- Email: [email protected]
The serine protease Sp7 is expressed in blood cells and regulates the melanization reaction in Drosophila
BBRC Biochemical and Biophysical Research Communications 338 (2005) 1075–1082 www.elsevier.com/locate/ybbrc
The serine protease Sp7 is expressed in blood cells and regulates the melanization reaction in Drosophila Casimiro Castillejo-Lo´pez *, Udo Ha¨cker Department of Experimental Medical Science and Lund Strategic Research Center for Stem Cell Biology and Cell Therapy, Lund University, BMC B13, Klinikgatan 26, 22184 Lund, Sweden Received 23 September 2005 Available online 21 October 2005
Abstract Serine proteases play a central role in defense against pathogens by regulating processes such as blood clotting, melanization of injured surfaces, and proteolytic activation of signaling pathways involved in innate immunity. Here, we present the functional characterization of the Drosophila serine protease Sp7 (CG3006) by inducible RNA interference. We show that Sp7 is constitutively expressed in blood cells during embryonic and larval stages. Silencing of the gene impairs the melanization reaction upon injury. Our data demonstrate that Sp7 is required for phenoloxidase activation and its activity is restricted to a subclass of blood cells, the crystal cells. Transcriptional up-regulation of Sp7 was observed after clean, septic injury and in flies expressing an activated form of Toll; however, mutations in the Toll or the IMD pathway did not abolish expression of Sp7, indicating the existence of other regulatory pathways and/or independent basal transcription. Ó 2005 Elsevier Inc. All rights reserved. Keywords: Drosophila; RNAi; Serine protease; Melanization; Phenoloxidase; Wound; Blood cells; Innate immunity
Innate immunity in insects consists of a humoral response based on antimicrobial peptides and a cellular response mediated by blood cells that promote phagocytosis, encapsulation of pathogens, and melanization [1–3]. The immediate immune response in Drosophila is the melanization reaction observed at the site of injury or on the surface of pathogens and parasites invading the hemocoel. Perforation of the cuticle by injury or by parasitic infection initiates a proteolytic reaction that leads to rapid hemolymph coagulation and deposition of melanin. This reaction plays an important role in wound healing where melanized tissue forms a plug that restricts bleeding [4,5]. The melanization reaction requires the activation of phenoloxidase (PO) that catalyzes the conversion of phenols to quinones, which spontaneously polymerize to form melanin (Fig. 7). In vitro studies have *
Corresponding author. Fax: +46 46 2220899. E-mail address: [email protected] (C. Castillejo-Lo´pez). 0006-291X/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2005.10.042
shown that PO exists as an inactive precursor, prophenoloxidase (pro-PO), which is activated by a stepwise process involving serine protease cascades [6–8]. As excessive melanin is deleterious to the host, the processing of pro-PO must be strictly regulated to restrict the melanization reaction. Inactive PO is cleaved into active PO by a protease known as prophenoloxidase-activating enzyme (pro-POAE). Pro-POAEs have been identified in the tobacco hornworm, Manduca sexta [9,10], in the silkworm, Bombyx mori [7], in the beetle, Holotrichia diomphalia [11], and in the crayfish, Pacifastus leniusculus [12] but so far, no such activator gene has been studied in vivo by mutational analysis in Drosophila. Negative regulators of the catalysis have been characterized and include members of the serine protease inhibitor (serpin) superfamily [13]. Serpins act as suicide substrate inhibitors of serine proteases by forming irreversible complexes. Biochemical and genetic characterization of one of them, serpin-27A (Spn27A) in Drosophila, demonstrates that the protein encoded by this gene regulates the
1076
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
melanization cascade through the specific inhibition of pro-PO processing by the terminal serine protease proPOAE [14,15]. In addition to the regulation at the biochemical level, melanization is controlled spacially and temporally by blood cells. Drosophila has three blood cell types; the plasmatocytes, phagocytic cells that comprise 95% of circulating hemocytes and synthesize antimicrobial peptides; the crystal cells, named for the crystalline inclusion bodies and lamellocytes, the largest cells normally not present in healthy larvae and involved in the encapsulation response against intruders that are too large to be phagocytized. Rizki et al. [16] suggested that crystal cells are the only source of pro-PO in the fruit fly. However, lamellocytes that appeared after parasitoid wasp infection could contribute to the melanization of the capsule that isolated the parasite eggs [17,18]. The completion of the genome sequence has shown that Drosophila melanogaster has three pro-PO genes, Bc (DoxA1, CG5779), DoxA3 (CG2952), and CG8193. A recent proteomic analysis of larval hemolymph has shown a reduction of the pro-PO protein spots upon immune challenge [19]. The reduction is attributed to proteolytic activation, indicating that these genes code for proteins that are processed upon challenge probably by serine proteases. The protein annotation in this study corresponds to both the genes CG8193 and CG2952, and no further discrimination could be done. Based on histological location in non challenged embryos it has been inferred that pro-POs are expressed in crystal cells [20,21]. Recently it has been demonstrated that this is true for two of them, while CG2952 is exclusively expressed in lamellocytes [17]. The relative contribution of crystal cells and lamellocytes to the melanotic encapsulation of parasitoid wasp eggs is not understood. It is recognized, however, that the melanization at injury sites is mediated exclusively by crystal cells mainly because lamellocytes are not present in healthy larvae. Furthermore, three classical blood mutants: Domino (dom) which has fewer blood cells, Black Cells (Bc) which has aberrant crystal cells, and lozenge (Lz) which lacks crystal cells, all have severely impaired hemolymph melanization [22,23]. Here, we present the functional characterization of the first serine protease in any arthropod species that is necessary for activation of the PO pathway and melanization upon injury. Sp7 is constitutively expressed in blood cells and up-regulated after injury, infection, and in immune deficient mutations. Silencing of transcription in crystal cells using the early transcription factor Lozenge is sufficient to impair hemolymph melanization.
driver lines Act-Gal4/CyO, Tub-Gal4/TM3, hem-Gal4-GFP, and Lz-Gal4GFP. RNAi transgenic lines were generated by P element-mediated transformation using w1118 as recipient stock. Flies were maintained on standard medium at 19, 25, or 29 °C, as necessary. Transgenenic RNA interference. A pUAST-vector containing genomic and cDNA sequences in opposite direction was constructed to produce dsRNA. All the described transcripts of the gene CG3066 were targeted. The genomic fragment encompassing 729 bp and including the last three introns was amplified with the primers: (5 0 - GGGCGGCCGCAACGA CACTGCTATTGACG-3 0 ) and (5 0 -GGCTCGAGGTGCTCTTGCGGG CTATAGA-3 0 ). The corresponding cDNA encompasses 425 bp and was amplified with the primers: (5 0 - GGGGTACCACTGCTTGAGTACG TGGATA-3 0 ) and (5 0 -GGCTCGAGTTCGGCCCCACCCGCTGACCA3 0 ) (Fig. 1B). Subcloning and transformation were done as described in [24]. The induced dsRNA is expected to form a 425-bp hairpin linked by a 148 nucleotide loop. Ten independent transgenic lines were recovered and the efficiency of the silencing was examined by RT-PCR in flies induced with the ubiquitous drivers Act-GAl4 or Tub-Gal4 (Fig. 1B). The endogenous transcript was reduced between 60% and 98% depending on the line, the number of transgenic copies, and rearing temperature. On average, the increase of temperature from 25 to 29 °C improved the silencing by at least 25%. Two insertions with strong silencing effect, one in chromosome II (line i3066-42a) and one in chromosome III (line i3066-17a), were selected for further characterization. Semi-quantitative RT-PCR. One microgram of total RNA extracted from adults flies was reverse transcribed using oligo-(dT) and 1/20 of the reaction was amplified for 29 or 32 cycles with the primers (5 0 -GA GCACCATAAAGCAGCGA-3 0 ) and (5 0 -CTCAGGGACGAATGGTC TCCA-3 0 ) targeting the 3 0 end of the gene excluded in the RNAi construct and the primers (5 0 -GACCATCCGCCCAGCATACAGG C-3 0 ) and (5 0 -GAGAACGCAGGCGACCGTTGG-3 0 ) targeting the housekeeping gene rp-49 as internal control. Primer concentrations were 200 nM for CG3066 and 40 nM for rp-49. Quantification of the bands was done by densitometry using the 1D program (Kodak). In situ hybridization. Single stranded DNA probes were generated by asymmetric PCR. 200 ng of the 746 bp band amplified with the forward primer: (5 0 -ACGACTACCAGTTCAAGTTCA-3 0 ) and the reverse primer: (5 0 -TCAAAACCCCTTCGCATCAGC-3 0 ) corresponding to the 3 0 of the CG3066 transcripts was labeled using a digoxigenin-labeling kit (Roche Applied Science). The reverse primer was used to generate the antisense probe and the forward primer to produce the sense probe and used as a negative control. In situ hybridizations were carried out according to Tautz and Pfeifle [25]. Melanization and phenol oxidase activity. The paper assay of phenoloxidase activity was done by dropping the hemolymph of a single dissected larva on a filter paper soaked with 10 mM phosphate buffer (pH 6.5) containing 10 mM L-DOPA [26]. Photographs of the spots were taken at timed intervals. Phenoloxidase activity in the total protein extract was quantified 4 h after injury or infection with a mixture of Gram-negative (Escherichia coli) and Gram-positive (Microccocus luteus) bacteria. Five L3 instar larvae were homogenized in 100 ll cold 10 mM phosphate buffer. After centrifugation at 4 °C, 50 ll of clear lysate was collected avoiding the floating fat tissue. The protein concentration was determined using the Coomassie Plus Protein Assay kit (Pierce), using BSA as a standard. Ten micrograms of protein was assayed as in [14] measuring the absorbance at 490 nm after 3 min. Larva cuticle melanization was monitored by dissection and exposure to air for 1 h [27].
Results Materials and methods Drosophila stocks. Fly stocks were obtained from the Bloomington Stock Center unless otherwise specified. OregonR and w1118 were used as wild type controls. Mutant flies were: spaetzlerm7/spaetzlerm7 (spz); relishE20/relishE20 (rel); key1/key1 (key); tub1/tub2 (tube); the constitutive active form of Toll, Tl10b; domino (dom); Black cells (Bc) and the Gal4
The gene CG3066 encodes a clip-domain serine protease with four splicing forms Searches of amino acid sequence databases showed that Sp7 has the highest similarity to arthropod proteins containing a carboxyl-terminal serine protease domain
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
A
Sp
Clip
1077
Protease RC RB RA RD
B
w-
3066i
rp49 CG3066
388bp 310bp
Fig. 1. Gene structure (A) and silencing of CG3066 by transgenic RNAi (B). (A) Prediction of alternative splicing and putative polypeptides based on available Drosophila ESTs. The deduced amino acid sequence is drawn in dashed lines and functional domains are indicated above. Boxes represent exons. The signal peptide (Sp) indicating a secreted protein is absent in the polypeptide deduced from RB. A double line at the bottom indicates the product of the RT-PCR for expression quantification. A single line shows the amplified cDNA used for in situ hybridization and an arrowed double line the genomic DNA-cDNA sequence expressed in the transgenic RNAi lines. (B) Agarose gel electrophoresis of the RT-PCR products amplified for 29 cycles (odd lanes) or 32 cycles (even lanes). The 310 bp band represents the CG3066 transcripts and the 388 bp band the ribosomal protein gene rp-49 used as internal control.
and an amino-terminal clip domain. In addition to orthologous genes in sibling Drosophila species, Sp7 is highly similar to pro-POAE from M. sexta [28] and H. diomphalia [11]. Based on Drosophila ESTs, four transcript variants can be distinguished for the gene Sp7 (Fig. 1A). Two predicted transcripts (RC and RA) encode for a 391 aa secreted protein of the trypsin protease family containing the catalytic triad of serine proteases, His182, Asp244, and Ser341, embedded in a region of highly conserved residues. The clip domain, characterized by six Cys residues at conserved positions, is found downstream of the signal peptide. Upon activation by cleavage, the polypeptide is converted into a two-chain active form composed of a light clip-domain chain and a heavy protease chain [29]. Sp7 harbors the pair of Cys residues between His182 and Asp244 that are characteristic of pro-POAEs [6]. It has been proposed that the clip-domain functions as regulator of the protease domain by shielding the catalytic site [30]. Recently, it has been shown that a recombinant form of the serine protease homolog Vm50 from the endoparasitoid venom of C. rubecula impairs the activation of PO in its host [31]. Because the protease catalytic triad is non-functional, the effect was attributed to inhibitory interaction with pro-POAEs probably mediated by the clip-domain. The transcript RD (Fig. 1A) encodes for a secreted peptide harboring the clip-domain and lacking the protease region while the predicted protein of the transcript RB lacks the clip domain and the signal peptide, suggesting a constitutively active protease that is localized in the cytoplasm.
Silencing of CG3066 impairs the melanization reaction In order to address the function of the gene CG3066, we generated several inducible lines (UAS-iCG3066) that express double strand RNA targeting all the predicted transcripts of the gene (Fig. 1A). We first tested the viability of these lines and found that ubiquitous silencing of Sp7 impairs viability. In the cross: Act-Gal-4/CyO x UASiCG3066-17a/UAS-iCG3066-17a, the number of emerged CG3066-depleted flies was 25% in contrast to the expected 50%. Similar results were obtained with the line i3066-42a. Because of the sequence similarity with pro-POAEs in other insect species, we tested whether Sp7 is involved melanization. Septic injuries in wild type flies produce intense melanization at the wound site (Fig. 2A). In CG3066-depleted flies, however we observed a reduction or lack of melanization (Fig. 2B) indicating that Sp7 is required for
A
B
Wt
3066i
Fig. 2. Melanization phenotype after septic injury. Adult flies pricket with a needle dipped in a concentrated mixture of E. coli and M. luteus. The injured sites are indicated by arrows. (A) Wild type fly showing intense melanization at the injured site. (B) Fly carrying UAS-iCG3066 driven by Tub-Gal4 showing severe reduction of melanization.
1078
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
Sp7 is expressed in hemocytes and in the larval lymph gland
hemolymph melanization after septic injury. Next, we estimated the susceptibility to microbial infection in adult flies due to the previously reported micro-array data that indicate upregulation of Sp7 upon infection [32,33]. The survival rate decreased slightly after septic injury with the fungi Aspergillus fumigatus and the Gram-positive bacteria Pseudomonas aurigenosa but never reached the level of lethality obtained with control mutants of the Toll or IMD pathway [34,35].
Sp7 transcripts are first detected at stage 11 in discrete cells localized closed to the gnathal region. At stage 13, Sp7 expressing cells have increased in number and appear dispersed throughout the embryo (Fig. 4A). The number of Sp7 positive cells varies between 50 and 150 cells per embryo at stage 13. During organogenesis the number does not increase and positive cells are more abundant under the cuticle, in the proventricular zone, and close to the hindgut. In L3 larvae the transcripts are expressed at low levels in the lymph glands (Fig. 4B). The staining is more intense at the primary lobe where the more mature larval pre-hemocytes are positioned [36]. In dissected larvae, clusters of Sp7 expressing cells are observed attached to diverse organs (Fig. 2C). Those blood cells have a round morphology with large nuclei indicating a crystal cell nature.
The phenoloxidase pathway is inhibited in CG3066 knockdowns To validate the melanization phenotype, we measured the phenoloxidase activity of CG3066-depleted flies using L-DOPA as substrate and monitored the conversion to melanin (Fig. 3). We observed a reduction in PO activity in the CG3066-depleted flies both by using hemolymph (Fig. 3A) or by using total protein extract of larvae (Fig. 3B). Spontaneous melanization was monitored in dissected larvae (Fig. 3C). While in wild type animals, melanization of the hemolymph accumulated onto the cuticle within 1 h (Wt), the deposition was much lower in the CG3066-depleted larvae (i) as well as in the melanization deficient mutant Black cells (Bc) in which the melanization is constrained to crystal cells.
Silencing of Sp7 in crystal cells is sufficient to impair the hemolymph melanization To test whether the depletion of CG3066 in blood cells is sufficient to impair the phenoloxidase reaction, we combined two Gal-4 lines with our UAS-iCG3066 lines. The Gal4 driver hem-Gal4-UAS-GFP is expressed in the majority of mature larval hemocytes [37] and the LZ-GAl4-UAS-
C
A
B
Wt i
Bc
Wt
Wt
i
0,300 0,250
Bc
Abs.490
0,200 0,150 0,100 0,050 0,000
Bc
Dom
Wt
3066i
Fig. 3. (A) Phenoloxidase activities in hemolymph. Staining after deposition of the hemolymph of a single larva on a paper soaked with L-3,4dihydroxyphenylalanine, the substrate of phenoloxidase. To the left, spots produced by Sp7-depleted mutant (broken line) versus wild type larvae (continuous line). To the right, Black cells (Bc) mutant versus wild type. (B) PO activity of total protein extract from L3 larvae of various genotypes. Each independent experiment was performed in triplicate. Activity levels in extracts from the Sp7-depleted larvae are similar to those obtained in Bc and Domino (Dom) mutants, and reduced by 50% compared to wild type larvae. (C) Larvae were dissected and exposed to air. Melanization of the hemolymph generates melanin deposition in wild type larvae (Wt) but is absent in Sp7, depleted larvae (i). In Bc mutants, the melanization reaction in the hemolymph is blocked and no deposition occurs. In these mutants, aberrant crystal cells are seen through the cuticle.
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
1079
C
A
B LG DV
Fig. 4. Localization of the Sp7 transcript by in situ hybridization. (A) Sp7 is expressed during embryogenesis in dispersed blood cells. (B) In larval tissues, expression is observed in the lymph glands (LG), the larval hematopoietic organ that flanks the dorsal vessel (DV), and in mature blood cells that appear in clusters attached to diverse tissues. In (C), stained blood cells attach to the dissected larval fat body.
GFP is specifically expressed in precursors of crystal cells [23]. Depletion of Sp7 using the LZ driver reduces the PO reaction to similar levels obtained with ubiquitous drivers (Fig. 5), indicating that depletion of Sp7 in crystal cell precursors is sufficient to impair the PO reaction. Surprisingly, the Hemese driver, a pan-hemocyte marker, did not show any reduction of the reaction. The explanation for these apparently contradictory results lies in the delay of the silencing effect. In the Hemese driver, the activation of Gal4 occurs at larval stages in mature blood cells. At this stage of development, the Sp7 protein is already present, at least in crystal cells of embryonic origin. In contrast, LZ-Gal4 is an embryonic driver and silencing occurs before translation of the protein.
Fig. 5. Silencing of Sp7 in crystal cells is sufficient to impair the melanization reaction. (A) Blood cells in hemolymph are marked with GFP induced by the Lz-Gal4 driver (above) and the Hem-Gal4 driver (below). The crystal cells marked by LZ-GFP refract due to their typical crystalline inclusion. The Hem driver marks the majority of larval circulating blood cells but some pro-hemocytes lack expression of the marker (arrows). (B) PO activity in larval hemolymph. Signals generated after bleeding single larvae on L-Dopa soaked paper. The stains were photographed 5 (left) and 8 min (righ) after deposition to illustrate the time course of the reaction.
Sp7 is up-regulated upon infection We observed an increase of transcription upon clean injury of adult wild type flies (Fig. 6). The up-regulation is enhanced by septic injury with a mixture of Gram-positive and Gram-negative bacteria. This up-regulation has been reported in a micro-array analysis of immune challenged flies [32]. Our own results show a lower up-regulation (2.3- versus 5.5-fold in the micro-array report). This difference could be attributed to the semi-quantitative nature of our assay in which we do not quantify maximal changes, instead, the assay is conservative, detecting suboptimal differences. Using flies deficient in components of the Toll and IMD pathways, we asked whether the transcription of Sp7 is dependent on these immune-pathways. As is shown, neither of the mutants blocked the transcription of CG3066. We observed instead a slight increase of transcription that reached the level of clean injured flies (w-p); this is probably due to their permanently challenged-state caused by the reduced immune-defense of these flies. Flies carrying a constitutively active form of Toll (Tl10b) show a consistent increase in Sp7 transcription. To confirm this observation, we performed in situ hybridization in these mutants. As is evident in Fig. 6C, the increase of total transcription seems to be the result of an increase in the total number of cells expressing CG3066. An increased number of circulating hemocytes have previously been reported in Tl10b larvae [38]. Discussion In this study, we have identified the first serine protease in Drosophila that is necessary for activation of the PO pathway. The serine protease Sp7 is expressed in blood cells, it is constitutively transcribed, and it is up-regulated
1080
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
w-
A
w-p
w-i
spz
key
rel
tube
Tl10b
rp49 CG3066
388bp 310bp
Relative mRNA
B
w-
w-p
w-i
spz
key
rel
tube
Tl1 0 b
C
Fig. 6. Transcriptional regulation of Sp7. (A) Estimation of the Sp7 mRNA in adult flies of different genotypes by RT-PCR. Total RNA was extracted 3.5 h after clean pricking (w-p), infection with a mixture of E. coli and M. luteus (w-i,) or untreated flies (w-spz, key, rel, tube, and Tl10b). The amounts of Sp7 RNA (310 bp band) and the rp49 internal control (388 bp band) were visualized on an agarose gel after 28 (left well) and 31 cycles (right well) of RT-PCR. (B) Relative amounts of Sp7 mRNA standardized against rp49. The mean values are the result of three independent experiments. The RNA level of uninfected flies was set to 1. (C) In situ hybridization of Tl10b embryos using the Sp7 probe. The number of blood cells expressing Sp7 is considerably increased in comparison to wild type embryos.
after perforation of the cuticle, septic injury or in flies with mutations in the major immune pathways (Toll and IMD). Transcription of SP7 is first observed during embryogenesis. Specific silencing of Sp7 in crystal cells using the early transcription factor Lozenge is sufficient to impair hemolymph melanization. This study gives new insight at a molecular and cellular level into the melanization reaction in the fruit fly. It has been proposed that protease cascades are key components for the activation of phenoloxidase zymogens (pro-PO) to the active PO. In Drosophila, an inhibitor of serine proteases, the serpin Spn27A has been reported to be involved in the regulation of the pro-PO [14,15]. Null mutants show constitutive melanization while injured larvae react by uncontrolled melanization that expands throughout the body cavity. The function of Spn27A is to inhibit the activation of the PO probably by blocking pro-POAE. Thus, hemolymph melanization requires either the increase of the pro-POAE or the elimination
of the serpin Spn27A, a process that is probably dependent on another protease. It is possible that Sp7 acts as a pro-POAE, promoting directly or through a protease cascade the cleavage activation of the pro-PO zymogens. The same effect, however, could be achieved by promoting proteolytic degradation or titration of the serpin. At this stage, we are unable to discriminate between those two possibilities. However, in Toll pathway mutants, depletion of the serpin Spn27A is inhibited and no activation of PO occurs [14]. We did not observe a block of Sp7 in Toll mutants as would be expected if Spn27a was cleared from the hemolymph by the serine protease Sp7. Therefore, it is not likely that Spn27A is degraded by the transcriptionally regulated serine protease. The alternatively spliced form Sp7-PB (Fig. 7) could, however act, in the rapid removal of the serpin. Our results show that the serine protease Sp7 acts as activator of the PO but does not allow us to discriminate between the function of the different spliced variants.
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
1081
Acknowledgments
Toll
We thank the ‘‘Crafoorsdka Stiftelsen’’ and the ‘‘Nilsson-Ehle Fonden’’ for support. We also thank Ylva Engstro¨m and Dan Hultmark for providing the tub1/tub2 and Tl10b Drosophila stocks, Sol Da Rocha for technical assistance, Edgar Pera, Ulrich Theopold, and Michael Williams for critical reading of the manuscript.
Sp7 Spn27A
Sp7-PB
References
? Sp7-PD Sp7-PC
?
Spn27A
Pro-PO
PO
Phenols
Quinones
Melanin
Fig. 7. Model of the action of the serine protease SP7 in hemolymph melanization of Drosophila. Rupture of crystal cells at an injury site releases the components of the crystalline inclusion (Pro-PO) and the cytosolic Sp7-PB protease. Sp7-PB activates pro-PO and titrates out the serpin Spn27A. Diffusible products of the reaction activate the Toll receptor in intact cells. Toll triggers the signal transduction pathway that modulates the transcriptionally regulated transcripts Sp7-PC and Sp7-PD close to damaged tissues or the inhibitor Spn27a in stabilized tissues. The inhibitory effect of the spliced variants Sp7-PB and Sp7-PD is speculative, therefore indicated with question marks.
At the cellular level, we suggest the occurrence of two types of melanization waves, an immediate reaction that is probably mediated by the rupture of crystal cells and a late reaction that is controlled at the transcriptional level. The perforation of the cuticle produces a rapid locally restricted melanization. It is likely that the first reaction is mediated by the burst of nearby crystal cells and is not transcriptionally regulated. Later, modulation of the reaction may be mediated by the same type of cells but using a transcriptionally regulated pathway probably via the Toll receptor. The existence of the different translational products of Sp7 could facilitate the tight spatial and temporal control of the reaction. In the case of injury infection, disruption of crystal cells could initiate an instantaneous melanization response by releasing the melanization components into the hemolymph (Fig. 7). We reasoned further that the cytoplasmic form of SP7 (SP7-PB) could be the first cleavage activator of the Pro-PO and/or the mediator of the degradation-titration of Spn27A. Both actions reinforce an instantaneous melanization at the damaged tissue. Then, activated surrounding crystal cells could regulate the melanization by the Sp7 secreted pathway. Restriction of the reaction to the injured sites is finally achieved by de novo synthesis and release of the inhibitor Spn27A from the fat body and nearby blood cells. The production of specific antibodies and transgenic flies expressing the alternative transcripts could help to test this hypothesis and further clarify the molecular events in the Drosophila melanization reaction.
[1] J.A. Hoffmann, The immune response of Drosophila, Nature 426 (2003) 33–38. [2] D. Hultmark, Drosophila immunity: paths and patterns, Curr. Opin. Immunol. 15 (2003) 12–19. [3] M.R. Kanost, H. Jiang, X.Q. Yu, Innate immune responses of a lepidopteran insect, Manduca sexta, Immunol. Rev. 198 (2004) 97– 105. [4] U. Theopold, O. Schmidt, K. Soderhall, M.S. Dushay, Coagulation in arthropods: defence, wound closure and healing, Trends Immunol. 25 (2004) 289–294. [5] M.J. Galko, M.A. Krasnow, Cellular and genetic analysis of wound healing in Drosophila larvae, PLoS Biol. 2 (2004) E239. [6] H. Jiang, Y. Wang, M.R. Kanost, Pro-phenol oxidase activating proteinase from an insect, Manduca sexta: a bacteria-inducible protein similar to Drosophila easter, Proc. Natl. Acad. Sci. USA 95 (1998) 12220–12225. [7] D. Satoh, A. Horii, M. Ochiai, M. Ashida, Prophenoloxidaseactivating enzyme of the silkworm, Bombyx mori. Purification, characterization, and cDNA cloning, J. Biol. Chem. 274 (1999) 7441–7453. [8] R. Wang, S.Y. Lee, L. Cerenius, K. Soderhall, Properties of the prophenoloxidase activating enzyme of the freshwater crayfish, Pacifastacus leniusculus, Eur. J. Biochem. 268 (2001) 895–902. [9] H. Jiang, Y. Wang, X.Q. Yu, M.R. Kanost, Prophenoloxidaseactivating proteinase-2 from hemolymph of Manduca sexta. A bacteria-inducible serine proteinase containing two clip domains, J. Biol. Chem. 278 (2003) 3552–3561. [10] Y. Tong, H. Jiang, M.R. Kanost, Identification of plasma proteases inhibited by Manduca sexta serpin-4 and serpin-5 and their association with components of the prophenol oxidase activation pathway, J. Biol. Chem. 280 (2005) 14932–14942. [11] S.Y. Lee, T.H. Kwon, J.H. Hyun, J.S. Choi, S.I. Kawabata, S. Iwanaga, B.L. Lee, In vitro activation of pro-phenol-oxidase by two kinds of pro-phenol-oxidase-activating factors isolated from hemolymph of coleopteran, Holotrichia diomphalia larvae, Eur. J. Biochem. 254 (1998) 50–57. [12] A. Aspan, T.S. Huang, L. Cerenius, K. Soderhall, cDNA cloning of prophenoloxidase from the freshwater crayfish Pacifastacus leniusculus and its activation, Proc. Natl. Acad. Sci. USA 92 (1995) 939– 943. [13] Y. Tong, M.R. Kanost, Manduca sexta serpin-4 and serpin-5 inhibit the prophenol oxidase activation pathway: cDNA cloning, protein expression, and characterization, J. Biol. Chem. 280 (2005) 14923– 14931. [14] P. Ligoxygakis, N. Pelte, C. Ji, V. Leclerc, B. Duvic, M. Belvin, H. Jiang, J.A. Hoffmann, J.M. Reichhart, A serpin mutant links Toll activation to melanization in the host defence of Drosophila, EMBO J. 21 (2002) 6330–6337. [15] E. De Gregorio, S.J. Han, W.J. Lee, M.J. Baek, T. Osaki, S. Kawabata, B.L. Lee, S. Iwanaga, B. Lemaitre, P.T. Brey, An immune-responsive Serpin regulates the melanization cascade in Drosophila, Dev. Cell 3 (2002) 581–592. [16] T.M. Rizki, R.M. Rizki, E. Grell, A mutant affecting the crystal cells in Drosophila melanogaster, Wilhelm RouxÕs Arch. Dev. Biol. 188 (1980) 91–99.
1082
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
[17] P. Irving, J.M. Ubeda, D. Doucet, L. Troxler, M. Lagueux, D. Zachary, J.A. Hoffmann, C. Hetru, M. Meister, New insights into Drosophila larval haemocyte functions through genome-wide analysis, Cell Microbiol. 7 (2005) 335–350. [18] M.J. Williams, I. Ando, D. Hultmark, Drosophila melanogaster Rac2 is necessary for a proper cellular immune response, Genes Cells 10 (2005) 813–823. [19] S. de Morais Guedes, R. Vitorino, R. Domingues, K. Tomer, A.J. Correia, F. Amado, P. Domingues, Proteomics of immune-challenged Drosophila melanogaster larvae hemolymph, Biochem. Biophys. Res. Commun. 328 (2005) 106–115. [20] B. Duvic, J.A. Hoffmann, M. Meister, J. Royet, Notch signaling controls lineage specification during Drosophila larval hematopoiesis, Curr. Biol. 12 (2002) 1923–1927. [21] L. Waltzer, G. Ferjoux, L. Bataille, M. Haenlin, Cooperation between the GATA and RUNX factors Serpent and Lozenge during Drosophila hematopoiesis, EMBO J. 22 (2003) 6516–6525. [22] A. Braun, J.A. Hoffmann, M. Meister, Analysis of the Drosophila host defense in domino mutant larvae, which are devoid of hemocytes, Proc. Natl. Acad. Sci. USA 95 (1998) 14337–14342. [23] T. Lebestky, T. Chang, V. Hartenstein, U. Banerjee, Specification of Drosophila hematopoietic lineage by conserved transcription factors, Science 288 (2000) 146–149. [24] C. Castillejo-Lopez, W.M. Arias, S. Baumgartner, The fat-like gene of Drosophila is the true orthologue of vertebrate fat cadherins and is involved in the formation of tubular organs, J. Biol. Chem. 279 (2004) 24034–24043. [25] D. Tautz, C. Pfeifle, A non-radioactive in situ hybridization method for the localization of specific RNAs in Drosophila embryos reveals translational control of the segmentation gene hunchback, Chromosoma 98 (1989) 81–85. [26] R.P. Sorrentino, C.N. Small, S. Govind, Quantitative analysis of phenol oxidase activity in insect hemolymph, Biotechniques 32 (2002) 815–816, 818, 820, 822–813. [27] R. Bettencourt, H. Asha, C. Dearolf, Y.T. Ip, Hemolymph-dependent and -independent responses in Drosophila immune tissue, J. Cell Biochem. 92 (2004) 849–863. [28] Z. Zou, Y. Wang, H. Jiang, Manduca sexta prophenoloxidase activating proteinase-1 (PAP-1) gene: organization, expression, and regulation by immune and hormonal signals, Insect. Biochem. Mol. Biol. 35 (2005) 627–636.
[29] T. Muta, T. Miyata, Y. Misumi, F. Tokunaga, T. Nakamura, Y. Toh, Y. Ikehara, S. Iwanaga, Limulus factor C. An endotoxin-sensitive serine protease zymogen with a mosaic structure of complement-like, epidermal growth factor-like, and lectin-like domains, J. Biol. Chem. 266 (1991) 6554–6561. [30] J. Ross, H. Jiang, M.R. Kanost, Y. Wang, Serine proteases and their homologs in the Drosophila melanogaster genome: an initial analysis of sequence conservation and phylogenetic relationships, Gene 304 (2003) 117–131. [31] G. Zhang, Z.Q. Lu, H. Jiang, S. Asgari, Negative regulation of prophenoloxidase (proPO) activation by a clip-domain serine proteinase homolog (SPH) from endoparasitoid venom, Insect. Biochem. Mol. Biol. 34 (2004) 477–483. [32] E. De Gregorio, P.T. Spellman, G.M. Rubin, B. Lemaitre, Genomewide analysis of the Drosophila immune response by using oligonucleotide microarrays, Proc. Natl. Acad. Sci. USA 98 (2001) 12590– 12595. [33] M. Boutros, H. Agaisse, N. Perrimon, Sequential activation of signaling pathways during innate immune responses in Drosophila, Dev. Cell 3 (2002) 711–722. [34] B. Lemaitre, E. Kromer-Metzger, L. Michaut, E. Nicolas, M. Meister, P. Georgel, J.M. Reichhart, J.A. Hoffmann, A recessive mutation, immune deficiency (imd), defines two distinct control pathways in the Drosophila host defense, Proc. Natl. Acad. Sci. USA 92 (1995) 9465–9469. [35] B. Lemaitre, E. Nicolas, L. Michaut, J.M. Reichhart, J.A. Hoffmann, The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults, Cell 86 (1996) 973–983. [36] S.H. Jung, C.J. Evans, C. Uemura, U. Banerjee, The Drosophila lymph gland as a developmental model of hematopoiesis, Development 132 (2005) 2521–2533. [37] E. Kurucz, C.J. Zettervall, R. Sinka, P. Vilmos, A. Pivarcsi, S. Ekengren, Z. Hegedus, I. Ando, D. Hultmark, Hemese, a hemocyte-specific transmembrane protein, affects the cellular immune response in Drosophila, Proc. Natl. Acad. Sci. USA 100 (2003) 2622–2627. [38] C.J. Zettervall, I. Anderl, M.J. Williams, R. Palmer, E. Kurucz, I. Ando, D. Hultmark, A directed screen for genes involved in Drosophila blood cell activation, Proc. Natl. Acad. Sci. USA 101 (2004) 14192–14197.
The serine protease Sp7 is expressed in blood cells and regulates the melanization reaction in Drosophila Casimiro Castillejo-Lo´pez *, Udo Ha¨cker Department of Experimental Medical Science and Lund Strategic Research Center for Stem Cell Biology and Cell Therapy, Lund University, BMC B13, Klinikgatan 26, 22184 Lund, Sweden Received 23 September 2005 Available online 21 October 2005
Abstract Serine proteases play a central role in defense against pathogens by regulating processes such as blood clotting, melanization of injured surfaces, and proteolytic activation of signaling pathways involved in innate immunity. Here, we present the functional characterization of the Drosophila serine protease Sp7 (CG3006) by inducible RNA interference. We show that Sp7 is constitutively expressed in blood cells during embryonic and larval stages. Silencing of the gene impairs the melanization reaction upon injury. Our data demonstrate that Sp7 is required for phenoloxidase activation and its activity is restricted to a subclass of blood cells, the crystal cells. Transcriptional up-regulation of Sp7 was observed after clean, septic injury and in flies expressing an activated form of Toll; however, mutations in the Toll or the IMD pathway did not abolish expression of Sp7, indicating the existence of other regulatory pathways and/or independent basal transcription. Ó 2005 Elsevier Inc. All rights reserved. Keywords: Drosophila; RNAi; Serine protease; Melanization; Phenoloxidase; Wound; Blood cells; Innate immunity
Innate immunity in insects consists of a humoral response based on antimicrobial peptides and a cellular response mediated by blood cells that promote phagocytosis, encapsulation of pathogens, and melanization [1–3]. The immediate immune response in Drosophila is the melanization reaction observed at the site of injury or on the surface of pathogens and parasites invading the hemocoel. Perforation of the cuticle by injury or by parasitic infection initiates a proteolytic reaction that leads to rapid hemolymph coagulation and deposition of melanin. This reaction plays an important role in wound healing where melanized tissue forms a plug that restricts bleeding [4,5]. The melanization reaction requires the activation of phenoloxidase (PO) that catalyzes the conversion of phenols to quinones, which spontaneously polymerize to form melanin (Fig. 7). In vitro studies have *
Corresponding author. Fax: +46 46 2220899. E-mail address: [email protected] (C. Castillejo-Lo´pez). 0006-291X/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2005.10.042
shown that PO exists as an inactive precursor, prophenoloxidase (pro-PO), which is activated by a stepwise process involving serine protease cascades [6–8]. As excessive melanin is deleterious to the host, the processing of pro-PO must be strictly regulated to restrict the melanization reaction. Inactive PO is cleaved into active PO by a protease known as prophenoloxidase-activating enzyme (pro-POAE). Pro-POAEs have been identified in the tobacco hornworm, Manduca sexta [9,10], in the silkworm, Bombyx mori [7], in the beetle, Holotrichia diomphalia [11], and in the crayfish, Pacifastus leniusculus [12] but so far, no such activator gene has been studied in vivo by mutational analysis in Drosophila. Negative regulators of the catalysis have been characterized and include members of the serine protease inhibitor (serpin) superfamily [13]. Serpins act as suicide substrate inhibitors of serine proteases by forming irreversible complexes. Biochemical and genetic characterization of one of them, serpin-27A (Spn27A) in Drosophila, demonstrates that the protein encoded by this gene regulates the
1076
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
melanization cascade through the specific inhibition of pro-PO processing by the terminal serine protease proPOAE [14,15]. In addition to the regulation at the biochemical level, melanization is controlled spacially and temporally by blood cells. Drosophila has three blood cell types; the plasmatocytes, phagocytic cells that comprise 95% of circulating hemocytes and synthesize antimicrobial peptides; the crystal cells, named for the crystalline inclusion bodies and lamellocytes, the largest cells normally not present in healthy larvae and involved in the encapsulation response against intruders that are too large to be phagocytized. Rizki et al. [16] suggested that crystal cells are the only source of pro-PO in the fruit fly. However, lamellocytes that appeared after parasitoid wasp infection could contribute to the melanization of the capsule that isolated the parasite eggs [17,18]. The completion of the genome sequence has shown that Drosophila melanogaster has three pro-PO genes, Bc (DoxA1, CG5779), DoxA3 (CG2952), and CG8193. A recent proteomic analysis of larval hemolymph has shown a reduction of the pro-PO protein spots upon immune challenge [19]. The reduction is attributed to proteolytic activation, indicating that these genes code for proteins that are processed upon challenge probably by serine proteases. The protein annotation in this study corresponds to both the genes CG8193 and CG2952, and no further discrimination could be done. Based on histological location in non challenged embryos it has been inferred that pro-POs are expressed in crystal cells [20,21]. Recently it has been demonstrated that this is true for two of them, while CG2952 is exclusively expressed in lamellocytes [17]. The relative contribution of crystal cells and lamellocytes to the melanotic encapsulation of parasitoid wasp eggs is not understood. It is recognized, however, that the melanization at injury sites is mediated exclusively by crystal cells mainly because lamellocytes are not present in healthy larvae. Furthermore, three classical blood mutants: Domino (dom) which has fewer blood cells, Black Cells (Bc) which has aberrant crystal cells, and lozenge (Lz) which lacks crystal cells, all have severely impaired hemolymph melanization [22,23]. Here, we present the functional characterization of the first serine protease in any arthropod species that is necessary for activation of the PO pathway and melanization upon injury. Sp7 is constitutively expressed in blood cells and up-regulated after injury, infection, and in immune deficient mutations. Silencing of transcription in crystal cells using the early transcription factor Lozenge is sufficient to impair hemolymph melanization.
driver lines Act-Gal4/CyO, Tub-Gal4/TM3, hem-Gal4-GFP, and Lz-Gal4GFP. RNAi transgenic lines were generated by P element-mediated transformation using w1118 as recipient stock. Flies were maintained on standard medium at 19, 25, or 29 °C, as necessary. Transgenenic RNA interference. A pUAST-vector containing genomic and cDNA sequences in opposite direction was constructed to produce dsRNA. All the described transcripts of the gene CG3066 were targeted. The genomic fragment encompassing 729 bp and including the last three introns was amplified with the primers: (5 0 - GGGCGGCCGCAACGA CACTGCTATTGACG-3 0 ) and (5 0 -GGCTCGAGGTGCTCTTGCGGG CTATAGA-3 0 ). The corresponding cDNA encompasses 425 bp and was amplified with the primers: (5 0 - GGGGTACCACTGCTTGAGTACG TGGATA-3 0 ) and (5 0 -GGCTCGAGTTCGGCCCCACCCGCTGACCA3 0 ) (Fig. 1B). Subcloning and transformation were done as described in [24]. The induced dsRNA is expected to form a 425-bp hairpin linked by a 148 nucleotide loop. Ten independent transgenic lines were recovered and the efficiency of the silencing was examined by RT-PCR in flies induced with the ubiquitous drivers Act-GAl4 or Tub-Gal4 (Fig. 1B). The endogenous transcript was reduced between 60% and 98% depending on the line, the number of transgenic copies, and rearing temperature. On average, the increase of temperature from 25 to 29 °C improved the silencing by at least 25%. Two insertions with strong silencing effect, one in chromosome II (line i3066-42a) and one in chromosome III (line i3066-17a), were selected for further characterization. Semi-quantitative RT-PCR. One microgram of total RNA extracted from adults flies was reverse transcribed using oligo-(dT) and 1/20 of the reaction was amplified for 29 or 32 cycles with the primers (5 0 -GA GCACCATAAAGCAGCGA-3 0 ) and (5 0 -CTCAGGGACGAATGGTC TCCA-3 0 ) targeting the 3 0 end of the gene excluded in the RNAi construct and the primers (5 0 -GACCATCCGCCCAGCATACAGG C-3 0 ) and (5 0 -GAGAACGCAGGCGACCGTTGG-3 0 ) targeting the housekeeping gene rp-49 as internal control. Primer concentrations were 200 nM for CG3066 and 40 nM for rp-49. Quantification of the bands was done by densitometry using the 1D program (Kodak). In situ hybridization. Single stranded DNA probes were generated by asymmetric PCR. 200 ng of the 746 bp band amplified with the forward primer: (5 0 -ACGACTACCAGTTCAAGTTCA-3 0 ) and the reverse primer: (5 0 -TCAAAACCCCTTCGCATCAGC-3 0 ) corresponding to the 3 0 of the CG3066 transcripts was labeled using a digoxigenin-labeling kit (Roche Applied Science). The reverse primer was used to generate the antisense probe and the forward primer to produce the sense probe and used as a negative control. In situ hybridizations were carried out according to Tautz and Pfeifle [25]. Melanization and phenol oxidase activity. The paper assay of phenoloxidase activity was done by dropping the hemolymph of a single dissected larva on a filter paper soaked with 10 mM phosphate buffer (pH 6.5) containing 10 mM L-DOPA [26]. Photographs of the spots were taken at timed intervals. Phenoloxidase activity in the total protein extract was quantified 4 h after injury or infection with a mixture of Gram-negative (Escherichia coli) and Gram-positive (Microccocus luteus) bacteria. Five L3 instar larvae were homogenized in 100 ll cold 10 mM phosphate buffer. After centrifugation at 4 °C, 50 ll of clear lysate was collected avoiding the floating fat tissue. The protein concentration was determined using the Coomassie Plus Protein Assay kit (Pierce), using BSA as a standard. Ten micrograms of protein was assayed as in [14] measuring the absorbance at 490 nm after 3 min. Larva cuticle melanization was monitored by dissection and exposure to air for 1 h [27].
Results Materials and methods Drosophila stocks. Fly stocks were obtained from the Bloomington Stock Center unless otherwise specified. OregonR and w1118 were used as wild type controls. Mutant flies were: spaetzlerm7/spaetzlerm7 (spz); relishE20/relishE20 (rel); key1/key1 (key); tub1/tub2 (tube); the constitutive active form of Toll, Tl10b; domino (dom); Black cells (Bc) and the Gal4
The gene CG3066 encodes a clip-domain serine protease with four splicing forms Searches of amino acid sequence databases showed that Sp7 has the highest similarity to arthropod proteins containing a carboxyl-terminal serine protease domain
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
A
Sp
Clip
1077
Protease RC RB RA RD
B
w-
3066i
rp49 CG3066
388bp 310bp
Fig. 1. Gene structure (A) and silencing of CG3066 by transgenic RNAi (B). (A) Prediction of alternative splicing and putative polypeptides based on available Drosophila ESTs. The deduced amino acid sequence is drawn in dashed lines and functional domains are indicated above. Boxes represent exons. The signal peptide (Sp) indicating a secreted protein is absent in the polypeptide deduced from RB. A double line at the bottom indicates the product of the RT-PCR for expression quantification. A single line shows the amplified cDNA used for in situ hybridization and an arrowed double line the genomic DNA-cDNA sequence expressed in the transgenic RNAi lines. (B) Agarose gel electrophoresis of the RT-PCR products amplified for 29 cycles (odd lanes) or 32 cycles (even lanes). The 310 bp band represents the CG3066 transcripts and the 388 bp band the ribosomal protein gene rp-49 used as internal control.
and an amino-terminal clip domain. In addition to orthologous genes in sibling Drosophila species, Sp7 is highly similar to pro-POAE from M. sexta [28] and H. diomphalia [11]. Based on Drosophila ESTs, four transcript variants can be distinguished for the gene Sp7 (Fig. 1A). Two predicted transcripts (RC and RA) encode for a 391 aa secreted protein of the trypsin protease family containing the catalytic triad of serine proteases, His182, Asp244, and Ser341, embedded in a region of highly conserved residues. The clip domain, characterized by six Cys residues at conserved positions, is found downstream of the signal peptide. Upon activation by cleavage, the polypeptide is converted into a two-chain active form composed of a light clip-domain chain and a heavy protease chain [29]. Sp7 harbors the pair of Cys residues between His182 and Asp244 that are characteristic of pro-POAEs [6]. It has been proposed that the clip-domain functions as regulator of the protease domain by shielding the catalytic site [30]. Recently, it has been shown that a recombinant form of the serine protease homolog Vm50 from the endoparasitoid venom of C. rubecula impairs the activation of PO in its host [31]. Because the protease catalytic triad is non-functional, the effect was attributed to inhibitory interaction with pro-POAEs probably mediated by the clip-domain. The transcript RD (Fig. 1A) encodes for a secreted peptide harboring the clip-domain and lacking the protease region while the predicted protein of the transcript RB lacks the clip domain and the signal peptide, suggesting a constitutively active protease that is localized in the cytoplasm.
Silencing of CG3066 impairs the melanization reaction In order to address the function of the gene CG3066, we generated several inducible lines (UAS-iCG3066) that express double strand RNA targeting all the predicted transcripts of the gene (Fig. 1A). We first tested the viability of these lines and found that ubiquitous silencing of Sp7 impairs viability. In the cross: Act-Gal-4/CyO x UASiCG3066-17a/UAS-iCG3066-17a, the number of emerged CG3066-depleted flies was 25% in contrast to the expected 50%. Similar results were obtained with the line i3066-42a. Because of the sequence similarity with pro-POAEs in other insect species, we tested whether Sp7 is involved melanization. Septic injuries in wild type flies produce intense melanization at the wound site (Fig. 2A). In CG3066-depleted flies, however we observed a reduction or lack of melanization (Fig. 2B) indicating that Sp7 is required for
A
B
Wt
3066i
Fig. 2. Melanization phenotype after septic injury. Adult flies pricket with a needle dipped in a concentrated mixture of E. coli and M. luteus. The injured sites are indicated by arrows. (A) Wild type fly showing intense melanization at the injured site. (B) Fly carrying UAS-iCG3066 driven by Tub-Gal4 showing severe reduction of melanization.
1078
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
Sp7 is expressed in hemocytes and in the larval lymph gland
hemolymph melanization after septic injury. Next, we estimated the susceptibility to microbial infection in adult flies due to the previously reported micro-array data that indicate upregulation of Sp7 upon infection [32,33]. The survival rate decreased slightly after septic injury with the fungi Aspergillus fumigatus and the Gram-positive bacteria Pseudomonas aurigenosa but never reached the level of lethality obtained with control mutants of the Toll or IMD pathway [34,35].
Sp7 transcripts are first detected at stage 11 in discrete cells localized closed to the gnathal region. At stage 13, Sp7 expressing cells have increased in number and appear dispersed throughout the embryo (Fig. 4A). The number of Sp7 positive cells varies between 50 and 150 cells per embryo at stage 13. During organogenesis the number does not increase and positive cells are more abundant under the cuticle, in the proventricular zone, and close to the hindgut. In L3 larvae the transcripts are expressed at low levels in the lymph glands (Fig. 4B). The staining is more intense at the primary lobe where the more mature larval pre-hemocytes are positioned [36]. In dissected larvae, clusters of Sp7 expressing cells are observed attached to diverse organs (Fig. 2C). Those blood cells have a round morphology with large nuclei indicating a crystal cell nature.
The phenoloxidase pathway is inhibited in CG3066 knockdowns To validate the melanization phenotype, we measured the phenoloxidase activity of CG3066-depleted flies using L-DOPA as substrate and monitored the conversion to melanin (Fig. 3). We observed a reduction in PO activity in the CG3066-depleted flies both by using hemolymph (Fig. 3A) or by using total protein extract of larvae (Fig. 3B). Spontaneous melanization was monitored in dissected larvae (Fig. 3C). While in wild type animals, melanization of the hemolymph accumulated onto the cuticle within 1 h (Wt), the deposition was much lower in the CG3066-depleted larvae (i) as well as in the melanization deficient mutant Black cells (Bc) in which the melanization is constrained to crystal cells.
Silencing of Sp7 in crystal cells is sufficient to impair the hemolymph melanization To test whether the depletion of CG3066 in blood cells is sufficient to impair the phenoloxidase reaction, we combined two Gal-4 lines with our UAS-iCG3066 lines. The Gal4 driver hem-Gal4-UAS-GFP is expressed in the majority of mature larval hemocytes [37] and the LZ-GAl4-UAS-
C
A
B
Wt i
Bc
Wt
Wt
i
0,300 0,250
Bc
Abs.490
0,200 0,150 0,100 0,050 0,000
Bc
Dom
Wt
3066i
Fig. 3. (A) Phenoloxidase activities in hemolymph. Staining after deposition of the hemolymph of a single larva on a paper soaked with L-3,4dihydroxyphenylalanine, the substrate of phenoloxidase. To the left, spots produced by Sp7-depleted mutant (broken line) versus wild type larvae (continuous line). To the right, Black cells (Bc) mutant versus wild type. (B) PO activity of total protein extract from L3 larvae of various genotypes. Each independent experiment was performed in triplicate. Activity levels in extracts from the Sp7-depleted larvae are similar to those obtained in Bc and Domino (Dom) mutants, and reduced by 50% compared to wild type larvae. (C) Larvae were dissected and exposed to air. Melanization of the hemolymph generates melanin deposition in wild type larvae (Wt) but is absent in Sp7, depleted larvae (i). In Bc mutants, the melanization reaction in the hemolymph is blocked and no deposition occurs. In these mutants, aberrant crystal cells are seen through the cuticle.
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
1079
C
A
B LG DV
Fig. 4. Localization of the Sp7 transcript by in situ hybridization. (A) Sp7 is expressed during embryogenesis in dispersed blood cells. (B) In larval tissues, expression is observed in the lymph glands (LG), the larval hematopoietic organ that flanks the dorsal vessel (DV), and in mature blood cells that appear in clusters attached to diverse tissues. In (C), stained blood cells attach to the dissected larval fat body.
GFP is specifically expressed in precursors of crystal cells [23]. Depletion of Sp7 using the LZ driver reduces the PO reaction to similar levels obtained with ubiquitous drivers (Fig. 5), indicating that depletion of Sp7 in crystal cell precursors is sufficient to impair the PO reaction. Surprisingly, the Hemese driver, a pan-hemocyte marker, did not show any reduction of the reaction. The explanation for these apparently contradictory results lies in the delay of the silencing effect. In the Hemese driver, the activation of Gal4 occurs at larval stages in mature blood cells. At this stage of development, the Sp7 protein is already present, at least in crystal cells of embryonic origin. In contrast, LZ-Gal4 is an embryonic driver and silencing occurs before translation of the protein.
Fig. 5. Silencing of Sp7 in crystal cells is sufficient to impair the melanization reaction. (A) Blood cells in hemolymph are marked with GFP induced by the Lz-Gal4 driver (above) and the Hem-Gal4 driver (below). The crystal cells marked by LZ-GFP refract due to their typical crystalline inclusion. The Hem driver marks the majority of larval circulating blood cells but some pro-hemocytes lack expression of the marker (arrows). (B) PO activity in larval hemolymph. Signals generated after bleeding single larvae on L-Dopa soaked paper. The stains were photographed 5 (left) and 8 min (righ) after deposition to illustrate the time course of the reaction.
Sp7 is up-regulated upon infection We observed an increase of transcription upon clean injury of adult wild type flies (Fig. 6). The up-regulation is enhanced by septic injury with a mixture of Gram-positive and Gram-negative bacteria. This up-regulation has been reported in a micro-array analysis of immune challenged flies [32]. Our own results show a lower up-regulation (2.3- versus 5.5-fold in the micro-array report). This difference could be attributed to the semi-quantitative nature of our assay in which we do not quantify maximal changes, instead, the assay is conservative, detecting suboptimal differences. Using flies deficient in components of the Toll and IMD pathways, we asked whether the transcription of Sp7 is dependent on these immune-pathways. As is shown, neither of the mutants blocked the transcription of CG3066. We observed instead a slight increase of transcription that reached the level of clean injured flies (w-p); this is probably due to their permanently challenged-state caused by the reduced immune-defense of these flies. Flies carrying a constitutively active form of Toll (Tl10b) show a consistent increase in Sp7 transcription. To confirm this observation, we performed in situ hybridization in these mutants. As is evident in Fig. 6C, the increase of total transcription seems to be the result of an increase in the total number of cells expressing CG3066. An increased number of circulating hemocytes have previously been reported in Tl10b larvae [38]. Discussion In this study, we have identified the first serine protease in Drosophila that is necessary for activation of the PO pathway. The serine protease Sp7 is expressed in blood cells, it is constitutively transcribed, and it is up-regulated
1080
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
w-
A
w-p
w-i
spz
key
rel
tube
Tl10b
rp49 CG3066
388bp 310bp
Relative mRNA
B
w-
w-p
w-i
spz
key
rel
tube
Tl1 0 b
C
Fig. 6. Transcriptional regulation of Sp7. (A) Estimation of the Sp7 mRNA in adult flies of different genotypes by RT-PCR. Total RNA was extracted 3.5 h after clean pricking (w-p), infection with a mixture of E. coli and M. luteus (w-i,) or untreated flies (w-spz, key, rel, tube, and Tl10b). The amounts of Sp7 RNA (310 bp band) and the rp49 internal control (388 bp band) were visualized on an agarose gel after 28 (left well) and 31 cycles (right well) of RT-PCR. (B) Relative amounts of Sp7 mRNA standardized against rp49. The mean values are the result of three independent experiments. The RNA level of uninfected flies was set to 1. (C) In situ hybridization of Tl10b embryos using the Sp7 probe. The number of blood cells expressing Sp7 is considerably increased in comparison to wild type embryos.
after perforation of the cuticle, septic injury or in flies with mutations in the major immune pathways (Toll and IMD). Transcription of SP7 is first observed during embryogenesis. Specific silencing of Sp7 in crystal cells using the early transcription factor Lozenge is sufficient to impair hemolymph melanization. This study gives new insight at a molecular and cellular level into the melanization reaction in the fruit fly. It has been proposed that protease cascades are key components for the activation of phenoloxidase zymogens (pro-PO) to the active PO. In Drosophila, an inhibitor of serine proteases, the serpin Spn27A has been reported to be involved in the regulation of the pro-PO [14,15]. Null mutants show constitutive melanization while injured larvae react by uncontrolled melanization that expands throughout the body cavity. The function of Spn27A is to inhibit the activation of the PO probably by blocking pro-POAE. Thus, hemolymph melanization requires either the increase of the pro-POAE or the elimination
of the serpin Spn27A, a process that is probably dependent on another protease. It is possible that Sp7 acts as a pro-POAE, promoting directly or through a protease cascade the cleavage activation of the pro-PO zymogens. The same effect, however, could be achieved by promoting proteolytic degradation or titration of the serpin. At this stage, we are unable to discriminate between those two possibilities. However, in Toll pathway mutants, depletion of the serpin Spn27A is inhibited and no activation of PO occurs [14]. We did not observe a block of Sp7 in Toll mutants as would be expected if Spn27a was cleared from the hemolymph by the serine protease Sp7. Therefore, it is not likely that Spn27A is degraded by the transcriptionally regulated serine protease. The alternatively spliced form Sp7-PB (Fig. 7) could, however act, in the rapid removal of the serpin. Our results show that the serine protease Sp7 acts as activator of the PO but does not allow us to discriminate between the function of the different spliced variants.
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
1081
Acknowledgments
Toll
We thank the ‘‘Crafoorsdka Stiftelsen’’ and the ‘‘Nilsson-Ehle Fonden’’ for support. We also thank Ylva Engstro¨m and Dan Hultmark for providing the tub1/tub2 and Tl10b Drosophila stocks, Sol Da Rocha for technical assistance, Edgar Pera, Ulrich Theopold, and Michael Williams for critical reading of the manuscript.
Sp7 Spn27A
Sp7-PB
References
? Sp7-PD Sp7-PC
?
Spn27A
Pro-PO
PO
Phenols
Quinones
Melanin
Fig. 7. Model of the action of the serine protease SP7 in hemolymph melanization of Drosophila. Rupture of crystal cells at an injury site releases the components of the crystalline inclusion (Pro-PO) and the cytosolic Sp7-PB protease. Sp7-PB activates pro-PO and titrates out the serpin Spn27A. Diffusible products of the reaction activate the Toll receptor in intact cells. Toll triggers the signal transduction pathway that modulates the transcriptionally regulated transcripts Sp7-PC and Sp7-PD close to damaged tissues or the inhibitor Spn27a in stabilized tissues. The inhibitory effect of the spliced variants Sp7-PB and Sp7-PD is speculative, therefore indicated with question marks.
At the cellular level, we suggest the occurrence of two types of melanization waves, an immediate reaction that is probably mediated by the rupture of crystal cells and a late reaction that is controlled at the transcriptional level. The perforation of the cuticle produces a rapid locally restricted melanization. It is likely that the first reaction is mediated by the burst of nearby crystal cells and is not transcriptionally regulated. Later, modulation of the reaction may be mediated by the same type of cells but using a transcriptionally regulated pathway probably via the Toll receptor. The existence of the different translational products of Sp7 could facilitate the tight spatial and temporal control of the reaction. In the case of injury infection, disruption of crystal cells could initiate an instantaneous melanization response by releasing the melanization components into the hemolymph (Fig. 7). We reasoned further that the cytoplasmic form of SP7 (SP7-PB) could be the first cleavage activator of the Pro-PO and/or the mediator of the degradation-titration of Spn27A. Both actions reinforce an instantaneous melanization at the damaged tissue. Then, activated surrounding crystal cells could regulate the melanization by the Sp7 secreted pathway. Restriction of the reaction to the injured sites is finally achieved by de novo synthesis and release of the inhibitor Spn27A from the fat body and nearby blood cells. The production of specific antibodies and transgenic flies expressing the alternative transcripts could help to test this hypothesis and further clarify the molecular events in the Drosophila melanization reaction.
[1] J.A. Hoffmann, The immune response of Drosophila, Nature 426 (2003) 33–38. [2] D. Hultmark, Drosophila immunity: paths and patterns, Curr. Opin. Immunol. 15 (2003) 12–19. [3] M.R. Kanost, H. Jiang, X.Q. Yu, Innate immune responses of a lepidopteran insect, Manduca sexta, Immunol. Rev. 198 (2004) 97– 105. [4] U. Theopold, O. Schmidt, K. Soderhall, M.S. Dushay, Coagulation in arthropods: defence, wound closure and healing, Trends Immunol. 25 (2004) 289–294. [5] M.J. Galko, M.A. Krasnow, Cellular and genetic analysis of wound healing in Drosophila larvae, PLoS Biol. 2 (2004) E239. [6] H. Jiang, Y. Wang, M.R. Kanost, Pro-phenol oxidase activating proteinase from an insect, Manduca sexta: a bacteria-inducible protein similar to Drosophila easter, Proc. Natl. Acad. Sci. USA 95 (1998) 12220–12225. [7] D. Satoh, A. Horii, M. Ochiai, M. Ashida, Prophenoloxidaseactivating enzyme of the silkworm, Bombyx mori. Purification, characterization, and cDNA cloning, J. Biol. Chem. 274 (1999) 7441–7453. [8] R. Wang, S.Y. Lee, L. Cerenius, K. Soderhall, Properties of the prophenoloxidase activating enzyme of the freshwater crayfish, Pacifastacus leniusculus, Eur. J. Biochem. 268 (2001) 895–902. [9] H. Jiang, Y. Wang, X.Q. Yu, M.R. Kanost, Prophenoloxidaseactivating proteinase-2 from hemolymph of Manduca sexta. A bacteria-inducible serine proteinase containing two clip domains, J. Biol. Chem. 278 (2003) 3552–3561. [10] Y. Tong, H. Jiang, M.R. Kanost, Identification of plasma proteases inhibited by Manduca sexta serpin-4 and serpin-5 and their association with components of the prophenol oxidase activation pathway, J. Biol. Chem. 280 (2005) 14932–14942. [11] S.Y. Lee, T.H. Kwon, J.H. Hyun, J.S. Choi, S.I. Kawabata, S. Iwanaga, B.L. Lee, In vitro activation of pro-phenol-oxidase by two kinds of pro-phenol-oxidase-activating factors isolated from hemolymph of coleopteran, Holotrichia diomphalia larvae, Eur. J. Biochem. 254 (1998) 50–57. [12] A. Aspan, T.S. Huang, L. Cerenius, K. Soderhall, cDNA cloning of prophenoloxidase from the freshwater crayfish Pacifastacus leniusculus and its activation, Proc. Natl. Acad. Sci. USA 92 (1995) 939– 943. [13] Y. Tong, M.R. Kanost, Manduca sexta serpin-4 and serpin-5 inhibit the prophenol oxidase activation pathway: cDNA cloning, protein expression, and characterization, J. Biol. Chem. 280 (2005) 14923– 14931. [14] P. Ligoxygakis, N. Pelte, C. Ji, V. Leclerc, B. Duvic, M. Belvin, H. Jiang, J.A. Hoffmann, J.M. Reichhart, A serpin mutant links Toll activation to melanization in the host defence of Drosophila, EMBO J. 21 (2002) 6330–6337. [15] E. De Gregorio, S.J. Han, W.J. Lee, M.J. Baek, T. Osaki, S. Kawabata, B.L. Lee, S. Iwanaga, B. Lemaitre, P.T. Brey, An immune-responsive Serpin regulates the melanization cascade in Drosophila, Dev. Cell 3 (2002) 581–592. [16] T.M. Rizki, R.M. Rizki, E. Grell, A mutant affecting the crystal cells in Drosophila melanogaster, Wilhelm RouxÕs Arch. Dev. Biol. 188 (1980) 91–99.
1082
C. Castillejo-Lo´pez, U. Ha¨cker / Biochemical and Biophysical Research Communications 338 (2005) 1075–1082
[17] P. Irving, J.M. Ubeda, D. Doucet, L. Troxler, M. Lagueux, D. Zachary, J.A. Hoffmann, C. Hetru, M. Meister, New insights into Drosophila larval haemocyte functions through genome-wide analysis, Cell Microbiol. 7 (2005) 335–350. [18] M.J. Williams, I. Ando, D. Hultmark, Drosophila melanogaster Rac2 is necessary for a proper cellular immune response, Genes Cells 10 (2005) 813–823. [19] S. de Morais Guedes, R. Vitorino, R. Domingues, K. Tomer, A.J. Correia, F. Amado, P. Domingues, Proteomics of immune-challenged Drosophila melanogaster larvae hemolymph, Biochem. Biophys. Res. Commun. 328 (2005) 106–115. [20] B. Duvic, J.A. Hoffmann, M. Meister, J. Royet, Notch signaling controls lineage specification during Drosophila larval hematopoiesis, Curr. Biol. 12 (2002) 1923–1927. [21] L. Waltzer, G. Ferjoux, L. Bataille, M. Haenlin, Cooperation between the GATA and RUNX factors Serpent and Lozenge during Drosophila hematopoiesis, EMBO J. 22 (2003) 6516–6525. [22] A. Braun, J.A. Hoffmann, M. Meister, Analysis of the Drosophila host defense in domino mutant larvae, which are devoid of hemocytes, Proc. Natl. Acad. Sci. USA 95 (1998) 14337–14342. [23] T. Lebestky, T. Chang, V. Hartenstein, U. Banerjee, Specification of Drosophila hematopoietic lineage by conserved transcription factors, Science 288 (2000) 146–149. [24] C. Castillejo-Lopez, W.M. Arias, S. Baumgartner, The fat-like gene of Drosophila is the true orthologue of vertebrate fat cadherins and is involved in the formation of tubular organs, J. Biol. Chem. 279 (2004) 24034–24043. [25] D. Tautz, C. Pfeifle, A non-radioactive in situ hybridization method for the localization of specific RNAs in Drosophila embryos reveals translational control of the segmentation gene hunchback, Chromosoma 98 (1989) 81–85. [26] R.P. Sorrentino, C.N. Small, S. Govind, Quantitative analysis of phenol oxidase activity in insect hemolymph, Biotechniques 32 (2002) 815–816, 818, 820, 822–813. [27] R. Bettencourt, H. Asha, C. Dearolf, Y.T. Ip, Hemolymph-dependent and -independent responses in Drosophila immune tissue, J. Cell Biochem. 92 (2004) 849–863. [28] Z. Zou, Y. Wang, H. Jiang, Manduca sexta prophenoloxidase activating proteinase-1 (PAP-1) gene: organization, expression, and regulation by immune and hormonal signals, Insect. Biochem. Mol. Biol. 35 (2005) 627–636.
[29] T. Muta, T. Miyata, Y. Misumi, F. Tokunaga, T. Nakamura, Y. Toh, Y. Ikehara, S. Iwanaga, Limulus factor C. An endotoxin-sensitive serine protease zymogen with a mosaic structure of complement-like, epidermal growth factor-like, and lectin-like domains, J. Biol. Chem. 266 (1991) 6554–6561. [30] J. Ross, H. Jiang, M.R. Kanost, Y. Wang, Serine proteases and their homologs in the Drosophila melanogaster genome: an initial analysis of sequence conservation and phylogenetic relationships, Gene 304 (2003) 117–131. [31] G. Zhang, Z.Q. Lu, H. Jiang, S. Asgari, Negative regulation of prophenoloxidase (proPO) activation by a clip-domain serine proteinase homolog (SPH) from endoparasitoid venom, Insect. Biochem. Mol. Biol. 34 (2004) 477–483. [32] E. De Gregorio, P.T. Spellman, G.M. Rubin, B. Lemaitre, Genomewide analysis of the Drosophila immune response by using oligonucleotide microarrays, Proc. Natl. Acad. Sci. USA 98 (2001) 12590– 12595. [33] M. Boutros, H. Agaisse, N. Perrimon, Sequential activation of signaling pathways during innate immune responses in Drosophila, Dev. Cell 3 (2002) 711–722. [34] B. Lemaitre, E. Kromer-Metzger, L. Michaut, E. Nicolas, M. Meister, P. Georgel, J.M. Reichhart, J.A. Hoffmann, A recessive mutation, immune deficiency (imd), defines two distinct control pathways in the Drosophila host defense, Proc. Natl. Acad. Sci. USA 92 (1995) 9465–9469. [35] B. Lemaitre, E. Nicolas, L. Michaut, J.M. Reichhart, J.A. Hoffmann, The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults, Cell 86 (1996) 973–983. [36] S.H. Jung, C.J. Evans, C. Uemura, U. Banerjee, The Drosophila lymph gland as a developmental model of hematopoiesis, Development 132 (2005) 2521–2533. [37] E. Kurucz, C.J. Zettervall, R. Sinka, P. Vilmos, A. Pivarcsi, S. Ekengren, Z. Hegedus, I. Ando, D. Hultmark, Hemese, a hemocyte-specific transmembrane protein, affects the cellular immune response in Drosophila, Proc. Natl. Acad. Sci. USA 100 (2003) 2622–2627. [38] C.J. Zettervall, I. Anderl, M.J. Williams, R. Palmer, E. Kurucz, I. Ando, D. Hultmark, A directed screen for genes involved in Drosophila blood cell activation, Proc. Natl. Acad. Sci. USA 101 (2004) 14192–14197.