Stage-Specific Transcriptome and Proteome Analyses of the Filarial Parasite Onchocerca volvulus and Its Wolbachia Endosymbiont

ABSTRACT Onchocerciasis (river blindness) is a neglected tropical disease that has been successfully targeted by mass drug treatment programs in the Americas and small parts of Africa. Achieving the long-term goal of elimination of onchocerciasis, however, requires additional tools, including drugs, vaccines, and biomarkers of infection. Here, we describe the transcriptome and proteome profiles of the major vector and the human host stages (L1, L2, L3, molting L3, L4, adult male, and adult female) of Onchocerca volvulus along with the proteome of each parasitic stage and of its Wolbachia endosymbiont (wOv). In so doing, we have identified stage-specific pathways important to the parasite’s adaptation to its human host during its early development. Further, we generated a protein array that, when screened with well-characterized human samples, identified novel diagnostic biomarkers of O. volvulus infection and new potential vaccine candidates. This immunomic approach not only demonstrates the power of this postgenomic discovery platform but also provides additional tools for onchocerciasis control programs.

biont (wOv), a process made possible by having a high-quality assembly and annotation of the entire O. volvulus and wOv genomes (9). Systematic comparisons across the parasite stages and across related nematodes have offered insights into the unique biology of this longlived filarial parasite, while an immunomic approach has leveraged this information to allow the identification of potential vaccine and diagnostic candidates.

Transcriptome and proteome of O. volvulus.
Transcriptome data obtained using RNA-seq obtained from most of the vectorand human-derived stages (see Fig. S1 in the supplemental material) of the parasite were not only used to help curate the O. volvulus genome annotation (9) but were used in the present study to understand stage-specific mRNA expression. Over 75% of the genes had 100% transcript coverage in all of the stages, with the exception of the adult female worm (http://exon.niaid.nih.gov /transcriptome/O_volvulus/v245/Ov-v245-web.xlsx), most of which, in contrast to the adult male, is composed of uterine tissue containing all of the intrauterine stages that range from early embryos through pretzel-like immature mf to mature mf (10). Because many transcripts with expression levels of Ͻ1 RPKM (reads per kilobase of transcript per million mapped reads) were found to encode proteins whose products were validated in separate proteomic analyses, they also have been included in the analyses (https://exon .niaid.nih.gov/transcriptome/O_volvulus/Additional_file_1.xlsx).
The median coverage over the length of any predicted protein by peptides derived from the proteome of each of the stages profiled ranged between~10 and 15% (see Fig. S2 in the supplemental material). A total of 7,774 O. volvulus proteins were identified across all of the stages, resulting in the validation of Ͼ64% of the predicted proteins (9). There was maximal proteomic coverage during the development of L3 to L4; adult male and female worms had the second most extensive coverage (Fig. 1a). We were also able to identify/validate the presence 465 of the 785 Wolbachia (wOv) proteins predicted by the genomic sequencing of wOv (Fig. 1a and the entire wOv data set at http://exon.niaid.nih.gov/ transcriptome/O_volvulus/v245/wOv_web/wOv_Web.xlsx). Overall, there was a high degree of correlation (P Ͻ 0.0001) between the RPKM (for the transcriptome) and the normalized protein abundance (for the proteome) (see Fig. S2 in the supplemental material), with r values that ranged from 0.25 to 0.39, values considered acceptable for global comparisons (11,12). Multivariate analysis using closely related/overlapping larval stages as replicates (r ϭ Ͼ0.9; see Fig. S3 in the supplemental material) revealed stage-specific transcript profiles that clearly segregated the L2 and L3 vector-derived stages (Vec1, Vec2), the early human developmental stages mimicked in vitro (L3 to L4 molting; M1, M2), microfilarial stages (MF1, MF2), adult males (AM1, AM2), and adult females (AF1, AF2) ( Fig. 1b and c). Proteomic analyses ( Fig. 1d and e) indicate a bias toward male-like expression profiles as the parasites develop from L3 to L4, suggesting a slight selective advantage for the males during early development. As expected from the principal-component analyses, there was a clear stage-specific clustering of the transcriptional and proteomic expression profiles ( Fig. 1c and e). Although relatively few transcripts were expressed only in a single stage, except for the females (Fig. 2a), the adult males appeared to have the highest number of differentially expressed transcripts (Fig. 2b). Comparative analysis based on Caenorhabditis elegans germline expression data (13) indicated that a majority (69%) of the male-associated differential gene expression is related to spermatogenesis (see Fig. S3 in the supplemental material and https://exon.niaid .nih.gov/transcriptome/O_volvulus/Additional_file_2.xlsx). Inferences, however, from C. elegans also appear to classify 6% of the O. volvulus male-enriched transcripts as oogenic. Of note, the mf have significantly increased expression of two hypothetical proteins (OVOC1851 and OVOC1852) that are unique to O. volvulus, along with OVOC1611, a member of the serine protease inhibitor (SPI) family (see Fig. S3 in the supplemental material).
Amino acid repeats make up 12 to 14% of most of the proteomes studied to date (14). A total of 1,590 O. volvulus proteins were found to harbor single amino acid repeats, tandem re- vector-derived stages, molting larvae, adult males, and adult females. (b) Genes differentially expressed in adult males compared to all other stages plotted using log 2 -fold changes on the x axis and adjusted P values (log 10 ) on the y axis. (c, d) Venn diagrams illustrating the numbers of stage-specific proteins identified by mass spectrometry in the major somatic stages (c) and during the L3-to-L4 molt (d).
peats, or sequence repeat regions (https://exon.niaid.nih.gov/ transcriptome/O_volvulus/Additional_file_3.xlsx). In addition to the transcription factors and signaling proteins, uncharacterized proteins had the highest number of amino acid repeats in O. volvulus (https://exon.niaid.nih.gov/transcriptome/O_volvulus/ Additional_file_4.tif). Proteins containing polyglutamine repeats are common in most organisms, including O. volvulus (https: //exon.niaid.nih.gov/transcriptome/O_volvulus/Additional_file_ 4.tif) and other nematodes. While their exact role in O. volvulus is not clear, the expression of polyglutamine repeats was shown to be neurotoxic in both C. elegans and Drosophila and is dynamically regulated throughout the lifetime of an organism, suggesting a link to longevity (15). Interestingly, one of the lead O. volvulus vaccine candidates (16), Ov-RAL-2 (OVOC9988), has 11 glutamine residues in its amino terminus. These residues are probably not important for the protective ability of this protein, as the Brugia malayi homologue (Bm2001), that is also protective in animals, does not have this repeat (17).
Stage-specific functional enrichment. Although many of the putative proteins were classified into functional categories,~20% were considered to be uncharacterized and unique (or divergent) (9). Clustering of these unique genes on the basis of transcript and protein abundances indicates not only that distinct subsets are enriched at specific stages (https://exon.niaid.nih.gov/transcriptome/O _volvulus/Additional_file_5.tif) but also that 7% have signatures (SignalP and SecretomeP 2.0) indicative of secretion (secreteddivergent) (see the wOv data set at http://exon.niaid.nih.gov /transcriptome/O_volvulus/v245/wOv_web/wOv_Web.xlsx). Although the functions of these uncharacterized gene products remain largely unknown, identifying the stages in which they are highly expressed may provide a method to infer their functions on the basis of associations with particular developmental processes, such as molting or embryogenesis.
There were no significant differences in the total number of genes or proteins identified per functional category across the various stages (https://exon.niaid.nih.gov/transcriptome/O _volvulus/Additional_file_6.tif). However, genes involved in metabolism, cytoskeletal elements, protein modification, protein export, proteasome, and transcription machinery had significantly lower expression (adjusted P Ͻ 0.001, analysis of variance [ANOVA]) in adult females than in all of the other stages. Gene set enrichment analyses (GSEA) indicated that selected functional categories were enriched on the basis of the stage (see Fig. S4 in the supplemental material). The adult female stage was associated with pathways linked to detoxification and the extracellular matrix (see Fig. S4). This enriched subset of extracellular-matrixrelated genes was primarily composed of collagens and gene products associated with chitin synthesis machinery. Although the mf are an integral part of the fertile adult female, genes corresponding to NADH dehydrogenase activity (gene ontology [GO] category 0008137) and cytochrome c oxidase activity (GO category 0004129) were enriched in the adult females compared to all other stages. In contrast to the adult female worms, the mf stages showed significant enrichment of genes associated with protein synthesis (ribosomal proteins) and protein modification, with cyclophilins and chaperones (heat shock proteins) as the major contributors. These are likely part of the machinery required for cellular morphogenesis that occurs after the parasite is ingested by the blackfly vector. The expression of genes representing nuclear regulation was enriched in adult males, an enrichment that could be attributed to DNA replication and spermatogenesis. In addition, molecular functions of nucleotide binding (GO category 0000166), peptidase activity (GO category 0070011), and phosphoprotein phosphatase activity (GO category 0004721) were the top GO processes enriched in adult male O. volvulus.
Insights into development. Analyses of protein abundance in each of the stages identified 363 proteins expressed in all of the somatic stages (Fig. 2c). Genes encoding proteins involved in metabolism, cytoskeletal elements, and protein modification made up more than 50% of these core genes. Proteins common in OvEMB and OvAF are likely to play a role in oogenesis and/or embryogenesis. Similarly, proteins identified exclusively during the L3-to-L4 molt highlight the machinery required during this important first developmental molt after adaption to the human host environment (Fig. 2d). On the basis of C. elegans RNA interference data, O. volvulus homologues of C. elegans that are associated with embryonic lethality (EMB), larval arrest (LVA), larval lethality (LVL), defective molting (MLT), and lethality (LET) phenotypes were observed to be enriched not only in the embryos, mf (and thereby adult females), and L3 larval stages (https: //exon.niaid.nih.gov/transcriptome/O_volvulus/Additional_file_ 7.tif) but also in adult males. Similarly, O. volvulus encodes orthologues of the most critical genes essential for molting (on the basis of C. elegans; https://exon.niaid.nih.gov/transcriptome/O _volvulus/Additional_file_8.xlsx) during the L3-to-L4 transition.
The analyses also highlight other proteins that have been shown to be essential for molting and/or other developmental processes of filarial parasites. For example, embryogenesis and molting of filarial parasites are dependent on the activity of cathepsin L-like cysteine proteases (CPLs). During molting, CPLs stored in the glandular esophagus are released and help in the breakdown of the old cuticle and possibly support the synthesis of a new cuticle (18,19). Analyses using EnsemblCompara (20) indicate an expansion of CPL-like enzymes in O. volvulus (9). Significantly increased expression of CPL and CPZ molecules (21) was observed in L2 and L3 larvae compared to that in other stages (P Ͻ 0.002, ANOVA). While their role during L3-to-L4 molting is known (19,22,23), their increased expression in L2 larvae suggests that CPLs and CPZ also play a role in the L2-to-L3 molt in the vector, similar to what was described in B. malayi (24). Interestingly, the GO categories of GTPase activity (GO category 0003924), oxidoreductase activity (GO category 0016491), procollagen-proline dioxygenase activity (GO categories 0004656 and 0019798), peptidyl-proline dioxygenase activity (GO categories 0031543 and 0031545), and protein disulfide oxidoreductase activity (GO category 0015035) were among the categories represented by the differentially expressed genes during both the L2to-L3 and L3-to-L4 molting events. GSEA also identified an immunologically important class of molecules as being enriched in L3 larval stages (see Fig. S4 in the supplemental material) and a set of extracellular-matrix-related genes distinct from those extracellular matrix proteins enriched in adult female worms. Some of these include collagens that can be regulated by prolyl-4 hydroxylases, part of an expanded family of genes based on the analysis of the O. volvulus genome (9) that also shows differential expression across the various stages (https://exon .niaid.nih.gov/transcriptome/O_volvulus/Additional_file_9.tif).
Gene families-protein kinases, G protein-coupled receptors (GPCRs), and nuclear hormone receptors (NHRs). Approximately 55% of the predicted proteins had significant matches (Ͻ1E-05) to Pfam databases, with a protein kinase domain (Pkinase, PF00069), an RNA recognition motif (RRM_1, PF00076), and a seven-transmembrane receptor (7TM_1, PF00001) being the top three represented categories. The Pkinase domain forms the catalytic center of protein kinases that are involved in a wide variety of cellular processes influencing apoptosis, differentiation, and embryonic development. Their increased stage-specific increased expression (see Fig. S5 in the supplemental material) likely reflects their role in germline differentiation (in adult worms) or in larval developmental stages (mf and L3). Comparative analyses of the O. volvulus kinome (see Table S1 in the supplemental material) with other filarial and nonfilarial control kinomes resulted in the identification of O. volvulus kinases that are unique or belong to nematode-specific families (25) (see Table S2 in the supplemental material). More interestingly, the expression of casein kinase 1 and its subfamily the Tau tubulin kinase-like (CK1/ TTBKL) and FER subfamily of tyrosine kinases (TK/FER) was enriched in adult males (https://exon.niaid.nih.gov/transcriptome /O_volvulus/Additional_file_10.xlsx). Both CK1 and Fer kinases have been implicated in mammalian Wnt signaling. The nonreceptor tyrosine kinase of the Fer family was also shown to be important in spermatogenesis (26). These nematode-specific kinases-clearly distinct from human, fruit fly, and yeast kinasesmay provide insights into comparative worm biology studies and are very tractable drug targets.
Among the seven-transmembrane receptors, the GPCR gene families are greatly expanded in O. volvulus (9). Hierarchical clus-tering of the transcriptional and proteomic expression data reveals clear patterns of stage-specific expression (Fig. 3). Some members of the Srx family show very high expression in the skin and nodular mf, while other families show increased expression in adult males. These patterns reveal unique chemosensory requirements for certain stages of O. volvulus, including migratory adult males and skin mf, compared to the more stationary female worms.
Secretome and host-parasite interactions. Approximately 20% of the genes in the O. volvulus genome are predicted to be secreted through classical secretion mechanisms, and about~42% are predicted to be secreted through nonclassical mechanisms. All filarial worms are known to release excretory/secretory (E/S) products that are critical components in the helminth arsenal of proteins that perform diverse functions (33). These include (i) modulation of the host immune response, (ii) remodeling of host tissue, (iii) alteration of host tissue nutritional status, and (iv) enhancement of larval tissue migration (34). The O. volvulus genome contains many of these immunologically relevant genes (9), among which are the L3-enriched or mf-enriched cystatins that have been shown to interfere with antigen processing and presentation to T cells (35), small SPIs (Ov-SPI-1, Ov-SPI-2), and serpins (https:// exon.niaid.nih.gov/transcriptome/O_volvulus/Additional_file _12.tif). SPIs also play an important role in controlling the molting process (36) and immune evasion (37). The analysis of the O. volvulus genome revealed the presence of 10 additional SPIs beyond the 2 previously identified, Ov-SPI-1 and Ov-SPI-2 (9). Eleven of these 12 are highly expressed during the L3-to-L4 molt, consistent with their potential role in molting or immune invasion during the first days of establishment in the human host (https://exon.niaid.nih.gov/transcriptome/ O_volvulus/Additional_file_12.tif). Interestingly, 2 of the 12 SPIs are highly expressed in adult males, consistent with a previous study demonstrating SPIs being localized in sperm within the adult male testis (36). The other immunologically relevant protein enrichment was the O. volvulus adult male-enriched expression of indoleamine 2,3-dioxygenase, an interferon-inducible enzyme that suppresses adaptive T-cell immunity, and the enrichment of the homologue of suppressor of cytokine signaling 7 (SOCS7; OVOC678) in L3/L4 developmental stages.
Stage-specific expression of serine, aspartic, cysteine, and metalloproteases reflects their roles in host invasion, molting and migration in nematodes (https://exon.niaid.nih.gov/transcriptome/O _volvulus/Additional_file_13.tif). Scanning of the O. volvulus genome, notably, identified the presence of 13 chitinases, 8 of which belong to glycosyl hydrolase family 18. We know that the larva-specific chitinase (Ov-CHI-1) plays an important role in parasite transmission, molting, and remodeling of the L4 cuticle (38), the inhibition of which inhibits larval molting (38,39). The relatively higher expression of Ov-CHI-1 in the L2 larval stages suggests a probable role in migration through insect tissues (https://exon.niaid.nih.gov/transcriptome/O_volvulus/ Additional_file_13.tif). The increased expression of what had been thought to be larva-specific chitinases in the O. volvulus adult male-seen in both transcriptomic and proteome data-raises the possibility that these glycosyl hydrolase 18 family molecules have another functional role not previously recognized.
Wolbachia proteome. Wolbachia, a genus of Alphaproteobacteria, infects an estimated two-thirds of all arthropod species. The members of this genus are probably the most abundantly and vertically transmitted organisms that exist in facultative or obligate endosymbiotic associations (40). In nematodes, their prevalence is restricted to two families of worms, Onchocercidae and Pratylenchidae (41), with Onchocercidae exhibiting widespread obligate mutualism, as evidenced by retarded larval growth (42), embryostasis (43), and death (44) when the Wolbachia bacteria are eliminated. Because we performed RNA-seq analyses with poly(A)-enriched transcripts, wOv transcript identification was limited. Proteomic analyses, however, identified 465 wOv proteins, some of which were common to or uniquely expressed in specific stages of O. volvulus (Fig. 1a and Fig. 4a). Comparative analyses with the Wolbachia symbiont of the bovine parasite Onchocerca ochengi (wOo) (45) was limited because of a lack of corresponding stage-specific protein identifications. However, there was a significant correlation with the number of peptides of wolbachial origin identified from adult O. ochengi females (OoAF) and OvAF and OvEMB (see Fig. S6 in the supplemental material). Similar to what was observed in the proteomes of wBm (46) and wOo (45), the GroEL (wOv00466), outer membrane protein (wOv00621), outer surface protein (WSP, wOv00566), and molecular chaperone DnaK (wOv00687) were among the most commonly identified abundant proteins (top 25) across most of the stages. Clustering of the normalized protein abundance values resulted in the identification of wOv clusters that were specifically OvAF, OvEMB, OvL3D2, or OvL3D3 abundant (Fig. 4b). Among the clusters overrepresented in OvL3-to-L4 development (Fig. 4b,  right), many could be mapped to functional categories, with the top five functions being (i) translation, ribosomal structure, and biogenesis; (ii) posttranslational modification, protein turnover, and chaperone; (iii) energy production and conversion; (iv) coenzyme metabolism and cell envelope biogenesis, and (v) outer membrane proteins (Fig. 4c).
The increase in proteins related to particular functions in certain developmental stages may reflect multiplication of Wolbachia bacteria within these defined stages. In these stages, such as the developing L3 and in the adult female (and thus during embryogenesis), when the metabolic needs are the greatest, it is possible that the increased wOv replication seen may reflect the symbiotic needs of the O. volvulus parasite during growth and development (47).
Biomarkers-an immunomic approach. Using an immunomic approach (48-50), we profiled host antibody responses to a subset of parasite stage-specific proteins (397 proteins). These selected proteins (https://exon.niaid.nih.gov/transcriptome/O _volvulus/Additional_file_14.xlsx) were printed on a protein array, quality checked, and assessed for isotype-specific antibody responses (IgG1, IgG3, IgG4, and IgE) (Fig. 5b and c) by using serum samples from 52 individuals comprising O. volvulus-infected, putatively immune, and control individuals from Ecuador (51), Guatemala (52), and Cameroon (53). After normalization, clusters specific for IgG4, IgG3 and/or IgG1, and IgE reactivities were identified. Further analyses led to the identification of OVOC10819, OVOC5395, OVOC11598, OVOC12235, OVOC8619, and OVOC7083 as potential novel vaccine candidates (Fig. 5d) on the basis of significant IgG1 and/or IgG3 reactivity (with little to no IgE reactivity [54]) or potentially allergenic proteins that induce elevated levels of IgE (OVOC9414, OVOC7138). The identification of proteins that are highly expressed by the mf (see Table S3 in the supplemental material and https://exon.niaid.nih.gov/transcriptome/O_volvulus/ Additional_file_15.tif) and are specifically recognized by serum samples from protected individuals opens up new possibilities for the development of a safe antitransmission or therapeutic vaccine (55). Because IgG4 reactivity to filarial antigens provides better serodiagnostic species specificity (56,57) than other isotypes, the data were also analyzed for IgG4-reactive proteins. On the basis of the IgG4 responses (Fig. 5e), we were able to identify heretofore unrecognized biomarkers of active infection (e.g., OVOC10469,  OVOC10602, OVOC11950, OVOC3261, OVOC5127, OVOC8491, and OVOC9988). Indeed, we have successfully tested and validated OVOC10469 in a luciferase immunoprecipitation system (LIPS) immunoassay that highlights not only the utility of this protein as a biomarker of active O. volvulus infection (https: //exon.niaid.nih.gov/transcriptome/O_volvulus/Additional_file_16 .tif) but also the usefulness of this type of immunomic approach.

DISCUSSION
Using a combination of transcriptomic and proteomic data, we performed stage-specific analyses that are, to our knowledge, the most comprehensive for any of the filarial parasites. Because of the limited availability of O. volvulus larval-stagespecific material, we used the transcriptomes of the closely related/overlapping larval stages as biological replicates, since their expression patterns were significantly similar (see Fig. S3 in the supplemental material and https://exon.niaid.nih.gov /transcriptome/O_volvulus/Additional_file_17.xlsx).
Further, for the proteomic analyses, we chose depth of coverage over distribution of the limited stage-specific sample for technical replicates, with the rationale that although a difference in transcript (RNA) and protein recovery from the various stages is expected, normalized data (RPKM and spectral abundance) provide provisional evidence of the relative abundance of any particular gene or protein at a given stage. This data set thus provides an in-depth resource for understanding and analyzing the biological pathways that are critical for the development of the various stages of the parasite in the vector and human hosts, in host-worm interactions, and for the identification of novel biomarkers (e.g., OVOC10469, OVOC3261, and OVOC12838) and targets for interventions. Comparative stage-specific proteomic analyses of the recently available O. ochengi (58) proteome and the corresponding O. volvulus stages demonstrate a relatively high degree of correlation (https://exon.niaid.nih.gov/transcriptome/O_volvulus /Additional_file_18.tif), albeit a weaker one than that with the more distant species B. malayi (46) (https://exon.niaid.nih.gov /transcriptome/O_volvulus/Additional_file_19.tif) that is likely due to differences in coverage depth (58).
Natural immunity to O. volvulus can be acquired in affected populations by a few individuals who have been qualified as being putatively immune (52,59,60). The hypothesis is that they are protected because they are able to mount an appropriate immune response against the L3 larvae; surface proteins of L3 larvae and/or the E/S products released by molting larvae are considered to be an important source of such protective antigens (59). The identification of O. volvulus unique proteins that are larva-specific or the identification of adult and/or mf stage-specific biomarkers suggests that tools are already available to help achieve the lofty goal of onchocerciasis elimination in the coming decades.

MATERIALS AND METHODS
Detailed materials and methods are available in Text S1 in the supplemental material.
Parasite samples. Parasite material used for RNA-seq and proteomic analyses was collected at the research facility at the Tropical Research Station, Kumba, Cameroon (53), or in Ecuador (51). Adult worm samples were obtained from nodules excised during nodulectomies. L3 larvae were obtained from Simulium damnosum flies 7 to 8 days after infection with skin mf as described previously (61). Fresh L3 larvae were also cultured in vitro at 37°C for 1 to 3 days (L3D1, L3D2, and L3D3) or for 6 days when they molt and L4 were then collected. Nodular and skin mf were purified as described previously (62,63). Embryonic stages were purified from mf and eggs that were extruded into the medium during the process of cleaning female adult worms.
Human serum samples. All human serum samples were obtained by using protocols approved by the Institutional Review Board of either the National Institute of Allergy and Infectious Diseases (Guatemala or Ecuador) or by the New York Blood Center and the Tropical Research Station, Kumba, Cameroon. All infected people were microfilaria positive on the basis of skin snip analyses.
RNA-seq, assembly, and analyses. RNA-seq libraries were prepared from parasite stages in accordance with the RNA-seq protocols of the Illumina mRNA-Seq Sample Prep kit and the Illumina TruSeq kit. The reads were analyzed as previously described (64). Genes that had blast scores of Ͻ30% of the maximum possible score (self-blast) in other nematodes with E values of Ͼ1E-05 were considered unique. The normalized transcriptome data were analyzed in JMP Genomics (SAS Inc., Cary, NC) for general assessment of distribution analyses, correlations, principalcomponent analyses, ANOVA, hierarchical clustering and heat map generation, and parallel coordinate plots.
Protein isolation. For proteomic analyses, additional stages of embryos (OvEMB) and L3D2 (OvL3D2) and L4 (OvL4) larvae were also analyzed. Total soluble proteins were extracted from all of the stages, dialyzed, desalted, and digested with trypsin.
Liquid chromatography-tandem mass spectrometry. Liquid chromatography was performed with the Easy-nLC 1000 UHPLC system (Thermo). The spectra were searched by using a combined database of O. volvulus, wOv, and human proteins. Proteins were required to have one or more unique peptides across the samples analyzed with E values of Յ0.0001.
Protein arrays. The genes of interest were cloned into T7 expression vector pXT7. Proteins were expressed with a coupled in vitro transcription and translation system, the Escherichia coli-based cell-free Rapid Translation System 100 High Yield kit (5 Prime), as 10ϫHis-tagged or hemagglutinin (HA)-tagged fusion proteins and printed onto arrays. The arrays were probed with serum samples, and isotype reactivity was detected with Cy5-labeled antibodies. The slides were scanned, and reactivity was quantified. The quantified data were normalized and analyzed for significant proteins.
LIPS assay. For evaluation of antibody titers, a standard antibodybased LIPS assay (65) was performed.
Accession number(s). The entire O. volvulus data set analyzed is available as a hyperlinked Excel workbook at http://exon.niaid.nih.gov /transcriptome/O_volvulus/v245/Ov-v245-web.xlsx; the wOv data set is available as a hyperlinked Excel workbook at http://exon.niaid.nih.gov /transcriptome/O_volvulus/v245/wOv_web/wOv_Web.xlsx. The RNAseq data are available under BioProject PRJEB2965 (O. volvulus) under the accession numbers shown in the supplemental material. The mass spectrometry proteomic data have been deposited in the ProteomeXchange Consortium via the PRIDE partner repository under the data set identifier PXD003585.