Cathepsin B Gene Disruption Induced Leishmania donovani Proteome Remodeling Implies Cathepsin B Role in Secretome Regulation

Leishmania cysteine proteases are potential vaccine candidates and drug targets. To study the role of cathepsin B cysteine protease, we have generated and characterized cathepsin B null mutant L. donovani parasites. L. donovani cathepsin B null mutants grow normally in culture, but they show significantly attenuated virulence inside macrophages. Quantitative proteome profiling of wild type and null mutant parasites indicates cathepsin B disruption induced remodeling of L. donovani proteome. We identified 83 modulated proteins, of which 65 are decreased and 18 are increased in the null mutant parasites, and 66% (55/83) of the modulated proteins are L. donovani secreted proteins. Proteins involved in oxidation-reduction (trypanothione reductase, peroxidoxins, tryparedoxin, cytochromes) and translation (ribosomal proteins) are among those decreased in the null mutant parasites, and most of these proteins belong to the same complex network of proteins. Our results imply virulence role of cathepsin B via regulation of Leishmania secreted proteins.


Introduction
Leishmaniasis is a spectrum of diseases ranging in severity from spontaneously healing and non-healing cutaneous lesions (caused by several species; L. tropica, L. major, L. aethiopica and L. mexicana) to fatal visceral disease (caused by L. donovani complex species; L. donovani, L. infantum, L. chagasi). There are over 500,000 cases of visceral leishmaniasis with 50,000 deaths annually. There is no vaccine against any form of leishmaniasis, available drugs have serious side effects, and drug resistant cases are on the rise [1][2][3][4][5].
Leishmania papain family cysteine proteases, cathepsin B and cathepsin L, are potential drug targets and vaccine candidates. Leishmania genome consists of multi copy Cathepsin L and a single copy cathepsin B cysteine protease genes [6][7][8]. The importance of L. mexicana cathepsin L and B cysteine proteases has been studied using null mutant parasites in mice. L. mexicana cathepsin L null mutant parasites are unable to induce lesion development in mice and protect mice against challenge with wild type parasites [9]. Unlike L. mexicana cathepsin L, L. mexicana cathepsin B null mutants induce lesion development in mice although they show reduced survival inside macrophages in vitro [10]. L. donovani cathepsin B null mutant parasites have not been generated yet, and their pathogenesis in vivo and potential as an attenuated vaccine candidate remains to be determined. In a closely related protozoan, Trypanosoma brucei, RNAi targeting of cathepsin B leads to clearance of parasites from the bloodstream and prevents lethal infection in mice [11]. Enzymatic assays and sequence analysis suggest difference in the function of cathpesin B of different Leishmania species. L. major cathepsin B, unlike L. mexicana cathepsin B, has no activity against Z-Arg-Arg-AMC [12], which was attributed to difference in S 2 sub-site amino acid [13,14]. L. major cathepsin B has glycine at S 2 sub-site, whereas L. mexicana cathepsin B has a serine. L. donovani cathepsin B, which has a glycine residue at S 2 sub-site and 90% identical with L. major cathepsin B, shares only 60% identity with L. mexicana cathepsin B [8].
By using antisense inhibition and specific protease inhibitors, studies from our lab and others have shown the importance of L. donovani cathepsin B for survival inside macrophages and activation of latent-TGF-b1 in vitro. L. donovani cathepsin B null mutant parasites have not been generated yet, hence the role of L. donovani cathepsin B and the mechanisms by which cathepsin B affect Leishmania intracellular survival in vivo remain unknown. Here, we report the generation of L. donovani cathepsin B null mutant parasites and cathepsin B gene disruption induced proteome remodeling. Proteome profiling of L. donovani cathepsin B wild type and null mutant parasites hints at virulence role of cathepsin B via regulation of Leishmania secreted proteins.

Leishmania, Cell lines, and Growth media
Leishmania donovani 1S2D was obtained from Dr. Greg Matlashewski, McGill University, Montreal, Quebec, Canada. Leishmania parasites were maintained in M199 medium. Leishmania selections were done on semi solid SDM-79 medium as described previously [15]. U937 cell lines were obtained from ATCC and maintained in a complete RPMI media.

Enzymes, Peptides, Antibodies, Plasmids, and Kits
Restriction enzymes were purchased from GE Healthcare or Invitrogen. AMC (7-Amino-4-Methyl-Coumarin), Z-Arg-Arg-AMC, and Z-Phe-Arg-AMC were purchased from Peptides International. Anti L. major cathepsin B antibody was kindly provided by Dr. Judy Sakanari, Sonoma State University, California, USA. Peroxidoxin-4 antibody was raised in mice [16]. Anti a-tubulin antibody was purchased from SIGMA and secondary antibodies were purchased from R&D Biosystems. Homologous recombination (pX63Hyg and pX63Neo) and episomal expression vectors (pXGSAT, pXNeo) were kindly provided by Dr. Stephen Beverley, University of Washington, St. Louis, Missouri, USA. Geneticin and hygromycin B were purchased from Invitrogen and nourseothricin was purchased from Sigma. Genomic and plasmid DNA preparation and gel purification kits were purchased from QIAGEN. DIG-High Prime DNA Labeling and Detection Starter Kit I was purchased from Roche Diagnostics. Western and southern blotting membranes and ECL western blot detection kits were purchased from GE Healthcare.

DNA preparations
Mid log stage Leishmania parasites were washed three times with PBS and DNA was extracted using DNAeasy Blood and Tissue kit (QIAGEN). Small-scale plasmids were prepared by using Spin Mini Prep Kit (QIAGEN) and large-scale plasmids were prepared by using Maxi Prep Kit (QIAGEN).
Generation of null mutant and complemented null mutant parasites L. donovani cathepsin B null mutants were generated by deleting part of cathepsin B coding sequence (529 bp) containing active site cysteine and occluding loop ( Figure S1). Knockout constructs were made by PCR amplifying and cloning 5' (819 bp) and 3' (765 bp) sequences flanking the 529 bp region. The 59 flanking sequence was amplified with primers carrying HindIII and XhoI restriction sites (59flankF, 59-CTGCACAAGCTTTTGCAACf-GTGCCCGCCGACGTG-39; 59flankR, 59-CACCTCCTC-GAGGCTCTTGCCGCTGACC-39). The 39 flanking sequence was amplified with primers carrying SmaI and BglII restriction sites (39flankF, 59-AGCGGTCCCGGGGATGGATCTGGTCAAG-TACAAGG-39; 39flankR, 59-AGCGGTAGATCTTTCT-CAGCCTCCTTTCCCG-39). The PCR amplified products were Figure 1. Homologous recombination based gene deletion strategy. Single allele cathepsin B mutant parasites were generated using hygromycin replacement construct and the remaining single allele was replaced with the neomycin construct. A, B, C, and D indicate probes used to screen the generated mutant parasites. 'A' binds to part of the mature region of L. donovani cathepsin B replaced by the hygromycin and neomycin markers. 'B' and 'C' are hygromycin and neomycin coding region probes respectively. 'D' hybridizes to the 3' flanking sequences of the homologous recombination constructs. doi:10.1371/journal.pone.0079951.g001 digested with restriction enzymes and gel purified. The purified PCR products were ligated with digested, dephosphorylated, and gel purified pX63-HYG and pX63-NEO plasmids. For complementation, cathepsin B coding sequence was amplified from pXNeo vector carrying a cDNA clone of L. donovani cathepsin B by using BamHI restriction site carrying primers (Forward, 59-CTAGATGGATCCGATGGCTACTC-CTG-39; Reverse, 59-GGCAGTGGATCCATGGCCCTCCG-CGCCAAG-39). The amplified product was gel purified and ligated into the BamHI site of digested, dephosphorylated, and gel purified pXGSAT vector. Ligation products were transformed into competent E. coli DH5a cells. E. coli DH5a competent cell preparations [17] and transformations [18] were done following standard procedures. Plasmids were isolated and insert sequences and orientations were confirmed by sequencing both strands of DNA at the University of Calgary UCDNA sequencing facility. The knockout constructs were linearized with HindIII and BglII restriction enzymes and the hygromycin construct was transfected into wild type parasites. The transfected parasites were plated on SDM-79 media containing 100 mg/ml hygromycin to select for heterozygous mutants. The single allele knockout parasites were transfected with the neomycin knockout construct and plated on SDM-79 media containing 100 mg/ml hygromycin and 50 mg/ml geneticin to select for homozygous mutants. Cathepsin B complemented null mutants were created by transfecting pXGSAT plasmids carrying cathepsin B coding sequence into the null mutant parasites. The transfected null mutant parasites were plated on SDM-79 media containing 100 mg/ml hygromycin, 50 mg/ml geneticin and 100 mg/ml nourseothricin. Leishmania log stage promastigotes (4610 7 parasites) were transfected with 5 mg linearized knockout construct DNA and 30 mg circular plasmid DNA as previously described [19]. Transfected parasites were re-suspended in drug-free media for 24 hours then drugs were added for selection. The generated double allele knockouts and complemented null mutant parasites were characterized by using PCR and southern and western blot analysis.
The PCR products were gel purified and 1 ug of the purified products were labeled with DIG. Leishmania genomic DNA was digested overnight with PstI enzyme. Ten micro gram of the digested DNA were run on 0.8% agarose gel and transferred onto positively charged nylon membranes. Blots were detected with anti-DIG antibody solution (75 mU/ml).

Leishmania lysate preparations
Leishmania parasites were washed twice in ice-cold PBS and resuspended in 5610 8 parasites per 500 ml lysis buffer. Two different lysis buffers were used to prepare lysates for protease assay (20 mM Tris pH 8.0, 10 mM EDTA, pH 8.0, 4 mM NaCl) and for western blot analysis (5 mM Tris pH 8.0, 2 mM EDTA pH 8.0, 0.5 mM PMSF). Lysates were kept on ice for 30 minutes, sonicated (3 cycles, 30 sec/cycle, 30 sec cooling in between; 45 mv), and centrifuged at 13400 xg for 20 minutes to remove cell debris. Protein concentrations were determined using BCA method and stored at -80 o C until use.

Western blotting
Leishmania protein lysates were run on SDS-Polyacrylamide gel and transferred on to Hybond P membranes (GE healthcare). ECL advanced western blot detection kit (GE healthcare) was used to detect proteins using primary antibodies and HRP conjugated secondary antibodies.

Cysteine protease activity assay
Leishmania lysates were incubated with fluorogenic substrates (Z-Arg-Arg-AMC or Z-Phe-Arg-AMC) in a protease assay buffer (100 mM sodium acetate, pH 5.5, 10 mM dithiothreitol, 0.1% Triton X-100, 1 mM EDTA and 0.5% DMSO; pH 6.0) in 96 well plate at 37uC for 10 minutes. AMCs liberated from fluorogenic substrates by protease activity were quantified by ELISA reader at 460 nm emission and an excitation wavelength of 355 nm.

Infection of U937 macrophages
Infections of U937 macrophages were done as previously described [20]. Cells were grown to confluency, washed, and cultured at a concentration of 2.5610 6 cells/cm 2 /well in a chamber slide in the presence of 150 nM PMA, and they were infected with stationary stage Leishmania parasites at a cell to parasite ratio of 1:20. Extracellular parasites were removed by washing in a complete RPMI media and infection was monitored by Diff-Quick (Baxter) staining.

Leishmania proteome profiling
Protein extraction. L. donovani cathepsin B wild type and null mutant proteins were extracted from stationary stage parasites as described previously [21] and transported to UVic Genome BC Proteomics Centre (University of Victoria, Victoria, BC, Canada) for iTRAQ based LC-MS/MS analysis.
Trypsin digestion and peptide labeling. Trypsin digestions and iTRAQ labelings were done following standard procedures (Applied Biosystems, Foster City, CA, USA). Protein samples (100 mg each) extracted from cathepsin B wild type and null mutant parasites were precipitated overnight in acetone at 4 o C followed by resolublization in 0.5 M triethylammonium bicarbonate (TEAB) and 0.2% SDS. The proteins were reduced with tris-(2-carboxyethyl) phosphine (TCEP), alkylated with methyl methanethiosulfonate (MMTS), and digested with trypsin (Promega). Peptdies of the wild type and the null mutant parasites were labeled with different iTRAQ reagents and combined for analysis with strong cation exchange (SCX) high performance liquid chromatography (HPLC) and liquid chromatographytandem mass spectrometry (LC-MS/MS). SCX-HPLC and LC-MS/MS Analysis. Peptides were fractionated by SCX-HPLC on a VISION Workstation (Applied Biosystems) equipped with Polysulfoethyl A SCX column, 100 mm (length)64.6 mm (inner diameter), 5 um (particle size), and 300Å (pore size; Poly LC, Columbia, MD). Samples in buffer A (10 mM KPO 4 (pH 2.7), 25% ACN) were injected onto a preequilibrated column under a flow rate of 0.5 ml minute 21 . A gradient of 0-35% buffer B (10 mM KH 2 PO 4 , 25% ACN, 0.5 M KCl) in 30 minutes was applied. Peptide fractions were analyzed by reverse phase (RP) LC-MS/MS system. LC separations were done on a 75 mm615 cm C18 PepMap Nano LC column (3 mm particle size, 100Å pore size, LC Packings, Amsterdam, The Netherlands) with A 300 mm65 mm C18 PepMap guard column (5 mm particle size, 100Å pore size, LC Packings, Amsterdam, The Netherlands). Solvent A [water/acetonitrile (98:2, v/v), 0.05% formic acid] was used for sample injection and equilibration on the guard column at a flow rate of 100 ml minute 21 . Solvent B [acetonitrile/water (98:2, v/v), 0.05% formic acid] was used for high-resolution chromatography and introduction into mass spectrometer at a flow rate of 200 nl minute 21 .
Tandem mass spectrometry was performed on a hybrid quadrupole-time of flight (TOF) mass spectrometer (QStar Pulsar I) equipped with a nano-electrospray ionization source (Proxeon, Odense, Denmark). LC-MS/MS data acquisitions were done as described previously [21,22]. Briefly, A 1 second TOFMS scan of mass range 400-1200 amu and two 2.5 second product ion scans of mass range 100-1500 amu were performed. The two most intense peaks over 20 counts with 2-5 charge state were selected for fragmentation. Fragmentation of the same isotopic cluster peaks was prevented by using a 6 amu (atomic mass unit) window. Ions selected for fragmentation were excluded from identification for 180 seconds. Protein identifications were done by searching the data against the L. infantum v.3.0 database (http://www.genedb. org/genedb/linfantum/) using ProteinPilot Version 2.0 [23] (Applied Biosystems). Proteins were considered to be confidently identified if they had an unused Prot Score . 1.3 (confidence $95%). ProteinPilot calculates relative abundance of a protein from average abundant values of all its peptides.

Gene ontology (GO) enrichment and Biological network analysis
To determine over represented gene ontology terms, enrichment analysis was performed using a cytoscape plugin BiNGO Version 2.44 [24]. Modulated protein sequences were used to retrieve their respective yeast orthologs (Table S3) from Saccharomyces cerevisiae S288c (taxid:559292) refseq database by using BLASTP 2.2.25 [25]. BLASTP was run using default parameters and Leishmania sequence IDs that found hit with E # 4.4E-02 and bit score of 32.7 were used for enrichment analysis in BinGO. Hypergeometric test (p , 0.05) with Benjamin-Hochberg FDR correction, whole annotation as a reference set, and S. cerevisiae as a test organism were used for the analysis. Interaction networks were generated by using Cytoscape MiMI plugin version 3.1 [26] and well-connected interactions were identified by MCODE [27].

Characterization of L. donovani cathepsin B null mutant parasites
We generated L. donovani cathepsin B null mutant parasites by deleting 529 bp region of the coding sequence containing the active site cysteine residue and the occluding loop (Figure 1). We confirmed proper integration of knockout constructs to the cathepsin B locus by southern blotting (Figure 2). Hygromycin (Figure 2A, Hyg) and Neomycin (Figure 2A, Neo) probes hybridize only to the null mutant parasite DNA resulting in two expected size bands. The 39 flanking region probe hybridizes to a 3.8 kb band in the wild type parasite and to 3.9 and 4.0 kb bands in the cathepsin B null mutant parasite (Figure 2A, 39). The 100 bp difference between the two bands is due to difference in a PstI restriction site position on the neomycin and hygromycin markers. We confirmed cathepsin B gene disruption by a probe designed based on the 529-bp region of cathepsin B, which hybridizes only to the wild type parasite DNA (Figure 2A, CatB). Specific primers of the deleted region amplify only from the wild type and  Figure  2B, Sat). Cathepsin B 30-kDa mature protein was detected in the wild type and the complemented null mutant parasites, but not in the null mutant parasites ( Figure 2C, CatB). We studied functional effect of cathepsin B disruption by testing protease activity and intracellular survival of the wild type and the null mutant parasites. L. donovani cathepsin B wild type lysates show significantly higher activity against Z-Phe-Arg-AMC, and complementation partially reverses protease activity of the null mutant parasites ( Figure 2D). We found no difference in the activity of L. donovani cathepsin B wild type and null mutant parasites against Z-Arg-Arg-AMC (data not shown). Cathepsin B null mutant parasites show significantly reduced number of amastigotes inside macrophages 24 hours after infection, and episomal complementation augments this effect (Figure 3). The reduced number of amastigotes could be due to attenuated multiplication and/or survival of the parasites. There is no difference in the survival of the wild type and the null mutant parasites in culture (data not shown).
Cathepsin B gene disruption modulates L. donovani proteome L. donovani cathepsin B wild type and null mutant protein samples pooled from three independent preparations were analyzed using iTRAQ based quantitative proteomics and data generated from two experiments by iTRAQ swapping (iTRAQ 115 and iTRAQ 116) are presented here. Proteins detected in both experiments and represented by at least one unique peptide with $ 95% confidence were considered identified by LC-MS/ MS. This resulted in the identification of a total of 463 proteins. The list of peptides used for identifications are given in Table S1. Each protein was identified on average by 8 peptides, and only four proteins were identified by a single unique peptide. Over 95% of the peptides had precursor delta masses less than 0.16 Da. The relative abundance of the 463 proteins was used to determine cathepsin B disruption induced modulation of L. donovani proteome (Table S2, Figure 4). Among the 463 proteins, 83 proteins were modulated at least by 1.5 fold change ( Table 1). Sixty-five proteins were increased in the L. donovani cathepsin B wild type and 18 proteins were increased in L. donovani cathepsin B null mutant parasites. Cathepsin B is the most significantly modulated protein with 9.12 fold increase in the wild type parasites ( Table 1). The level of cathepsin B detected in the null mutant parasites comes from peptides belonging to the non-deleted region of cathepsin B. Western blotting of peroxidoxin-4 (LinJ.23.0050), which was increased by 1.6 fold in the wild type parasites, and alpha tubulin (LinJ.13.1460), which was increased by 1.73 fold in the null mutant parasites, validate the proteome data ( Figure 5). The fold change for peroxidoxin obtained by western blot is larger than that obtained by iTRAQ analysis. This is possibly due to underestimation of protein ratios by iTRAQ [28][29][30][31].
Cathepsin B gene disruption modulates Lesihmania secreted proteins L. donovani secreted proteins reported in [32] were used to examine the effect of cathepsin B disruption on L. donovani secretome. Sixty six percent (55/83) of the proteins modulated by L. donovani cathepsin B gene disruption were found to be Leishmania secretome proteins and 18% (15/83) of these were identified as actively secreted proteins [32].
Five molecular function GO terms are significantly enriched in the modulated proteins (Figure 7). The list of modulated proteins representing molecular function categories are given in Table S5. Catalytic activity (GO3824, 58%) is the most enriched followed by structural molecule activity (GO5198, 26.8%), structural constituent of ribosome activity (GO3735, 23.8%), oxido-reductase activity (GO16491, 19.4%) and rRNA binding activity (GO19843, 5.9%). Among 13 modulated proteins with oxido-reductase activity, all except for one (LinJ.34.0170, Malate dehydrogenase) are increased in the wild type parasites. All the proteins with ribosome structural constituent activity (GO3735) are increased in the wild type parasites. Only one of the modulated proteins with structural molecule activity (GO5198), alpha tubulin (TUB1, LinJ.13.1460), is increased in the null mutant parasites, and it is known to have structural role distinct from the 18 identified proteins with the same GO term. Among four proteins with rRNA binding activity (GO19843), only elongation factor 2 (LinJ.36.0210) is increased in the null mutant parasites and the remaining three are ribosomal proteins. The results indicate that significant number of proteins that are involved in oxido-reductase and structural constituents of ribosome activity are increased in cathepsin B wild type parasites.
Enrichment analysis of the site of function of the modulated proteins indicates macromolecular complex (GO32991, 50.7%) and mitochondrion (GO-5739, 38.8%) among the most significantly enriched GO terms (Figure 8). The modulated proteins representing cellular component categories are given in Table S6. Among the enriched macromolecular complex structures are ribonucleoprotein complex (GO30529, 31.3%), respiratory chain complex III (GO45275, 3%) and proteasome core complex (GO9773, 3%). Proteins residing in the mitochondrion include those involved in oxido-reductase activity (GO16491) and cell death. Among proteins involved in cell death are cytochrome c1 (LinJ.07.0210, CyT1) and cytochrome c (LinJ. 16.1390, CyT7). The cytochrome proteins are differentially modulated and unlike cytochrome c, cytochrome c1 is part of the respiratory chain complex III (GO45275). L. donovani proteins modulated by Cathepsin B gene disruption form molecular complex networks As shown above, cathepsin B gene disruption affects proteins with related function and those forming macromolecular complex structures. In order to gain more insight, we performed interaction network analysis on the modulated proteins ( Figure 9A). Interaction analysis of the 67 proteins provided three different size networks made of 33, 4, and 2 nodes. More than 50% of the tested modulated proteins interact with one another and most of them belong to the same interaction network. The 33-node network consists of 28 nodes that were increased in the wild type parasites, whereas the 4-node and 2-node networks have an equal number of differentially modulated nodes. Ribosomal proteins ( Figure 9B) and respiratory chain complex proteins ( Figure 9C) form highly connected network of proteins increased in the wild type parasites.

Discussion
We have generated and characterized L. donovani cathepsin B null mutant parasites by deleting part of cathepsin B coding region containing the active site cysteine residue and the occluding loop. Cathepsin B null mutants show significantly reduced activity against Z-Phe-Arg-AMC substrate indicating loss of cathepsin B Figure 9. Network analysis of proteins modulated by cathepsin B gene disruption. Interaction network of 68 modulated proteins mapped to 67 yeast proteins (A). Highly connected network of ribosomal proteins (B). Highly connected network of oxidation-reduction proteins (C; TSA1, Peroxidoxin; GLR1, trypanothione reductase; TRX2, tryparedoxin). Proteins increased and decreased in the null mutant parasites are shown in green and red respectively. Node color intensity is proportional to the extent of modulation. doi:10.1371/journal.pone.0079951.g009 Table 2. Modulated proteins involved in oxido-reductase activity, translation, and proteolysis.

Id
Description Proteins involved in oxido-reductase activity activity. L. donovani cathepsin B null mutant parasites show ,66% reduction in activity, whereas antisense cathepsin B inhibited L. chagasi had ,35% reduction in activity [8]. We found no difference in the activity of L. donovani cathepsin B wild type and null mutant parasites against Z-Arg-Arg-AMC (data not shown). This is consistent with the predicted role of the S 2 sub-site glycine residue in L. donovani and L. mexicana cathepsin B [12][13][14]. Therefore, like T. brucei cathepsin B, which happens to have a glycine residue at S 2 sub-site, L. major and L. donovani cathepsin B may have indispensable role in survival and pathogenesis [11]. In agreement with previous study done by antisense inhibition of cathepsin B [8], L. donovani cathepsin B null mutants show significantly reduced virulence inside macrophages (Figure 3).
To gain insight on the mechanism of cathepsin B role in Leishmania survival and pathogenesis, the global effect of cathepsin B gene disruption on L. donovani proteome was studied by using iTRAQ based quantitative proteomics. We identified 463 proteins, of which 83 (18%) are modulated. We validated the proteome data by western blotting alpha tubulin and peroxidoxin-4 proteins. Most (66%) of the modulated proteins, including alpha tubulin and peroxidoxin, are Leishmania secreted proteins [32]. Some of these proteins, particularly those involved in protein turnover and cell death, are enriched in vesicle derived exoproteomes of Leishmania promastigotes [33]. Leishmania vesicle derived exosomes target virulence factors into host cell and modulate signalling pathways of macrophages to inhibit host defense system allowing the parasite to establish infection [34][35][36]. Hence, cathepsin B gene disruption induced modulation of secreted proteins and the significantly reduced virulence of the null mutants inside macrophage is consistent with the role of Leishmania secreted proteins. Leishmania secreted proteins potentiate anti-inflammatory immune response to suppress the host defense during infection [37][38][39]. Some of the modulated Leishmania secreted proteins, alpha-tubulin, ribosomal protein S4, and elongation factor 2, have been implicated in immune modulation [40]. We are currently investigating the effect of cathepsin B on the host immune system, and preliminary evidence suggests differential modulation of the host immune response by L. donovani cathepsin B wild type and null mutant parasites (unpublished data).
Significant number of proteins modulated by cathepsin B gene disruption play role in translation process (GO6412, 38.8%; Table S4). Most of these proteins are ribosomal proteins, which are increased in the wild type parasites and involved in posttranscription regulation of gene expression (GO10608 ; Table  S4). Leishmania ribosomal proteins are stage-regulated with less protein level in amastigotes and believed to be responsible for decline of protein synthesis in the amastigotes [41]. They form highly connected interaction network ( Figure 9B), which are likely to be co-regulated. Recent findings in Trypanosoma suggest post-transcriptional regulon mediated co-regulation of functionally related proteins [42][43][44][45]. Post-transcriptional regulon mediated regulation of functionally related proteins and the effect of ribosomal proteins on protein synthesis remains to be determined in Leishmania. Post-transcriptional regulation of gene expression GO term is also represented by RNA binding protein, glycyl tRNA synthetase, and isoleucyl-tRNA synthetase (GO10608 ;  Table S4). Their role in post-transcriptional regulation of gene expression of Leishmania is unknown. Oxidation-reduction process (GO 55114, Table S4) is another significantly enriched GO term represented by proteins involved in cell death (cytochrome c and cytochrome c1) and in oxido-reductase activity ( Table 2). Cytochrome c and c1 proteins are differentially modulated in cathepsin B wild type and null mutant parasites, and unlike cytochrome c, cytochrome c1 is part of the mitochondrial respiratory chain complex III (GO5750 , Table S4). Cytochrome c1 cleavage by caspase 3 amplifies release of cytochrome c resulting in mammalian cell death [46]. Cathepsin B has been implicated in Leishmania cell death [47]. Since there are no caspases in Leishmania, the effect of cathepsin B on Leishmania cytochromes deserves an investigation. Most of the proteins with oxido-reductase activity are increased in cathepsin B wild type parasites ( Table 2) and some are complex forming proteins ( Figure 9C). Peroxidoxins or trypareodxin peroxidase, trypanothione reductase, and tryparedoxin are virulence factors of Leishmania known to have role in defense against oxygen and nitrogen free radicals [48][49][50][51]. Mitochondrial peroxidoxin (peroxidoxin 4) has been shown to prevent Leishmania cell death [52]. The decreased level of these proteins might explain the attenuated virulence of L. donovani cathepsin B null mutant parasites inside macrophages.
In summary, we have generated L. donovani cathepsin B null mutant parasites and revealed cathepsin B disruption induced modulation of L. donovani proteome. The results implicate cathepsin B in the regulation of Leishmania secreted proteins, particularly those involved in oxidation-reduction, cell death, and protein turnover. The generated null mutant parasites will be useful to elucidate the role of cathepsin B in Leishmania protein secretion and its effect on the host immune system in vivo. Figure S1 Cathepsin B amino acid sequence alignment. LdCatB, LmajCatB, LmexCatB, HumanCatB, RatCatB and BovineCatB are L. donovani, L. major, L. mexicana, Human, Rat, and Bovine cathepsin B cysteine proteases respectively. The filled and unfilled arrows indicate pre and pro cleavage sites respectively. The Cysteine and Histidine active site residues are indicated with red arrows. D1, D2 and D3 represent deletion sites in Leishmania cathepsin B cysteine proteases. The red line underlines the occluding loop. The histidine residues of the occluding loop are shown with double green arrows. The conserved amino acid regions are colour shaded. The blue arrow indicates the S 2 sub-site amino acids. (TIFF) Table S1 Unique peptides used for the identification of L. donovani cathepsin B wild type and knock out proteome. Among the 463 proteins identified, 459 proteins had at least two unique peptides that were detected in two independent experiments. The peptides corresponding to each protein are shown. (XLS) Table S2 Relative abundance of L. donovani cathepsin B wild type and knock out proteins identified in two experiments. L. donovani cathepsin B wild type (WT) and knock out (KO) protein ratios determined in two independent experiments (Exp. 1 and Exp. 2), mean of protein ratios of the two experiments, standard deviation of the means, and log 2 transformed mean of the protein ratios are shown.