TY - JOUR
AU - Fischer, Peter U.
AB - Introduction Onchocerca volvulus is a filarial nematode parasite and the agent of onchocerciasis, also known as river blindness. Onchocerciasis is a neglected tropical disease that is targeted for global elimination. Recent data from the World Health Organization (WHO) estimated that over 220 million people live in areas at risk of onchocerciasis transmission, almost exclusively in sub-Saharan Africa [1]. Adult worms reside in subcutaneous nodules (onchocercomas) and have a reproductive life span of up to 15 years. The female worms release millions of microfilariae that migrate through the skin and sometimes invade the eye and may cause eye disease. Male worms grow up to 8 cm and migrate from one nodule to another, while females can grow up to 60 cm and are permanently coiled up within the nodules. Nodules vary in size from less than 1 cm to up to 5 cm in diameter and can contain up to 12 females, although about 30% of the nodules contain only a single female [2]. Currently there is no drug available that efficiently kills all adult worms that can be used for mass drug administration (MDA) [3]. The main strategy to eliminate onchocerciasis is annual or semiannual MDA with ivermectin alone or ivermectin plus albendazole in areas co-endemic for lymphatic filariasis. Ivermectin efficiently kills microfilariae and after repeated doses it also has a limited effects on adult worms [4]. A small percentage of older, adult female O. volvulus are known to develop pleomorphic neoplasms (PN) and it appears that this percentage increases after several doses of ivermectin and can be as high as 10% [5,6]. The induction of PN has been considered as one mechanism how ivermectin contributes to the death of worms as the filarial origin of the neoplasms and the pleomorphism of the tumor cells have been addressed in previous studies [5,7]. Morphological studies have shown that female worms with PN have abnormal embryogenesis, are often sterile and do not produce viable microfilariae [5,6]. This sterility is independent of the location of the PN, and it is independent whether they are confined to the pseudocoelomic cavity or are within the reproductive system. PN tissue can be detected in distal and proximal parts of a female, indicating that neoplasms can reach many cm in length. Immunohistochemical studies have revealed that PN are comprised of several different cell types, which can be labeled using various antibodies targeting worm proteins (such as those against hypodermis, endothelia, muscles, oocytes, spermatocytes, or embryos of the worms). The differential localization of proteins within the tumorous cell types, highlight the pleomorphism of these tumors [5]. Although it was assumed that the ovaries are the origin of the PN, the detailed origin and the protein inventory of PN is unknown. Technical advances in proteome analysis by liquid chromatography mass spectrometry (LC/MS) and increased genome information of O. volvulus together with improved laser capture microdissection (LCM) have made visual proteomics of O. volvulus possible. Visual proteomics combines morphology and digital image analysis with LCM and ultra-high-sensitivity LC/MS [8]. If a standard amount or area of tissue is analyzed, the method can provide semi-quantitation of proteins. Several studies describe the proteome of whole adult female O. volvulus worms, and between 2,100 and 3,900 expressed proteins have been identified in this lifecycle stage [9–11]. However, these studies used adult female worms of different ages that contain various reproductive stages, and no proteomic information is available at the tissue or organ level of individual worms. The objective of the current study was to establish a protocol for efficient visual proteomic analysis of paraffin-embedded O. volvulus nodules. We compared the proteomic inventory of PN tissue to the inventory of the body wall (hypodermis, lateral chords, muscles), uterus and intestine from three worms with PN and three healthy worms without neoplasms with intact embryogenesis. We were able to show that ethanol fixed, paraffin embedded nodules are a rich source for proteomic analysis of adults O. volvulus; more than 2,200 proteins were identified in the analyzed tissue types. We identified a specific proteomic inventory of proteins in PN that could be associated with the development of tumors in O. volvulus. Materials and methods Ethics statement The protocol for the clinical trial was reviewed and approved by ethical review committees at the University of Health and Allied Sciences (UHAS) in Ho, Ghana, the Ghana Health Service, The Ghana Food and Drug Authority, Case-Western Reserve University (Cleveland, OH, USA) and Washington University School of Medicine (St. Louis, MO, USA) (IRB ID: 201910085) [12]. The use of de-identified nodule sections for further analysis is not considered human subjects research according to the institutional review of Washington University School of Medicine. Nodule samples preparation for laser capture microdissection O. volvulus nodules were obtained from a clinical trial that compared the tolerability and efficacy of IDA (ivermectin, diethylcarbamazine, albendazole) versus a comparator treatment (ivermectin plus albendazole) in persons with onchocerciasis in the Volta region, Ghana [12]. As part of that study, nodules were evaluated by two independent readers who had to agree on the number of worms in a nodule, the sex of the worms, their embryogenesis, and whether a worm was alive or dead at the time of nodulectomy. All nodules, whether containing females with PN or healthy controls (HC), were excised from patients who have completed ivermectin combination treatment. PN female worms represented 5.6% (24/428) of all female adult worms found in participants from the Ghana trial (Table 1, see ref 12 Table 4). For each parasite, four body regions were selected: body wall (including hypodermis, lateral chord, and muscles), gut/intestine, uterus and PN tissue or embryos (morulae, coiled and stretched microfilariae) (Fig 1). Download: PPT PowerPoint slide PNG larger image TIFF original image Table 1. Summary of females in excised nodule and nodules with females with PN by treatment group. The table shows the total number of nodules evaluated 18 months after treatment, summarizing the total number of alive females and number of “live” PN females in each treatment arm. I: ivermectin, D: diethylcarbamazine, A: albendazole, IDA3: IDA treatment for 3 days and PN: pleomorphic neoplasm. https://doi.org/10.1371/journal.pntd.0012929.t001 Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 1. H&E stained cross-sections of O. volvulus females. A: HC female with coiled embryos in the uterus. B: PN worm with neoplasm in both uterus branches. C: PN female worm with neoplasm in one uterus branch and in the pseudocoelomic cavity. The second uterus branch is filled with degenerated embryos and was not used for LCM. D: PN female with neoplasm in one uterus branch and the pseudocoelomic cavity. The second uterus branch is empty. E: PN cells in one uterus branch and in the pseudocoelomic cavity. HC: healthy control, PN: pleomorphic neoplasm, b= body wall, co= coiled embryos i = intestine, lc = lateral chord, mu= muscle, ut =uterus, scale bar= 50 μm. https://doi.org/10.1371/journal.pntd.0012929.g001 For LCM, six nodules with a single female worm (three with PN, three HC) were selected to provide biological triplicate samples of each condition (Table 2). Nodules with a single “live” female worm were used to avoid mix-up of different females. In each nodule section, at least ten worm sections needed to be available for each female. Three slides with three 10 μm sections were prepared on PEN (polyethylene naphthalate) slides for each nodule and one consecutive slide with a 5 µm section. This slide was stained either with hematoxylin and eosin (H&E) or O. volvulus aspartic protease (Ov-APR). The slide was photographed, and the tissue of interest was color coded for each worm tissue. The image was then used as guide during LCM to avoid potential contamination with sperm or degenerate embryos, which can also be found in the uteri (S1 Fig.) [13]. Tissue was collected from three nodule sections to assure sufficient material using the LCM System LMD7000 (Leica Microsystems GmbH, Wetzlar, Germany). The system automatically recorded the size of the excised sample. Tissue was compiled into the cap of a 0.5 mL PCR tube (Axygen, Union City, CA, USA) containing 30 μL of 8M urea buffer. Collection tubes were spun down and ready for subsequent preparation for mass spectrometry analysis. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 2. Sample information for the analyzed paraffin-embedded O. volvulus nodules. Three different PN and three healthy control (HC) worms were selected from the IDA trial, Ghana. Four body regions were selected for laser capture microdissection: body wall (including hypodermis, lateral chord, and muscles), intestine, uterus wall and PN tissue or embryos (morulae, coiled and stretched mf). In parenthesis is the area excised from the worm for each tissue in µm2. I: ivermectin, D: diethylcarbamazine, A: albendazole and PN: pleomorphic neoplasm, *average, assessed in technical triplicates. https://doi.org/10.1371/journal.pntd.0012929.t002 Sample preparation for LC-MS/MS The collected LCM cells were lysed in 8M urea-containing buffer. Samples were then reduced with 4 mM dithiothreitol (DTT) at 50°C for 30 minutes, followed by alkylation of cysteine residues with 18 mM iodoacetamide for 30 minutes in the dark. Proteins were digested with 2 µg of trypsin overnight at 37°C. The resulting peptides were desalted using solid-phase extraction with a C18 spin column and eluted with 0.1% trifluoroacetic acid (TFA) in 50% acetonitrile (ACN). Peptides from biological replicates (three PN and three HC worms) were reconstituted in 0.1% formic acid (FA) in water and analyzed by LC-MS/MS using a Vanquish Neo UHPLC system coupled to an Orbitrap Eclipse Tribrid Mass Spectrometer with a FAIMS Pro Duo interface (Thermo Fisher Scientific, San Jose, CA). Samples were loaded onto a Neo trap cartridge coupled with an analytical column (75 µm ID x 50 cm PepMap Neo C18, 2 µm) and separated using a linear gradient of solvent A (0.1% FA in water) and solvent B (0.1% FA in ACN) over 120 minutes. For MS acquisition, FAIMS switched between CVs of −35 V and −65 V with cycle times of 1.5 s per CV. MS1 spectra were acquired at 120,000 resolution, with a scan range from 375 to 1500 m/z, AGC target set at 300%, and maximum injection time set at Auto mode. Precursors were filtered using monoisotopic peak determination set to peptide, charge state 2 to 7, dynamic exclusion of 60 s with ±10 ppm tolerance. For the MS2 analysis, the isolated ions were fragmented by assisted higher-energy collisional dissociation (HCD) at 30% and acquired in an ion trap. The AGC and maximum IT were Standard and Dynamic modes, respectively. Data were searched using Mascot (v.2.8.3 Matrix Science) against two sets of combined databases. The first set is including Homo sapiens, cRAP, and O. volvulus databases, and the second set is Wolbachia and virus databases only. Trypsin was selected as the enzyme, and the maximum number of missed cleavages was set to 3. The precursor mass tolerance was set to 10 ppm, and the fragment mass tolerance was set to 0.6 Da for the MS2 spectra. Carbamidomethylated cysteine was set as a static modification, and dynamic modifications were set as oxidized methionine, deamidated asparagine/glutamine, and protein N-term acetylation. The search results were validated with 1% FDR of protein threshold and 90% of peptide threshold using Scaffold v5.3.0 (Proteome Software Portland, OR, USA). Proteins containing indistinguishable peptides based on MS/MS analysis alone were grouped to satisfy the principles of parsimony. Two independent Scaffold files were generated for O. volvulus and Wolbachia data sets. Spectra counts and peptide count for each peptide sequence are provided in Excel spreadsheets (Microsoft, Redmond, WA, USA) (S1 and S2 Tables). The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [14] partner repository with the dataset identifier PXD056237 and 10.6019/PXD056237 Protein functional annotation and graphic visualization A sequence search against a H. sapiens database was performed to discard any peptides that exactly matched host peptide sequences (considering leucine/isoleucine to be equivalent). Using H. sapiens and O. volvulus databases, a similar search was completed for Wolbachia and virus results, ensuring that no peptide spectrum matches (PSMs) were assigned to more than one species. The processed spectra and peptide count results are available for each condition and sample in S1 and S2 Tables, respectively. Functional annotations for all O. volvulus or Wolbachia proteins were assigned using results from InterProScan v5.59-91.0 to identify gene ontology classification and InterPro functional domains, and GhostKOALA v2.2 to assign KEGG (Kyoto Encyclopedia of Genes and Genomes) annotations [15–19]. Additional protein annotation was performed using PANNZER and Sma3s [20,21]. Potentially secreted proteins were identified using SignalP v6.0, where any protein with a predicted signal peptide and fewer than 2 transmembrane domains was classified as secreted [22]. Protein conservation data across nematodes, bacteria and host were quantified using BLAST [23]. Functional enrichment was performed using the website http://webgestalt.org/ for KEGG and InterPro domains while GOSTATS v2.50 was used for GO (Gene ontology) “molecular function” child term enrichment. Enrichment was considered significant if the FDR-corrected p-values were ≤ 0.05, and at least 3 proteins were represented in the enriched group. Protein abundance as normalized spectral abundance factors (NSAFs) were used to reduce bias in quantification toward larger proteins [24,25]. NSAF data was plotted in Excel spreadsheets (Microsoft) to compare protein abundances between tissues and conditions, and to perform data analysis. Protein and antibody production The sequence for OVOC8391 and its filarial orthologues were retrieved from WormBase Parasite [26]. MegAlign Pro from DNA Star was used to align the sequences with OVOC8391 as the reference. Two portions of this large gene with low similarity to other filarial species were identified (S2 Fig). Primers were designed for one of the low similarity regions and purchased from Integrated DNA Technology (IDT) (Coralville, IA, USA) (S3 Table). PCR amplification was completed using blunt end PCR using Phusion High Fidelity DNA Polymerase (Thermo Fisher Scientific, Waltham, MA, US) with an annealing temperature of 50°C. To amplify, O. volvulus adult worm complementary DNA (cDNA) was used, and manufacturer instructions were followed for all kits used. Following PCR, the results were visualized on a 1% agarose gel. Once confirmed, the gene fragment was ligated into linearized pET100D plasmid (Invitrogen, Waltham, MA, USA) and transformed into the TOP10 Escherichia coli strain. Colony PCR was performed to check if the fragment was transformed correctly. Plasmid preparations were completed using Qiagen Plasmid Kit. Plasmid preparations were sent to be sequenced (Azenta, Burlington, MA, USA) with sequence analysis being completed using SeqManPro v17.4 (DNAStar). After sequence confirmation, the plasmid was transformed into the BL21 E. coli strain for protein production. Cultures were grown at 37º C, 180 rpm in Luria Broth (Sigma, St. Louis, MO) with ampicillin at 50 ug/mL (GoldBio). Once they reached an OD600 over 0.6, the culture was induced with 1mM Isopropyl β-D-1-thiogalactopyranoside (IPTG) (GoldBio) and grown for an additional three hours. Cells were collected with centrifugation at 10,000 g for 15 minutes. Pellets were stored at -80°C until purification. Pellets were suspended in a urea buffer with sodium phosphate at pH 8. The cell suspension rocked for 1 hour at room temperature. The suspension was sonicated to further break up the cells at 30 second intervals a total of three times. Benzonase (Sigma, St. Louis, MO) was added at 1μL for every 10mL of buffer. The suspension was centrifuged at 10000g for 15 minutes to collect the lysate. The lysate was poured over a column with cobalt His-affinity beads (Sigma, St. Louis, MO). The column was washed with the suspension buffer and protein was eluted in a urea and sodium phosphate buffer pH8 with 500mM imidazole. Protein was further purified using an Electro-Eluter (Bio-Rad, Hercules CA, USA). The collected elute was concentrated using an Amicon Ultra 3.5 kD MWCO cutoff filter (Millipore-Sigma, Burlington MA, USA). Protein purity was evaluated by SDS-PAGE electrophoresis followed by staining with SimplyBlue SafeStain (Invitrogen). Protein was quantified using a Qubit. Three BALB/c mice were immunized with 20 µg of recombinant OVOC8391-Fragment 2 in complete Freund’s adjuvant. After 2 weeks, the mice were boosted with 20 µg of recombinant protein in incomplete Freund’s adjuvant. An additional boost was completed 2 weeks later with 20 µg of recombinant protein in incomplete Freund’s adjuvant. Sera was collected from the mice six days later. Immunohistochemical localization of OVOC8391-1 in adult O. volvulus. Paraffin embedded nodules with adult O. volvulus were sectioned at 5 µm and mounted on slides. After deparaffinization, the sections were blocked with 10% bovine serum albumin (BSA) solution for 30 minutes at room temperature. The first polyclonal mouse antibody OVOC 8391-2 was applied in a 1:100 dilution in 0.1% BSA and left at room temperature for one hour or at 4° C overnight. Polyclonal anti-mouse IgG produced in rabbit was used at a 1:1000 dilution for 30 minutes at room temperature. Alkaline phosphatase-anti-alkaline phosphatase antibody (APAAP) produced in mouse was added to the section at a 1:40 dilution and left at room temperature for 30 minutes. Fast Red TR/Naphthol AS was used as chromogen and Mayer’s hematoxylin solution as counterstain. All reagents were provided by Millipore Sigma, St. Louis, MO, USA. Nodule section images were taken using Olympus DP70 microscope digital camera (Olympus, Tokyo, Japan). Ethics statement The protocol for the clinical trial was reviewed and approved by ethical review committees at the University of Health and Allied Sciences (UHAS) in Ho, Ghana, the Ghana Health Service, The Ghana Food and Drug Authority, Case-Western Reserve University (Cleveland, OH, USA) and Washington University School of Medicine (St. Louis, MO, USA) (IRB ID: 201910085) [12]. The use of de-identified nodule sections for further analysis is not considered human subjects research according to the institutional review of Washington University School of Medicine. Nodule samples preparation for laser capture microdissection O. volvulus nodules were obtained from a clinical trial that compared the tolerability and efficacy of IDA (ivermectin, diethylcarbamazine, albendazole) versus a comparator treatment (ivermectin plus albendazole) in persons with onchocerciasis in the Volta region, Ghana [12]. As part of that study, nodules were evaluated by two independent readers who had to agree on the number of worms in a nodule, the sex of the worms, their embryogenesis, and whether a worm was alive or dead at the time of nodulectomy. All nodules, whether containing females with PN or healthy controls (HC), were excised from patients who have completed ivermectin combination treatment. PN female worms represented 5.6% (24/428) of all female adult worms found in participants from the Ghana trial (Table 1, see ref 12 Table 4). For each parasite, four body regions were selected: body wall (including hypodermis, lateral chord, and muscles), gut/intestine, uterus and PN tissue or embryos (morulae, coiled and stretched microfilariae) (Fig 1). Download: PPT PowerPoint slide PNG larger image TIFF original image Table 1. Summary of females in excised nodule and nodules with females with PN by treatment group. The table shows the total number of nodules evaluated 18 months after treatment, summarizing the total number of alive females and number of “live” PN females in each treatment arm. I: ivermectin, D: diethylcarbamazine, A: albendazole, IDA3: IDA treatment for 3 days and PN: pleomorphic neoplasm. https://doi.org/10.1371/journal.pntd.0012929.t001 Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 1. H&E stained cross-sections of O. volvulus females. A: HC female with coiled embryos in the uterus. B: PN worm with neoplasm in both uterus branches. C: PN female worm with neoplasm in one uterus branch and in the pseudocoelomic cavity. The second uterus branch is filled with degenerated embryos and was not used for LCM. D: PN female with neoplasm in one uterus branch and the pseudocoelomic cavity. The second uterus branch is empty. E: PN cells in one uterus branch and in the pseudocoelomic cavity. HC: healthy control, PN: pleomorphic neoplasm, b= body wall, co= coiled embryos i = intestine, lc = lateral chord, mu= muscle, ut =uterus, scale bar= 50 μm. https://doi.org/10.1371/journal.pntd.0012929.g001 For LCM, six nodules with a single female worm (three with PN, three HC) were selected to provide biological triplicate samples of each condition (Table 2). Nodules with a single “live” female worm were used to avoid mix-up of different females. In each nodule section, at least ten worm sections needed to be available for each female. Three slides with three 10 μm sections were prepared on PEN (polyethylene naphthalate) slides for each nodule and one consecutive slide with a 5 µm section. This slide was stained either with hematoxylin and eosin (H&E) or O. volvulus aspartic protease (Ov-APR). The slide was photographed, and the tissue of interest was color coded for each worm tissue. The image was then used as guide during LCM to avoid potential contamination with sperm or degenerate embryos, which can also be found in the uteri (S1 Fig.) [13]. Tissue was collected from three nodule sections to assure sufficient material using the LCM System LMD7000 (Leica Microsystems GmbH, Wetzlar, Germany). The system automatically recorded the size of the excised sample. Tissue was compiled into the cap of a 0.5 mL PCR tube (Axygen, Union City, CA, USA) containing 30 μL of 8M urea buffer. Collection tubes were spun down and ready for subsequent preparation for mass spectrometry analysis. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 2. Sample information for the analyzed paraffin-embedded O. volvulus nodules. Three different PN and three healthy control (HC) worms were selected from the IDA trial, Ghana. Four body regions were selected for laser capture microdissection: body wall (including hypodermis, lateral chord, and muscles), intestine, uterus wall and PN tissue or embryos (morulae, coiled and stretched mf). In parenthesis is the area excised from the worm for each tissue in µm2. I: ivermectin, D: diethylcarbamazine, A: albendazole and PN: pleomorphic neoplasm, *average, assessed in technical triplicates. https://doi.org/10.1371/journal.pntd.0012929.t002 Sample preparation for LC-MS/MS The collected LCM cells were lysed in 8M urea-containing buffer. Samples were then reduced with 4 mM dithiothreitol (DTT) at 50°C for 30 minutes, followed by alkylation of cysteine residues with 18 mM iodoacetamide for 30 minutes in the dark. Proteins were digested with 2 µg of trypsin overnight at 37°C. The resulting peptides were desalted using solid-phase extraction with a C18 spin column and eluted with 0.1% trifluoroacetic acid (TFA) in 50% acetonitrile (ACN). Peptides from biological replicates (three PN and three HC worms) were reconstituted in 0.1% formic acid (FA) in water and analyzed by LC-MS/MS using a Vanquish Neo UHPLC system coupled to an Orbitrap Eclipse Tribrid Mass Spectrometer with a FAIMS Pro Duo interface (Thermo Fisher Scientific, San Jose, CA). Samples were loaded onto a Neo trap cartridge coupled with an analytical column (75 µm ID x 50 cm PepMap Neo C18, 2 µm) and separated using a linear gradient of solvent A (0.1% FA in water) and solvent B (0.1% FA in ACN) over 120 minutes. For MS acquisition, FAIMS switched between CVs of −35 V and −65 V with cycle times of 1.5 s per CV. MS1 spectra were acquired at 120,000 resolution, with a scan range from 375 to 1500 m/z, AGC target set at 300%, and maximum injection time set at Auto mode. Precursors were filtered using monoisotopic peak determination set to peptide, charge state 2 to 7, dynamic exclusion of 60 s with ±10 ppm tolerance. For the MS2 analysis, the isolated ions were fragmented by assisted higher-energy collisional dissociation (HCD) at 30% and acquired in an ion trap. The AGC and maximum IT were Standard and Dynamic modes, respectively. Data were searched using Mascot (v.2.8.3 Matrix Science) against two sets of combined databases. The first set is including Homo sapiens, cRAP, and O. volvulus databases, and the second set is Wolbachia and virus databases only. Trypsin was selected as the enzyme, and the maximum number of missed cleavages was set to 3. The precursor mass tolerance was set to 10 ppm, and the fragment mass tolerance was set to 0.6 Da for the MS2 spectra. Carbamidomethylated cysteine was set as a static modification, and dynamic modifications were set as oxidized methionine, deamidated asparagine/glutamine, and protein N-term acetylation. The search results were validated with 1% FDR of protein threshold and 90% of peptide threshold using Scaffold v5.3.0 (Proteome Software Portland, OR, USA). Proteins containing indistinguishable peptides based on MS/MS analysis alone were grouped to satisfy the principles of parsimony. Two independent Scaffold files were generated for O. volvulus and Wolbachia data sets. Spectra counts and peptide count for each peptide sequence are provided in Excel spreadsheets (Microsoft, Redmond, WA, USA) (S1 and S2 Tables). The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [14] partner repository with the dataset identifier PXD056237 and 10.6019/PXD056237 Protein functional annotation and graphic visualization A sequence search against a H. sapiens database was performed to discard any peptides that exactly matched host peptide sequences (considering leucine/isoleucine to be equivalent). Using H. sapiens and O. volvulus databases, a similar search was completed for Wolbachia and virus results, ensuring that no peptide spectrum matches (PSMs) were assigned to more than one species. The processed spectra and peptide count results are available for each condition and sample in S1 and S2 Tables, respectively. Functional annotations for all O. volvulus or Wolbachia proteins were assigned using results from InterProScan v5.59-91.0 to identify gene ontology classification and InterPro functional domains, and GhostKOALA v2.2 to assign KEGG (Kyoto Encyclopedia of Genes and Genomes) annotations [15–19]. Additional protein annotation was performed using PANNZER and Sma3s [20,21]. Potentially secreted proteins were identified using SignalP v6.0, where any protein with a predicted signal peptide and fewer than 2 transmembrane domains was classified as secreted [22]. Protein conservation data across nematodes, bacteria and host were quantified using BLAST [23]. Functional enrichment was performed using the website http://webgestalt.org/ for KEGG and InterPro domains while GOSTATS v2.50 was used for GO (Gene ontology) “molecular function” child term enrichment. Enrichment was considered significant if the FDR-corrected p-values were ≤ 0.05, and at least 3 proteins were represented in the enriched group. Protein abundance as normalized spectral abundance factors (NSAFs) were used to reduce bias in quantification toward larger proteins [24,25]. NSAF data was plotted in Excel spreadsheets (Microsoft) to compare protein abundances between tissues and conditions, and to perform data analysis. Protein and antibody production The sequence for OVOC8391 and its filarial orthologues were retrieved from WormBase Parasite [26]. MegAlign Pro from DNA Star was used to align the sequences with OVOC8391 as the reference. Two portions of this large gene with low similarity to other filarial species were identified (S2 Fig). Primers were designed for one of the low similarity regions and purchased from Integrated DNA Technology (IDT) (Coralville, IA, USA) (S3 Table). PCR amplification was completed using blunt end PCR using Phusion High Fidelity DNA Polymerase (Thermo Fisher Scientific, Waltham, MA, US) with an annealing temperature of 50°C. To amplify, O. volvulus adult worm complementary DNA (cDNA) was used, and manufacturer instructions were followed for all kits used. Following PCR, the results were visualized on a 1% agarose gel. Once confirmed, the gene fragment was ligated into linearized pET100D plasmid (Invitrogen, Waltham, MA, USA) and transformed into the TOP10 Escherichia coli strain. Colony PCR was performed to check if the fragment was transformed correctly. Plasmid preparations were completed using Qiagen Plasmid Kit. Plasmid preparations were sent to be sequenced (Azenta, Burlington, MA, USA) with sequence analysis being completed using SeqManPro v17.4 (DNAStar). After sequence confirmation, the plasmid was transformed into the BL21 E. coli strain for protein production. Cultures were grown at 37º C, 180 rpm in Luria Broth (Sigma, St. Louis, MO) with ampicillin at 50 ug/mL (GoldBio). Once they reached an OD600 over 0.6, the culture was induced with 1mM Isopropyl β-D-1-thiogalactopyranoside (IPTG) (GoldBio) and grown for an additional three hours. Cells were collected with centrifugation at 10,000 g for 15 minutes. Pellets were stored at -80°C until purification. Pellets were suspended in a urea buffer with sodium phosphate at pH 8. The cell suspension rocked for 1 hour at room temperature. The suspension was sonicated to further break up the cells at 30 second intervals a total of three times. Benzonase (Sigma, St. Louis, MO) was added at 1μL for every 10mL of buffer. The suspension was centrifuged at 10000g for 15 minutes to collect the lysate. The lysate was poured over a column with cobalt His-affinity beads (Sigma, St. Louis, MO). The column was washed with the suspension buffer and protein was eluted in a urea and sodium phosphate buffer pH8 with 500mM imidazole. Protein was further purified using an Electro-Eluter (Bio-Rad, Hercules CA, USA). The collected elute was concentrated using an Amicon Ultra 3.5 kD MWCO cutoff filter (Millipore-Sigma, Burlington MA, USA). Protein purity was evaluated by SDS-PAGE electrophoresis followed by staining with SimplyBlue SafeStain (Invitrogen). Protein was quantified using a Qubit. Three BALB/c mice were immunized with 20 µg of recombinant OVOC8391-Fragment 2 in complete Freund’s adjuvant. After 2 weeks, the mice were boosted with 20 µg of recombinant protein in incomplete Freund’s adjuvant. An additional boost was completed 2 weeks later with 20 µg of recombinant protein in incomplete Freund’s adjuvant. Sera was collected from the mice six days later. Immunohistochemical localization of OVOC8391-1 in adult O. volvulus. Paraffin embedded nodules with adult O. volvulus were sectioned at 5 µm and mounted on slides. After deparaffinization, the sections were blocked with 10% bovine serum albumin (BSA) solution for 30 minutes at room temperature. The first polyclonal mouse antibody OVOC 8391-2 was applied in a 1:100 dilution in 0.1% BSA and left at room temperature for one hour or at 4° C overnight. Polyclonal anti-mouse IgG produced in rabbit was used at a 1:1000 dilution for 30 minutes at room temperature. Alkaline phosphatase-anti-alkaline phosphatase antibody (APAAP) produced in mouse was added to the section at a 1:40 dilution and left at room temperature for 30 minutes. Fast Red TR/Naphthol AS was used as chromogen and Mayer’s hematoxylin solution as counterstain. All reagents were provided by Millipore Sigma, St. Louis, MO, USA. Nodule section images were taken using Olympus DP70 microscope digital camera (Olympus, Tokyo, Japan). Immunohistochemical localization of OVOC8391-1 in adult O. volvulus. Paraffin embedded nodules with adult O. volvulus were sectioned at 5 µm and mounted on slides. After deparaffinization, the sections were blocked with 10% bovine serum albumin (BSA) solution for 30 minutes at room temperature. The first polyclonal mouse antibody OVOC 8391-2 was applied in a 1:100 dilution in 0.1% BSA and left at room temperature for one hour or at 4° C overnight. Polyclonal anti-mouse IgG produced in rabbit was used at a 1:1000 dilution for 30 minutes at room temperature. Alkaline phosphatase-anti-alkaline phosphatase antibody (APAAP) produced in mouse was added to the section at a 1:40 dilution and left at room temperature for 30 minutes. Fast Red TR/Naphthol AS was used as chromogen and Mayer’s hematoxylin solution as counterstain. All reagents were provided by Millipore Sigma, St. Louis, MO, USA. Nodule section images were taken using Olympus DP70 microscope digital camera (Olympus, Tokyo, Japan). Results Morphological description of nodules with live female worms with neoplasms Our study is based on an accurate histological analysis of worm tissue. Stained nodules slides retrieved were digitalized and evaluated by two independent microscopists. Both microscopists agreed on worm numbers, viability, and fertility of worms, but detection of PN was only indirectly included in the analysis [12]. For the current study, only nodules with a single “live” female worm with or without PN were used, to avoid potential contamination by other worms. O. volvulus females without neoplasm (HC) showed well-delimited organs like uterus, intestine, lateral cord and muscles and the presence of normal embryogenesis. This is defined as the presence of intact morula-, coiled-, pretzel- and stretched mf (Fig 1A, red staining). PN female worms had cells invading the pseudocoelom with a high cytoplasm/nuclei ratio (Fig 1B-E). Important parasite organ systems were still clearly recognizable, for example two uterine branches, the uterus wall, and the intestine (Fig 1B). One of the uterus branches can be present but is difficult to identify because of heavy colonization with the invasive PN cells (Fig 1C-E). The high affinity for the H&E stain in the PN tissue is indicative of high, condensed DNA content, because of rapid cell division. Proteins detected in neoplasms and other worm tissues In a first step, we determined the proteins present in the main tissue types (PN, uterus, body wall, intestine and embryos) from cross-sections of PN or HC female O. volvulus worms. In total, 3,150 eukaryote proteins (2,618 proteins from O. volvulus; 443 proteins from H. sapiens; 29 proteins from cRap; and 60 decoy proteins) and 165 proteins from Wolbachia were identified (S1 and S2 Tables). 2,325 O. volvulus derived proteins were detected in any sample but only 1,534 proteins were present in at least two replicates that also had two peptides in at least one tissue. Among these confidently identified proteins, 1,390 were present in the PN tissue with two peptides in two replicates (Fig 2A). The total number of proteins present only in the PN tissue was 1,221, but if the proteins that were also detected in HC parasite tissues are disregarded, this number goes down to 594 PN unique proteins (Fig 2A). In the HC worms (Fig 2B), 929 O. volvulus-derived proteins were detected, with 133 shared among all tissues. Embryo tissues (morula-, coiled-, pretzel- and stretched mf stages) of HC have with 574 unique proteins. Embryo tissue of HC worms contained 109 unique proteins that were not detected in PN worms. Because viruses can be linked to tumor development and an RNA virus that elicits an antibody response in humans was reported from O. volvulus [27] we searched our peptide results using a virus database. However, no matches to viral proteins were found. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 2. Venn diagram representing the number of O. volvulus-derived proteins for each tissue in the PN (A) and HC (B) worms. The overlap region between the circles shows proteins present in two or more stages. Numbers in parenthesis represents the total protein number found for the tissue/ the number of unique proteins not detected in the other sample type (i.e., detected in PN worms but not in HC worms in (A) and in HC worms but not PN worms in (B)). PN: pleomorphic neoplasm, HC: healthy control. https://doi.org/10.1371/journal.pntd.0012929.g002 Correlation between analyzed tissue areas and spectra count across samples One problem that we had to overcome during this study was to quantify minute amounts of samples from different tissues. We used the dissected area (in µm2) of the 10 µm thick sections as proxy for sample amount (see Table 2). While the dissected areas from each tissue differ, this variation does not fully account for the variations in spectra count or total protein number. For example, as illustrated in Figs 2 and 3, PN tissue (average analyzed tissue area 383,236 µm2) contains twice the number of proteins compared to HC embryos (average analyzed tissue area 357,919 µm2) and 42% of the spectra counts (16,567 vs 7018, respectively), despite having similar analyzed areas. A similar pattern is observed with HC gut and HC uterus tissues. The PN uterus, despite having 9% of the cut area, only accounts for 3% of the spectra count. All percentage values are relative to the PN tissue. Additionally, PN tissue has ten times more proteins than body wall tissue, yet only 2.6 times more area was cut and they have only 10% spectra counts. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 3. Comparison of protein detection in tissues from PN and HC O. volvulus. Venn diagrams represent the total number of proteins found for each tissue. In parenthesis: average tissue area cut (in µm²)/average number of spectra counts for the sample. Percentages are from the total protein PN: pleomorphic neoplasm, HC: healthy control. https://doi.org/10.1371/journal.pntd.0012929.g003 Proteins associated with adult worm neoplasms. In order to identify proteins or groups of proteins associated with PN, we compared the protein signature of PN with the proteins found in the intestine, body wall and uterus of the same worm and the corresponding tissues in HC worms (Fig 2). From the 1,390 O. volvulus-derived proteins found in PN tissue, 594 were not detected in any tissue of HC worms (Fig 2A and S1 Table). An interesting finding was that several tissues looked healthy and were easily recognizable by shape in the PN worms (Fig 2B-E). However, despite cutting a similar area for both tissue types under all conditions, fewer proteins were detected in each PN tissue compared to their counterpart in the HC worms (Fig 3). When compared, fewer proteins were identified in the intestine tissue from PN worms than in HC worms (97 vs 222, Fig 2) and they have only 66 proteins in common, representing 26.1% of all the proteins found in intestine tissues. Similar observations were made with the tissues of the uterus and body wall, where the number of overlapping proteins between PN and HC tissues was 12.6% and 43.3%, respectively (Fig 3). To functionally characterize the PN tissue we performed over-representation enrichment analysis, using Gene Ontology [16] for functional term enrichment, KEGG [19] for pathway enrichment and InterPro [17] for functional domain enrichment. Table 3 shows the 10 most significantly enriched pathways identified for each database for the PN tissue. The complete list of enriched PN tissue proteins is presented in S4 Table. Concerning the KEGG pathways, the most significant are for the ribosome, citrate cycle, glycolysis, and proteasome pathways. Protein enrichment analysis traced 10 InterPro domains; the categories with lowest FDR were nucleotide-binding alpha-beta plait domain superfamily, RNA-binding domain superfamily and RNA recognition motif domain. GO analysis identified 141 terms for cellular components with FDR-adjusted P values < 0.05, including the most significant terms “intracellular anatomical structure” and “cytoplasm”. Likewise, 75 different molecular functions GO terms were identified being RNA binding, structural molecule activity structural constituent of ribosome the domains with the most significant p-values. Similarly, 228 biological processes were significantly enriched and cellular amide metabolic, amide biosynthetic and peptide metabolic processes were the principal categories with lowest FDR-adjusted P value and more than 250 proteins included for each process. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 3. Significantly enriched gene ontology pathways, InterPro and KEGG domains. O. volvulus-derived proteins detected in the PN tissue, only proteins that were supported by at least two unique peptides in two biological replicates were used. The top 10 enriched terms for each functional category are shown. https://doi.org/10.1371/journal.pntd.0012929.t003 Although the number of proteins found in the PN tissue was high, the abundance of these proteins was lower compared to other tissues where they were detected (S1 Table). The most abundant protein was laminin (OVOC10067) which was also detected in all other tissue types. It was more abundant in the intestine, body wall and uterus for PN and HC worms but was less abundant in the embryos than in the NP tissue. Among the 10 most abundant proteins in the PN tissue, only two proteins did not share substantial sequence conservation with the host and are nematode specific: a pepsin inhibitor (OVOC9984) and a calcium binding protein (OVOC8391). When PN and HC embryo tissues are compared, as they have similar areas cut from the worm slides (Fig 1), they share 745 proteins (85.7% of the total HC embryo tissue proteins). In terms of abundance, 296 proteins are more abundant in the PN tissue and 182 are only present in PN (Fig 4, yellow dots). PN and HC embryo tissues share 420 proteins (Fig 4, red dots). Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 4. Protein abundance correlation between PN and HC worm tissues. The relative abundance of each detected protein in HC embryo tissue were plotted against the relative abundance in the PN tissue. Yellow dots represent proteins that are more abundant in the PN tissue. Red dots are proteins that are only highly abundant in PN and HC embryo tissue when compared to all other HC tissues proteins. PN: pleomorphic neoplasm, HC: healthy control. https://doi.org/10.1371/journal.pntd.0012929.g004 Immunolocalization of the immunoglobulin-like protein OvDig-1 To confirm our proteomic results using an independent method, we decided to clone, express, and generate antibodies to the calcium binding protein OVOC8391 (OvDig-1). This large protein has a well-characterized ortholog in Caenorhabditis elegans (CeDig-1) in which this giant immunoglobulin-like protein (13,100 amino acids) is predicted to enable calcium ion binding activity [28]. CeDig-1 is expressed in several structures, including glutamate-like receptors, body wall musculature, head mesodermal cells, male sex myoblasts, and non-striated muscles. In HC O. volvulus worms the protein was commonly detected by proteomics in all analyzed tissue and in PN worms it was found only in the body wall and PN tissues (S1 Table). Immunolocalization confirmed these results, with OvDig-1 being highly expressed in intrauterine and pseudocoelomic PN, but distribution was linked to specific cell types (Fig 5). OvDig-1 was highly expressed in larger cells with high plasma content (Fig 5C-F). The protein was absent in the pseudocoelomic PN with a large number of small, dividing cells as indicated by a strong nuclear DNA stain (Fig 5E). Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 5. Immunhistolocalization of OvDig-1 (OVOC8391) in a PN female O. volvulus. A: Cross-section of a O. volvulus female with a few stretched Mf in the vagina, negative control using pre-immune serum. B: Same section as in A but stained (red) with an antiserum raised against OvDig-1. The protein is localized in the hypodermis, connective muscle tissue and in the neoplasm. Stretched Mf are also OvDig-1 positive (arrow). C: Cross-section of a female with PN cells in both uterus branches. The outer uterus wall, the connective tissue, muscle and PN cells are OvDig-1positive (red). D: PN in the pseudocoelom cavity with strong staining for OvDig-1. Both uterus branches are empty, constricted and not labeled. E: PN in the uterus strongly labeled for OvDig-1. PN in the pseudocoelomic cavity shows a more homogenous cell type with weak labeling. HC: healthy control, PN: pleomorphic neoplasm, i = intestine, mu= muscle, ut =uterus, scale bar= 50 μm. https://doi.org/10.1371/journal.pntd.0012929.g005 In female O. volvulus with intact embryogenesis and without PN, OvDig-1 was expressed in primary oocytes (Fig 6B), stretched microfilaria (Fig 6C), coiled microfilariae surrounded by eggshells and “pretzel” (Fig 6E). Also stained were connective tissue, outer uterus membrane and in the fibrillar zone of muscle cells (Fig 6). Red labelling for the protein was not observed in the rest of the muscle cells, the hypodermis or lateral chord. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 6. Immunohistolocalization of OvDig-1 (OVOC8391) in a healthy female O. volvulus. A: Non stained longitudinal section with primary oocytes in the ovary. B: Consecutive section as in A stained for OvDig-1 with positive primary oocytes (red). C: Longitudinal section of a uterus branch with stretched Mf that are strongly positive for OvDig-1. D: Cross-section of a female with different morulae stage embryos in the uterus. The outer uterus membrane is stained, some muscle cells and connective tissue. E: Cross-Section of a female with coiled (pretzel) stages in the uterus. More advanced coiled mf are OVOC8391 positive, while unmature mf are negative, like the morulae. co= coiled embryos i = intestine, lc = lateral chord, mo= morulae, mu= muscle, ov=oocytes, ut =uterus, scale bar= 20 μm. https://doi.org/10.1371/journal.pntd.0012929.g006 Wolbachia-derived proteins in healthy and pleomorphic neoplasm worms Wolbachia proteins were curated the same way O. volvulus proteins and all peptides that may have originated from the mammalian host or the parasite were excluded from the final protein list (Tables 4 and S2). Overall, relatively few Wolbachia proteins were detected in 2 of the 3 worms with PN. In the PN worms most Wolbachia-derived proteins were detected in the body wall (six proteins), followed by PN tissue (four proteins) and only one protein in the uterus. In the HC worms, eight Wolbachia proteins were detected in the body wall and seven in the embryos with an ATP-dependent chaperone (WP_025264048.1) found in both. Two Wolbachia proteins were found in the intestine but also in the embryo and body wall tissues. Body wall tissue from the PN and HC worms have all six proteins shared. Blast for all these hits match Rickettsiales bacteria and several arthropods (S2 Table). Download: PPT PowerPoint slide PNG larger image TIFF original image Table 4. A summary of Wolbachia derived proteins detected in the O. volvulus samples. Presence of Wolbachia-derived proteins in the O. volvulus samples. https://doi.org/10.1371/journal.pntd.0012929.t004 Morphological description of nodules with live female worms with neoplasms Our study is based on an accurate histological analysis of worm tissue. Stained nodules slides retrieved were digitalized and evaluated by two independent microscopists. Both microscopists agreed on worm numbers, viability, and fertility of worms, but detection of PN was only indirectly included in the analysis [12]. For the current study, only nodules with a single “live” female worm with or without PN were used, to avoid potential contamination by other worms. O. volvulus females without neoplasm (HC) showed well-delimited organs like uterus, intestine, lateral cord and muscles and the presence of normal embryogenesis. This is defined as the presence of intact morula-, coiled-, pretzel- and stretched mf (Fig 1A, red staining). PN female worms had cells invading the pseudocoelom with a high cytoplasm/nuclei ratio (Fig 1B-E). Important parasite organ systems were still clearly recognizable, for example two uterine branches, the uterus wall, and the intestine (Fig 1B). One of the uterus branches can be present but is difficult to identify because of heavy colonization with the invasive PN cells (Fig 1C-E). The high affinity for the H&E stain in the PN tissue is indicative of high, condensed DNA content, because of rapid cell division. Proteins detected in neoplasms and other worm tissues In a first step, we determined the proteins present in the main tissue types (PN, uterus, body wall, intestine and embryos) from cross-sections of PN or HC female O. volvulus worms. In total, 3,150 eukaryote proteins (2,618 proteins from O. volvulus; 443 proteins from H. sapiens; 29 proteins from cRap; and 60 decoy proteins) and 165 proteins from Wolbachia were identified (S1 and S2 Tables). 2,325 O. volvulus derived proteins were detected in any sample but only 1,534 proteins were present in at least two replicates that also had two peptides in at least one tissue. Among these confidently identified proteins, 1,390 were present in the PN tissue with two peptides in two replicates (Fig 2A). The total number of proteins present only in the PN tissue was 1,221, but if the proteins that were also detected in HC parasite tissues are disregarded, this number goes down to 594 PN unique proteins (Fig 2A). In the HC worms (Fig 2B), 929 O. volvulus-derived proteins were detected, with 133 shared among all tissues. Embryo tissues (morula-, coiled-, pretzel- and stretched mf stages) of HC have with 574 unique proteins. Embryo tissue of HC worms contained 109 unique proteins that were not detected in PN worms. Because viruses can be linked to tumor development and an RNA virus that elicits an antibody response in humans was reported from O. volvulus [27] we searched our peptide results using a virus database. However, no matches to viral proteins were found. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 2. Venn diagram representing the number of O. volvulus-derived proteins for each tissue in the PN (A) and HC (B) worms. The overlap region between the circles shows proteins present in two or more stages. Numbers in parenthesis represents the total protein number found for the tissue/ the number of unique proteins not detected in the other sample type (i.e., detected in PN worms but not in HC worms in (A) and in HC worms but not PN worms in (B)). PN: pleomorphic neoplasm, HC: healthy control. https://doi.org/10.1371/journal.pntd.0012929.g002 Correlation between analyzed tissue areas and spectra count across samples One problem that we had to overcome during this study was to quantify minute amounts of samples from different tissues. We used the dissected area (in µm2) of the 10 µm thick sections as proxy for sample amount (see Table 2). While the dissected areas from each tissue differ, this variation does not fully account for the variations in spectra count or total protein number. For example, as illustrated in Figs 2 and 3, PN tissue (average analyzed tissue area 383,236 µm2) contains twice the number of proteins compared to HC embryos (average analyzed tissue area 357,919 µm2) and 42% of the spectra counts (16,567 vs 7018, respectively), despite having similar analyzed areas. A similar pattern is observed with HC gut and HC uterus tissues. The PN uterus, despite having 9% of the cut area, only accounts for 3% of the spectra count. All percentage values are relative to the PN tissue. Additionally, PN tissue has ten times more proteins than body wall tissue, yet only 2.6 times more area was cut and they have only 10% spectra counts. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 3. Comparison of protein detection in tissues from PN and HC O. volvulus. Venn diagrams represent the total number of proteins found for each tissue. In parenthesis: average tissue area cut (in µm²)/average number of spectra counts for the sample. Percentages are from the total protein PN: pleomorphic neoplasm, HC: healthy control. https://doi.org/10.1371/journal.pntd.0012929.g003 Proteins associated with adult worm neoplasms. In order to identify proteins or groups of proteins associated with PN, we compared the protein signature of PN with the proteins found in the intestine, body wall and uterus of the same worm and the corresponding tissues in HC worms (Fig 2). From the 1,390 O. volvulus-derived proteins found in PN tissue, 594 were not detected in any tissue of HC worms (Fig 2A and S1 Table). An interesting finding was that several tissues looked healthy and were easily recognizable by shape in the PN worms (Fig 2B-E). However, despite cutting a similar area for both tissue types under all conditions, fewer proteins were detected in each PN tissue compared to their counterpart in the HC worms (Fig 3). When compared, fewer proteins were identified in the intestine tissue from PN worms than in HC worms (97 vs 222, Fig 2) and they have only 66 proteins in common, representing 26.1% of all the proteins found in intestine tissues. Similar observations were made with the tissues of the uterus and body wall, where the number of overlapping proteins between PN and HC tissues was 12.6% and 43.3%, respectively (Fig 3). To functionally characterize the PN tissue we performed over-representation enrichment analysis, using Gene Ontology [16] for functional term enrichment, KEGG [19] for pathway enrichment and InterPro [17] for functional domain enrichment. Table 3 shows the 10 most significantly enriched pathways identified for each database for the PN tissue. The complete list of enriched PN tissue proteins is presented in S4 Table. Concerning the KEGG pathways, the most significant are for the ribosome, citrate cycle, glycolysis, and proteasome pathways. Protein enrichment analysis traced 10 InterPro domains; the categories with lowest FDR were nucleotide-binding alpha-beta plait domain superfamily, RNA-binding domain superfamily and RNA recognition motif domain. GO analysis identified 141 terms for cellular components with FDR-adjusted P values < 0.05, including the most significant terms “intracellular anatomical structure” and “cytoplasm”. Likewise, 75 different molecular functions GO terms were identified being RNA binding, structural molecule activity structural constituent of ribosome the domains with the most significant p-values. Similarly, 228 biological processes were significantly enriched and cellular amide metabolic, amide biosynthetic and peptide metabolic processes were the principal categories with lowest FDR-adjusted P value and more than 250 proteins included for each process. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 3. Significantly enriched gene ontology pathways, InterPro and KEGG domains. O. volvulus-derived proteins detected in the PN tissue, only proteins that were supported by at least two unique peptides in two biological replicates were used. The top 10 enriched terms for each functional category are shown. https://doi.org/10.1371/journal.pntd.0012929.t003 Although the number of proteins found in the PN tissue was high, the abundance of these proteins was lower compared to other tissues where they were detected (S1 Table). The most abundant protein was laminin (OVOC10067) which was also detected in all other tissue types. It was more abundant in the intestine, body wall and uterus for PN and HC worms but was less abundant in the embryos than in the NP tissue. Among the 10 most abundant proteins in the PN tissue, only two proteins did not share substantial sequence conservation with the host and are nematode specific: a pepsin inhibitor (OVOC9984) and a calcium binding protein (OVOC8391). When PN and HC embryo tissues are compared, as they have similar areas cut from the worm slides (Fig 1), they share 745 proteins (85.7% of the total HC embryo tissue proteins). In terms of abundance, 296 proteins are more abundant in the PN tissue and 182 are only present in PN (Fig 4, yellow dots). PN and HC embryo tissues share 420 proteins (Fig 4, red dots). Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 4. Protein abundance correlation between PN and HC worm tissues. The relative abundance of each detected protein in HC embryo tissue were plotted against the relative abundance in the PN tissue. Yellow dots represent proteins that are more abundant in the PN tissue. Red dots are proteins that are only highly abundant in PN and HC embryo tissue when compared to all other HC tissues proteins. PN: pleomorphic neoplasm, HC: healthy control. https://doi.org/10.1371/journal.pntd.0012929.g004 Proteins associated with adult worm neoplasms. In order to identify proteins or groups of proteins associated with PN, we compared the protein signature of PN with the proteins found in the intestine, body wall and uterus of the same worm and the corresponding tissues in HC worms (Fig 2). From the 1,390 O. volvulus-derived proteins found in PN tissue, 594 were not detected in any tissue of HC worms (Fig 2A and S1 Table). An interesting finding was that several tissues looked healthy and were easily recognizable by shape in the PN worms (Fig 2B-E). However, despite cutting a similar area for both tissue types under all conditions, fewer proteins were detected in each PN tissue compared to their counterpart in the HC worms (Fig 3). When compared, fewer proteins were identified in the intestine tissue from PN worms than in HC worms (97 vs 222, Fig 2) and they have only 66 proteins in common, representing 26.1% of all the proteins found in intestine tissues. Similar observations were made with the tissues of the uterus and body wall, where the number of overlapping proteins between PN and HC tissues was 12.6% and 43.3%, respectively (Fig 3). To functionally characterize the PN tissue we performed over-representation enrichment analysis, using Gene Ontology [16] for functional term enrichment, KEGG [19] for pathway enrichment and InterPro [17] for functional domain enrichment. Table 3 shows the 10 most significantly enriched pathways identified for each database for the PN tissue. The complete list of enriched PN tissue proteins is presented in S4 Table. Concerning the KEGG pathways, the most significant are for the ribosome, citrate cycle, glycolysis, and proteasome pathways. Protein enrichment analysis traced 10 InterPro domains; the categories with lowest FDR were nucleotide-binding alpha-beta plait domain superfamily, RNA-binding domain superfamily and RNA recognition motif domain. GO analysis identified 141 terms for cellular components with FDR-adjusted P values < 0.05, including the most significant terms “intracellular anatomical structure” and “cytoplasm”. Likewise, 75 different molecular functions GO terms were identified being RNA binding, structural molecule activity structural constituent of ribosome the domains with the most significant p-values. Similarly, 228 biological processes were significantly enriched and cellular amide metabolic, amide biosynthetic and peptide metabolic processes were the principal categories with lowest FDR-adjusted P value and more than 250 proteins included for each process. Download: PPT PowerPoint slide PNG larger image TIFF original image Table 3. Significantly enriched gene ontology pathways, InterPro and KEGG domains. O. volvulus-derived proteins detected in the PN tissue, only proteins that were supported by at least two unique peptides in two biological replicates were used. The top 10 enriched terms for each functional category are shown. https://doi.org/10.1371/journal.pntd.0012929.t003 Although the number of proteins found in the PN tissue was high, the abundance of these proteins was lower compared to other tissues where they were detected (S1 Table). The most abundant protein was laminin (OVOC10067) which was also detected in all other tissue types. It was more abundant in the intestine, body wall and uterus for PN and HC worms but was less abundant in the embryos than in the NP tissue. Among the 10 most abundant proteins in the PN tissue, only two proteins did not share substantial sequence conservation with the host and are nematode specific: a pepsin inhibitor (OVOC9984) and a calcium binding protein (OVOC8391). When PN and HC embryo tissues are compared, as they have similar areas cut from the worm slides (Fig 1), they share 745 proteins (85.7% of the total HC embryo tissue proteins). In terms of abundance, 296 proteins are more abundant in the PN tissue and 182 are only present in PN (Fig 4, yellow dots). PN and HC embryo tissues share 420 proteins (Fig 4, red dots). Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 4. Protein abundance correlation between PN and HC worm tissues. The relative abundance of each detected protein in HC embryo tissue were plotted against the relative abundance in the PN tissue. Yellow dots represent proteins that are more abundant in the PN tissue. Red dots are proteins that are only highly abundant in PN and HC embryo tissue when compared to all other HC tissues proteins. PN: pleomorphic neoplasm, HC: healthy control. https://doi.org/10.1371/journal.pntd.0012929.g004 Immunolocalization of the immunoglobulin-like protein OvDig-1 To confirm our proteomic results using an independent method, we decided to clone, express, and generate antibodies to the calcium binding protein OVOC8391 (OvDig-1). This large protein has a well-characterized ortholog in Caenorhabditis elegans (CeDig-1) in which this giant immunoglobulin-like protein (13,100 amino acids) is predicted to enable calcium ion binding activity [28]. CeDig-1 is expressed in several structures, including glutamate-like receptors, body wall musculature, head mesodermal cells, male sex myoblasts, and non-striated muscles. In HC O. volvulus worms the protein was commonly detected by proteomics in all analyzed tissue and in PN worms it was found only in the body wall and PN tissues (S1 Table). Immunolocalization confirmed these results, with OvDig-1 being highly expressed in intrauterine and pseudocoelomic PN, but distribution was linked to specific cell types (Fig 5). OvDig-1 was highly expressed in larger cells with high plasma content (Fig 5C-F). The protein was absent in the pseudocoelomic PN with a large number of small, dividing cells as indicated by a strong nuclear DNA stain (Fig 5E). Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 5. Immunhistolocalization of OvDig-1 (OVOC8391) in a PN female O. volvulus. A: Cross-section of a O. volvulus female with a few stretched Mf in the vagina, negative control using pre-immune serum. B: Same section as in A but stained (red) with an antiserum raised against OvDig-1. The protein is localized in the hypodermis, connective muscle tissue and in the neoplasm. Stretched Mf are also OvDig-1 positive (arrow). C: Cross-section of a female with PN cells in both uterus branches. The outer uterus wall, the connective tissue, muscle and PN cells are OvDig-1positive (red). D: PN in the pseudocoelom cavity with strong staining for OvDig-1. Both uterus branches are empty, constricted and not labeled. E: PN in the uterus strongly labeled for OvDig-1. PN in the pseudocoelomic cavity shows a more homogenous cell type with weak labeling. HC: healthy control, PN: pleomorphic neoplasm, i = intestine, mu= muscle, ut =uterus, scale bar= 50 μm. https://doi.org/10.1371/journal.pntd.0012929.g005 In female O. volvulus with intact embryogenesis and without PN, OvDig-1 was expressed in primary oocytes (Fig 6B), stretched microfilaria (Fig 6C), coiled microfilariae surrounded by eggshells and “pretzel” (Fig 6E). Also stained were connective tissue, outer uterus membrane and in the fibrillar zone of muscle cells (Fig 6). Red labelling for the protein was not observed in the rest of the muscle cells, the hypodermis or lateral chord. Download: PPT PowerPoint slide PNG larger image TIFF original image Fig 6. Immunohistolocalization of OvDig-1 (OVOC8391) in a healthy female O. volvulus. A: Non stained longitudinal section with primary oocytes in the ovary. B: Consecutive section as in A stained for OvDig-1 with positive primary oocytes (red). C: Longitudinal section of a uterus branch with stretched Mf that are strongly positive for OvDig-1. D: Cross-section of a female with different morulae stage embryos in the uterus. The outer uterus membrane is stained, some muscle cells and connective tissue. E: Cross-Section of a female with coiled (pretzel) stages in the uterus. More advanced coiled mf are OVOC8391 positive, while unmature mf are negative, like the morulae. co= coiled embryos i = intestine, lc = lateral chord, mo= morulae, mu= muscle, ov=oocytes, ut =uterus, scale bar= 20 μm. https://doi.org/10.1371/journal.pntd.0012929.g006 Wolbachia-derived proteins in healthy and pleomorphic neoplasm worms Wolbachia proteins were curated the same way O. volvulus proteins and all peptides that may have originated from the mammalian host or the parasite were excluded from the final protein list (Tables 4 and S2). Overall, relatively few Wolbachia proteins were detected in 2 of the 3 worms with PN. In the PN worms most Wolbachia-derived proteins were detected in the body wall (six proteins), followed by PN tissue (four proteins) and only one protein in the uterus. In the HC worms, eight Wolbachia proteins were detected in the body wall and seven in the embryos with an ATP-dependent chaperone (WP_025264048.1) found in both. Two Wolbachia proteins were found in the intestine but also in the embryo and body wall tissues. Body wall tissue from the PN and HC worms have all six proteins shared. Blast for all these hits match Rickettsiales bacteria and several arthropods (S2 Table). Download: PPT PowerPoint slide PNG larger image TIFF original image Table 4. A summary of Wolbachia derived proteins detected in the O. volvulus samples. Presence of Wolbachia-derived proteins in the O. volvulus samples. https://doi.org/10.1371/journal.pntd.0012929.t004 Discussion This study is the first proteomic analysis of O. volvulus worms with PN. PN in O. volvulus female worms were first reported by Duke et al. in 1990 and were more closely described in several subsequent studies [5,6,29]. PN can occur in the pseudocoelomic cavity and uterus branches where they can be easily differentiated in histological sections from normal and degenerated embryos by their distinct morphology (Fig 1). PN were also found in adult male parasites, but they are less common than in females [5]. Previous immunohistological studies of nodule sections indicated that PN are caused by filarial tissues and not by an invasion of human host cells. All eight filarial proteins examined by immunohistology previously were also detected in PN in our comprehensive LCM-LC/MS study (S1 Table). In addition, we selected the nematode-specific calcium binding protein OvDig-1 (OVOC8391), that has a well-characterized ortholog in C. elegans, for expression, generation of antibodies and immunolocalization and were able to localize this protein in a specific cell type within the PN. PN cells in O. volvulus adults appear first at the posterior end and spread from there to the anterior end. It was postulated that their origin could be in either ovarian or testis cells [5,6]. We studied only female worms, and ovaries are very short so that no cross-sections were found in PN worms that were suitable for proteomic analysis. However, we observed that 86% of the proteins found in HC embryo tissue (745 proteins) were also detected in PN tissue, while the HC embryo tissue area was only 7% smaller than the analyzed PN tissue area. Among these proteins, 296 were more abundant in PN and 420 proteins were abundant in both tissues (Fig 4, yellow and red dots). The 296 more abundant proteins in PN include mostly members of the protein synthesis pathways (ribosomes, proteins processing and RNA transport). Interestingly, these were the same pathways enriched for the proteins that are abundant in the embryo tissue. Regulation of protein expression is a fundamental biological process during metazoan development and enriched regulatory proteins indicate rapid cell division. Regulatory proteins including transcription factors and the components of various signaling pathways are well-known for their roles in orchestrating organism development through transcriptional, translational, or posttranslational control. KEGG pathways, InterPro domains and GO terms enrichment for all the proteins present in the PN tissue indicated a strong enrichment of proteins associated with protein production specially with results related to ribosome, spliceosome, RNA transport and chaperones and folding catalysis (Table 3). Processes involved in protein degradation as proteasome and ubiquitin proteins are also enriched in this tissue. The PI3K-Akt signaling pathway is one of the most significantly enriched pathways, represented by three proteins in the PN, of the 5 proteins inferred from the O. volvulus genome, is one of the enriched proteins found in the neoplasm. This signaling pathway is considered mammals as a master regulator for cancer and is responsible for the regulation of cell growth, motility, survival, metabolism, and angiogenesis [30]. Chromatin metabolism is frequently altered in cancer cells and facilitates cancer development and this affects mRNA transcription level, which involve histone-dependent chromatin access [31,32]. Alteration in histones were seen in the PN tissue, in terms of different protein detection; 12 histone proteins were found in the PN tissue while only six, four and one histone proteins were found in other tissues (HC embryo, intestine, body wall, respectively; S1 Table). These findings showed that PN is a dysregulated tissue. All tissues but PN have lower protein numbers and abundance reduced when compared with the HC worms (Fig 2). This is expected as the worms are in the process of dying as the PN tissues are expanding through the other, normal tissues. Overall, we identified a substantial number of different proteins in PN, protein abundance detection was generally lower (S1 Table). A long list of proteins was found in this study but some of them, were only found in PN tissue, and caught our attention as potential biomarkers (S1 Table). Peroxidasin homolog (OVOC4906) is one of the most abundant proteins in PN tissue. It is part of a family of heme-containing peroxidases that catalyze the oxidation of various substrates, primarily using reactive oxygen species (ROS). Studies in cell cultures indicate that this protein is essential for extracellular matrix assembly, as well as for survival and growth signaling [33]. Proteins in this family are associated with cancer progression through the regulation of metabolic and oxidative stress pathways, specifically by inhibiting oxidative stress, which leads to decreased apoptosis [34]. Secreted Protein Acidic and Rich in Cysteine (SPARC, OVOC4177) is a matricellular glycoprotein involved in a wide range of physiological and pathological conditions marked by significant remodeling and plasticity. In cancer, this protein plays diverse and context-dependent roles depending on the cancer type, cell of origin, and surrounding environment. In ovarian cancer it functions as a negative regulator by reducing macrophage recruitment and decreasing the associated inflammation [35]. In melanomas, SPARC expression has been reported to increase with tumor progression and its overexpression enhanced vascular leakiness, extravasation and metastasis [35–37]. Neural cell adhesion molecules such as neuronal glia related cell adhesion molecule (NrCAM, OVOC3931) play crucial roles in development and regeneration of the central nervous system. However, they are also associated with early tumor development of neuroblastoma [38]. Calsequestrin is a calcium binding protein that modulate muscle contraction. When overexpressed in breast cancer cells led to more aggressive phenotypes, disrupted intracellular signaling pathways, increased tumorigenesis, and remodeling of collagen structures [39]. To confirm our proteomic findings using an independent method, the localization of OvDig-1 in PN and HC worms was studied in more detail (Figs 5 and 6). The calcium binding protein CeDig-1 (OVOC8391 ortholog) is well-studied in C. elegans is a component of the basement membrane that facilitates specific interactions between cellular surfaces and their environment through its interaction with a cell-type-specific set of other maintenance factors [40]. In C. elegans the expression in muscles surrounding the gonads was described previously. The distinct labeling for OvDig-1 of a plasma rich cell type in PN is a new finding and its function is not clear. It also shows the different cell types that can give rise to PN tissue, since not all PN tissue is stained with the antibody (Fig 5C-E). This can also be seen on the HC worm (Fig 6) when we can see the two uterus branches have different developmental stages as stained mature microfilariae (Fig 6C), non-stained morula stages (Fig 6D), stained coiled microfilaria surrounded by eggshells and non-stained “pretzel” stage (Fig 6E) [41]. Calcification of adult O. volvulus related to drug treatment or old age has been frequently observed previously [4,6,29,42,43]. It can be hypothesized that OvDig-1 also plays a role in pathogenesis of worm calcification and PN are a risk factor for calcification. The combination of morphological and proteomic techniques can also lead to interesting and unexpected findings. A morphological difference between worms with PN and HC without is obvious, but a close comparison (see Figs 5 and 6) shows no distinct morphological difference between structure such as the intestine that are not directly affected by PN. However, our proteomic analysis shows a significant difference in protein composition in PN and HC worms also for those structures (Fig 3 and S1 Table). For example, in the intestine of worms with PN only 97 proteins were detected with 31 unique intestinal proteins, while in the intestine of HC worms 222 proteins were detected with 156 unique intestinal proteins. This suggests that PN affect the entire worm and causes a systemic dysregulation of protein expression and is not confined to PN. This phenomenon is similar to early tumor metastasis in mammals, where primary-tumor-driven systemic processes that occur before metastasis have been found to dictate the site where subsequently disseminated cancer cells extravasate into other tissues [44]. Wolbachia is the most widespread genus of endosymbiotic bacteria in the animal kingdom, infecting a diverse range of arthropods and nematodes. Unlike many insects, filarial nematodes require Wolbachia for survival [45]. In a female adult worm, Wolbachia can be observed in the hypodermis, lateral chord, oocytes, and embryos. However, distribution in the hypodermis and the lateral chord is often uneven and Wolbachia density can vary. Brattig et al. observed Wolbachia frequently in worms with PN but not directly in the PN cells [5]. In the present study we detected a small number of Wolbachia proteins directly in the PN cells (Table 4 and S1 Fig). These proteins included the chaperonin GroEL, a surface protein, an outer membrane protein and a chaperone DnaK. The first 100 BLAST hits for these proteins shows a high similarity with the closely related Rickettsiacae and some arthropods species (S2 Table). It is possible that these Rickettsiales sequences were mislabeled as arthropods/hosts sequences in the databases [46]. The authors (us) discuss the possibility that these Wolbachia proteins were remnants of HC tissue cells or even off cuts from the adjacent tissues, but spectra count and peptide levels between PN and HC worms were too similar to consider that Wolbachia are in the sample by chance (S2 Table). Overall, relatively few Wolbachia proteins were constantly detected in the PN worms and the heathy control, but LCM was not targeting Wolbachia and the specific distribution of the endobacteria was not known. One of the primary objectives of this study was to explore the potential of combining LCM with LC-MS/MS for tissue protein localization. This is particularly important for samples from parasites that cannot be dissected or maintained in animal models, such as O. volvulus, as it allows the use of frozen or fixed tissue samples to overcome these limitations. LCM techniques allow scientists to obtain isolated tissue samples, or even a single cell, effectively addressing the issue of tissue heterogeneity. However, the method has drawbacks, including limited sensitivity for detecting protein post-translational modifications and challenges in data quantification or normalization. The parasite tissue material is insufficient to allow BCA quantification, making it challenging to compare the different tissue samples. Although the sectioned areas from each tissue differ, this does not fully explain the variations in spectra count or total protein number observed in this study. For instance, as shown in Table 2, PN tissue contains twice the number of proteins and spectra counts as embryos, despite having similar cut areas. A similar pattern is observed with the spectral counts and protein numbers for gut and uterus tissues from PN and HC worms, although the spectral count numbers for PN uterus tissue are noticeably lower in comparison. Furthermore, PN tissue has ten times more proteins than body wall tissue, with only 2.6 times more area cut, and 80% more spectra counts numbers. We would like to emphasize that despite varying dissected areas among different tissues, the cut areas within the same tissue type are similar, which rules out that area size differences are a significant cause of the observed variations. Conclusions LCM combined with LC/MS is a powerful technology to assess the protein inventory of O. volvulus on a microscopic level. PN have a diverse proteome that is similar to the proteome of developing embryos in HC worms. The presence of PN is associated with a systemic dysregulation of protein expression also in organ systems that are not directly affected by neoplasms. Additionally, we identified proteins exclusively present in PN tissue that are associated with cancer in mammals. Supporting information S1 Table. Excel spreadsheet with the functional annotation, peptides, spectra count and NSAF values of the Onchocerca volvulus parasite-related-proteins and classification of the matched proteins by BLASTP searches against Homo sapiens and O. volvulus databases. https://doi.org/10.1371/journal.pntd.0012929.s001 (XLSX) S2 Table. Excel spreadsheet with the functional annotation, peptides, spectra count and NSAF values of the Onchocerca volvulus parasite-related-proteins and classification of the matched proteins by BLASTP searches against Wolbachia database. https://doi.org/10.1371/journal.pntd.0012929.s002 (XLSX) S3 Table. Primers for OVOC8391 amplification. Primers were designed for one of the low similarity with other parasites regions. https://doi.org/10.1371/journal.pntd.0012929.s003 (XLSX) S4 Table. Complete list of enriched gene ontology pathways, InterPro and KEGG domains. O. volvulus-derived proteins detected in the polymorphic neoplasm tissue, only proteins that were supported by at least two unique peptides in two biological replicates were used. Each tab represents one complete analysis. https://doi.org/10.1371/journal.pntd.0012929.s004 (XLSX) S1 Fig. Overview of H&E stained O. volvulus sections. Worm sections which were used for LCM and which material was dissected using an image which had been manually color-coded to help keeping track which tissue should be dissected with the laser. A is a pleomorphic neoplasm worm and C is a healthy female. B and D are examples of images used at the LCM. Ut= uterus, b= bodywall, neo= neoplasm, m= male mo=morulae, Orange= neoplasm or embryos, green= gut; yellow= body wall, periwinkle= uterus wall. https://doi.org/10.1371/journal.pntd.0012929.s005 (PDF) S2 Fig. Alignment for with filarial orthologues. The sequence for OVOC8391 and its filarial orthologues were retrieved from WormBase Parasite. MegAlign Pro from DNA Star was used to align the sequences using OVOC8391 as reference. https://doi.org/10.1371/journal.pntd.0012929.s006 (PDF) Acknowledgments The authors would like to express their gratitude to John Martin for his expert bioinformatics support during the proteomic analysis and Sarah Greene MD, PhD for valuable discussion during manuscript revision. The work at the Mass Spectrometry Technology Access Center at the McDonnell Genome Institute (MTAC@MGI) at Washington University School of Medicine, was supported in part by the Diabetes Research Center/NIH grant P30 DK020579, Institute of Clinical and Translational Sciences/NCATS CTSA award UL1 TR002345, and Siteman Cancer Center/NCI CCSG grant P30 CA091842.
TI - Spatial proteomics of Onchocerca volvulus with pleomorphic neoplasms shows local and systemic dysregulation of protein expression
JF - PLoS Neglected Tropical Diseases
DO - 10.1371/journal.pntd.0012929
DA - 2025-03-31
UR - https://www.deepdyve.com/lp/public-library-of-science-plos-journal/spatial-proteomics-of-onchocerca-volvulus-with-pleomorphic-neoplasms-CKo0HWiqZh
SP - e0012929
VL - 19
IS - 3
DP - DeepDyve
ER -