation of the nervous system that allows a fine tuned homeostasis of ions, peptides and other small molecules(Stork et al., 2008), soluble NEP4 secreted from glia cells is likely to remain within the nervous system instead of diffusing into the hemolymph.

Independent of the isoform distribution, the mere expression of NEP4 in testes (Fig. 4M) suggests an involvement in reproductive physiology. A potential role in reproduction has already been stated for Drosophila NEP2 which is also expressed in testes (Thomas et al., 2005)and shown for mammalian peptidases, where male mice lacking germinal angiotensin converting enzyme reveal strongly impaired fertility, while female knockouts behave like wild-type (Krege et al., 1995). This sex specific relevance can also be anticipated with respect to Drosophila NEP4 as protein expression was found to occur only in male but not female reproductive organs(Fig. 4M, Fig. 3B). The result that in testes the membrane bound isoform of NEP4 is expressed(Fig. 3B) together with the apparently soluble NEP2 (Thomas et al.,2005) is another indication for distinct but concerted functions of different neprilysins in certain tissues.

With respect to the expression in EPCs, a possible physiological relevance for NEP4 could be the processing or degradation of signals sent from the pericardium to the dorsal vessel. Recently, it was shown that the heart function in Drosophila is likely to be modulated non-autonomously by secreted molecules from neighboring cells, eventually the EPCs(Buechling et al., 2009). The central position of the pericardial cells together with the capability of NEP4 to hydrolyze the tissue hormone angiotensin I(Fig. 6A) furthermore suggests that NEP4 expressed in EPCs might be responsible for the homeostasis of different signaling peptides circulating the hemolymph and passing the heart of Drosophila with the respective pericardial cells functioning as a checkpoint for peptide clearance. Such a physiological function would be the first ever attributed to this cell type. Indeed, in addition to the apparent occurrence of nep4A mRNA in EPCs, at high magnifications a plasma membrane bound immunosignal is visible in these cells, strongly indicating expression of isoform A. In addition to this signal, tiny round shaped structures are also stained, presumably vesicles of the secretory pathway containing NEP4A (not shown). Because the applied antisera do not discriminate between NEP4A and NEP4B, we obviously cannot exclude the possibility that the stained vesicles also contain isoform B which could eventually be exocytosed and serve its purpose as a soluble enzyme in the hemolymph. In any case, based on the current data we propose that NEP4 expression in the pericardial cells is required for the homeostasis of signaling peptides circulating the hemolymph of Drosophila.

A noteworthy observation is the transient expression of at least NEP4A in a particular set of dorsal somatic muscle founders. Muscle founders are crucial for the formation of the somatic body wall musculature and a pivotal step during myogenesis is myoblast fusion, a process that is initiated by the muscle founder cells. Expression of NEP4A in founder cells rather than in multinucleated myofibers indicates an early role for the protein in these cells. One possibility is that all NEP4A expressing cells in the dorsal mesoderm, muscle founders and EPCs are involved in peptide clearance. On the other hand, it could be that the enzyme harbors a specific but as yet unknown function in myogenesis. However, future investigations on a specific NEP4 mutant will be crucial for the identification of the physiological processes that NEP4 is involved in.

LIST OF ABBREVIATIONS

     
  • CNS

    central nervous system

  •  
  • eGFP

    enhanced green fluorescent protein

  •  
  • EPC

    even-skipped positive pericardial cell

  •  
  • Eve

    even-skipped

  •  
  • Kr

    Krüppel

  •  
  • NEP

    neprilysin

  •  
  • Pa

    phosphoramidon

  •  
  • PDF

    pigment dispersing factor

  •  
  • Repo

    reversed polarity

  •  
  • Tp

    thiorphan

We thank Eva Haß-Cordes for excellent technical assistance and Dr Stefan Walter for giving mass spectrometry support. We also thank R. Bodmer,G. Technau, the Bloomington Stock Center and the DSHB for antibodies, fly stocks and cloned material. This research was supported by grants from the DFG to A.P. (SFB 431:Membranproteine – Funktionelle Dynamik und Kopplung an Reaktionsketten)and the AFM (French Association against Myopathies; no. 13209) to K.J. Support was obtained from the European Program of Excellence MYORESto K.J., M.Z. and A.P.

Alvarez, A. D., Shi, W., Wilson, B. A. and Skeath, J. B.(
2003
). pannier and pointedP2 act sequentially to regulate Drosophila heart development.
Development
130
,
3015
-3026.
Barnes, K., Turner, A. J. and Kenny, A. J.(
1993
). An immunoelectron microscopic study of pig substantia nigra shows co-localization of endopeptidase-24.11 with substance P.
Neuroscience
53
,
1073
-1082.
Barolo, S., Castro, B. and Posakony, J. W.(
2004
). New Drosophila transgenic reporters: insulated P-element vectors expressing fast-maturing RFP.
BioTechniques
36
,
436
-440, 442.
Beckervordersandforth, R. M., Rickert, C., Altenhein, B. and Technau, G. M. (
2008
). Subtypes of glial cells in the Drosophila embryonic ventral nerve cord as related to lineage and gene expression.
Mech. Dev.
125
,
542
-557.
Bland, N. D., Thomas, J. E., Audsley, N., Shirras, A. D.,Turner, A. J. and Isaac, R. E. (
2006
). Expression of NEP2, a soluble neprilysin-like endopeptidase, during embryogenesis in Drosophila melanogaster.
Peptides
28
,
127
-135.
Buechling, T., Akasaka, T., Vogler, G., Ruiz-Lozano, P., Ocorr,K. and Bodmer, R. (
2009
). Non-autonomous modulation of heart rhythm, contractility and morphology in adult fruit flie.
Dev. Biol.
328
,
483
-492.
Carmena, A., Gisselbrecht, S., Harrison, J., Jimenez, F. and Michelson, A. M. (
1998
). Combinatorial signaling codes for the progressive determination of cell fates in the Drosophilaembryonic mesoderm.
Genes Dev.
12
,
3910
-3922.
Carpentier, M., Guillemette, C., Bailey, J. L., Boileau, G.,Jeannotte, L., DesGroseillers, L. and Charron, J. (
2004
). Reduced fertility in male mice deficient in the zinc metallopeptidase NL1.
Mol. Cell. Biol.
24
,
4428
-4437.
Doe, C. Q., Chu-LaGraff, Q., Wright, D. M. and Scott, M. P.(
1991
). The prospero gene specifies cell fates in the Drosophila central nervous system.
Cell
65
,
451
-464.
Duan, H., Skeath, J. B. and Nguyen, H. T.(
2001
). Drosophila Lame duck, a novel member of the Gli superfamily, acts as a key regulator of myogenesis by controlling fusion-competent myoblast development.
Development
128
,
4489
-4500.
Freeman, M. R. and Doherty, J. (
2006
). Glial cell biology in Drosophila and vertebrates.
Trends Neurosci.
29
,
82
-90.
Fujioka, M., Wessells, R. J., Han, Z., Liu, J., Fitzgerald, K.,Yusibova, G. L., Zamora, M., Ruiz-Lozano, P., Bodmer, R. and Jaynes, J. B.(
2005
). Embryonic Even Skipped-dependent muscle and heart cell fates are required for normal adult activity, heart function, and lifespan.
Circ. Res.
97
,
1108
-1114.
Ghaddar, G., Ruchon, A. F., Carpentier, M., Marcinkiewicz, M.,Seidah, N. G., Crine, P., Desgroseillers, L. and Boileau, G.(
2000
). Molecular cloning and biochemical characterization of a new mouse testis soluble-zinc-metallopeptidase of the neprilysin family.
Biochem. J.
347
,
419
-429.
Halter, D. A., Urban, J., Rickert, C., Ner, S. S., Ito, K.,Travers, A. A. and Technau, G. M. (
1995
). The homeobox gene repo is required for the differentiation and maintenance of glia function in the embryonic nervous system of Drosophila melanogaster.
Development
121
,
317
-332.
Han, Z. and Bodmer, R. (
2003
). Myogenic cells fates are antagonized by Notch only in asymmetric lineages of the Drosophila heart, with or without cell division.
Development
130
,
3039
-3051.
Hersh, L. B. and Morihara, K. (
1986
). Comparison of the subsite specificity of the mammalian neutral endopeptidase 24.11 (enkephalinase) to the bacterial neutral endopeptidase thermolysin.
J. Biol. Chem.
261
,
6433
-6437.
Ikeda, K., Emoto, N., Raharjo, S. B., Nurhantari, Y., Saiki, K.,Yokoyama, M. and Matsuo, M. (
1999
). Molecular identification and characterization of novel membrane-bound metalloprotease, the soluble secreted form of which hydrolyzes a variety of vasoactive peptides.
J. Biol. Chem.
274
,
32469
-32477.
Isaac R. E., Parkin E. T., Keen J. K., Nässel D. R.,Siviter R. J. and Shirras A. D. (
2002
) Inactivation of a tachykinin-related peptide: identification of four neuropeptide-degrading enzymes in neuronal membranes of insects from four different orders.
Peptides
23
,
725
-733.
Isaac, R. E., Johnson, E. C., Audsley, N. and Shirras, A. D.(
2007
). Metabolic inactivation of the circadian transmitter,pigment dispersing factor (PDF), by neprilysin-like peptidases in Drosophila.
J. Exp. Biol.
210
,
4465
-4470.
Iwata, N., Tsubuki, S., Takaki, Y., Watanabe, K., Sekiguchi, M.,Hosoki, E., Kawashima-Morishima, M., Lee, H. J., Hama, E., Sekine-Aizawa, Y. et al. (
2000
). Identification of the major Abeta1-42-degrading catabolic pathway in brain parenchyma: suppression leads to biochemical and pathological deposition.
Nat. Med.
6
,
143
-150.
Jagla, K., Frasch, M., Jagla, T., Dretzen, G., Bellard, F. and Bellard, M. (
1997
). ladybird, a new component of the cardiogenic pathway in Drosophila required for diversification of heart precursors.
Development
124
,
3471
-3479.
Kerr, M. A. and Kenny, A. J. (
1974
). The purification and specificity of a neutral endopeptidase from rabbit kidney brush border.
Biochem. J.
137
,
477
-488.
Kosman, D., Small, S. and Reinitz, J. (
1998
). Rapid preparation of a panel of polyclonal antibodies to Drosophilasegmentation proteins.
Dev. Genes Evol.
208
,
290
-294.
Krege, J. H., John, S. W., Langenbach, L. L., Hodgin, J. B.,Hagaman, J. R., Bachman, E. S., Jennette, J. C., O'Brien, D. A. and Smithies,O. (
1995
). Male-female differences in fertility and blood pressure in ACE-deficient mice.
Nature
375
,
146
-148.
Malfroy, B., Swerts, J. P., Guyon, A., Roques, B. P. and Schwartz, J. C. (
1978
). High-affinity enkephalin-degrading peptidase in brain is increased after morphine.
Nature
276
,
523
-526.
Matsas, R., Fulcher, I. S., Kenny, A. J. and Turner, A. J.(
1983
). Substance P and [Leu]enkephalin are hydrolyzed by an enzyme in pig caudate synaptic membranes that is identical with the endopeptidase of kidney microvilli.
Proc. Natl. Acad. Sci. USA
80
,
3111
-3115.
McIntyre, L. M., Bono, L. M., Genissel, A., Westerman, R., Junk,D., Telonis-Scott, M., Harshman, L., Wayne, M. L., Kopp, A. and Nuzhdin, S. V. (
2006
). Sex-specific expression of alternative transcripts in Drosophila.
Genome Biol.
7
,
R79
.
Mentlein, R. and Dahms, P. (
1994
). Endopeptidases 24.16 and 24.15 are responsible for the degradation of somatostatin, neurotensin, and other neuropeptides by cultivated rat cortical astrocytes.
J. Neurochem.
62
,
27
-36.
Molinaro, G., Rouleau, J. L. and Adam, A.(
2002
). Vasopeptidase inhibitors: a new class of dual zinc metallopeptidase inhibitors for cardiorenal therapeutics.
Curr. Opin. Pharmacol.
2
,
131
-141.
Oefner, C., D'Arcy, A., Hennig, M., Winkler, F. K. and Dale, G. E. (
2000
). Structure of human neutral endopeptidase(Neprilysin) complexed with phosphoramidon.
J. Mol. Biol.
296
,
341
-349.
Oefner, C., Roques, B. P., Fournie-Zaluski, M. C. and Dale, G. E. (
2004
). Structural analysis of neprilysin with various specific and potent inhibitors.
Acta Crystallogr.
60
,
392
-396.
Roques, B. P., Noble, F., Dauge, V., Fournie-Zaluski, M. C. and Beaumont, A. (
1993
). Neutral endopeptidase 24.11: structure,inhibition, and experimental and clinical pharmacology.
Pharmacol. Rev.
45
,
87
-146.
Rose, C., Voisin, S., Gros, C., Schwartz, J. C. and Ouimet,T. (
2002
). Cell-specific activity of neprilysin 2 isoforms and enzymic specificity compared with neprilysin.
Biochem. J.
363
,
697
-705.
Ruiz-Gómez, M., Romani, S., Hartmann, C., Jäckle, H. and Bate, M. (
1997
). Specific muscle identities are regulated by Krüppel during Drosophila embryogenesis.
Development
124
,
3407
-3414.
Schwartz, J. C., de la Baume, S., Malfroy, B., Patey, G.,Perdrisot, R., Swerts, J. P., Fournie-Zaluski, M. C., Gacel, G. and Roques, B. P. (
1980
). “Enkephalinase”, a newly characterised dipeptidyl carboxypeptidase: properties and possible role in enkephalinergic transmission.
Int. J. Neurol.
14
,
195
-204.
Sellin, J., Drechsler, M., Nguyen, H. T. and Paululat, A.(
2009
). Antagonistic function of Lmd and Zfh1 fine tunes cell fate decisions in the Twi and Tin positive mesoderm of Drosophila melanogaster.
Dev. Biol.
326
,
444
-455.
Stacey, S. M., Thomas, G. B., Labbe, A. and Van Meyel, D. J.(
2007
). Longitudinal glia in the fly CNS: pushing the envelope on glial diversity and neuron-glial interactions.
Neuron Glia Biol.
3
,
27
-33.
Stork, T., Engelen, D., Krudewig, A., Silies, M., Bainton, R. J. and Klämbt, C. (
2008
). Organization and function of the blood-brain barrier in Drosophila.
J. Neurosci.
28
,
587
-597.
Thomas, J. E., Rylett, C. M., Carhan, A., Bland, N. D., Bingham,R. J., Shirras, A. D., Turner, A. J. and Isaac, R. E. (
2005
). Drosophila melanogaster NEP2 is a new soluble member of the Neprilysin family of endopeptidases with implications for reproduction and renal function.
Biochem. J.
386
,
357
-366.
Turner, A. J. and Tanzawa, K. (
1997
). Mammalian membrane metallopeptidases: NEP, ECE, KELL, and PEX.
FASEB J.
11
,
355
-364.
Turner, A. J., Isaac, R. E. and Coates, D.(
2001
). The neprilysin (NEP) family of zinc metalloendopeptidases: genomics and function.
Bioessays
23
,
261
-269.
Vilijn, M. H., Das, B., Kessler, J. A. and Fricker, L. D.(
1989
). Cultured astrocytes and neurons synthesize and secrete carboxypeptidase E, a neuropeptide-processing enzyme.
J. Neurochem.
53
,
1487
-1493.
Voisin, S., Rognan, D., Gros, C. and Ouimet, T.(
2004
). A three-dimensional model of the neprilysin 2 active site based on the X-ray structure of neprilysin. Identification of residues involved in substrate hydrolysis and inhibitor binding of neprilysin 2.
J. Biol. Chem.
279
,
46172
-46181.
Whitworth, J. A. (
2003
). Emerging drugs in the management of hypertension.
Expert Opin. Emerg. Drugs
8
,
377
-388.
Whyteside, A. R. and Turner, A. J. (
2008
). Human neprilysin-2 (NEP2) and NEP display distinct subcellular localisations and substrate preferences.
FEBS Lett.
582
,
2382
-2386.