Monocytes were isolated from fresh sterile buffy coats (Jena University Hospital, Germany) or from fresh blood collected from healthy volunteers after informed consent according to guidelines from the local ethics committee. The study was conducted in accordance with the Declaration of Helsinki, all the protocols were approved by the Ethics Committee of the Jena University Hospital. Human PBMCs were isolated by Ficoll-Paque PLUS (GE healthcare, Freiburg, Germany) (density 1.077 g/ml) density gradient centrifugation. Lymphocytes were removed from the isolated PBMCs using 46% Percoll (GE healthcare, Freiburg, Germany) density gradient centrifugation. Monocytes were further purified according the manufacturer’s protocol provided with pan monocyte isolation kit (Miltenyi Biotec, Teterow, Germany). Isolated monocytes were suspended in growth medium (RPMI medium supplemented with 10% heat-inactivated FCS and 2 mM ultra glutamine 1) and used for experiments within 1 h. Purity of isolated cells was confirmed by identifying CD14 (FITC anti-human CD14 antibody, Biolegend, London, UK) on the cells using flow cytometry.

Monocytic cells THP-1 (ATCC 16, Manassas, VA, USA) were maintained in growth medium at 37°C and 5% CO₂. Cells were passaged every 2 days and used for experiments until 30th passage. Complement active human serum was prepared from fresh whole blood, which was immediately centrifuged (10 min, 13,200 rpm, 4°C) mixed in a pooled stock and stored in aliquots at −80°C.

Candida albicans wild type (SC5314) (19) and GFP-expressing C. albicans (20) were grown overnight in yeast extract–peptone–dextrose medium (2% d-glucose, 1% peptone, 5% yeast extract in water) at 30°C, reseeded in yeast–peptone–dextrose medium, grown for 4 h at 30°C into the midlog-phase. Staphylococcus aureus (USA 300 LAC) and Escherichia coli (XL1-Blue) were grown overnight in tryptic soy broth and Luria–Bertani broth at 37°C, reseeded in same broth and grown until OD 1 at 37°C into the midlog-phase.

GFP-expressing C. albicans (5 × 10⁶) were added to monocytes (5 × 10⁵) in a 24-well plate in growth medium or in growth medium with complement active normal human serum (NHS, 10%). Monocytes were pre-stained with Vybrant DiD Cell-Labeling solution (5 µM) (Thermo Fisher Scientific, Dreieich, Germany). After 0–2 h of coincubation at 37°C with 5% CO₂, C. albicans were collected by removing the supernatant and monocytes were collected by detachment of cells using accutase (GE healthcare, Freiburg, Germany). Phagocytosis of C. albicans by monocytes was analyzed by flow cytometry. In parallel, C. albicans cells were labeled with mouse monoclonal Pra1 antibodies, and phagocytosis was measured as described above. Monoclonal antibodies (mAb) were generated by standard methods and specificity of the mAb was confirmed by Western blotting (data not shown).

Previously described procedure was followed for this quantification (21). Shortly, monocytes (10⁵ cells/well) were seeded into 96-well black plates. For coincubation, C. albicans (5 × 10⁵ cells/well) was added to each well at a MOI of 1:5 in growth medium alone or in growth medium with 10% NHS or with 10% heat-inactivated NHS (56°C for 1 h). No C. albicans was seeded for control monocytes. SYTOX green (5 µM) was added to all probes and incubated at 37°C with 5% CO₂. DNA release and ROS formation was quantified by measuring fluorescence intensity of SYTOX green (485/538 nm) every hour using a Safire plate reader (Tecan, Männedorf, Switzerland) and subtracting the values of fluorescence intensity of control monocyte. Similarly, DNA was quantified from neutrophils. The experiments were repeated three to four times.

Extracellular traps from monocytes were isolated according to a modified protocol by Barrientos et al. (22). Briefly, monocytes (1.5 × 10⁶) were seeded into 12-well plates and incubated with C. albicans cells (7.5 × 10⁶) in growth medium with 10% NHS for 4 h at 37°C and 5% CO₂. Monocytes alone were used as controls. Following 4 h incubation period, the wells were washed twice by DPBS, and probes were incubated with AluI (4 U/ml) (Thermo Fisher Scientific, Dreieich, Germany) for 20 min at 37°C to digest the extracellular traps into fragments. The supernatants were harvested by centrifugation (10,000 rpm for 5 min at 4°C) and the amount of extracellular DNA was quantified after staining with SYTOX green (5 µM). For standardization, Lambda DNA (Thermo Fisher Scientific, Dreieich, Germany) was similarly stained and measured (485/538 nm). Neutrophils were incubated with C. albicans or 100 nM phorbol 12-myristate 13-acetate (PMA) (Sigma-Aldrich, Taufkirchen, Germany) in growth medium and extracellular traps were isolated. Presence of citrullinated H3 was also assessed by Western blot analysis of isolated MoETs using rabbit anti-histone H3 (citrulline R2 + R8 + R17) antibody (Abcam, Cambridge, UK) (1:1,000) and polyclonal goat anti-rabbit antibody/HRP (Dako, Hamburg, Germany) (1:2,000).

Freshly cultured wild-type C. albicans cells (5 × 10³) were incubated with 50 ng of isolated MoET DNA or 50 ng of THP-1 chromosomal DNA or 50 ng of monocyte chromosomal DNA in growth medium at 37°C for 30 h inside Spectra Max 190 plate reader. Chromosomal DNA was isolated using PAXgene blood DNA kit (Qiagen, Venlo, Netherlands). C. albicans growth was measured every 10 min for 30 h by measuring absorbance at 604 nm using the SOFTmax PRO software. The experiment was repeated five times.

Immobilization of NETs on 96-well plates and blocking was performed according to a modified protocol by Caudrillier et al. (23). Briefly, rabbit anti-human MPO antibodies (Sigma-Aldrich, Taufkirchen, Germany) (1:100) or rabbit anti-human histone H3 (citrulline R2 + R8 + R17) antibodies (Abcam, Cambridge, UK) (1:200) were coated onto 96-well plates overnight at 4°C. Blocking was performed with 5% BSA and 0.1% human serum albumin in DPBS. Isolated NETs were added and incubated at room temperature for 1 h. NHS- or C3-depleted sera or alternative pathway activated NHS (by addition of 10 mM EGTA) or complement inhibited NHS (by the addition of 10 mM EDTA) or purified FH (5–20 μg/ml) and/or C3b (20 µg/ml) (both Comptech., TX, USA) were added and incubated for 1 h at 37°C. FH and C3b were detected on the extracellular traps using goat anti-human FH or goat anti-human C3 antiserum (Comptech, TX, USA) (1:5,000) followed by rabbit anti-goat antibody/HRP (Dako, Hamburg, Germany) (1:5,000). The experiments were repeated three to five times.

Interleukin-1 beta (IL-1β) generation by monocytes was measured with IL-1β ready-set-go ELISA kit (eBioscience, Frankfurt, Germany) according to the manufacturer’s protocol. Blood-derived monocytes were incubated with similar amounts of isolated NETs in growth medium (also with 10% NHS) for 20 h and IL-1β was determined in the supernatants by the ELISA kit. The amount of IL-1β produced by monocytes in presence of NETs with FH (12.5 µg/ml) or together with C3b (12.5 µg/ml) was also determined. The experiment was repeated three times.

Neutrophil extracellular traps were immobilized on a microtiter plate as described before (23) and incubated with C3b (20 µg/ml) and FH (5–20 µg/ml) or NHS (10%) for 30 min at RT. After washing off unbound FH, C3b (10 µg/ml) and factor I (7 µg/ml) (Comptech, TX, USA) were added and incubated at 37°C for 1 h. The reaction was stopped with the addition of Roti-Load 1 (Carl Roth, Karlsruhe, Germany), and proteins were separated by SDS-PAGE. C3b cleavage was assessed by Western blot analysis using goat complement C3 antiserum (Comptech, TX, USA) (1:1,000) followed by rabbit anti-goat antibody/HRP (1:2,500). The experiment was repeated three times.

All animal experiments were conducted in compliance with European and German regulations. Protocols were approved by the responsible Federal State authority and ethics committee (Thüringer Landesamt für Verbraucherschutz, permit number: 03-006/09). Female BALB/c mice (Charles River, Germany) weighing 18–20 g were housed in groups of five in individually ventilated cages with free access to water and food. For infection, C. albicans was grown for 12 h at 30°C in YPD medium, washed three times in sterile PBS, and diluted to the desired concentrations. The infection dose was confirmed by plating. On day 0, the mice were infected via the lateral tail vein with 2.5 × 10⁴ C. albicans cfu/g body weight. The health status of the mice was examined at least twice daily. Liver was collected 6 and 24 h post-infection, fixed in 10% neutral buffered formalin (Histofix, Carl Roth, Karlsruhe, Germany), embedded in paraffin, and sectioned at 4 µm thickness.

Paraffin-embedded mice liver sections were deparrafinized by placing them consecutively in Roticlear (Carl Roth, Karlsruhe, Germany), 100% ethanol, and 95% ethanol. Sectioned tissues were boiled in 10 mM Na citrate buffer (pH 6.5) for antigen retrieval and blocked in 1% BSA supplemented PBS for 30 min. For the detection of CD115, Ly-6C, and Ly-6G in tissues from 6 h post-infected mice, samples were treated with Alexa Fluor 488 anti-mouse CD115 (CSF-1R) Antibody (1:100), Alexa Fluor 700 anti-mouse Ly-6C Antibody (1:100), and Brilliant Violet 421 anti-mouse Ly-6G Antibody (1:100), respectively (all three antibodies are from Biolegend, London, UK). About 0.1% saponin supplemented PBS was used to dilute antibodies, and SYTOX orange was used to detect DNA in the tissue samples. Alexa Fluor 350 labeled anti-mouse FH antibody (1:1,000) (Bioss antibodies, MA, USA) diluted in 0.1% saponin supplemented PBS was used for detection of FH deposition in 24 h post-infected tissues and SYTOX green (5 µM) was used to detect DNA in the probes. Images were captured using LSM 710 with ZEN 2011 (401/421 nm for Alexa Fluor 350 and Brilliant Violet 421, 488/525 nm for Alexa Fluor 488, 504/523 nm for SYTOX green, 547/570 nm for SYTOX orange, and 679/702 nm for Alexa Fluor 700). The experiment was repeated three times.

Significant differences between two groups were analyzed using the Student’s two-tailed t-test or one way ANOVA. Values of *p < 0.05, **p < 0.01, and ***p < 0.001 were considered statistically significant.

To follow the direct response of human monocytes to C. albicans in vitro, live cell imaging was performed using laser scanning microscopy. Human monocytes were isolated from human blood using MACS (purity about 95% according to CD14 expression). These monocytes were incubated with GFP-expressing C. albicans at a ratio of 1:1 in growth medium with NHS in the presence of SYTOX blue, a nucleic acid dye that does not cross the intact cell wall membrane. Interaction of monocytes with C. albicans was followed over 12 h. Monocytes migrated toward the fungal pathogen and engulfed C. albicans cells within minutes (Figure 1A). Phagocytosis in 10% NHS, as evaluated by flow cytometry, was enhanced (about 30% after 15 min) compared with that in growth medium alone (Figures 1B,C).

Within the first minutes of contact with C. albicans, monocytes were highly active and engulfed C. albicans cells. During the next 2–4 h of cultivation, several monocytes showed morphological changes. The nuclei enlarged, and faint staining of nuclear DNA appeared. This staining is explained by increased membrane permeability and influx of the DNA dye SYTOX blue (Figures 1D,E). During the following 2 h, nuclear DNA decondensed and in several cases the DNA was suddenly released from the monocyte and appeared in the extracellular compartment around the ruptured membrane and on C. albicans (Video S1 in Supplementary Material; Figures 1D,E). In single cases, membrane rupture and release of extracellular DNA was detected even earlier, after 20–40 min. As cell remnants appeared in the medium, this type of DNA release resulted in monocyte cell death.

To compare this DNA release by monocytes with the previously described extracellular trap formation by neutrophils (24), the response of neutrophils to C. albicans was followed under identical experimental conditions. Neutrophils that were challenged with live C. albicans cells migrated toward the fungus, engulfed the pathogen, and also released the DNA in about the same time frame as monocytes (Figure S1 in Supplementary Material). Thus, both human innate immune cells, monocytes and neutrophils, release DNA within the first hours upon contact with C. albicans.

To determine whether release of DNA in response to C. albicans is specific to one subtype of monocytes or whether both types of monocytes (CD14⁺⁺CD16⁻/CD14⁺CD16⁺) burst and release nuclear DNA, blood monocytes were again incubated with C. albicans in the presence of complement active human serum together with the DNA marker SYTOX blue. Monocytes were stained for CD14 and CD16 to allow identification of the specific subtype. Upon contact with C. albicans, both types of monocytes (CD14⁺⁺CD16⁻ and CD14⁺CD16⁺) released DNA within 2–3 h (Figure 1F).

Having shown that, upon contact with C. albicans, monocytes respond to the fungal pathogen by undergoing nuclear changes and releasing their DNA, the extracellular DNA was further characterized. Monocytes were cultivated with C. albicans in growth medium with 10% NHS for 4 h then fixed and stained for DNA with SYTOX and evaluated by microscopy. The released extracellular DNA appeared as fibrous structures, which were formed between several monocytes and C. albicans (Figure 2A). The extracellular DNA release was triggered by C. albicans but not after addition of WGP or LPS. To exclude DNA release of monocytes by piercing of C. albicans formed hyphae monocytes were also incubated with hyphae locked mutant C. albicans strain (cph1Δ/efg1Δ). Monocytes also formed extracellular DNA traps within 2–4 h coincubation with mutant C. albicans cells (Figure S2 in Supplementary Material). Upon DNase treatment, the extracellular traps disappeared, confirming DNA comprising the traps (Figure 2B). The shape and appearance of this extracellular DNA was further determined by electron microscopy. The extracellular DNA released by monocytes (Figure 2C) formed cluster-like structures where groups of C. albicans were intertwined (Figure 2D). Furthermore, extracellular DNA formed dense nets, which were comprehensively deposited onto the surface of C. albicans cells and hyphae (Figures 2E,F). In addition, bead-like structures appeared on the DNA strings (Figure 2G), which most likely represent multiple proteins. These extracellular DNA covered the fungal surfaces (Figure 2H). These results demonstrate that monocytes liberate DNA to form extracellular trap-like structures, which we call MoETs that immobilize C. albicans cells and hyphae.

Extracellular trap formation is originally described for neutrophils and represents a specific response of suicidal cell death, which is distinct from apoptosis (25). To rule out the possibility that monocytes undergo rapid apoptosis and secondary necrosis in response to C. albicans, cellular morphology was also studied in monocytes undergoing apoptosis. Upon induction of apoptosis, blebs appeared at around 2.5 h on the cell surface. In contrast to trap forming monocytes nuclear DNA condensed followed by secondary necrosis and extracellular DNA after 5 h (Figure S3 in Supplementary Material). In summary, DNA release by monocytes via so-called MoETosis is a specific reaction by monocytes to combat microbes and is morphologically different to apoptosis.

In neutrophils, ROS trigger the dissociation of neutrophil elastase (NE) from a membrane-associated complex into the cytosol and activate its proteolytic activity in a MPO-dependent manner (26). NE is stored in azurophilic granules and contributes to antimicrobial activity in the phagosome. During NET formation, NE translocates from the granules to the nucleus and cleaves histones to promote chromatin decondensation. In this process, histone H3 becomes citrullinated and is liberated together with the DNA. To investigate whether MPO and elastase are active in MoETosis and whether histones are citrullinated in MoETs, extracellular DNA of monocytes was stained for serine protease elastase, MPO, and cit H3. MoETs were positively stained for all three proteins (Figures 3A–C). Similarly, MoETs were positively stained for LAC, a granule protein with antimicrobial activity (27) (Figure 3D). MPO, elastase, cit H3, and LAC within the extracellular traps covered the surface of C. albicans cells and hyphae. These results demonstrate that the DNA released from monocytes in response to C. albicans represents nuclear DNA and comprises antimicrobial proteins. Thus, the mechanism of MoETosis described here for monocytes is related to NETosis in neutrophils.

To clarify whether DNA release by monocytes is a specific reaction to C. albicans or a more general response to infectious microbes, monocytes were incubated with either the Gram-positive bacterium S. aureus or the Gram-negative bacterium E. coli. Using the same experimental setup as employed with C. albicans, monocytes were fixed after 4 h of incubation and DNA release was determined by laser scanning microscopy. Monocytes engulfed both the Gram-positive and Gram-negative bacteria and also released nuclear DNA (Figures 3E,F). Thus, human blood monocytes respond to the microbes C. albicans, S. aureus, and E. coli by spontaneous DNA release. Furthermore, the liberated DNA contained cit H3, a specific marker for extracellular traps.

To confirm monocyte origin of released DNA in response to C. albicans, monocytes and released DNA was stained for presence of CD14, a surface marker which is abundant on human monocytes and much less expressed on human neutrophils (Figure S4 in Supplementary Material). The monocyte and the released DNA showed strong CD14 staining after 4 h of coincubation with C. albicans (Figure 4A), confirming the monocytic origin. In order to confirm DNA release by monocytes in vivo formalin-fixed liver tissues from 6 h C. albicans-infected mice were stained with the monocytic markers CD115 and Ly6c, the granulocyte marker Ly6g and DNA stain SYTOX orange. Extracellular DNA was identified in the liver tissue which stained for CD115 and Ly6C (both markers for monocytes) and absence of Ly6G (marker for neutrophils) (Figure 4B). These data confirm in vivo trap formation.

Having shown that monocytes release DNA upon contact with the pathogenic fungus C. albicans, we asked whether this extracellular DNA affects pathogen survival. To this end, extracellular DNA released by monocytes upon incubation for 4 h with C. albicans was isolated. The isolated DNA and the associated proteins were separated by SDS-PAGE, proteins were immunoblotted, and the presence of cit H3 was assayed with anti-cit H3 antibodies. The antibody detected a specific band at 15 kDa, confirming the presence of cit H3 and MoETs (Figure 5A). cit H3 was not detected in human chromosomal DNA isolated by standard methods from THP-1 cells or from human blood monocytes. To test the antifungal activity of the isolated MoETs, the isolated MoETs were added to C. albicans and fungal growth was followed over 30 h by assaying the optical density at 604 nm. Treatment with this isolated extracellular trap DNA resulted in fungal growth inhibition. Fungal growth was reduced by about 45% after 6 h and by 35% after 24 h compared with C. albicans incubated in medium lacking DNA or incubated with monocyte or THP-1 chromosomal DNA (Figure 5A). Thus, MoETs isolated from human monocytes have antifungal activity and reduce C. albicans growth.

To determine the influence of serum factors on the extracellular trap formation, monocytes were incubated with C. albicans in 10% NHS or heat-inactivated 10% NHS, the released extracellular DNA was stained with SYTOX green, and the amount of released DNA was determined every hour for a total of 5 h. Monocytes cocultivated with C. albicans in the presence of NHS and inactive NHS released DNA into the medium, although presence of NHS aided the process. DNA release increased 5-fold after 1 h of incubation and 58-fold after 4 h of incubation compared with monocytes incubated in active serum but without C. albicans (Figure 5B). Neutrophils under the same conditions released 11-fold more DNA after 4 h in active NHS than neutrophils in active serum without C. albicans, but released 70–80% less DNA in total than monocytes. DNA release by neutrophils did not significantly differ in active and inactive NHS (Figure 5B). In summary, when challenged with C. albicans, monocytes released fourfold more DNA into the medium than neutrophils.

To investigate whether MoETs activate complement in human serum, monocytes were incubated with C. albicans and active human serum. After incubation and immobilization on coverslips, the samples were stained with C3b and C5b–9 antibodies and investigated by laser scanning microscopy analysis. C3b and C5b–9 were deposited on MoETs as seen by colocalization of SYTOX green staining of the MoETS with C3b (Figure 6A) and C5b–9 (Figure 6B). Similarly, NETs showed C3b deposition (Figure 6C). To find out whether complement regulator FH recognizes and binds C3b on the DNA, the traps were immunostained with FH antibodies. Both extracellular traps from monocytes and neutrophils bound FH from the serum (Figures 6D,E) indicating restriction of complement activation by FH on the surfaces of the traps. Having shown that human extracellular DNA traps bind FH in vitro, we aimed at identifying DNA traps with FH in tissue of C. albicans-infected mice. Therefore, ultra-thin sections of embedded liver derived from C. albicans-infected mice (24 h post-infection) were stained to detect DNA traps with FH. Extracellular DNA was identified in the sections which also stained with FH antibodies (Figure 6F). These results confirmed that FH bound to released nuclear DNA in the liver tissue.

To confirm in vitro binding of FH to the extracellular traps, NETs were isolated and immobilized to a microtiter plate via coated antibodies to citrullinated H3. The isolated immobilized NETs were incubated in NHS (10%) or with FH (10 µg/ml) and binding of FH was determined by ELISA. FH alone bound weakly to the NETs but attached to the NETs when recruited from NHS (Figure 7A). Complement inhibition by the addition of EDTA or EGTA to NHS substantially reduced FH binding to the NETs and suggested that FH was recruited by C3b to the NETs. Therefore, NETs were incubated in NHS activated via the alternative pathway and this time stained for C3b binding. C3b deposited on the NETs and binding was blocked by adding the complement inhibitor EDTA. Also, purified C3b bound to the NETs (Figure 7B). FH binding to the NETs from human serum was substantially reduced when NETs were incubated in C3-depleted serum compared to NHS (Figure 7C). FH alone did not bind to the NETs but together with C3b FH bound in a dose-dependent manner. Thus, FH binding to NETs is mediated by deposited C3b. To determine whether bound FH on the NETs is active, cofactor activity of bound FH was followed by Western blot analysis. Again NETs were immobilized to the microtiter surface via anti-cit H3 and incubated with C3b and FH. After washing off the unbound FH and C3b, NETs were incubated with C3b and factor I. C3 cleavage fragments in the supernatant were then detected by Western blot analysis. NETs bound FH-mediated cleavage of C3b by factor I as seen by C3b cleavage products at 68, 46, and 43 kDa (Figure 7D). Similarly, C3b was cleaved to iC3b, when NETs were incubated with NHS. These results demonstrate that NETs bound FH is functionally active and can mediate complement cofactor activity.

Extracellular DNA traps often stay for a period of time in the blood before they are cleaved by DNase and are engulfed by phagocytes (28). Complement opsonization enhances the uptake of microbes and apoptotic cells by phagocytes (29). However, in contrast to microbes apoptotic self-cells do not induce the inflammasome and do not secrete inflammatory cytokines like IL-1β (30). Clearance of extracellular DNA traps by macrophages was previously shown not only to induce inflammatory cytokines (31) but also to act non-inflammatory (28). The presence of microbes, antimicrobial peptides, or microbial substances like LPS in the traps induces an inflammatory response in macrophages. Here, we isolated NETs from PMA induced neutrophils, incubated these NETs with human monocytes and IL-1β concentrations were determined after 20 h in the supernatants by ELISA. IL-1β production by PMA isolated with the NETs was excluded (Figure S5 in Supplementary Material). NETs induced the inflammasome in monocytes as determined by higher IL-1β concentrations in the supernatants compared to monocytes alone. Binding of FH via C3b on the NETs reduced IL-1β secretion in monocytes (Figure 7E). In contrast, FH alone added to the NHS did not modulate the IL-1β response in monocytes (data not shown). Altogether, the data demonstrate that binding of FH to NETs inhibits the inflammatory response of monocytes.