Functional Proteomics of the Active Cysteine Protease Content in Drosophila S2 Cells*

The fruit fly genome is characterized by an evolutionary expansion of proteases and immunity-related genes. In order to characterize the proteases that are active in a phagocytic Drosophila model cell line (S2 cells), we have applied a functional proteomics approach that allows simultaneous detection and identification of multiple protease species. DCG-04, a biotinylated, mechanism-based probe that covalently targets mammalian cysteine proteases of the papain family was found to detect Drosophila polypeptides in an activity-dependent manner. Chemical tagging combined with tandem mass spectrometry permitted retrieval and identification of these polypeptides. Among them was thiol-ester motif-containing protein (TEP) 4 which is involved in insect innate immunity and shares structural and functional similarities with the mammalian complement system factor C3 and the pan-protease inhibitor alpha2-macroglobulin. We also found four cysteine proteases with homologies to lysosomal cathepsin (CTS) L, K, B, and F, which have been implicated in mammalian adaptive immunity. The Drosophila CTS equivalents were most active at a pH of 4.5. This suggests that Drosophila CTS are, similar to their mammalian counterparts, predominantly active in lysosomal compartments. In support of this concept, we found CTS activity in phagosomes of Drosophila S2 cells. These results underscore the utility of activity profiling to address the functional role of insect proteases in immunity.

Recognition, uptake, and destruction of pathogens by phagocytic cells are fundamental principles of innate immunity that are conserved in organisms from insects to humans (1). An important step after the uptake of material into the endosomal/phagosomal pathway is proteolysis to destroy and clear pathogens (2). Proteolysis along the endocytic pathway has been extensively studied in the mammalian system (3)(4)(5). Endocytic proteases, in particular cysteine proteases of the cathepsin (CTS) 1 family, are actively involved in MHC class II-restricted antigen presentation and, therefore, play a key role in adaptive immunity (6). Members of the CTS family, such as CTS B (7), CTS S (8,9), CTS F (10), CTS L (11), and, more recently, CTS Z (12), have been implicated in this process (13,14).
Little is known about the function of CTS proteases in other biological processes that might explain both their evolutionary conservation and expansion in multicellular organisms. Several CTS-deficient strains of mice show dramatic phenotypes that indicate involvement in bone remodeling, development, and apoptosis (15,16). In addition, dysregulation of CTS activity is observed in various cancers and correlates with tumor progression (17,18). Cysteine proteases have been mainly studied in mice and humans (13,14); however, functional redundancy and significantly overlapping specificities of CTS make it difficult to explore individual CTS function in mice and humans. Therefore, less complex organisms such as the fruit fly, Drosophila melanogaster, may be helpful in gaining further insight into the functional roles of members of this protease family.
In the fruit fly, and relative to the genome as a whole, proteases and immunity-related genes are characterized by specific gene expansions (19 -21). Drosophila melanogaster possesses specialized blood cells, or hemocytes, that phagocytose microbes in a manner similar to their mammalian counterparts (1,22,23). In particular, the hemocyte-like Schneider's Drosophila line 2 (S2) has emerged as a model system to study phagocytosis and for the identification of phagocytic receptors that recognize and mediate the engulfment of microbes (24,25). S2 cells, therefore, are a relevant model for the application of a functional proteomic approach that permits simultaneous detection and identification of multiple protease activities.
Several approaches are now available to assess protease activity on a broad scale. Chemistry-based functional proteomics is a promising method used to assign putative gene products to an enzyme family (26). The rationale behind the design of such a functional proteomics approach is to assess the activity profiles of protease species in complex biological samples rather than merely their presence or absence. Identification and activity measurement can be performed without the need for purification of the individual enzymes under study. In addition, this approach allows activity measurement of several enzyme species at the same time, a task that would be more cumbersome using classical biochemical approaches. The success of this approach has been demonstrated for serine proteases (27), deubiquitinating enzymes (28), and cysteine proteases (29). In this study, we have used DCG-04, a chemical probe that specifically and covalently targets mammalian cysteine proteases. Our objective was to identify DCG-04-reactive polypeptides in Drosophila phagocytes and to characterize and monitor their activities.

EXPERIMENTAL PROCEDURES
Cell, Culture Conditions, and Reagents-A particularly phagocytic sub-line of S2 cells was obtained from A. Pearson (Department of Pediatrics, Massachusetts General Hospital, Boston, MA) (23,30). S2 cells were grown at 26-27°C in Schneider's Insect Medium (Sigma-Aldrich, St. Louis, MO) or Schneider's Drosophila Medium (Invitrogen Life Technologies, Carlsbad, CA) supplemented with 10% heat-inactivated fetal bovine serum (tested to support insect cell growth; Invitrogen Life Technologies) and maintained at a density of 2 ϫ 10 5 to 8 ϫ 10 6 cells/ml in 12.5 ml per T75 plastic tissue culture flask to ensure exponential growth. Under these conditions, S2 cells divided every 16 -18 h. Chemicals were obtained from Sigma-Aldrich, unless indicated otherwise.
Active Site Labeling of Cysteine Proteases in Cell Lysates and Detection by Streptavidin Blotting-JPM-565 and DCG-04 were synthesized and purified as previously described (31,32). Cells were harvested by centrifugation at 4°C, washed in Robb's Drosophila PBS pH 6.8 (52), and cell pellets corresponding to 5 ϫ 10 7 cells were frozen at Ϫ80°C. Cell pellets were thawed on ice and lysed in 100 l lysis buffer, pH 5.0 (50 mM sodium acetate, pH 5.0, 5 mM magnesium chloride, 0.5% Nonidet-P40), incubated for 30 min and centrifuged for 15 min at 13,000 ϫ g to remove nuclei. Protein concentration of the cell extract was measured using the Bio-Rad Protein Assay (Hercules, CA) with BSA as standard (average yield, 1-2 mg protein per 5 ϫ 10 7 cells). Cell lysate (25 g protein) was incubated with DCG-04 for 60 min at 37°C. When JPM-565 was used as competitor, cell lysates were pre-incubated with JPM-565 for 15 min before addition of DCG-04. The reaction was stopped by the addition of double-concentrated reducing Laemmli sample buffer and boiling for 10 min. Samples were analyzed by SDS-PAGE and transferred to polyvinylidene membrane (Immobilon P; Millipore, Billerica, MA). After blocking overnight in PBS, pH 7.2, containing 10% non-fat dry milk, the membrane was incubated with streptavidin-horseradish peroxidase (1: 2,500; Amersham Pharmacia Biotech, Uppsala, Sweden) in PBS containing 0.2% Tween 20 for 60 min at room temperature followed by extensive washing in PBS-Tween. DCG-04-reactive polypeptides were detected by enhanced chemiluminescence (Western Lightning; PerkinElmer, Wellesley, MA).
Isolation of DCG-04-Reactive Enzymes for Ms-Based Identification-Cell lysate from 5.4 ϫ 10 9 cells was prepared at pH 5.0, as described above. The lysate was divided into five samples, each corresponding to approximately 35 mg of protein. The samples were pre-cleared with 150 l bead volume streptavidin-agarose (Pierce, Rockford, IL) to decrease nonspecific background activities. Three samples were incubated with 5 M DCG-04 for 60 min at 37°C, one sample was pre-treated with 25 M JPM-565 before addition of DCG-04, and one sample received neither DCG-04 nor JPM-565. To stop the reaction, SDS was added to a final concentration of 0.5%, and the samples were incubated for 5 min at 95°C (in order to denature the DCG-04-tagged proteins to make the biotin moiety accessible). Affinity enrichment of DCG-04-reactive polypeptides was performed using streptavidin-agarose, as previously described (12,32). Briefly, the buffer was exchanged to pull-down buffer (50 mM Tris, pH 7.4, 150 mM NaCl) using a PD-10 column (Amersham Pharmacia Biotech), and each sample was incubated with 150 l bed volume streptavidin-agarose for 60 min at room temperature. After washing with excess pull-down buffer, bound polypeptides were eluted from the beads by addition of 100 l reducing Laemmli sample buffer and boiling for 10 min. DCG-04-reactive proteins were separated by SDS-PAGE (12.5%). One DCG-04-reacted sample and the control samples were stained with silver (see Fig. 3), 5% of each of these samples was used for streptavidin blotting (not shown), and the two remaining DCG-04-reacted samples were pooled and stained with Coomassie Brilliant Blue G as follows: the gel was fixed in H 2 O/25% isopropanol/ 10% acetic acid for 45 min, stained with 10% acetic acid/0.006% Coomassie Brilliant Blue G overnight, and destained with 10% acetic acid for 2h. Visible bands from the Coomassie-and silver-stained gels were excised and processed for tandem mass spectrometry analysis.
Tryptic Digestion and Analysis by Electrospray Tandem Mass Spectrometry-In-gel tryptic digestion was performed essentially as described (33). The samples were subjected to a nanoflow liquid chromatography system (CapLC; Waters, Medford, MA) equipped with a picofrit column (75-m ID ϫ 9.8 cm; NewObjective, Woburn, MA), at a flow rate of ϳ150 nl/min using a nanotee (Waters), 16:1 split (initial flow rate 5.5 l/min). The liquid chromatography system was directly coupled to a quadrupole time-of-flight micro-tandem mass spectrometer (Micromass, Manchester, UK). Analysis was performed in survey scan mode, and parent ions with intensities greater than 7 were sequenced in MS/MS mode using MassLynx 4.0 software (Micromass). MS/MS data were processed and subjected to database searches using ProteinLynx Global Server 1.1 software (Micromass) against Swiss-Prot TrEMBL/New (www.expasy.ch), or Mascot (Matrixscience, www.matrixscience.com/, (34)) against the National Center for Biotechnology Information (NCBI) (www.ncbi.nlm.nih.gov/) non-redundant database (NCBInr). Alternatively, the Drosophila protein database from NCBI was also used for Drosophila-specific protein identification. In all searches, oxidation of methionine and carbamidomethylation of cysteine residues were considered as modifications. Matches for proteins were accepted as significant if scores were more than 75 using Protein Lynx Global Server 1.1, or based on the Mascot Probability Mowse Score. At least two peptides were found for each identified protein species. Information about the Drosophila gene products were obtained from the FlyBase database (FlyBase.bio.indiana.edu) and NCBI (www.ncbi.nlm.nih.gov/entrez/ query.fcgi) Active Site Labeling of Cysteine Proteases in Phagosomes of Live Cells Using Latex Beads-Phagosomal cysteine protease labeling has been adapted from Lennon-Dumenil et al. (12). S2 cells were plated on 12-well plates (10 6 cells/well) 1 day before the experiment. Streptavidin-coated latex beads (2-m diameter; Polysciences, Warrington, PA) were incubated with DCG-04 for 60 min at room temperature. Beads were washed three times with PBS and resuspended in complete culture medium. DCG-04-loaded beads were added to the cells and incubated for 30 min at 27°C (pulse). Phagocytosis was halted at 4°C, cells were collected and pelleted at 2,000 rpm (325 ϫ g) for 3 min, resuspended in complete medium, and non-internalized beads were removed by repeated centrifugation after layering on a cushion of heat-inactivated fetal bovine serum (latex beads remain in interphase). Cells were resuspended in complete medium and incubated in 12-well plates at 27°C for various periods of time (chase). At the desired time points, cells were harvested on ice, pelleted and lysed in hot SDS-sample buffer containing 100 M JPM-565, boiled for 10 min, cooled to room temperature, and passed through a 22.5-gauge hypodermic needle to shear DNA before pelleting the released (phagocytosed) latex beads at 13,000 rpm (13,000 ϫ g) for 5 min. Samples were analyzed by SDS-PAGE and streptavidin-blotting. Phagocytic uptake of streptavidin beads was controlled by light microscopy and uptake of deep-blue dyed latex beads (0.8-m diameter; Sigma-Aldrich) by light and electron microscopy (data not shown).

Protease Activity Profiling in Drosophila S2 Cells Using the Mechanism-Based, Epoxide Inhibitor-Derived Chemical
Probe DCG-04 -The epoxide inhibitor JPM-565 has been used as an irreversible inhibitor, with broad reactivity toward cysteine proteases (31,35). Based on the structure of this compound, a derivative was developed to include a biotin affinity tag (DCG-04; Fig. 1A) (32). This compound permits targeting of active cysteine proteases present in crude extracts via covalent attachment to the cysteine active site residue (Fig. 1B, (29)). In order to test whether DCG-04 reacts with Drosophila proteins, we incubated this probe with crude pH 5.0-level cell extracts prepared from S2 cells. DCG-04 reactive polypeptides were separated by SDS-PAGE and visualized by streptavidin blotting. As shown in Fig. 2, we observed labeling in the molecular mass ranges of 26 -29 and 33-37 kDa. Preheating of the extract before addition of DCG-04 abolished labeling (Fig. 2). Competition for labeling with increasing amounts of non-biotinylated probe (JPM-565) resulted in decreased recovery of these proteins (Fig.  2). Thus, DCG-04 appeared to react specifically in a doseand conformation-dependent manner with several Drosophila polypeptides.
Identification of DCG-04-Reactive Polypeptides in Drosophila S2 Cell Extracts-In order to identify the polypeptides labeled by DCG-04, we used a strategy based on streptavidin-agarose affinity purification and tandem mass spectrometry as outlined in Fig. 1B. Drosophila S2 cell extracts were either left untreated or incubated with DCG-04 in the presence or absence of non-biotinylated competitor, and DCG-04-reactive polypeptides were purified on streptavidin-agarose. SDS-PAGE of the eluted material, followed by silver staining, revealed several DCG-04-reactive polypeptides whose abundance was reduced when a five-fold excess of non-biotinylated competitor was included (Fig. 3). Subsequent analysis using tandem mass spectrometry (liquid chromatography-MS/MS) identified a total of 20 protein species, for which at least two peptide matches were found. We grouped these polypeptides into two categories, based on the absence (Table I) or presence (Table II) of a catalytically active thiol group. Due to the absence of such a thiol group, the polypeptides listed in Table I are likely to be nonspecific contaminants, co-purified in our isolation procedure. They appear to be mostly metabolic enzymes but also include cytoskeleton proteins and streptavidin. Less frequently observed background proteins included catalase, involved in oxidative stress and aging.
We found six polypeptides that might exert reactivity toward DCG-04. All had an N-terminal signal sequence indicating export into a secretory compartment (Table II). One of these was protein disulfide isomerase (PDI), an enzyme containing two thiols as active site residues. PDI catalyzes disulfide formation in newly synthesized polypeptides in the endoplasmic reticulum in a two-step reaction, which includes two alternative intermediate, mixed disulfides between the enzyme and substrate (36). Therefore, epoxysuccinyl compounds such as DCG-04 and JPM-565 may specifically bind to reactive thiol groups of such nature.
Two polypeptides at ϳ120 and 160 kDa were identified as thiol ester-containing protein (TEP) 4 ( Fig. 3 and Table II). TEP 4 belongs to a group of five thiol ester-containing Drosophila proteins that show substantial structural and functional similarities, including a highly conserved thiol ester motif, to both a central component of mammalian complement system, factor C3, and to a pan-protease inhibitor, alpha2-macroglobulin (37). TEP 4 contains an internal beta-cysteinyl-gamma-glutamyl thiol ester (38). It is possible that DCG-04 forms a covalent adduct with the proteolytically activated nascent state of the thiol ester (39), although the details of this reaction mechanism remain unclear. TEP 4 appears to be expressed constitutively in Drosophila larvae but is significantly up-regulated after challenge with pathogens (37). In a mosquito hemocyte-like cell line, a related protein, TEP 1, serves as a complement-like opsonin and promotes phagocytosis of some Gram-negative bacteria. This activity is dependent on its internal thiol ester bond (40).
Active Cysteine Proteases with Homology to Mammalian Cathepsins-We identified multiple polypeptide species that showed significant homologies to mammalian CTS proteases L, B, F and K, respectively (Table II and Fig. 3). The CTS F-like protease has not been previously described as an active entity but is found annotated in the D. melanogaster database FlyBase as a putative protease. The CTS B-and K-like proteases have never been isolated and characterized as proteolytic enzymes from Drosophila, although expression of the  CTS B-like protease has been observed in Drosophila embryos (41) and a K-like enzyme in the flesh fly Sarcophaga peregrina (42). A CTS L-like protease has been reported to be expressed in embryonal and larval midgut (43) and was found in granules in the hemocyte-like Drosophila cell line mbn-2, suggesting lysosomal localization (44). The latter three enzymes have been implicated in general digestive processes, yolk degradation, and immunity, and are strongly conserved in other non-vertebrate species (see FlyBase and Refs. 45 and 46)). Sequence alignment and comparison of these CTS-like proteases and their corresponding mouse homologs revealed a high degree of conservation centering on two regions in the mature parts of the molecules, around the active site residues cysteine (C), histidine (H), and asparagine (N) (Fig. 4A). The N-terminal pro-regions were less well conserved but of similar length, except in the case of the Drosophila CTS K equivalent (26 -29-kDa protease), which contains an insertion of significant length not found in the pro-region of the mammalian homolog CTS K. A dendrogram constructed on the basis of overall sequence similarities shows that, with the exception of CTS K, evolutionary diversification of the CTS subgroups occurred before the ancestors of mammalian and invertebrate CTS proteases diverged from each other (Fig. 4B). These data suggest an evolutionary conserved function for these proteases and only later recruitment of these molecules for adaptive immunity in vertebrates.
Drosophila Cathepsins Are Most Active at pH 4.5 and Can Be Detected in Phagosomes of Live Cells-We used activity profiling to determine under which pH conditions these enzymes are active. Extracts from S2 cells were prepared at a pH level ranging from 4.5-7.5. DCG-04 labeling resulted in distinct polypeptide profiles in a highly pH-dependent manner (Fig. 5A, third panel). At a pH of 4.5, two major bands of 26 -29 and 33-37 kDa were observed, consistent with the experiments described above for pH 5.0. This suggests that these polypeptides correspond to the Drosophila CTS homologs identified by tandem mass spectrometry. Labeling of these polypeptides was specific, because they were not observed in heat-treated samples (Fig. 5A, second panel) and were competed by an excess of non-biotinylated compound (Fig. 5A,  fourth panel). The marked pH-dependence of these enzymes indicates a function within the endocytic compartment.
In order to address a possible phagosomal localization of these Drosophila CTS proteases in live cells, we used an approach recently pioneered by our laboratory that monitors the proteolytic environment in phagosomes of mammalian macrophages and dendritic cells (12). Streptavidin-latex beads loaded with DCG-04 are efficiently internalized by phagocytic cells and allow sampling of the proteolytic environment during phagosome formation and maturation. Drosophila S2 cells were incubated for 0.5 h with DCG-04-loaded beads. The non-internalized beads were removed, and cells were lysed after 1-6 h of incubation (chase). To prevent post-lysis reactivity of DCG-04, cells were lysed by the addition of boiling SDS sample buffer containing an excess of non-biotinylated competitor. Streptavidin blotting revealed labeling in the 26 -29 kDa range, which increased over time (Fig.  5B). This labeling profile was dependent on DCG-04-loaded beads, because no labeling was observed in the absence of latex beads or in the presence of latex beads that had not been loaded with DCG-04. These results suggest that CTS activity can be detected by activity profiling in phagosomal compartments of living Drosophila S2 cells. DISCUSSION Chemical targeting of cysteine proteases by means of a mechanism-based probe combined with tandem mass spectrometry has allowed rapid identification of multiple active protease species in a Drosophila cell line with phagocytic properties. Four of 12 total CTS-like proteases encoded in the Drosophila genome were identified (Tables II and III). Exami- nation of S2 cell-specific expressed sequence tags available in the FlyBase database revealed that only six of these 12 CTS-like genes appear to be expressed in S2 cells, including the four proteases identified by tandem mass spectrometry (Table III and Fig. 4C). Inspection of the sequence of the other two genes revealed that the active site residue cysteine was FIG. 4. Structure-based amino acid sequence alignment of Drosophila and mouse CTS proteases. A, Sequence alignments were performed using ClustalW (www.ch.embnet.org/software/ClustalW.html) and Vector NTI (Informax, Frederick, MD) software. Conserved residues are indicated under the sequence by: *, conserved; :, strongly homologous residues; ⅐, homologous residues. The catalytic cysteine and histidine residues are indicated by dots above the sequence, and the conserved region around the catalytic cysteine is boxed. The beginning of the mature enzymes, i.e. beginning and end of single chains or heavy and light chains (by similarity to human enzymes), is indicated above the sequence by angles. not conserved, precluding reaction with DCG-04 (Fig. 4C). The Drosophila genome contains a CTS H-like protease that is not expressed in S2 cells and lacks the catalytic cysteine residue (Table III and Fig. 4C). This is in contrast to the mammalian system in which CTS H was shown to react with DCG-04 (12).
We isolated not only cysteine proteases but also other enzymes that catalyze thiol-based reactions, such as PDI and TEP 4 (Table II). Although it is possible that DCG-04 directly reacted with these polypeptides (see "Results"), PDI and TEP 4 may have been isolated as a consequence of their association with cysteine proteases that were targeted by DCG-04. The observation that DCG-04 can be used to isolate insect TEP will be useful in monitoring their activity. Insect TEP have attracted considerable interest lately because they play a role in innate immunity in the mosquito vector Anopheles by limiting the multiplication of malaria parasites inside the vector organism (40). 3 We observed two forms of Drosophila CTS L and F with molecular masses of 26 -29 and 33-37 kDa that reacted with DCG-04 ( Fig. 3 and Table II). Glycosylation has been described for mammalian CTS proteases (47) and could account for this difference in molecular mass. However, it seems more likely that 33-37-kDa forms correspond to pro-forms of these enzymes that have retained all (CTS L) or part (CTS F) of their pro-peptides, as suggested for human and mouse CTS L (47). In line with this possibility, we found that the 33-37-kDa form of Drosophila CTS L contained the tryptic peptide AADESFKGVTFISPAHVTLPK 106 -126 , which is located N-terminally of the predicted cleavage site for the mature enzyme ( Fig. 4A and see Ref. 13). Labeling of CTS pro-forms by epoxide-based probes such as DCG-04 has been observed previously in mammalian cells (48). This phenomenon can be attributed to the small molecular size of the chemical probe and to its covalent reaction mechanism. Although pro-peptide efficiently prevents the access of large polypeptide substrates to the active site, it may not be bound tightly enough to prevent the small-sized DCG-04 molecule from entering the active site and from irreversibly attaching to the proenzyme. This interpretation may also explain the results obtained by activity profiling in S2 cell phagosomes (Fig. 5B). In this subcellular compartment, DCG-04-reactive polypeptides were exclusively found in the 26 -29 kDa range, suggesting that only fully processed, mature CTS proteases are exposed to phagocytosed material.
Whereas the Drosophila CTS L-, B-, and F-like enzyme species shared significant overall sequence identities with their mouse counterparts (Fig. 4, A and B), CTS K (26 -29 kDa protease) appears to have unique properties. This enzyme showed only 28% overall sequence identity to mouse CTS K, 3 E. A. Levashina and F. C. Kafatos, personal communication.

FIG. 5. Drosophila CTS proteases show a narrow activity optimum under acidic conditions and can be detected in the phagosome.
Proteins were separated by SDS-PAGE on 12.5% gels, and labeled bands were visualized by streptavidin blotting. A, Cytoplasmic extracts from Drosophila S2 cells were prepared at the indicated pH level. Samples were incubated in the presence or absence of DCG-04 and a 10-fold excess of non-biotinylated competitor JPM-565, followed, or not, by preheating for 5 min at 100°C. Labeled polypeptides were separated by SDS-PAGE (12.5%) and visualized by streptavidin blotting. B, Labeling of phagosomal cysteine proteases by DCG-04, immobilized on streptavidin beads, and phagocytosed by live Drosophila S2 cells. Streptavidin latex beads (0.2 m) were loaded with 10 M DCG-04. S2 cells were incubated with DCG-04-loaded or unloaded latex beads for the indicated times (pulse). Non-phagocytosed beads were removed, and cells were incubated for the indicated times (chase). and 27% to mouse CTS L (Fig. 4). From a single precursor, it is processed to two separate 26 and 29 kDa polypeptides, which then dimerize via disulfide bonds, as shown for its close homologue in the flesh fly (42). The 29-kDa subunit corresponds to a single chain CTS such as mammalian CTS F or K (Swiss-Prot TrEMBL annotation, www.expasy.ch), whereas the 26-kDa subunit has no homolog in mammals, suggesting a specific role of the 26 -29-kDa protease in invertebrates.
DCG-04 activity profiling at different pH levels was used to show that Drosophila CTS proteases are most active under acidic conditions, suggesting that these enzymes exert their biological function in late endocytic or lysosomal cell compartments (Fig. 5). In contrast, some mammalian CTS proteases, such as CTS L, B, and K are also active at a neutral pH level (49), which may reflect a role in a wider range of biological functions in higher vertebrates. However, the predominant presence of CTS in phagosomes of both classes of organisms (insects and mammals), as well as the phylogenetically conserved branching into CTS subgroups L, F, and B (Fig. 4B), is consistent with a general role in lysosomal prote-olysis, which precedes their function in antigen processing and presentation in mammalian cells. Drosophila, which lacks an adaptive immune system, is an appropriate organism to investigate whether CTS proteases have a function in innate immunity during a challenge with bacterial or fungal pathogens. However, examination of published genome-wide microarray data indicated no alteration of Drosophila CTS L, K, B, and F mRNA expression levels in response to immune stimuli (50,51). In addition, preliminary experiments with crude extracts from S2 cells exposed to immune stimulators revealed no significant differences in DCG-04-labeling profiles (data not shown). It will therefore be interesting to assess Drosophila CTS activity profiles in a more refined way, for example, by immunoprecipitation of specific DCG-04-reactive CTS proteases from phagosomes of living cells exposed to such stimuli.
Taken together, our results demonstrate the usefulness of this functional proteomics approach and provide a basis for the simultaneous monitoring of multiple protease activities in insect models at different developmental stages or during an immune challenge.