Ethics statement

Ethical approval for this study was obtained from the Cameroonian National Ethics Committee for Health Research (N^o 2015/01/543/CE/CNERSH/SP) and administrative authorization was obtained from the Cameroonian Ministry of Public Health. A written consent was obtained from all participants employed in the study. For minors, the consent forms were validated by parents or guardians. Participation was entirely voluntary and individuals were free to opt out at their discretion without fear of community leaders or health practitioners. Ethical concerns working with rabbit are addressed by the GenScript Institutional Animal Care and Use Committee (IACUC), licence # SYXK(SU) 2008–0021. The IACUC is accredited by the AAALAC (approval date: 6/19/2009).

Onchocerca parasite material

O. volvulus and O. ochengi parasite material were collected at different stages of the parasites. O. volvulus nodules were obtained from patients in the Kombone Health area of the South-Western Region of Cameroon by a trained medical doctor and O. volvulus worms were obtained from these nodules as previously described [42]. Briefly, average nodular masses were digested in 0.5 mg/ml collagenase for 9 h at 37 ^oC shaking at 90 rpm after which the male and female worms were cultured in incomplete RPMI (Gibco, USA) supplemented with 0.25 mg/ml gentamycin and L-glutamine. Bloodless skin snips were also collected from the left and right knee of heavily infected individuals (microfilaria (Mf) load greater than 150 Mfs per mg of skin) and placed in incomplete culture medium for 12 h for Mf to emerge. The worms were pipetted and washed thrice with 20% percoll. L2 and L3 larval stages were obtained from infected and/or infective blackflies as described elsewhere [43]. Briefly, infected Simulium flies were captured and grown in the laboratory for seven days after which the thorax and head regions were excised to obtain L2s and L3s respectively.

Infected cattle skins containing O. ochengi nodules were obtained from the abattoir in Douala and immediately brought to the laboratory where O. ochengi worms were obtained from these nodules as previously described [44]. These infected cattle skins were selected following a search on available skins obtained after regular slaughtering and skinning procedures carried out at the abattoir.

Only motile worms were used for the extraction of RNA, protein crude extracts and ESPs. Protein crude extracts were obtained by crushing worm samples in lysis buffer (150 mM NaCl, 1.0% Triton X-100 and 50 mM Tris pH 8.0) supplemented with 1 mM Phenylmethylsulfonyl fluoride (PMSF). ESPs were obtained by culturing parasites in vitro using incomplete RPMI medium (Gibco, USA). After 16 h of culture of parasites (4 adult females per ml of RPMI medium, 50 adult males and 800 L3s per 300 μl of RPMI medium) in vitro, the medium was collected, quantified by Bradford assay and used for SDS-PAGE and western blot analyses.

Serum samples

Following examination of participants involved in the study by trained medical personnel, blood was collected from patients living in the endemic region of Kombone Health Area of the Mbonge Health district, South Western Region, Cameroon. O. volvulus is known to be specific in this region at the exclusion of other filarial infections. These patients (OVS) were selected on the basis of an established presence of Mf in skin snip biopsies and/or presence of clinical manifestation of onchocerciasis. The clinical and demographic data of these patients are indicated in S1 Table.

Sera were obtained from blood samples by employing an established protocol [45], diluted 1:2 in glycerol and stored at -20 ^oC. Sera obtained from individuals residing in an onchocerciasis hypo-endemic region (HES) of Huye, Rwanda, and from European subjects (ESC), were used as controls. Sera were also collected from individuals in Bandjoun (Bandjoun, Cameroon), a region which has been on constant ivermectin administration for over two decades. Sera were collected from individuals whose infection status had previously been well characterized [46]. Loa loa serum (LLS) samples were collected from the loiasis patients in the endemic regions of the Mvila division in the rain forest of the southern region of Cameroon in areas with high loiasis endemicity but <20% prevalence of onchocerciasis and with no ongoing CDTI programs. These subjects had been previously recruited for a study on the effects of albendazole on L. loa microfilariae [47]. These samples were obtained from the Centre for Research on Filariasis and other tropical diseases (CRFILMT), Yaounde, Cameroon. In summary, all the different sera types used have been extensively characterized in a related study [23]. Serum samples for subjects infected with other nematode infections including Brugia malayi, Wuchereria bancrofti, Ascaris lumbricoides and Mansonella perstans were obtained from the filarial repository, courtesy of the laboratory of Dr. Steven Williams.

Sequence analysis and phylogenetic analysis

The amino acid sequence analysis of OvGM2AP was done using Protparam [48] and the sequence alignment of OvGM2AP and homologous sequences from selected nematode species was done using PROMALS3D [49]. Structural prediction analysis was done using the Phyre2 online tool [50]. The phylogenetic tree of OvGM2AP sequences was generated by neighbor joining using PHYLIP 3.695.

Cloning, expression and purification of OvGM2AP

The full-length OvGM2AP cDNA (WormBase ID: OVOC1952) fused to a C-terminal 8x-His tag was cloned into pAC8_MF, a modified version of the pAC8_H transfer vector [51]. The resulting plasmid (pAC8_MF_OvGM2AP) was co-transfected in insect cells with AcMNPV viral DNA (Bac10:KO_1629,Δv-cath/chiA-LoxP:DsRed) linearized by Bsu36I to generate the recombinant baculovirus which was subsequently amplified and used to infect large scale cultures for protein production following standardized procedures [52]. Briefly, 1 μg of pAC8_MF_OvGM2AP mixed with 2 μg of linearized viral DNA in 750 μl of the buffer (25 mM HEPES, 140 mM NaCl, 125 mM CaCl₂, pH 7.1) was added dropwise to 750 μl of Grace’s medium supplemented with 10% of fetal bovine serum (FBS) to obtain a calcium-phosphate precipitate which was allowed to form during 15 min. The precipitate was layered onto 2 x 10⁶ Sf9 cells (Novagen) grown in Grace’s Medium (G8142-SIGMA) previously seeded in a 25 cm² flask. After 4 h incubation at 27°C, the medium was changed and cells further incubated for 5 days at 27°C. The resulting culture supernatant constitutes the initial virus stock (V₀), which was used to obtain the first (V₁) and second (V₂) amplification.

For large scale OvGM2AP production, Sf21 insect cells (IGBMC, Strasbourg) cultivated in the serum free Sf-900 II (Gibco) medium were infected at a density of 1.0x10⁶ cells/ml and MOI of 5 with the V₂ stock. After 3 days of infection, cells were pelleted by centrifugation (1000 x g for 10 min) and the culture medium was incubated with Ni Sepharose 6Fast Flow (GE Healthcare) at 4 ^oC for 2 h using 1 mL of resin for 1 L of culture. Following three washes using a 35 mM Imidazole solution in 10% glycerol, 20 mM HEPES pH 7.0, 300 mM NaCl, 5 mM 2-mercaptoethanol (Buffer A) and, bound proteins were eluted with the same buffer in the presence of 500 mM Imidazole. Eluted proteins were dialyzed against Buffer A and snap frozen for storage at -80°C.

Purification of OvGM2AP_8His contained in the clarified lysate was also tested. Cell pellets were re-suspended in a buffer containing 5 mM Imidazole, 10% glycerol, 20 mM HEPES pH 7.0, 300 mM NaCl, 5 mM 2-mercaptoethanol and cells were lysed by sonication at 4°C, 40% Amplitude for 10 sec. The lysate was centrifuged at 20000 g for 30 min and incubated with Ni Sepharose 6Fast Flow (GE Healthcare).

Western blot analysis of the poly-histidine tag from recombinant OvGM2AP_8His was performed using the mouse anti-His antibody 1D11, (IGBMC antibody facility, Strasbourg, France) for primary detection, the donkey F(ab')2 anti-mouse IgG (H+L) conjugated with Horse Raddish Peroxidase (HRP) as secondary antibody (Interchim SA) and the Super Signal West Pico Chemiluminescent substrate for detection of HRP (Thermo Fisher scientific). Chemiluminescence was detected using the Amersham imager 600 QC (GE Healthcare, Sweden).

Enzymatic deglycosylation of OvGM2AP and mass spectrometry

For glycosylation analysis, purified OvGM2AP was treated with Peptide:N-glycosidase F (PNGase F) (Promega, USA) using a protein/glycosidase ratio of 19/1 (w/w) for 2 h at 37°C. Deglycosylated and undigested OvGM2AP proteins were resolved by SDS PAGE followed by either Coomassie staining or western blot to analyze a band displacement. For mass spectrometry analysis, the proteins were digested by trypsin (after reduction with 10 mM DTT for 1 h at 57°C and alkylation for 45 min in the dark with 55 mM iodoacetamide). Tryptic digests were analyzed using an Ultimate 3000 nano-RSLC (Thermo Scientific, San Jose, California) coupled with a linear trap Quadrupole (LTQ)-Orbitrap ELITE mass spectrometer via a nano-electrospray ionization source (Thermo Scientific). Peptide mixtures were loaded on a C18 Acclaim PepMap100 trap-column (75 μm IDx 2cm, 3μm, 100 Å, Thermo Fisher Scientific) equilibrated with 3% acetonitrile and 0.1% formic acid in H₂O, and then separated on a C18 Accucore nano-column (75 mm internal diameter (ID)x50 cm, 2.6 mm, 150 Å, Thermo Fisher Scientific) with a 120 min linear gradient from 3 to 80% acetonitrile in the same buffer. Peptides were analyzed by Top 20-CID (collision induced dissociation) data-dependent MS. Spectra were processed with Proteome Discover 2.1 against a protein sequence database for Sf21 cells which include recombinant OvGM2AP. The minimum peptide length required was set to six residues and a minimum of two peptides were required to consider a protein as identified. The protein identification list was filtered at a False Discovery Rate below 1%.

Peptide synthesis and generation of Polyclonal antibodies (GenScript, USA)

A synthetic peptide was generated following bioinformatics analysis of the OvGM2AP protein using OptimumAntigen design tool. The best peptide was selected on the basis of low host homology and high Antigenicity/Surface/Hydrophilicity index. The synthetic peptide corresponded to amino acid positions 62–72 of the full-length protein with the following sequence from N to C terminal ‘SSKSDGVKFTAEKS’. This synthetic peptide was conjugated to Keyhole Limpet Hemocyanin (KLH) and used to immunize rabbits (GenScript, USA); after a secondary booster, immune serum was collected from rabbits and the antibodies were further purified from the immune serum by affinity chromatography using the synthetic peptide. Approximately 3.5 mg of antibodies was supplied with a titre greater than 1:64,000, affinity purified from 2 rabbits.

Stage specific expression of OvGM2AP

Total RNA was extracted from the different O. volvulus parasite stages using the Ambion RecoverAll Total Nucleic Acid Isolation kit (Thermo Fisher Scientific, Belgium). The RNA was reverse-transcribed to cDNA using the iScript cDNA synthesis kit (BioRad, USA). All primers were purchased from Integrated DNA Technologies (IDT, Belgium). Primers were designed targeting OvGM2AP cDNA and amplifying a 165 bp product (Forward primer 5’ GCCGAACAGCTCTGGAATTTG 3’, Reverse primer 5’ TCGGTGACATGCGATCAGAC 3’) and these primers were used to amplify the OvGM2AP cDNA sequence from the parasite total cDNA as well as the genomic DNA. The designed primer set spans exon 5 to exon 6 (S1 Fig). O. volvulus Glutaraldehyde-3-Phosphate Dehydrogenase (OvGAPDH) was used as a reference gene and a 192 bp fragment was also amplified from cDNA and genomic DNA preparations using the following primers; Forward primer 5’ GAAGGGTGGCGCTAAGAAAG 3’; Reverse primer 5’ GTTGTTGCATGTACGGTGGT 3’. Amplification was carried out using the SensoQuest thermocycler (Göttingen, Germany) under the following conditions: 94 ^oC for 2 min, 1 cycle; 94 ^oC for 10 s, 55 ^oC for 10 s and 72 ^oC for 1 min, 40 cycles; 72 ^oC for 10 min. The reaction mix was prepared using the ReadyMix REDTaq PCR reaction mix (Sigma) according to manufacturer’s protocol. The PCR products were run on a 1% agarose gel stained with SYBR safe and visualized under UV light using the Gel Doc XR+ (BioRad, USA).

A western blot analysis of ESPs of different parasite stages was performed using anti-OvGM2AP peptide antibodies (GenScript, USA). Approximately 10 μg of proteins were loaded onto SDS-PAGE gels and run at 200 V. The proteins were transferred to nitrocellulose membranes and blocked with 10% skimmed milk (Régilait, France) overnight at 4 ^oC followed by incubation with primary antibody at a dilution of 1:2000 for 1 h. After three changes of washing buffer (PBS + 0.05% Tween 20) at 5 min interval each, goat anti-rabbit IgG (whole molecule) HRP conjugate (Sigma) was used as secondary antibody and incubated at a dilution of 1:5000 for 45 min at 37 ^oC. The membranes were again washed thrice with wash buffer, revealed by chemiluminescence using the ECL substrate and visualized using a C-Digit chemiluminescence scanner (LI-COR, USA).

Serological characterization of OvGM2AP

IgG response to OvGM2AP was investigated by western blot and indirect ELISA using patient and control sera. For ELISA, optimal antigen concentration was determined by the checkerboard titration method. Maxisorp 96 well microtiter plates (Nunc, Denmark) were coated with 2 μg/ml of purified OvGM2AP overnight at 4 ^oC. Plates were washed thrice with wash buffer, 5 min between each wash and blocked with 10 mg/ml of Bovine Serum Albumin (BSA) for 1 h 30 min at 37 ^oC. The plates were washed as above and incubated with the various serum samples as primary antibody at a dilution of 1:2000 for 1 h at 37 ^oC. Following three rounds of washes at 5 min interval each, the plates were incubated with goat anti-human IgG (Fc Specific) peroxidase conjugate (Sigma) as secondary antibody at a dilution of 1:5000 for 1 h at 37 ^oC. After a final wash, 3,3’,5,5’ tetramethylbenzidine (TMB) was added as substrate for 10 min at 37 ^oC. The reactions were stopped with 2M sulfuric acid after which Optical Densities (OD) were read at 450 nm using the iMark microplate reader (BIORAD, USA). All washes and antibody dilutions were done in wash buffer (PBS+0.05%Tween-20).

IgG subclass responses were analyzed as described above with the exception of mouse anti-human IgG subclasses (Sigma) as secondary antibody and an additional incubation step of rabbit anti-mouse IgG peroxidase conjugate (Sigma) as tertiary antibody.

For western blot, approximately 700 ng of purified OvGM2AP was transferred from SDS-PAGE gels to nitrocellulose membranes. The rest of the experimental procedures were carried as described for ELISA above except for the use of 10% skimmed milk (Régilait, France) for blocking and antibody dilution, the use of the Enhanced chemiluminescent (ECL) substrate and revelation using the chemiluminescence scanner.

GM2AP activity assay

For the assessment of GM2AP activity of OvGM2AP, preparation of human glycosylated recombinant GM2AP with hexahistidine-tag was done as previously described [53]. The enzyme Hex A was purified from human placenta as described for the purification of sphingomyelinase [54]. [¹⁴C] GM2 was synthesized from its corresponding lyso-lipid, following published procedures [55]. This radiolabeled [¹⁴C] GM2 was incorporated into large unilamellar vesicles as described before [53,56]. Liposomes contained 20 mol% bis(monoacylglycero)phosphate (BMP) (Sigma, Germany), 5 mol% cholesterol (Sigma, Germany), 1 mol% [¹⁴C] GM2 and 1,2-Dioleoyl-sn-glycerol-3-phosphocholine (DOPC) (Avanti polar lipids, Alabaster, USA) as a host lipid in 20 mM sodium citrate buffer, pH 4.2. Total lipid concentration was measured to be 50 mM.

The actual liposomal activity assay of GM2AP with Hex A was carried out as previously described [53]. Briefly, 40 μl of liposome dispersion was mixed with 6 mU β hexosaminidase A and for BMP containing vesicles with 4 μg OvGM2AP or 4 μg recombinant human glycosylated GM2AP with a hexahistidine tag made up to 80 μl with 20 mM sodium citrate buffer pH 4.2. The samples were incubated at 37°C for 30 min. Afterwards, the assay was put on ice and stopped by the addition of 20 μl chloroform/ methanol (1/1, v/v). Quantification of the generated [¹⁴C] GM3 from [¹⁴C] GM2 was done by thin layer chromatography. The preparations were dried under a stream of nitrogen, re-dissolved with 20 μl chloroform/methanol (1/1, v/v), vortexed and sonified for 15 min after which the solution of lipids was applied to a high-performance thin-layer chromatography plate (Merck, Darmstadt, Germany). Lipids were separated in chloroform/methanol/0.22% CaCl₂ (55/45/10, v/v/v). Radioactive bands were visualized with a Bio Imaging Analyzer 1000 (Fuji, Japan), and the quantification was performed with the image analysis software ‘‘Tina” (Raytest, Staubenhardt, Germany).

MUG and MUGS competitive inhibition assay

To assess if OvGM2AP has functional domains analogous to the human orthologue, we employed the use of the artificial fluorogenic substrate 4-methylumbelliferyl-2-acetamido-2-deoxy-β-D-glucopyranoside (MUG) (Sigma, Germany) and its sulfated derivative 4-methylumbelliferyl-2-acetamido-2-deoxy-6-sulfo-β-D-glucopyranoside (MUGS) (Sigma, Germany) in competitive inhibition with OvGM2AP. The assays were performed as previously described [57]. Briefly, a 40 μl reaction mix was constituted for each assay set comprising of 10 mM citrate buffer, pH 4.2, 2 mM MUG or 2 x 10⁻⁵ M MUGS, 0.250 μg human β-hexosaminidase A (R&D systems, USA), 6 μg BSA (Sigma, Germany) and varying amounts of OvGM2AP. The reaction was incubated for 30 min at 37 ^oC after which the fluorescence of the released 4-umbelliferone was determined following excitation and emission at 320 nm and 430 nm respectively using a Tecan scanner (Infinite F200 Pro, Austria).

Statistical analyses

Normality of distributions was assessed using a Shapiro-Wilk test. Normally distributed data are expressed as mean +/- standard deviation and were compared using parametric tests. For non-Gaussian distributions, data are expressed as median with interquartile ranges and were compared using non-parametric tests. Comparisons of more than two groups were made using a one-way analysis of variance (ANOVA) or a Kruskal-Wallis test (with Dunn’s or Tukey’s correction for multiple comparisons) for independent groups as appropriate. The discriminatory performance of total IgG, IgG1, IgG2, IgG3 and IgG4 was assessed using receiver operating curve (ROC) analyses. Areas under the ROC curves (AUCs) were evaluated using the trapezoid method. Standard errors of AUCs were calculated as previously described [58]. Exact confidence intervals for the AUCs were determined using a binomial approach. A p-value < 0.05 was considered statistically significant. Calculations were performed using SigmaPlot for Windows, version 12.5 (Systat Software Inc., Chicago, IL, USA). Diagnostic sensitivity and specificity as well as other diagnostic accuracy parameters were calculated as previously described [59]. Scatter plots were generated using Graph Pad Prism 5.0 (La Jolla, CA, USA).

Accession numbers

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

https://doi.org/10.1371/journal.pntd.0007591.t001

Structural features of OvGM2AP, a putative Onchocerca volvulus GM2AP activator protein

The complete cDNA sequence coding for OvGM2AP was retrieved from NCBI non-redundant database by blasting the reverse complementary of the partial sequence referred to as OvL3.C1 [60]. The full-length protein is assigned a WormBase ID of OVOC1952. It contains 259 amino acids comprising 10 cysteine residues, 6 of which occupy the same positions in other species such as Homo sapiens, Loa loa, Trichinella pseudospiralis, and the free-living Caenorhabditis elegans (Fig 1A). INTERPRO analysis of this protein revealed that it belongs to the superfamily of the GM2 activator protein and has a lipid binding domain. SUPERFAMILY, a member database of InterPro which is a library of profile hidden Markov models representing all proteins of known structure, was used to construct the entry. This same analysis holds true for orthologues of the protein in other nematode species and remains uncharacterized in all these species, except for T. pseudospiralis. Gene Ontology suggest it may be involved in biological processes such as nematode larval development, body morphogenesis, molting cycle, growth and locomotion. The degree of relatedness of OvGM2AP with orthologues in other nematodes depicts a conservation of the protein across species and further supports O. ochengi as the closest relative of O. volvulus (Fig 1B). The protein sequence was found to contain a signal peptide (amino acids 1–27 for OvGM2AP and 1–23 for human GM2AP) in most orthologues. The amino acid sequences corresponding to the β-hexosaminidase A binding domain for the human GM2AP as well as the putative β-hexosaminidase A binding domain for OvGM2AP are illustrated in Fig 2A. Phyre2, a structure prediction online tool not affiliated to the INTERPRO consortium, equally predicted OvGM2AP as a GM2AP with greater than 90% confidence. The complete list of hits obtained from the Phyre2 blast have been compiled in S2 Table. A model of the structure of the human GM2AP template is indicated in Fig 2B

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Sequence analysis of OvGM2AP.

OvGM2AP protein sequences of the various nematodes were retrieved from WormBase and aligned using PROMALS3D (A). Ov = Onchocerca volvulus (OVOC1952), Ts = Trichinella pseudospiralis (T4C_8078.1) 16%, Ll = Loa loa (EN70_10408) 72%, Ce = Caenorhabditis elegans (C34E7.4) 65%. Sequence colour representation; red = predicted alpha helices, blue = predicted beta strands; Consensus Structure (ss) symbols: e = beta strand, h = alpha helix. Consensus amino acid symbols: conserved amino acids = bold and uppercase letters; aliphatic (I, V, L): l; aromatic (Y, H, W, F): @; hydrophobic (W, F, Y, M, L, I, V, A, C, T, H): h; alcohol (S, T): o; polar residues (D, E, H, K, N, Q, R, S, T): p; tiny (A, G, C, S): t; small (A, G, C, S, V, N, D, T, P): s; bulky residues (E, F, I, K, L, M, Q, R, W, Y): b; positively charged (K, R, H): +; negatively charged (D, E): -; charged (D, E, K, R, H): c. Bold residues in the consensus sequence represent greater than 80% consensus. Numbers in the first row represent a level of conservation above 4. Schematic representation of the structure features of OvGM2AP. Phylogenetic analysis of OvGM2AP orthologues (B) of closely related nematodes: Onchocerca volvulus (OVOC1952), O. ochengi (nOo.2.0.1.g11554) 99%, Dirofilaria immitis (nDi.2.2.2.g07380) 78%, Brugia malayi (Bm2577) 72%, Wuchereria bancrofti (J9F8Z0) 73%, Loa loa (EN70_10408) 72%, Ascaris suum (ASU_07674) 66%, Toxocara canis (TCNE_0001193501) 69%, Caenorhabditis elegans (C34E7.4) 65% and A. lumbricoides (ALUE_0002125701) 65%. The Jones-Taylor-Thornton model was used to estimate protein distances for neighbor joining. Bootstrap values are indicated on the nodes.

https://doi.org/10.1371/journal.pntd.0007591.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Structural annotations of the GM2AP.

Representation of human GM2AP domain organization (A(i)). Predicted lipid binding domain and disulfide bond positions for OvGM2AP are indicated as well (A(ii)). Disulphide bonds were predicted using the DISULFIND online prediction tool. 3D structure of human GM2AP (PDB access code 1G13) (B).

https://doi.org/10.1371/journal.pntd.0007591.g002

In terms of amino acid composition of the protein, serine was the most abundant, with 23 residues (8.9%). Other amino acids with structural implications present in the protein include Glycine (7.3%) and Proline (5.8%).

Transcriptional and translational expression patterns of OvGM2AP gene

In order to determine the timing of OvGM2AP transcriptional expression during the various life cycle stages of the parasite, total RNA was purified from the larval stage 1 (L1) or Mf, stage 2 (L2), stage 3 (L3), adult male (AM) as well as adult female (AF) stages and analyzed by RT-PCR using primers targeting OvGM2AP exon 5 (forward primer) and exon 6 (reverse primer) and generating a 165 bp product. Results obtained reveal that OvGM2AP mRNA is present in L1, L2, L3, adult male (AM) and adult female (AF) stages of O. volvulus (Fig 3A). To confirm that the detected bands of interest were not resulting from DNA contamination of the samples, the same primer pairs were also used to analyze O. volvulus genomic DNA (gDNA). As shown in Fig 3A (right panel), our primers detected signals with higher sizes, consistent with the genomic position of the targeted exons separated by intron 5 (S1 Fig). The size difference between RNA and gDNA samples was also visible using O. volvulus GAPDH primers as control (Fig 3A, lower panels), again consistent with the genomic positions of the targeted exons. The primer set designed to amplify OvGM2AP in O. volvulus samples was also successfully used to amplify the gene in O. ochengi species with exact precision as in O. volvulus further supporting the genetic closeness between the two species. In conclusion, the presence of the transcript in all the analyzed parasite stages suggests its permanent expression, which required confirmation at the translation level. In order to address this translational expression experimentally and since in silico analysis of OvGM2AP protein sequence suggests that it is a secreted protein, L3, AM and AF O. volvulus worms were cultured in-vitro for 16 h and their corresponding ESPs were analyzed by western blotting using an anti-OvGM2AP peptide antibody. OvGM2AP was detected in 16 h in-vitro ESPs of third stage larvae (L3), AM and AF with the anti-OvGM2AP peptide antibodies but not with the pre-immune serum (Fig 3B). This suggest that OvGM2AP is expressed and secreted by the parasite in its host at the tested life cycle stages and likely at all the different stages as revealed by RT-PCR analysis above.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Expression profile of OvGM2AP.

Transcriptional expression of OvGM2AP and OvGAPDH was evaluated from L1, L2, L3, adult male (AM) and adult female (AF) parasite stages by RT-PCR (A). PCR products were run on 1% agarose gels. Primers were designed for the specific amplification of 165 bp sequences of OvGM2AP and 192 bp sequences of OvGAPDH. OvGAPDH was used as a normalization control. OvGM2AP was also amplified from genomic DNA (gDNA) of the various parasite stages as control. Arrow indicates 957 bp target amplicon from gDNA while arrow head indicates a possible non-specific PCR product of approximately 600 bp. Mr = Promega 1Kb DNA ladder, C = No Template Control. OvGM2AP was also detected in 16 h ESP by western blot analysis using anti-OvGM2AP peptide antibody as primary antibody (B). Arrows indicate signals of OvGM2AP indicating the 28.8 KDa protein and possible oligomeric forms at approximately 85 and 110 KDa. The recombinant OvGM2AP expressed in Sf21 cells was analyzed by western blot after purification from IMAC Ni-NTA resin with 500 mM Imidazole (C). The flow-through (FT), the two wash fractions (W1 and W2) and the eluate (E) were analyzed by electrophoresis on a coomassie blue stained polyacrylamide gel (iii). Recombinant OvGM2AP was detected using serum from a rabbit immunized against an OvGM2AP peptide (anti-OvGM2AP) or a rabbit pre-immune serum (PI) (i). Sf21 cells, un-infected (lane 1) or infected by the recombinant virus expressing OvGM2AP were harvested 24 or 48 hours (lanes 2 and 3) after infection. The corresponding lysates were analysed by western blot using a mouse anti-His monoclonal antibody (1D11) (ii). Total IgG responses to the purified recombinant protein was analysed (D) using the indicated serum types (ESC = European serum control, HES = Hypoendemic serum and OVS = O. volvulus serum) as primary antibodies by western blot. Arrow heads indicate positions of OvGM2AP around 37 kDa and possible oligomers of around 85 and 110 kDa. A 12% SDS-PAGE gel was prepared in every instance of gel preparation using either a 40% acrylamide stock (C) or a 30% acrylamide stock (B and D).

https://doi.org/10.1371/journal.pntd.0007591.g003

Mass expression and purification of recombinant OvGM2AP from insect cells

A constant and reliable source of recombinant antigen is required for in depth analysis of the OvGM2AP antigen. Attempts to produce soluble recombinant OvGM2AP in E. coli were only partially successful as mass spectrometry analysis of the purified protein showed the dominance of chaperones (S2 Fig), we therefore switched to the insect cell baculovirus expression system. A recombinant virus encoding the OvGM2AP in fusion with a C-terminal 8His-tag under the control of the late pH promoter was generated by homologous recombination and was tested for its capacity to produce soluble OvGM2AP. This led to the production of the full-length protein including the TEV cleavage site and 8x His expected with a MW of 32 KDa (Fig 3C(iii)). Western blot analysis of the clarified lysate from cells infected by the recombinant virus detected a protein of approximately 35 kDa that reacted with the serum of a rabbit immunized against an OvGM2AP peptide but not with the pre-immune serum (Fig 3C(i)).

As OvGM2AP is a secreted protein with a putative classical signal peptide, IMAC affinity purification of the histidine-tagged recombinant protein was assessed from cell lysates as well as from the culture medium. The latter purification strategy was successful and purification of OvGM2AP_8His from the culture medium using a single Ni-NTA affinity step yielded a pure protein (Fig 3C) with a concentration close to 1 mg per liter of culture. Mass spectrometry confirmed the identity of the purified protein with 23 unique peptides detected and a sequence coverage of 78.88% (S3 Table). We could not detect any peptide corresponding to the N-terminal signal sequence of OvGM2AP as identified by sequence analysis. This suggests the peptide was cleaved and could constitute the signal peptide sequence of the protein.

We next investigated the antigenic potential of the insect cell expressed and purified recombinant OvGM2AP. We first asked if the protein, new for analysis in Onchocerca, could distinguish between onchocerciasis patients and onchocerciasis non-infected individuals. When a 14 sera pool of O. volvulus Serum (OVS) and Hypoendemic Serum (HES) control, as well as a 3 sera pool of European serum control (ESC) were employed for western blot analysis of 700 ng of purified OvGM2AP, OvGM2AP was found to react specifically with patient sera (OVS) and not with HES and ESC controls (Fig 3D). The presence of other bands of higher molecular weight suggests a possible polymerization of the native protein, as also observed with the in vitro ESP analysis above (Fig 3B).

The asparagine at position 173 of the recombinant OvGM2AP is glycosylated

The Homo sapiens GM2AP is known to be glycosylated at asparagine 63 [61]. There is experimental evidence of glycosylation of the C. elegans orthologous protein at asparagine 85 [62]. It therefore appears that the glycosylated asparagine amino acid residue differs across species and also between different members of the family. We herein analyzed the recombinant OvGM2AP produced in insect cell for the possible presence of glycosylation. To achieve this, Peptide:N-glycosidase F (PNGase F) was used to digest the recombinant OvGM2AP and the digestion product was resolved by SDS-PAGE. As shown (see arrow head) in Fig 4A(i), a band shift was observed between the digested and undigested purified protein. This band shift was confirmed by western blot analysis using the anti-OvGM2AP peptide antibody (Fig 4A(ii)). The band corresponding to the enzyme used in the digestion (see arrow, Fig 4A(i)) failed to show a signal by western blot, further testifying to the specificity of our anti-OvGM2AP peptide antibody.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. PNGase F Digestion analysis of OvGM2AP.

OvGM2AP was subjected to PNGase F digestion and analyzed on Coomassie blue stained SDS-PAGE gel and on nitrocellulose membranes by western blot (A). Arrow represents PNGase on coomassie staining which could not be detected using anti-OvGM2AP peptide antibodies on western blot. Protein exist predominantly in glycosylated form in the undigested fraction with a relatively small part of it unglycosylated (arrow head). MS analysis of the corresponding peptides revealed the presence of a deamidation at asparagine 173 in the deglycosylated (digested) protein but not in the glycosylated (undigested) protein (B); red circles indicate point of deamidation.

https://doi.org/10.1371/journal.pntd.0007591.g004

To verify if the shift was associated with enzymatic deglycosylation, mass spectrometry analysis was performed on peptides obtained from trypsin digestion of PNGase F treated and untreated OvGM2AP samples. The asparagine at position 173 in the native recombinant OvGM2AP was found to be deamidated to aspartate in the digested protein (Fig 4B), a consequence of PNGase F digestion of N-linked glycans from glycoproteins. Two different glycosylation sites were observed by mass spectrometry with 100% and 86% confidence corresponding to the peptide “NISLRICLPTK” and peptide “NLEPGKYKNISLR” respectively. This PNGase F treatment therefore suggests that the purified recombinant OvGM2AP is glycosylated predominantly at Asn 173 but with the possibility of a second glycosylation site at Asn 165. The ion exchange chromatogram corresponding to the deamidation peaks of the identified glycosylated peptides is provided in S3 Fig. Altogether, this supports the glycosylation of the protein at asparagine 173 following the pattern N-X-S where X is any amino acid except proline.

The humoral immune response to OvGM2AP divulges its immunogenicity

As an attempt to understand the immune response to OvGM2AP, total IgG responses were measured by ELISA in infected and non-infected individuals. Results obtained indicate, consistent with the western blot analysis reported above (Fig 3D), a discriminatory immune response to OvGM2AP between infected and non-infected individuals. The mean OD_450nm for patient sera (OVS) was significantly different (P<0.05) from that of normal African sera (HES) and European sera (ESC) with p-values lower than 0.001 for both HES and ESC (Fig 5A). The area under the ROC curve (AUC) was found to be high, with a value of 0.9863, and a p-value < 0.0001 (Table 1; S4 Fig) indicating both high sensitivity and specificity.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Analysis of humoral immune response to OvGM2AP.

Purified 8X His-fused OvGM2AP was used to coat microtiter plates. Plates were blocked and later incubated with serum from indicated individuals (OVS, HES, or ESC), followed by secondary antibody of either goat anti-human IgG peroxidase conjugate (a) or mouse anti-human IgG1-4 (b-e) which was followed with goat anti-mouse IgG, peroxidase conjugate as tertiary antibody. The plates were revealed using TMB. Optical density of stopped reactions was read at 450 nm and OD values were plotted against the different serum types. OVS = O. volvulus Serum (n = 30), HES = Hypo-endemic Serum (n = 14), ESC = European serum control (n = 03). A one-way ANOVA was used to compare groups. Error bars represent median with range. Dotted horizontal lines represent cut-off at mean + 3 SD of HES.

https://doi.org/10.1371/journal.pntd.0007591.g005

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Table of ROC values for IgG and IgG subclass responses to OvGM2AP.

ROC values were calculated for IgG and IgG subclass. Other parameters such as standard error (SE), confidence interval (CI), and P-values were also calculated for the IgG and IgG subclass immune responses to OvGM2AP. AUC = Area under the ROC curve.

https://doi.org/10.1371/journal.pntd.0007591.t002

We further investigated IgG subclass responses to the recombinant OvGM2AP by ELISA. All IgG subclasses (with the exception of IgG3) were detected at significantly different levels between infected individuals (OVS) and uninfected African Controls (HES) (Fig 5B–5E). The IgG4 subclass was found to have a higher AUC value (0.9451) compared to the other subclasses with a p-value < 0.0001 (Table 1). Diagnostic accuracy parameters investigated for the IgG and IgG subclass responses also revealed IgG4 as the favorite (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Diagnostic accuracy parameters for OvGM2AP.

Cut off values were obtained at mean + 3 standard deviations of the hypoendemic serum (HES). Sensitivity refers to the probability of getting a positive test result in subjects with the disease; Specificity is the probability of a negative test result in a subject without the disease; Positive predictive value defines the probability of having the disease of interest in a positive result; Negative predictive value is the probability of not having the disease in a subject with a negative test result; Likelihood ratio for positive test results indicates how much more likely the positive test result is to occur in subjects with the disease compared to those without the disease; Likelihood ratio for negative test results represents the ratio of the probability that a negative test result will occur in subjects with the disease to the probability that the same result will occur in subjects without the disease.

https://doi.org/10.1371/journal.pntd.0007591.t003

Levels of OvGM2AP antibodies are inversely correlated with duration of ivermectin treatment but cross-reacts with serum samples from L. loa patients

In a bid to assess OvGM2AP as a possible marker of a decrease of infection related to ivermectin treatment, we decided to investigate the correlation between the humoral immune response and rounds of ivermectin (IVM) intake. To this end, sera from Bandjoun in the Western region of Cameroon where onchocerciasis is almost eliminated [46] were grouped (10 samples per group) according to the number of rounds of ivermectin intake (expressed as years of treatment) and analyzed by ELISA using OvGM2AP as a bait. As shown in Fig 6A, results obtained suggest a negative correlation between the anti-OvGM2AP response and rounds of IVM intake. Consistently, samples from this site were compared to those from Kombone in the South-Western region where onchocerciasis remains actively endemic. In the analysis, sera from Rwanda where this disease is hypo-endemic (HES) were used as negative controls together with European sera (ESC). As shown in Fig 6B, OvGM2AP significantly discriminated between sera from the endemic region of Kombone (OVS) and the hypoendemic region of Bandjoun. This suggests that monitoring anti-OvGM2AP immune response would allow the discrimination of endemic and hypoendemic regions as well as the evaluation of the state of infection following ivermectin treatment.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. ELISA-analysis of IgG responses of treated individuals to OvGM2AP.

Purified 8X His-fused OvGM2AP was used to coat microtiter plates. Plates were blocked and later incubated with the indicated categories of serum followed by secondary antibody (goat anti-human IgG peroxidase conjugate). The plates were revealed using TMB. Optical density of stopped reactions was read at 450 nm and OD values were plotted against ivermectin treatment rounds (a) or different the different serum types (b). a) IgG responses to OvGM2AP in patients (n = 40) with different rounds of ivermectin intake including groups of ivermectin naïve patients, 0 (n = 10), 1–6 (n = 10), 7–12 (n = 10) and >13 (n = 10) rounds of ivermectin administration. b) IgG responses to OvGM2AP in different patient groups. OVS = O. volvulus Serum (n = 30), Bandjoun = serum samples from Bandjoun (n = 30), an onchocerciasis hypo-endemic zone, HES = Hypoendemic Serum (n = 14), ESC = European serum control (n = 03). A one-way ANOVA was used to compare groups. Error bars represent median with range.

https://doi.org/10.1371/journal.pntd.0007591.g006

In order to investigate a possible cross-reaction with OvGM2AP in individuals infected with other nematodes, we analyzed sera from patients infected with Loa loa, Brugia malayi, Wuchereria bancrofti, Mansonella perstans and Ascaris lumbricoides. With this in mind, OvGM2AP was used to coat ELISA plates as earlier described and IgG response was measured in L. loa infected serum (LLS), W. bancrofti infected serum (WBS), B. malayi infected serum (BMS), M. perstans infected serum (MPS), A. lumbricoides infected serum (ALS) and compared with OVS and controls. Results obtained indicated a very strong response from related nematodes (Fig 7). This suggests consistently with the high sequence identity shown above (Fig 1A), that epitopes are common to Ov and Ll GM2AP, as well as the other nematodes. This is a drawback for the use of OvGM2AP in the specific diagnosis of the disease.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. Analysis of humoral immune response of related nematode sera to OvGM2AP.

Purified 8X His-fused OvGM2AP was used to coat microtiter plates. Plates were blocked and later incubated with serum from indicated individuals (OVS, HES, LLS, WBS, BMS, MPS, ALS or ESC), followed by incubation with goat anti-human IgG peroxidase conjugate as secondary antibody. The plates were revealed using TMB. Optical density of stopped reactions was read at 450 nm and OD values were plotted against the different serum types. OVS = O. volvulus Serum (n = 30), HES = Hypo-endemic Serum (n = 14), ESC = European serum control (n = 03), LLS = L. loa serum (n = 39), WBS = W. bancrofti serum (n = 06), BMS = B. malayi serum (n = 03), MPS = M. perstans serum (n = 06), ALS = A. lumbricoides serum (n = 06). A one-way ANOVA was used to compare groups. Error bars represent median with range. Dotted horizontal lines represent cut-off at mean + 3 SD of HES.

https://doi.org/10.1371/journal.pntd.0007591.g007

OvGM2AP could not hydrolyze the GM2 to GM3

The ability of OvGM2AP to mediate the hydrolysis of GM2 by β-hexosaminidase A was investigated in an in vitro liposomal assay using negatively charged, BMP-containing liposomes. Liposomal composition was Cholesterol (5 mol%), BMP (20 mol%) and [¹⁴C] GM2 (1 mol%), made up to 100 mol% by DOPC.

Turnover of [¹⁴C] GM2 in the liposomal activity assay was stimulated by human recombinant GM2AP or tauridesoxicolate (TDC) (50 μg), both used as positive control. No hydrolysis of [¹⁴C] GM2 was detected in the presence of OvGM2AP (Fig 8A).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 8. Analysis of GM2AP activity of OvGM2AP.

The ability of OvGM2AP to hydrolyze GM2 to GM3 was assessed in a liposomal setting (A). Liposomes were incubated with the following accordingly to the numbers; 1 = human recombinant GM2AP (negative control), 2 = OvGM2AP (negative control), 3 = β-hexosaminidase A (negative control), 4 = β-hexosaminidase A and tauridesoxicolate (positive control), 5–7 = β-hexosaminidase A and human recombinant GM2AP (positive control), 8–10 = β-hexosaminidase A and OvGM2AP. Varying concentrations of OvGM2AP were incubated with recombinant Hex A and MUG or MUGS artificial substrate (B) and levels of degradation of the substrates was assessed using a fluorimeter.

https://doi.org/10.1371/journal.pntd.0007591.g008

OvGM2AP competitively inhibits MUG degradation by recombinant β-hexosaminidase A but not MUGS

The human GM2AP has been reported to competitively inhibit the degradation of the artificial fluorogenic MUGS substrate but not the neutral MUG substrate. In a bid to understand if a parasitic GM2AP like OvGM2AP could share the same characteristics like the human GM2AP, we employed MUG and MUGS in a competitive inhibition assay with OvGM2AP. As shown in Fig 8B, OvGM2AP was found to competitively inhibit the degradation of MUG, but not MUGS as previously reported for the human GM2AP [63].