Human Leukocytes Kill Brugia malayi Microfilariae Independently of DNA-Based Extracellular Trap Release

Abstract

Background: Wuchereria bancrofti, Brugia malayi and Brugia timori infect over 100 million people worldwide and are the causative agents of lymphatic filariasis. Some parasite carriers are amicrofilaremic whilst others facilitate mosquito-based disease transmission through blood-circulating microfilariae (Mf). Recent findings, obtained largely from animal model systems, suggest that polymorphonuclear leukocytes (PMNs) contribute to parasitic nematode-directed type 2 immune responses. When exposed to certain pathogens PMNs release extracellular traps (NETs) in the form of chromatin loaded with various antimicrobial molecules and proteases.

Principal findings: In vitro, PMNs expel large amounts of NETs that capture but do not kill B. malayi Mf. NET morphology was confirmed by fluorescence imaging of worm-NET aggregates labelled with DAPI and antibodies to human neutrophil elastase, myeloperoxidase and citrullinated histone H4. A fluorescent, extracellular DNA release assay was used to quantify and observe Mf induced NETosis over time. Blinded video analyses of PMN-to-worm attachment and worm survival during Mf-leukocyte co-culture demonstrated that DNase treatment eliminates PMN attachment in the absence of serum, autologous serum bolsters both PMN attachment and PMN plus peripheral blood mononuclear cell (PBMC) mediated Mf killing, and serum heat inactivation inhibits both PMN attachment and Mf killing. Despite the effects of heat inactivation, the complement inhibitor compstatin did not impede Mf killing and had little effect on PMN attachment. Both human PMNs and monocytes, but not lymphocytes, are able to kill B. malayi Mf in vitro and NETosis does not significantly contribute to this killing. Leukocytes derived from presumably parasite-naïve U.S. resident donors vary in their ability to kill Mf in vitro, which may reflect the pathological heterogeneity associated with filarial parasitic infections.

Conclusions/significance: Human innate immune cells are able to recognize, attach to and kill B. malayi microfilariae in an in vitro system. This suggests that, in vivo, the parasites can evade this ability, or that only some human hosts support an infection with circulating Mf.

Fig 1. B. malayi Mf entangled within neutrophil extracellular traps.

Fluorescence imaging of a worm-NET aggregate labelled with DAPI (blue) and antibodies to myeloperoxidase (MPO; purple), citrullinated histone H4 (histone; green) and human neutrophil elastase (HNE; red). The co-localization of markers with Mf is shown in the merged image with transmitted light (top left panel) and the overlap of MPO, histone and HNE with DNA shown in the merged image (bottom right panel). Live Mf were incubated in the presence of human neutrophils for 18 hours at 37°C and 5% CO₂. Scale bar = 10μm.

Fig 2. Extracellular DNA release from human PMNs.

A-C: Images of SYTOX Orange-labelled extracellular DNA present within Mf-treated (A), zero Mf (negative control) (B) and PMA-treated (positive control) (C) wells. The images were taken 7 hours post-assay set up. These wells did not contain serum. D-G: Changes in mean SYTOX Orange intensity were used to monitor the release of extracellular DNA from PMNs incubated for 7 hours in vitro. In each panel, the dotted lines indicate the Standard Error of the mean fluorescence intensity. D: 25nM phorbol myristate acetate (PMA)- and Mf-induced DNA release in the absence of serum was measured as described in Materials and Methods (n = 7). E: 10μM diphenyleneiodonium (DPI) and 30μg/ml DNAse I inhibited Mf-induced DNA release in the absence of serum (n≥4). F: Extracellular DNA release in Mf- and PMA-treated wells that contained 5% autologous serum (n = 6). G: 5% autologous serum and 5% autologous heat treated serum (HTS) inhibited Mf induced DNA release (n≥6).

Fig 3. The attachment of PMNs to B. malayi Mf.

A: The percentage of Mf that had at least one PMN adhered to their surface or indirectly fastened by extracellular DNA at 24 hours post-experimental set up (n≥6). White, yellow, red, blue, and black bars represent no serum, 5% autologous serum, 5% autologous heat-treated (55°C, 30 mins) serum, 10μM diphenyleneiodonium (DPI) and 30μg/ml DNase I treated wells respectively. Error bars represent standard error of the mean; *P<0.05, **P<0.01, ***P<0.001. B: Image of PMNs adhered to a Mf. Image taken after 24 hours of incubation in 5% autologous serum at 37°C and 5% CO₂. The preparation was stained with modified Wrights’ stain.

Fig 4. Mf survival in the presence of PMNs and PBMCs.

The percentage of Mf that survived to days 1 (A), 2 (B) and 5 (C) post-experimental set up (n = 5). White, blue, yellow, and red bars represent zero cell, 1500 PMN/Mf, 1500 PBMC/Mf and PMN plus PBMC treated wells respectively. All wells contained either 25% autologous serum or 25% heat treated serum (HTS). DNase I denotes the addition of 30μg/ml DNase I. Error bars represent standard error of the mean; *P<0.05, **P<0.01, ***P<0.001. D. Scatter plot comparing leukocyte attachment and Mf survival in the presence of PMNs and PBMCs. The percentage of Mf that had at least one PMN/PBMC adhered to their surface or indirectly fastened by extracellular DNA at 1 hour post-experimental set up plotted against the percentage of Mf that survived to day 5 (n = 15). Each point represents an individual biological replicate derived from distinct human blood donors. All wells contained 25% autologous serum, ~100 Mf, ~150,000 PMNs and ~150,000 PBMCs.

Fig 5. Compstatin-mediated inhibition of complement activation has little effect on Mf survival.

A: The percentage of Mf that had at least one PMN adhered to their surface or indirectly fastened by extracellular DNA at 24 hours post-experimental set up (n = 3). Orange and red bars represent 5% autologous serum and 25% autologous serum treated wells respectively. The autologous serum was either untreated (no peptide control), treated with 100μM inactive compstatin analogue (control peptide) or treated with 100μM active compstatin. Error bars represent standard error of the mean. B: The percentage of Mf that survived 5 days in the presence of both PMNs and PBMCs after treatment of serum with control peptide or compstatin (n = 3). C: PMNs were incubated with blocking antibodies for 2 hours prior to incubation with Mf for 5 days. PMN-mediated killing occurred to the same extent in control (no antibody) wells and those treated with anti-ICAM-1 and anti-Cd11b (p = <0.0001). No significant differences were observed between PMN incubated with Mf with or without the anti-ICAM-1 or anti-Cd11b (n = 3).

Fig 6. Mf survival in the presence of human PMNs and monocytes.

A: The percentage of Mf that survived to day 5 (n≥7). White, blue, green, and light green bars represent no cell, 1500 PMN/Mf, 1500 monocytes/Mf and 1500 PMN/Mf plus 1500 monocytes/Mf treated wells respectively. Error bars represent standard error of the mean; *P<0.05, **P<0.01, ***P<0.001. B: CD14⁺ and CD16⁺ cells attach to B. malayi Mf in culture. Mf were co-incubated with both PMNs and monocytes for 5 days and processed for immunofluorescence microscopy as described in Materials and Methods. Monocytes were isolated using negative immunoselection for CD16 and therefore are CD14 positive and CD16 negative. These are detected with anti-CD14-Alexa fluor 598 (red). PMNs express both CD14 and CD16 so are identified by both anti-CD14 (red) and CD16 (green) labelling. DAPI staining of Mf nuclei is indicated by a yellow arrow, highlighting the position of the Mf. Bar = 20 μM. C: The percentage of mf that survived to day 5 in the absence of immune cells (white bar), or in the presence of 1500 PMN (blue bar), 1500 lymphocytes (purple bar) or 1500 PMN plus 1500 lymphocytes (light purple bar). Error bars represent standard error of the mean; *P<0.05 (n = 3).