The endosymbiotic bacteria, Wolbachia, induce neutrophilic responses to the human helminth pathogen Onchocerca volvulus. The formation of Neutrophil Extracellular Traps (NETs), has been implicated in anti-microbial defence, but has not been identified in human helminth infection. Here, we demonstrate NETs formation in human onchocerciasis. Extracellular NETs and neutrophils were visualised around O. volvulus in nodules excised from untreated patients but not in nodules from patients treated with the anti-Wolbachia drug, doxycycline. Whole Wolbachia or microspheres coated with a synthetic Wolbachia lipopeptide (WoLP) of the major nematode Wolbachia TLR2/6 ligand, peptidoglycan associated lipoprotein, induced NETosis in human neutrophils in vitro. TLR6 dependency of Wolbachia and WoLP NETosis was demonstrated using purified neutrophils from TLR6 deficient mice. Thus, we demonstrate for the first time that NETosis occurs during natural human helminth infection and demonstrate a mechanism of NETosis induction via Wolbachia endobacteria and direct ligation of Wolbachia lipoprotein by neutrophil TLR2/6.
Onchocerciasis (river blindness) is a parasitic disease affecting 37 million people worldwide, causing debilitating skin and eye pathology and accounting for an estimated global loss of 1 million Disability-Adjusted-Life-Years1. It is caused by the migrating progeny (microfilariae-mf) of the filarial nematode Onchocerca volvulus, released by adult females residing in subcutaneous nodules (onchocercomata). The inflammatory response to dying mf in the skin and eyes, and the release of their bacterial endosymbiont Wolbachia, is the basis of Onchocerca dermatitis and ocular keratitis (river blindness) immunopathology2.
Neutrophils are the major component of the early inflammatory infiltrate around damaged mf in the cornea and skin3,4,5,6. Their recruitment and activation, with subsequent development of tissue pathology or systemic adverse reactions to microfilaricidal drug treatment, depend on the presence of Wolbachia3,4,5,7,8,9,10. The diacylated N-terminal polypeptide (WoLP) of the Wolbachia surface peptidoglycan-associated lipoprotein (wPAL) has been identified as the main trigger of the neutrophil inflammatory response during ocular keratitis via the activation of Toll-like Receptor (TLR)-2/6 on immune and stromal cells11. In addition, recent data illustrates that the WoLP TLR2/6 ligation motif is sufficient to directly induce a range of activation phenotypes in human neutrophils in vitro, demonstrating a direct functional role in neutrophil-mediated inflammatory responses12.
In onchocercomata, neutrophils are an abundant leukocytic infiltrate and are frequently found adjacent or attached to Wolbachia-containing adult worms13. Neutrophil numbers are dramatically decreased within nodules derived from doxycycline treated patients, where Wolbachia are effectively depleted from nematode tissues, or in animals parasitized by Onchocerca spp. naturally devoid of the symbiont13,14. Nevertheless, neutrophils do not appear to be detrimental to the survival of O. volvulus adults, although defensin and calgranulins have been found on their surface, suggestive of neutrophil degranulation13,15,16,17,18.
A novel effector mechanism defined first in neutrophils19 has been identified as the formation of “neutrophil extracellular traps” (NETs), in a process referred to as NETosis. NET formation consists of the extrusion of nuclear content, decorated with granular and cytoplasmic proteins19,20,21. Extracellular traps (ETs) have latterly been observed from other granulocytes, such as eosinophils as well as mast-cells, and macrophages22. Although the range of biological functions of NETs continues to grow, a consensus has emerged that in infectious disease, NETs serve to limit microbial spread, by the entrapment of live microorganisms22. Moreover, it has been suggested that NETs may also limit collateral tissue damage in the context of inflammation by entrapping and degrading soluble pro-inflammatory cytokines and chemokines23.
In parasitological infectious disease, NET formation can be induced by a range of protozoan pathogens such as Leishmania, Plasmodium, Toxoplasma, Eimeria, Besnoitia and Trypanosoma24,25. The induction of NETs by multi-cellular, macroparasites has been reported in vitro and in vivo in experimental murine infection by larvae of the human gut nematode, Strongyloides stercoralis26,27, and eggs of the human blood fluke, Schistosoma japonicum28 where NETs are observed within the core of hepatic egg granulomas29. In filariasis, NETs release in vivo and in vitro has also been recently reported in the Litomosoides sigmodontis murine model with NETs observed in mouse skin at 6 hours post-infection30.
Because of the well-defined relationship between neutrophil recruitment and the Onchocerca endobacteria, Wolbachia, we hypothesised that NET formation was a possible phenotypic outcome of neutrophil recruitment in human onchocerciasis. Availability of nodule tissues from a placebo-controlled clinical trial31 where Wolbachia had effectively been depleted, allowed us to examine both evidence of NET production and dependency on Wolbachia. From our histological analyses we conclude that NETs are produced adjacent to the nematode in human onchocerciasis and their presence is strongly associated with neutrophils and Wolbachia. Our in vitro experiments demonstrate that whole Wolbachia or latex microspheres decorated with the WoLP TLR2/6 ligand are sufficient to induce NETosis in human and mouse neutrophils, dependent on the TLR6 receptor.
NETs release occur within human onchocercomata coincident with Wolbachia and neutrophil infiltrates
All investigated onchocercomata from placebo (P) matching doxycycline +/− ivermectin-treated patients contained worms positive for Wolbachia, as assessed using immunohistochemistry (IHC) with a purified anti-wPAL IgG11,32, whilst all nodules derived from doxycycline (D) or doxycycline + ivermectin (DI) treated patients were Wolbachia-negative (Fig. 1A,B). This corroborated previous data from the same sample group examining the effective depletion of Wolbachia following doxycycline treatment via Wolbachia surface protein (WSP) IHC and wsp DNA copy number31 but in addition, confirmed the absence of the major Wolbachia neutrophil activating molecule, wPAL, from antibiotic-treated onchocercomata. In all P, but not D/DI onchocercomata, haematoxylin and Hoechst- or DAPI-positive extra-cellular material with a NET-like structure could be observed in zones adjacent to nematode cuticle (Figs 1C–J and 2). Neutrophils were frequently observed in placebo-treated (Wolbachia+) onchocercomata, around and in close contact with adult worms. These neutrophils were often found immersed in the NET-like material (Fig. 1C,E,G–I). A lack of eosinophilic granules on standard Haematoxylin-Eosin (H&E) staining was observed in the extracellular space around these cells, suggesting it was highly unlikely that those NET-like structures were degranulated eosinophils or mast cells (Fig. 1C,G,I). Absence of co-staining of NET-like structures with eosinophil major basic protein (MBP) and tryptase (TRYPT) (Supplementary Fig. S1), confirmed this hypothesis. On the contrary, these NET-like structures co-stained with myeloperoxidase (MPO), neutrophil elastase (ELA) and citrullinated histones (CITH3) indicating the presence of neutrophils undergoing NETosis in the NET-like DAPI-positive areas neighbouring the nematode cuticle (Fig. 2). Neutrophil number was dramatically decreased in doxycycline treatment groups, with virtual disappearance of these cells (p < 0.001 P vs D and DI, Fig. 3 and Table 1). In sections derived from doxycycline-treated onchocercomata, mononucleated cells were virtually the only cells surrounding worm tissues, although these were present in lower numbers compared to nodules from placebo patients, and there was no evidence of Hoechst-positive extracellular material (Fig. 1H). In particular, macrophages, identified as CD68+ cells, were present both in the centre and in the periphery of nodules, (p = 0.003 P vs D) and CD4+ T cells were mostly found in the centre of the nodules around the worms but not in close contact with them (p < 0.001 P vs D; p = 0.022 P vs DI). Eosinophils, although significantly reduced in D and DI compared to P nodules, were scant in all samples (Fig. 3 and Table 1).
Whole Wolbachia induce NETosis in human neutrophils in vitro
As the presence of neutrophils and NETs were associated with the presence of Wolbachia in onchocercomata, we investigated whether whole Wolbachia bacteria freshly isolated from the culture supernatant of a Wolbachia infected Aedes albopictus cell line (C6/36 [wAlbB]) induced NETosis when co-cultured with human neutrophils. Detection and quantification of wAlbB within supernatants was verified using qPCR of the single copy Wolbachia 16S rRNA gene. We found that neutrophils incubated for 2 h with cell supernatant containing Wolbachia bacteria at a 1:25 ratio optimally induced NETs, which were typically observed in 10–15% of the total culture area. NETosis was reliant on the presence of Wolbachia as parental uninfected C6/36 cell line supernatant (wAlbB-) added to neutrophil cultures failed to induce NETs. Corroborating the visualisation of NETs post wAlbB co-incubation, increased extracellular DNA was quantified when stripped from neutrophil cultures (p = 0.036 cf wAlbB- exposed cultures, Fig. 4).
Wolbachia lipopeptide coated micro-particles are sufficient to induce NETosis in human neutrophils in vitro
The diacylated N-terminus of Wolbachia lipoprotein, WoLP, has been defined as the predominant Wolbachia Pathogen-Associated Molecular Pattern, interacting with the host immune system via the TLR2/6-MyD88 signalling pathway11. Two native Wolbachia surface-associated lipoproteins have been characterised in both filarial and insect Wolbachia11,32. Filarial wPAL is abundantly expressed in female O. volvulus tissues (Fig. 1A and11) and has also been identified as a secreted Wolbachia protein in the secretome of the related human lymphatic filaria, Brugia malayi33,34. The WoLP motif of wPAL is sufficient to activate multiple functions of neutrophils in vitro12. Therefore, we investigated whether WoLP was a candidate Wolbachia molecule that interacted with human neutrophils to induce NETosis. Soluble WoLP did not activate NETs in human neutrophils at all doses tested (0.001–5 μg/ml, data not shown). Because whole Wolbachia could induce NET formation, we hypothesised that particulates were necessary for NETosis, putatively via phagocytosis. To test this, 5 μm, fluorescently-labelled latex beads were coated with WoLP and WoLP+ or uncoated WoLP- beads were used as a stimulus in human neutrophil assays. NETosis could be observed in approximately 10% of the culture area when isolated human neutrophils were exposed to WoLP+ beads incubated for 2 h at ratios of between 10:1 and 5:1 beads:neutrophil (Fig. 5). No NETs were induced in neutrophil cultures incubated with WoLP- control beads up to a ratio of 10:1 beads:neutrophil (Fig. 5).
Wolbachia mediated NETosis is TLR6 dependent
Because filarial Wolbachia and WoLP innate inflammatory activity, including neutrophil recruitment in vivo, has been shown to be TLR2/6 dependent11,35,36, we tested whether NETosis induction by Wolbachia also required an intact TLR2/6 receptor. For this we utilised TLR6 knock out mice that are non-responsive to cognate bacterial diacyl-lipoprotein ligands, including WoLP11. Purified, isolated bone marrow mouse neutrophils from wild type (WT) C57Bl/6 inbred mice or C57BL/6 mice deficient in the TLR6 gene (TLR6−/−) were cultured with PMA, Wolbachia stimuli or controls for 2 h. NETs were visualised by DAPI staining of extracellular DNA and, where indicated, co-localised with anti-neutrophil elastase staining (Fig. 6). PMA induced vigorous NET production in both WT and TLR6−/− neutrophils (mean culture area covered by NETs = 36%, WT vs 34%, TLR6−/−; Fig. 6A,D). Confirming WoLP-specific NETosis in human neutrophils, we observed significant NET production in murine WT neutrophils (mean culture area covered by NETs = 18%, ratio of 10 WoLP beads per neutrophil, p < 0.001 vs control bead cultures; Fig. 6B,D). We verified that the WoLP-mediated NETosis visualised in human neutrophils was via the TLR2/6 receptor by determining that NET production was almost completely ablated in TLR6−/− neutrophil cultures (mean culture area covered by NETs = 0.4%, p < 0.001 vs corresponding WT cultures; Fig. 6B,D). Corroborating data produced in human neutrophil cultures, whilst NETs were not observed when cultured with Wolbachia- supernatant, Wolbachia+ supernatant mediated significant NETs production in WT neutrophils (mean culture area covered by NETs = 7%, 50:1 Wolbachia:neutrophil ratio, p < 0.001 vs control; Fig. 6C,D). However, TLR6 deficiency made murine neutrophils unresponsive to Wolbachia+ supernatant in terms of NET production (p < 0.001 WT vs TLR6, 50:1 Wolbachia: neutrophil ratio).
We show, for the first time, that neutrophil infiltrates surrounding O. volvulus adult parasites in nodules are frequently contained within a DNA net-like material decorated with neutrophil-derived granule molecules and citrullinated histones as markers of NETosis. These structures were completely absent in Wolbachia-depleted nodules where the neutrophil infiltrate was replaced by NET-free mononuclear cells. These results are suggestive that neutrophil derived NETosis is a consistent feature of adult O. volvulus nematode infection of humans. Although other cell types may be a potential source of ETs, eosinophils appear to be virtually absent from these nodules, and ETs did not contain eosinophil or mast cell specific markers. To further investigate the ability of Wolbachia to drive NETosis, we exposed purified, peripheral blood human and bone marrow murine neutrophils to whole Wolbachia bacteria, soluble WoLP, and latex beads coated with a synthetic lipopeptide, WoLP, of the major Wolbachia TLR2/TLR6 ligand, peptidoglycan associated lipoprotein (PAL). Both whole Wolbachia and WoLP-coated beads induced NETosis in a TLR6-dependent manner, whereas soluble WoLP failed to trigger NETosis. Our results are consistent with a TLR2/6, phagocytosis-mediated induction of NETosis by liberated Wolbachia endobacteria, similar to that reported in response to Staphylococcus aureus infection in mice and described as “rapid vital NETosis”37,38. S. aureus whole bacteria induce rapid NETosis in vivo in a timeframe of < 30 min-2 h. S. aureus vital NETosis is dependent on both phagocytosis and TLR2 because phagocytosis is not sufficient to induce NETosis in the absence of TLR2, and triacylated lipopeptides do not mimic the effect of the whole bacteria. Interleukin (IL)-8 has also been reported to induce NETosis39. Our previous work12 showed that WoLP is able to induce IL-8 from human neutrophils, and so a contribution from autocrine IL-8 stimulation to Wolbachia-induced NETosis may further amplify and sustain NETosis.
Neutrophils play a pivotal role in Wolbachia-mediated pathogenesis of onchocercal disease manifestations. The TLR2/6 ligand, synthetic N-terminal diacylated lipopeptide WoLP, of Wolbachia lipoprotein wPAL which constitutes the major proinflammatory molecule expressed by the endosymbiont, induces corneal neutrophil recruitment and activation in a mouse model of keratitis11. WoLP is sufficient to activate multiple functions of human neutrophils such as chemotaxis, cytokine secretion, modulation of expression of surface adhesion molecules, oxidative burst and survival12. However, neutrophils do not appear to be directly detrimental to living adult worms in onchocercomata13 and their role in this stage of filarial parasitism is still unresolved. It has been suggested that Wolbachia-dependent recruitment of neutrophils around adult parasites in onchocercomata may facilitate nutrient uptake for reproductive function or to block the possibly deleterious recruitment and activation of eosinophils proximal to the worm cuticle13,16,40.
Our results from human onchocercomata tissues corroborate the identification of filarial-induced NETosis in skin-stage infections30. Filarial NET formation may be a host protective response provoked by endobacteria liberated with uterine and other secretion/excretions, which may entrap and limit the dissemination of Wolbachia. Also, appropriate NETosis enveloping viable female worms may aid, at the site of adult infection, in prevention of inflammatory damage induced by Wolbachia released from mf. Because NETs have been shown to effectively trap nematode larvae in vitro27, NETosis triggered by Wolbachia may also comprise an anti-parasitic response to limit the density of the tens of thousands of uterine-released mf produced daily by each female worm. Any decreases in migratory mf densities would limit the potential for immunopathology in the skin and eye and may also impact on the transmission potential of the parasite to infect the intermediate black fly vector. It has recently been shown that skin resident neutrophils can elicit anti-filarial responses30 and that NETs modify inflammation by entrapping soluble inflammatory mediators23. Therefore, Wolbachia-induced NETs may also have direct anti-inflammatory properties in modifying skin inflammation during onchocerciasis, and their release after drug-induced mf death deserves investigation.
The presence of NETs may also serve as a ‘cloaking device’ to decrease the penetrance of more damaging immune cells toward the cuticle surface of adult Onchocerca. In support of this, in closely related bovine onchocerciasis, Wolbachia depletion by antibiotic treatment triggers a replacement of neutrophils with degranulating eosinophils adjacent to the worm surface16,40. The alteration in granulocyte recruitment following loss of the endosymbiont precedes the time point when adult filariae lose viability16. Further, it is possible that the formation of NETs, at vulnerable points along the worm body, such as the uterine pore, may be important in preventing access by immune cells and the cytotoxic granule products they release. In addition to serving as a physical barrier to cellular penetration of the parasite cuticle, NETs may possibly down-regulate an orchestrated protective immune attack through the degradation of cytokine and chemokine paracrine signals23. These hypotheses promote the phenomenon of filarial NETs production via Wolbachia symbionts in a role consistent with a balanced, permissive parasitism whereby adult worms survive to reproduce yet immunopathology is limited.
Immunohistochemistry of onchocercomata
O. volvulus infected patients from North West Province, Cameroon were enrolled into a double-blind placebo-controlled randomized trial of doxycycline (6 weeks ± ivermectin 4 months after the start of treatment) and placebo31. Onchocercomata were surgically removed 21 months after treatment, fixed in 80% ethanol and embedded in paraffin. Sections of 4 μm mounted on Poly-L-lysine slides were rehydrated through serial dilutions of xylene and ethanol to water. After heat-induced antigen retrieval in 1 mM EDTA pH 8.0, presence of Wolbachia was assessed using rabbit polyclonal affinity-purified IgG against wBmPAL11 with Ultra-Vision ONE Detection System AP Polymer & Fast Red Chromogen (Thermo Scientific) following the manufacturer’s instructions. For CD68 and CD4 staining, after re-hydration and antigen retrieval, endogenous peroxidase was quenched by incubation in 3% H2O2 in methanol, before sections were blocked in TNB blocking buffer (Perkin Elmer). Primary antibodies were mouse anti-human CD4 IgG1 (clone 1 F6, Novocastra) and mouse anti-human CD68 IgG1 (clone KP1, Dako); secondary antibody was goat anti-mouse IgG-HRP conjugated secondary antibody (NEF822, Perkin Elmer). The Tyramide Signal Amplification (TSA) Plus FITC System (Perkin Elmer) was used as the revealing system. Slides were mounted in Pro-Long Gold anti-fade reagent or Vectashield Mounting Medium with DAPI. Slides from a nodule from placebo-treated patients were stained omitting the primary antibody as control for non-specific binding of the secondary antibody. A slide from a placebo-treated nodule stained omitting the primary antibody was included as a control. Hoechst 2 μg/ml (Invitrogen) was used to visualise DNA and cell nuclear shape. Sections were mounted with 1:1 glycerol:PBS and visualized using an Olympus BX60 microscope supporting Nikon DS-Fi1c camera with NIS Elements Imaging software (Nikon). To assess whether polymorphonucleated cells were neutrophils or eosinophils, standard Haematoxylin-Eosin (H&E) staining was carried out by serial passages in Harris Haematoxylin (Raymond A Lamb), 1% acid alcohol (1% v/v HCl, 70% v/v ethanol), Scott’s tap water (238 mM NaHCO3, 29 mM MgSO4), and 1% aqueous Eosin (Raymond A Lamb). To identify neutrophils as the polymorphonucleated cells in NET-like structures, after acquisition of images, the coverslip was carefully removed and on the same sections H&E staining was applied. Sections were mounted with Low Viscosity DPX mountant (Bios Europe) and images of the same areas were captured and merged with corresponding images of Hoechst-stained sections. To confirm the presence and the cellular origin of ETs in Wolbachia-containing onchocercomata, immunofluorescent staining was performed, using a specific marker of NETosis (i.e. citrullinated histone H3), specific markers of neutrophil ETs (neutrophil elastase and myeloperoxidase), and markers of mast cells (tryptase) and eosinophils (major basic protein) ETs. After re-hydratation and heat-induced antigen retrieval in target retrieval solution (Dako), onchocercomata sections were blocked in PBS-BSA 5%. Primary antibodies were polyclonal rabbit anti-neutrophil elastase (Abcam), polyclonal goat anti-myeloperoxidase (R&D), polyclonal rabbit anti-citrullinated histone H3 (Abcam), polyclonal rabbit anti-major basic protein (Abcam) and mouse anti-mast cell tryptase (clone AA1, Abcam). Secondary antibodies were goat anti-rat (AF594, Life Technology), goat anti-rabbit (AF488, Life Technology), donkey anti-goat IgG (AF594, Abcam) and goat anti-mouse IgG (Hparticular, macrophages, identified as L) Texas-Red (Thermo Fisher). Slides were mounted in Vectashield Mounting Medium with DAPI (Vector Labs). Non-specific binding control staining was performed by omitting primary antibodies in one slide per batch. Sections were visualized with a confocal laser-scanning microscope (Leica DM2500) and images captured with 40x or 63x objectives.
Whole Wolbachia purification
Aedes albopictus mosquito cell lines C6/36 and C6/36 (wAlbB) were passaged in Leibovitz’s L-15 medium (Life Technologies) supplemented with 20% fetal bovine serum, 2% tryptose phosphate broth and 1% non-essential amino acids (Sigma Aldrich), as previously described41. Extracellular Wolbachia are routinely observed in culture supernatants of C6/36 (wAlbB) cells. To harvest Wolbachia-enriched supernatant, cultures were maintained without passage for 14–21 days. Supernatants were removed and centrifuged at 1000 g for 10 min to remove whole cells and debris. The resulting supernatant was stored on ice until use in the neutrophil culture experiments. Supernatant from the uninfected parental C6/36 cell line was collected using the same protocol and acted as a negative control. Wolbachia numbers present in each supernatant sample used in the neutrophil experiments were quantified using qPCR targeting the single copy 16S rRNA gene42. DNA was extracted from 100 μl aliquots of supernatant using QIAamp DNA mini-kit reagents (Qiagen), eluting in a final volume of 50 μl and qPCR was performed as previously described43. Supernatant from the parental C6/36 line was confirmed to be Wolbachia free using this method.
WoLP microparticle preparation
Fluoresbrite 2 μm carboxylated latex microparticles (Polyscience) were covalently conjugated to WoLP as per manufacturer’s instructions. Briefly, 12.5 mg beads were activated by water-soluble carbodiiamide (Polyscience), transferred into PolyLink coupling buffer (Polyscience) and incubated either with 500 μg WoLP diluted in coupling buffer or matching volume of buffer (negative bead control). Beads were mixed for 1 h at room temperature before pelleting by centrifugation. Supernatant was retained to determine binding efficiency. Beads were transferred into storage buffer (Polyscience) and stored at 4 °C. Efficiency of WoLP binding was determined by subtracting residual protein concentration in binding buffer from initial concentration, using colorimetric total protein estimation (BCA assay, Pierce). Efficiency of binding was typically >80%.
Human peripheral blood neutrophil isolation
Peripheral blood was collected by venipuncture in lithium-heparin from healthy volunteers and neutrophils were isolated by Polymorphprep (Axis Shield) following manufacturer’s instructions. Contaminating red blood cells were lysed with 9:1 ammonium chloride lysis buffer (13.4 mM KHCO3, 155 mM NH4Cl, 96.7 μM EDTA) in RPMI 1640 culture media (Gibco). Cell viability was assessed by 0.2% Trypan Blue staining (Sigma Aldrich) and was always ≥98%. The purity of isolated neutrophils was assessed by rapid Romanowsky stain (HD Supplies) of cytospins (Cytospins 3, Shandon) followed by differential count of ≥700 cells by optical microscopy. The purity of isolated neutrophils was ≥97%.
Mouse bone marrow neutrophil isolation
Homozygous Toll-like Receptor (TLR)-6−/− mice on the C57BL/6 inbred background (a gift from Professor Akira, Osaka University, Japan) were re-derived by embryonic transfer into specific pathogen free (SPF) recipient female C57BL/6. The homozygous colony was maintained by sibling mating in SPF barrier facility at designated animal facility (Liverpool University of Biological Services Unit) as licensed procedure under the UK Animals Scientific Procedures Act. C57BL/6 wild type (WT) mice were purchased from Charles River UK. All experiments were performed in accordance with relevant guidelines and regulations. Approval was obtained for all animal experiments from the ethical committees of the University of Liverpool and LSTM. Experiments were conducted according to Home Office (UK) requirements. Bone marrow cells were collected from both hind legs’ femurs of 8–9-week-old WT or TLR6−/− mice and were resuspended in PBS. Red blood cells were lysed by osmotic shock using 0.2% and 1.6% NaCl solutions. Cells were then subjected to a discontinuous 72–64% Percoll® density gradient centrifugation (450 g) in 15 mL Falcon tubes for 30 minutes at 4 °C. Mature neutrophils were collected at the 72–64% interface (purity >93%, assessed by flow cytometry and by optical microscopy), washed three times in cold PBS then resuspended in PBS at the working concentration of 106 cells/mL.
For the whole Wolbachia bacteria-induced NETosis assay, neutrophil cultures were performed in 24-wells cell culture plates (Costar) at a density of 4×105 cells/ml at 37 °C with 5% CO2 in RMPI media with 2% FCS (Sigma Aldrich). Cells were allowed to adhere for 1 h to a sterile round glass coverslip inserted in each well and followed by incubation for 2 h in the presence stimuli. Phorbol myristate acetate (PMA, 600 nM, Sigma Aldrich) was used as the positive control and culture media alone as the negative control. Wolbachia-containing supernatants of infected mosquito cells were used at a range of 2:1 to 300:1 ratio bacteria:neutrophil, while soluble WoLP (synthetic 20-mer of the N-terminal region of wBmPAL di-palmitoylated at the N-terminal cysteine residue11, EMC Microcollections) was tested at a concentration range of 0.001–5 μg/ml. Equivalent volumes of supernatant from uninfected mosquito cell culture or culture media containing equivalent volumes of the WoLP vehicle, Dimethyl Sulphoxide (DMSO, Sigma Aldrich) were also included as negative controls. WoLP-conjugated fluorescent beads or negative control beads were added to neutrophils cultures at a concentration range of 5:1 to 100:1 microparticles per cell.
After incubation with stimuli or controls, neutrophil-coated coverslips were fixed in 4% paraformaldehyde in PBS, washed in PBS, permeabilized in 0.05% Tween20 in TBS for 1 min, and washed with TBS. For mouse derived neutrophils which were not stimulated with control or WoLP-conjugated beads, cells were then stained for neutrophil elastase (1:200, Abcam), incubated 1 h at room temperature, washed three times 5 min with 1xTBS, and then incubated for 1 h with a secondary antibody (Donkey anti-rabbit AlexaFluor488 IgG (Hparticular, macrophages, identified as L), 1:400, Life technologies). For all human or mouse derived neutrophils, DNA was then stained with 1 μg/ml 2-(4-Amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI) (Sigma-Aldrich), for 3 min. Cells were washed with TBS and viewed with a confocal laser-scanning microscope (Leica DM2500). Images were captured with a 40x objective.
Neutrophil cultures were digested using micrococcal nuclease (Sigma Aldrich) to dismantle the NET scaffold. The culture was incubated for 10 min at 37 °C in the presence of 0.5 nuclease units and enzymatic digestion was terminated by using 5 mM EDTA. After centrifugation at 200 g for 8 min, 100 μL of the cell-free supernatant was transferred into a flat-bottom 96-well plate for the quantification of double stranded DNA using the Quant-iT Picogreen assay (Invitrogen, Carlsbad, CA). One hundred microliters of Picogreen reagent was added to the samples, which were then incubated at room temperature in the dark for 4 min. Extracellular DNA was measured with a spectrofluorometer at 480-nm excitation and 520-nm emission.
For the statistical analysis of cellular infiltration in onchocercomata, cells were counted in 20 randomly selected 100x (neutrophils, eosinophils and CD68+ cells) and 60x (CD4+ cells) magnification fields of view of 5 nodules from 5 doxycycline-treated patients, 3 nodules from 3 doxycycline + ivermectin treated patients and 5 nodules from 5 placebo treated patients. The distribution of the mean number of cells per field was positively skewed, therefore a Poisson regression model was applied and the corresponding p-value adjusted for clustering of replicates within nodules. Significant differences between quantities of extracellular DNA released following Wolbachia-free and Wolbachia-containing cell supernatant stimulation of human neutrophils were evaluated by Mann-Whitney tests. Significant differences in NETs occurrence in murine neutrophil cultures were evaluated by 1-way ANOVA with Tukey post-hoc tests. A p-value ≤ 0.05 was considered significant. Computations were carried out in SPSS Statistics 20.0 (IBM) and Prism 5 (GraphPad).
Human parasite material was obtained from patients enrolled in a double-blind placebo-controlled randomized clinical trial conducted in Cameroon. The experimental protocol for this study was designed in accordance with the general ethical principles outlined in the Declaration of Helsinki. The trial was approved by ethics committees of the Tropical Medicine Research Station, Kumba, Cameroon, and the Research Ethics Committee of The Liverpool School of Tropical Medicine, Liverpool, UK. Written informed consent was obtained from all participants, with the exception of those who were illiterate, where a literate witness signed on behalf of the participant and the participant added a thumbprint. The trial is registered with the current controlled trials registry, no: ISRCTN48118452. Ethical approval for the storage and experimentation on nodule tissues was obtained from NHS National Research Ethics Service (09/H1001/81, Northwest 4 REC) and for the use of blood neutrophils from adult healthy volunteers by the Research Ethics Committee of the University of Liverpool, UK.
How to cite this article: Tamarozzi, F. et al. Wolbachia endosymbionts induce neutrophil extracellular trap formation in human onchocerciasis. Sci. Rep. 6, 35559; doi: 10.1038/srep35559 (2016).
This work was supported by the A-WOL Consortium through a grant (OPP1045261) of the Bill and Melinda Gates Foundation to MJT, a BMGF Global Health Grand Challenges Explorations grant to JDT and MJT (OPP10867) and an MRC New investigator award to JDT (MR/L018756/1). We are grateful to Prof. Brian Faragher and Dr. Stephen Cross, Department of Parasitology, Liverpool School of Tropical Medicine, for statistics and technical assistance respectively.