Paramyosin of canine Onchocerca lupi: usefulness for the diagnosis of a neglected zoonotic disease

Background Of increasing importance to the medical and veterinary communities is the zoonotic filarioid nematode Onchocerca lupi. Onchocercosis, thus far found in wolves, dogs, cats and humans, is diagnosed via skin snips to detect microfilariae and surgical removal of adults from the eye of the host. These methods are time-consuming, laborious and invasive, highlighting the need for new tools for the diagnosis of O. lupi in susceptible hosts. Symptoms related to the presence of the adults in the eye can range from none apparent to severe, including blindness. No reliable chemotherapeutic protocols are available, as yet, to eliminate the infection. Paramyosin, an invertebrate-specific protein, has been well-studied as an allergen, diagnostic marker and vaccine candidate. The aim of this study, therefore, was to isolate and characterise paramyosin from O. lupi to assess its suitability for the development of a serological diagnostic assay. Methods The adult and microfilarial stages of O. lupi were isolated from the eyes and skin of a 3-year-old male dog. Total RNA was extracted and reverse transcribed into single stranded cDNA. Reverse-transcription PCR was used to isolate a full-length paramyosin cDNA from adult worms and to investigate the temporal expression patterns of this gene. All amplicons were sequenced using dideoxy chain termination sequencing. Bioinformatics was used to predict the amino acid sequence of the gene, to compare the DNA and protein sequences with those available in public databases and to investigate the phylogenetic relationship of all molecules. Antibody binding sites were predicted using bioinformatics and mapped along with published antigenic epitopes against the O. lupi paramyosin protein. The native protein, and three smaller recombinantly expressed peptides, were subjected to western blot using serum from dogs both positive and negative for O. lupi. Results Paramyosin of O. lupi was herein molecularly characterized, encoded by a transcript of 2,643 bp and producing a protein of 881 amino acids (101.24 kDa). The paramyosin transcript was detected, by reverse transcription PCR, in adults and microfilariae, but not in eggs. Phylogenetic analysis indicates that this molecule clusters with paramyosins from other filarioids to the exclusion of those from other taxa. A total of 621 unique antibody binding epitopes were predicted for this protein and another 28 were conserved in other organisms. This information was used to design three peptides, for recombinant expression, to identify the antibody binding epitope(s) and reduce potential cross-reactivity with serum from dogs infected with other filarioid nematodes. Native paramyosin, purified from microfilariae and adults, was detected by antibodies present in serum from dogs with known O. lupi infections. Conclusions Data provided herein may assist in the development of a serological diagnostic test, based on antibodies to O. lupi paramyosin, for the diagnosis of this infection, in order to gain more information on the real distribution of this little known filarioid of zoonotic concern. Electronic supplementary material The online version of this article (doi:10.1186/s13071-016-1783-z) contains supplementary material, which is available to authorized users.


Background
Ocular parasitism in humans and animals may be caused by a range of zoonotic helminths, including strongylids, ascarids, thelaziids and filarioids [1]. The parasitic nematode Onchocerca lupi (Spirurida: Onchocercidae) localises to the connective tissue of the sclera of dogs, wolves and cats [2]. In dogs, subconjunctival granulomas are the most commonly reported manifestation of infection by this filarioid [3]; however, clinical signs may vary from none apparent [4] to severe (i.e. blindness) [2]. Affected dogs have periodically been reported from Hungary, Greece, Germany, Portugal and Spain [3,[5][6][7][8][9][10]. More recently, reports have also emerged of infections of dogs and cats in the United States of America [11,12].
There is a paucity of information on the biology and epidemiology of O. lupi (e.g. vector insects and geographical distribution). The only report available, to date, on the prevalence of O. lupi came from apparently healthy dogs sampled in Greece (8.7 %) and Portugal (8.3 %) [4]. In addition, the DNA sequence data available are limited to single ribosomal and mitochondrial genes, or parts thereof, which have been used for molecular and phylogenetic analyses [23][24][25]. Importantly, the frequency, distribution and full zoonotic potential of this parasite has not yet been investigated, probably due to the difficulty in achieving a diagnosis of the infection, which is based on the detection of microfilariae in skin sediments or surgical removal and identification of the adult worm from the eye [26]. Some preliminary scientific evidence indicates that an ELISA, prepared with somatic paramyosin antigens from Onchocerca gibsoni (Og4C3), which infects cattle [27][28][29], cross-reacts with serum from dogs infected with O. lupi [30].
Paramyosin is a large structural protein (98-101 kDa) component of invertebrate muscle and found in endo-and ectoparasites of medical importance, but not in vertebrates [31]. By virtue of its major antigenic properties, this protein has been used for the development of diagnostic assays for filarioids of medical concern such as Wuchereria bancrofti and Brugia malayi [27,32,33]. Many studies of the paramyosin protein, as a potential vaccine candidate, have focussed on platyhelminths [34][35][36][37][38][39], with fewer studies on nematodes [40,41] and ticks [42]. An immunomodulatory effect of this protein has also been identified, inhibiting components of the complement cascade [43,44], binding of immunoglobulins and collagen [45,46]. This suggests paramyosin is an important part of the mechanism by which a parasite invades and maintains itself in the host [31], making this protein of potential usefulness for serological diagnosis of, and protective immunity against, O. lupi.
Therefore, the research herein describes the isolation and characterisation of a paramyosin transcript from O. lupi, bioinformatic analysis of the predicted protein and preliminary assessment of the potential for the full protein and recombinant pieces for the development of a specific serological diagnostic assay.

Biological material
Three adult female and 2 adult male nematodes were isolated from the eye of an infected dog (3-year-old, male) that died accidentally, from an area (Algarve, southern Portugal) where O. lupi is known to occur [4]. Microfilariae were collected by sedimentation from the skin of the dorsal region of the dog and eggs by cutting the uterus of the adult females. All samples were preserved in RNAlater® (Life Technologies, California, USA) and stored at -80°C until used. Serum samples were obtained from this dog and from O. lupi positive (n = 5) and negative dogs (n = 3) from previous research [30].

Nucleic acid isolation and single-stranded cDNA synthesis
Total RNA was isolated from adult and microfilarial O. lupi, following homogenization using a pestle and 400-600 μm glass beads (Sigma-Aldrich, Missouri, USA) employing the TriPure® isolation reagent according to the manufacturer's instructions (Roche Molecular Biochemicals, Basel, Switzerland). RNase inhibitor (RNasin®, Promega, Wisconsin, USA) was added to total RNA before quantification and storage. Due to the tiny amounts of RNA able to be extracted from these nematodes, no DNase treatment was performed. Nucleic acids were quantified using a Bioanalyzer (Agilent, California, USA) and all extracted material was stored at -80°C until used. Firststrand cDNA synthesis was performed using the Super-Script Reverse Transcriptase II kit (Invitrogen, California, USA), 0.5 g oligo dT primer (n = 12-18 primer, Promega, Wisconsin, USA) and 100 ng of total RNA from adult males and females, microfilariae and eggs according to the manufacturer's instructions. Each completed reaction was diluted to 250 ng/l with TE (10 mM Tris-HCl, 1 mM EDTA, pH 8.0) before use in reverse transcription PCR.
PCR of the paramyosin cDNA PCR of the paramyosin cDNA, using 250 ng of single stranded cDNA from a single adult female, was performed using Phusion High-Fidelity DNA polymerase (Life Technologies, California, USA) and primers (PARAdegF 5′-ATG TCC GGT TCA TTG TAC CGT AG-3′; PAR-AdegR 5′-CTA TTG CTC ATC TTC GAG AAC GT-3′), designed from available filarial nematode paramyosin sequences (B. malayi NCBI Accession no. U77590.1, Onchocerca volvulus NCBI Accession no. M95813.1 and Acanthocheilonema viteae nAv_1_0_scaf01040, www.nematodes.org). Reactions consisted of 1× GC reaction buffer, 2 pmol of each primer, 0.2 mM of dNTPs and 1 U of Phusion Taq DNA polymerase in a volume of 50 μl. Cycling conditions were 98°C/10 s for initial denaturation, then 35 cycles of 98°C for10 s, 55°C for 20 s and 72°C for 3 min for DNA amplification and 72°C for 10 min for the final extension. Products were resolved on 1 % agarose stained with 0.5× GelRed (Biotium, California, USA) and visualized on a GelLogic 100 gel documentation system (Kodak, New York, USA). Amplicon sequencing employed the PCR primers and Big Dye Terminator v.3.1 chemistry in a 3130 genetic analyzer (Applied Biosystems, California, USA).

Gene cloning and sequencing
The full-length cDNA isolated above was cloned into pGEM®-T Easy, employing T4 DNA ligase (3 U), and transformed into Escherichia coli (MACH-1 strain, Life Technologies, California, USA) using an established method [47]. Colonies were picked, resuspended in 20 μl of water, with 2 μl used as a PCR template for insert screening and the remaining grown overnight in Luria Bertani/ampicillin (100 g/ml) broth. Plasmid DNA from positive colonies was isolated using the PureYield™ Plasmid Miniprep System (Promega, Wisconsin, USA) and inserts were sequenced in both directions using the T7 and SP6 primers of pGEM®-T Easy [47].
Temporal expression of O. lupi paramyosin in adult males and females, microfilariae and eggs The entire paramyosin (2,643 bp; PARA_degF, PARA_degR primers) and elongation factor 1α (as a control gene, designed from the O. volvulus ELF1α (1,776 bp; GenBank M64333.1, OvELF1α_F 5′-ATT GAG ATT TCG GGA TTA AGT GAA T-3′, OvELF1α_R 5′-TTC AGT GTA GCA GGA GCA TAT G-3′) genes were amplified using Phusion High-Fidelity DNA polymerase enzyme (Life Technologies, California, USA) according to the method used above for cDNA amplification (53°C annealing for ELF1α). Reactions used 125 ng of single-stranded cDNA template for adults and eggs and 500 ng for microfilariae. Two reactions were performed for microfilariae using template from the first reaction, including the negative controls, in order to amplify both genes. The experiment was repeated 3 times (i.e. 3 replicates per gene) on 3 separate days. Products were sequenced using the PCR primers as described above.

Protein expression and western blots
Native paramyosin was isolated from microfilariae and adults using a differential ammonium sulphate precipitation method [58]. Protein was measured using the Qubit® Protein Assay Kit according to the manufacturer's instructions (Thermo Fisher Scientific, Massachusetts, USA). Subsequently, in order to define antigenic epitope(s) recognized by antibodies from dogs infected with O. lupi, bioinformatics was used to design 3 peptides, corresponding to both the presence of potential antigenic epitopes and natural methionine residues (to maintain the correct reading frame). These sections were designated pieces 1 (1,272 bp, 48.2 kDa), 2 (908 bp, 33.5 kDa) and 3 (513 bp, 19.65 kDa) Onchocerca volvulus T-cell D176-L192 [66] Trichinella spiralis Antibody E88-E107 N-terminal [67] from the N-to the C-terminal part of the protein respectively. Primers used were: Piece 1 -PARAdegF, Piece 1R 5′-CGC CTT CTC GTA CAA ATT CTT CAT-3′; Piece 2 -PIECE2F 5′-ATG AAG AAT TTG TAC GAG AAG GCG-3′, PIECE2R 5′-GAT TTC CTG AGA GCA TCA ATT TTC AT-3′; Piece 3 -PIECE3F 5′-ATG AAA ATT GAT GCT CTC AGG AAA TC-3′, PARAdegR using Phusion Taq polymerase under the conditions described above. All cDNAs were ligated into the SmaI restriction enzyme site of the pGEX3X glutathione S-transferase fusion protein expression vector (GE Healthcare Bio-Sciences, Pennsylvania, USA) using T4 DNA ligase (3 U) and transformed into E. coli (BL21 strain). Clones were induced using 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) for 3 h and then analyzed via SDS PAGE (10 % gels) using an established protocol [47]. Expressed peptides were purified using EZview™ Red Glutathione Affinity Gel (Sigma Aldrich, Missouri, USA) according to the manufacturer's instructions, without elution of the peptide from the resin. Peptides were then cleaved from GST using Factor Xa protease (0.5 U, Promega, Wisconsin, USA) and the cleaved peptides isolated from the supernatant. Following SDS PAGE (10 % gels) of total nematode protein (positive control), purified native paramyosin (< 0.5 ng -microfilariae, 500 ng -adult), purified recombinant peptide (500 ng), uninduced bacterial culture (5 μl) and the Precision Plus Protein™ All Blue Prestained Protein Standards (BioRad, California, USA), proteins were transferred to PVDF membrane (0.5A/45 min; Roche Applied Sciences, Penzburg, Germany). Following blocking, membranes were incubated with serum from uninfected dogs (negative control) or those infected with O. lupi (1:1000 dilutions). Dog antibodies were detected using rabbit anti-dog IgG horseradish peroxidase conjugate (1:3000; Sigma Adrich, Missouri, USA) and detected using the femto-Chromo kit (G Biosciences, Missouri, USA). Documentation was via scanning on an Epson Perfection 2450 scanner.

Results
The  Reverse transcription PCR (RT-PCR) amplified Ol-para from adult males and females, microfilariae, but not from eggs, whereas ELF1α was amplified from all stages (Fig. 1 A total of 874 antibody binding epitopes, predicted using bioinformatic algorithms, and 14 published T-cell and 14 B-cell epitopes were mapped against Ol-PARA ( Fig. 2 and Additional file 1: Figure S1). Of the 874 predicted antibody epitopes, when compared with those predicted for A. viteae, B. malayi, D. immitis, L. loa, O. volvulus and W. bancrofti, 621 were unique to O. lupi.
Forty-six full-length paramyosin molecules from a range of nematodes and platyhelminths were identified, some species having more than one form of paramyosin (Table 3). Alignment and phylogenetic analysis show distinct clusters  corresponding to nematodes and platyhelminths (Fig. 3). In the nematode branch, free-living nematodes such as Caenorhabditis spp. grouped together, whilst the parasitic nematodes formed two separate groups. The Platyhelminthes consist of both digeneans and cestodes, each forming a separate group. Further phylogenetic analysis of the nematode paramyosin molecules separates the nematode proteins according to distinct taxonomic groups (Fig. 4).
Western blot analysis revealed that serum from dogs infected with O. lupi readily detected native paramyosin purified from microfilariae (< 0.5 ng, lower than the detection limit of the Qubit® Protein Assay Kit) and adults (< 1 μg) (Fig. 5). An additional co-purifying protein (~50 kDa) was detected, but was also detected by negative serum (data not shown). The microfilarial protein was detected, whilst the western blot of adults contained many additional proteins, most likely from the dog, as the adults are often embedded within the eye tissue, which cannot be completely removed. As DNA, RNA and protein are extracted from each sample simultaneously, enzyme digestion to facilitate removal of dog tissue was not performed due to the concurrent risk of RNA degradation. None of the purified, Factor Xa cleaved peptides, which were of the expected sizes, were detected by serum from dogs with O. lupi.

Discussion
Onchocerca lupi paramyosin (Ol-para, 2,643 bp cDNA) was herein isolated and characterised in adults, both males and females, and microfilariae. The predicted Ol-PARA protein (i.e. 881 amino acids, 101.24 kDa) represents a unique set of data for filarioids nematodes and, most specifically, for this little known vector-borne helminth. Indeed, although O. lupi is known to cause morbidity in both humans and animals, very little molecular information is available. The length of Ol-para is consistent with those of other filarioid nematodes, which range from the smallest 2,508 bp (L. loa, NCBI Accession no. XM_003141046.1) to 2,640 bp (B. malayi, NCBI Accession no. Q01202.2). The nucleotide identity levels  also reflect the relatedness of Ol-para compared with paramyosin sequences from other organisms, with high homology to other filarioid nematodes (91-98 %) compared with non-filarioid nematodes (~77 %) and the platyhelminths (~52 %). The AT-richness of Ol-para (i.e. 60.23 %) was consistent with other filarioid nematode sequences, including O. volvulus (AT 61 %), L. loa (AT 61 %) and B. malayi (AT 61 %). Some, but not all, of the platyhelminth paramyosins (Schistosoma spp.) are also AT-rich (65-67 %), whilst the remaining platyhelminths and non-filarioid nematodes have percentages of AT nucleotides in their sequences of between 44 % and 54 % ( Table 2). Whether there is any functional significance of this difference in AT proportions in paramyosin molecules is yet to be determined. Unlike some other species of nematodes, however, only one transcript encoding paramyosin was detected in O. lupi adults and microfilariae, whilst no paramyosin transcript was detected in eggs. The detection of transcription of paramyosin in O. lupi microfilariae is consistent with what has been found in other nematodes. Indeed, two forms of paramyosin have been detected in B. malayi (NCBI Accession no. P13392.2), which differ in size by only three base pairs from adults (NCBI Accession no. P13392). Paramyosin transcripts in other species, such as the free-living nematode Caenorhabditis elegans, differ significantly in size (WormBase Accession no. F07A5.7a and b, 2,616 and 1,659 bp, respectively). In O. lupi it is possible that additional transcripts exist, but may be expressed at levels not able to be detected by the methods used. Upon deep transcriptomic sequencing of B. malayi life-cycle stages, paramyosin and ELF1α were identified across the life-cycle of the parasite [59]. Similarly, in Dictyocaulus viviparus (cattle lungworm) and Clonorchis sinensis (liver fluke) paramyosin is expressed throughout the life-cycle [60,61]. Undoubtedly, next generation sequencing of the O. lupi transcriptome could be useful in determining the number of transcripts and confirming the expression patterns of paramyosin across the parasite life-cycle.
The Ol-PARA sequence is also consistent with that of other filarioid nematodes, with highest amino acid identity with the O. volvulus, B. malayi and L. loa, and within the range of lengths of proteins (90-93 % identity, 836-880 amino acids). Like other filarioid proteins, it has no signal , has a proton donor site. This motif is commonly associated with enzyme catalysis of hydrolysis of 1,3-or 1,4-bonds in β-D-glucans (prosite.expasy.org), but the functional significance of this activity in paramyosin is not yet known and is, as yet, unreported in the available literature. There is evidence from some nematode species to suggest that there is a short and a long isoform of paramyosin, hence the differences in size (Table 3). However, there is not enough information available for many species to confirm this. Phylogenetic analysis of platyhelminth and nematode proteins indicated that, whilst these proteins share conserved Tand B-cell epitopes, they are distinctly different proteins. Based on the phylogenesis of paramyosin molecules, five groups of nematodes were recognized: the rhabditids, including Caenorhabditis spp., the strongylids (Haemonchus contortus and A. caninum), the metastrongylids (Dictyocaulus viviparous), the ascarids (Anisakis spp. and Ascaris suum) and the filarioids (Onchocherca spp., B. malayi, A. viteae, L. loa and D. immitis). The W. bancrofti paramyosin does not closely cluster with other filarioids (Fig. 4). It is possible that for W. bancrofti that only one of a number of transcripts has been identified. Fig. 4 Phylogenetic analysis of nematode proteins. Thirty-two full-length paramyosin proteins from a range of nematodes were aligned and subjected to phylogenetic analysis. The topology the of trees from Bayesian inference and maximum parsimony analyses were in consensus, the representative tree displayed here with bootstrap values assigned to each branch (Bootstrap cut-off of 50). The protein sequence from Meloidogyne hapla (www.wormbase.org; Accession number: Contig185) was assigned as the outgroup Proteomic analysis of Ol-PARA has indicated that this protein, like other paramyosins, has the potential to elicit both humoral and cellular immune responses in the host [31]. The bioinformatic approach in this study, along with recombinant peptide expression, was however unable to identify the antibody binding epitope(s) of this protein. Nonetheless, 28 published epitopes (14 T-and 14 B-cell) experimentally determined in platyhelminths, ticks and O. volvulus, were conserved in Ol-PARA [38][39][40][41][42][43][62][63][64][65][66][67].
In addition, paramyosin is a vaccine candidate, predominantly in platyhelminths and also in Trichinella spiralis [41], and it has been listed by the World Health Organization as one of six antigen candidates for vaccines against Schistosoma spp. [64]. This, including the positive antibody binding of very small amounts of microfilarial and adult native paramyosin detected herein, suggests that Ol-PARA should be tested as a potential vaccine candidate. Accordingly, Olpara could also be tested as a DNA vaccine, particularly given the greater ease with which DNA can be manipulated when compared with proteins. To date, four studies have assessed paramyosin DNA-based vaccination (100 μg per rat/mouse) against Schistosoma japonicum [68,69], B. malayi [70] and T. solium [71], with evidence of some protective effect. Nonetheless, for its high antigenicity, Ol-PARA could also be exploited for the development of a serological diagnostic test. Indeed, this antigen (Og4C3) has been used for the preparation of a number of diagnostic kits based on the detection of O. gibsoni paramyosin in cattle, which may also cross-react with filarioids infecting humans such as W. bancrofti and B. malayi [27][28][29]32] and, to some extent, with O. lupi in serum from dogs that scored positive for microfilariae [30]. In addition, the identification of 621 antibody binding epitopes, unique to O. lupi when compared with other filarioid nematodes, provides a dataset invaluable for the development of a serological diagnostic assay through specific targeting of one or more of these epitopes. Antibody detection of O. lupi would not only assist in the detection of subclinically infected animals, gaining a better understanding of the true distribution and prevalence of O. lupi, but would also be amenable to the rapid screening of potential insect vectors. Vaccine studies would determine whether this molecule can provide a protective effect against infection by a zoonotic nematode for which the full zoonotic potential is yet to be determined. Future work will involve recombinant expression (eukaryotic) of the full protein, production of antibodies and testing of the serum against a range of filarial and other nematodes infecting humans and animals to ensure specificity. Serum from animals with infections of other filarioid nematodes will also be tested against the recombinant protein. If there is no cross-reactivity with serum from animals with other filarioid nematode infections this would provide a strong basis for the development of a specific serological diagnostic test.

Conclusions
This study indicates that paramyosin is a suitable candidate for the development of a specific diagnostic test and as a potential vaccine candidate, avoiding in the future a need for invasive, time-consuming tests. The development of a specific diagnostic test will enable further studies on the distribution and prevalence of this little known zoonotic filarioid.

Additional file
Additional file 1: Figure S1. Epitopes predicted by bioinformatics and those reported in the literature for other organisms mapped to Onchocerca lupi paramyosin. Residues underlined indicate T-cell epitopes whilst those in bold indicate B-cell epitopes. Of the 874 predicted antibody epitopes, when compared with those predicted for Acanthocheilonema viteae, Brugia malayi, Dirofilaria immitis, Loa loa, Onchocerca volvulus and Wuchereria bancrofti, 621 were unique to Onchocerca lupi. (DOCX 32 kb)

Acknowledgments
The first author would like to take the opportunity to thank all past and present members of the Laboratorio di Parassitologia e Micologia e Infettive, Dipartimento di Medicina Veterinaria, Università degli Studi di Bari "Aldo Moro", Italy. Special mentions for Rafael Antonio do Nascimento Ramos and Riccardo Paolo Lia. Maria João Vila-Viçosa (Victor Caeiro Laboratory of Parasitology, Universidade de Évora, Portugal) is thanked for her skilled technical support in the dissection of adult worms.

Funding
This study was partially funded by Merial.

Availability of data and material
The full-length paramyosin cDNA (Accession no. KJ699378) and protein (Accession no. AJO15918.1) sequences are available in the NCBI database. All predicted antibody binding epitopes are available in Additional file 1: Figure S1.
Authors' contributions BC, AG and DO conceived the research. HC, LC, BC and AG collected the samples. BC and GA performed the experiments. AP performed the DNA sequencing. BC performed the data analysis and wrote the manuscript, with contributions from DO for interpretation of the data and revision of the manuscript. AG, SL and FDT contributed to revision of the manuscript. All authors read and approved the final version of the manuscript.

Competing interests
The authors declare that they have no competing interests.

Consent for publication
Not applicable.

Ethics approval and consent to participate
The study was conducted according to the Guideline on Good Clinical Practices (The European Agency for the Evaluation of Medicinal Products, Veterinary Medicines and Information Technology Unit, VICH Topic GL9; www.emea.eu.int/pdfs/vet/vich/059598en.pdf) and procedures were approved by the ethical commission at the University of Évora (identification number: AE02Fila2013), complying with Portuguese legislation for the protection of animals (Decree-Law no. 113/2013). An owner consent agreement was obtained before sample collection.