An impaired neutrophil response to pathogenic fungi puts patients at risk for fungal infections with a high risk of morbidity and mortality. Acquired neutrophil dysfunction in the setting of iatrogenic immune modulators can include the inhibition of critical kinases such as spleen tyrosine kinase (Syk). In this study, we used an established system of conditionally immortalized mouse neutrophil progenitors to investigate the ability to augment Syk-deficient neutrophil function against Candida albicans with TLR agonist signaling. LPS, a known immunomodulatory molecule derived from Gram-negative bacteria, was capable of rescuing effector functions of Syk-deficient neutrophils, which are known to have poor fungicidal activity against Candida species. LPS priming of Syk-deficient mouse neutrophils demonstrates partial rescue of fungicidal activity, including phagocytosis, degranulation, and neutrophil swarming, but not reactive oxygen species production against C. albicans, in part due to c-Fos activation. Similarly, LPS priming of human neutrophils rescues fungicidal activity in the presence of pharmacologic inhibition of Syk and Bruton’s tyrosine kinase (Btk), both critical kinases in the innate immune response to fungi. In vivo, neutropenic mice were reconstituted with wild-type or Syk-deficient neutrophils and challenged i.p. with C. albicans. In this model, LPS improved wild-type neutrophil homing to the fungal challenge, although Syk-deficient neutrophils did not persist in vivo, speaking to its crucial role on in vivo persistence. Taken together, we identify TLR signaling as an alternate activation pathway capable of partially restoring neutrophil effector function against Candida in a Syk-independent manner.
Invasive fungal disease, and in particular invasive candidiasis (IC), affects people on a global scale, with >700,000 cases of IC annually (1). Of the Candida spp., C. albicans is the most common and accounts for ∼50% of the clinical isolates, depending on the geographic area (2, 3). Development of IC may be attributed to intrinsic patient risk factors including diabetes mellitus, inflammatory bowel diseases, increased age, or sepsis, as well as iatrogenic risk factors that increase the risk of IC, including the use of empiric antibiotics, central venous catheters, solid organ or stem cell transplant, chronic corticosteroid usage, and long-term hospital or intensive care unit stays (2, 4–6). Considering the substantial mortality rate of IC (6, 7), patient risk prediction and risk reduction are critical and evolving challenges.
C. albicans in healthy individuals can exist as a commensal organism on skin, oral, gastrointestinal, and/or vaginal microbiota. Risk factors associated with invasive Candida infection tend to coincide with alterations in the host immune system. The innate immune system, composed primarily of myeloid cells capable of rapid pathogen engagement, are the first cell types to interact with C. albicans. The most abundant of these innate cells are neutrophils, which provide first-line protection against many infections, including IC. Neutropenia is a major risk factor for IC, demonstrating the crucial role that neutrophils play in the prevention of candidemia and other fungal infections (8–12). In addition, several primary immune deficiencies linked to impaired neutrophil killing function, such as NADPH oxidase and CARD9 deficiency, often result in recurrent fungal infections (13, 14).
Neutrophils play a central part in executing an early and effective response to fungal infections. This process involves the initial recognition of C. albicans through the use of host pattern recognition receptors and binding of pathogen-associated molecular patterns. β-Glucan, a fungal cell wall carbohydrate, is one of the several pathogen-associated molecular patterns recognized by neutrophils, principally through the lectin receptor, Dectin-1 (15). Additional receptors known to contribute to the detection of C. albicans include TLRs, nucleotide-binding oligomerization domain (Nod)–like receptors, retinoid-inducible gene 1 protein (RIG1)–like receptors, as well as the complement receptor (16–19).
The binding of C. albicans carbohydrate cell wall Ags through lectin receptors promotes an intracellular protein signaling cascade predominantly through spleen tyrosine kinase (Syk), an integral step in the initiation of intracellular signaling and cell stimulation (20). This activation leads to downstream effector functions including phagocytosis of C. albicans, reactive oxygen species (ROS) production, degranulation, cytokine production, and expulsion of neutrophil extracellular traps (19, 21). Our previous work demonstrated a critical dependence on Syk in mouse neutrophils for a well-coordinated response to C. albicans (22).
Small-molecule kinase inhibitors (SMKIs) targeting Syk, Btk, and other tyrosine kinases are a family of emerging therapeutics efficacious in the treatment of a broad spectrum of diseases including lymphomas, leukemias, and immune thrombocytopenia; however, paired with these positive disease outcomes have been an increased rate of invasive fungal infections (23–26).
We sought to define the interplay of Syk-dependent and Syk-independent alternate signaling pathways capable of enhancing neutrophil fungicidal activity in the face of genetic Syk deletion or use of small-molecule kinase inhibitors. In this study, we investigated whether TLR signaling is capable of augmenting Syk-deficient neutrophil function. As a model, we use LPS, which is recognized by TLR4–MyD88 signaling, with a resulting downstream response leading to NF-κB activation and proinflammatory gene expression (27–29). Our results demonstrate partial restoration of Syk-deficient neutrophil function through TLR signaling, highlighting the existence of alternate pathways able to circumvent crucial tyrosine kinases such as Syk in neutrophil responses to the human fungal pathogen, C. albicans.
Materials and Methods
LPS (InvivoGen, San Diego, CA) was used at a concentration of 400 ng/ml. Cytochalasin D (Sigma-Aldrich, St. Louis, MO) was used at a concentration of 20 nM. For assays utilizing a TLR panel of ligands for TLR1 through TLR9 activation was purchased from InvivoGen and used as suggested by the manufacturer. R406, a selective Syk inhibitor (Selleck Chemicals, Houston, TX), was used at a concentration of 20 µM. Ibrutinib, a Btk inhibitor (Cayman Chemical, Ann Harbor, MI), was used at a concentration of 1 µM. Nonidet P-40 was purchased from American Bioanalytical (Natick, MA). Flow cytometry staining buffer (FACS buffer) was composed of 2% heat-inactivated FBS (Life Technologies, Dun Laoghaire, Ireland) and 1 mM EDTA (Life Technologies) in PBS (Corning, Corning, NY). Formalin (10%) was purchased through Sigma-Aldrich. To culture C. albicans, YPD (yeast extract, peptone, dextrose) broth was used, composed of 1% yeast extract (Acros Organics, Fair Lawn, NJ), 2% peptone (BD Biosciences, San Jose, CA), and 2% dextrose (Sigma-Aldrich). Complete RPMI (cRPMI) was made up of RPMI 1640 (Corning) with 2 mM l-glutamine (Life Technologies), 10% heat-inactivated FBS and 1% penicillin-streptomycin. RPMI-MOPS contained RPMI 1640 with 2% glucose (Sigma-Aldrich) and 0.165 M MOPS (Thermo Fisher Scientific) buffered to pH 7. β-Estradiol positive (E2+) medium is composed of cRPMI plus stem cell factor added as a conditioned medium from stem cell factor–overexpressing Chinese hamster ovary (CHO) cells to a final concentration of 1% with 0.5 mM β-estradiol (Sigma-Aldrich). Swarming medium is composed of IMDM plus 20% FBS. Alexa Fluor 647 (AF647) or Alexa Fluor 405 (AF405) succinimidyl ester was purchased through Thermo Fisher Scientific for labeling of C. albicans. Anti-CD11b-PE Ab (CD11b-PE) and anti-CD45.2-AF647 (CD45.2-AF647) were purchased through BioLegend, (San Diego, CA). Hoechst 33342 stain and Sytox Green were both purchased from Thermo Fisher Scientific.
C. albicans preparation
SC5314 wild-type (WT) C. albicans was purchased from the American Type Culture Collection (Manassas, VA). Far-red fluorescent protein–expressing SC5314 (iRFP C. albicans) was a gift from Robert Wheeler (University of Maine, Orono, ME) (30). SC5314 was cultured in YPD broth overnight at 30°C on a shaking platform, washed, and resuspended in PBS and then counted using a LUNA automated fluorescence cell counter (VitaScientific, Beltsville, MD). C. albicans hyphae or pseudo hyphae (roughly twice the length of a yeast body) were produced by incubating the washed yeast in cRPMI for up to 2 h at 37°C, 5% CO2. Heat-killed (HK) C. albicans is generated by incubating C. albicans in PBS in a heat block at 95°C for 10 min. C. albicans was fluorescently labeled with AF647 or AF405 at 1 ng/ml for 30 min in the dark at room temperature while shaking and then washed and resuspended in PBS.
Primary human neutrophils
Healthy donors were consented under a Massachusetts General Hospital Institutional Review Board–approved protocol and whole blood was drawn the same day of experiments. In brief, whole blood was drawn into an EDTA-coated blood tube (BD Biosciences) and centrifuged at 1500 × g for 15 min at room temperature. The top plasma layer was aspirated, and the buffy coat was subjected to neutrophil isolation by negative selection using the EasySep direct human neutrophil isolation kit (STEMCELL Technologies, Seattle, WA). To confirm the purity of neutrophils at the end of the isolation process, Wright–Giemsa staining (Thermo Fisher Scientific) was performed as previously reported and viabilities were verified via live/dead staining with acridine orange/propidium iodine staining and counting on the LUNA automated fluorescent cell counter at a 1:10 dilution. Both were reported at or above 87% (31, 32).
Murine neutrophil progenitor cell lines
WT and Syk-deficient (Syk knockout [KO]) conditionally immortalized neutrophil progenitor cell lines were established and characterized as previously described (22). In brief, bone marrow cells from Cas9 transgenic mice (33) were transduced using a retrovirus encoding ER-Hoxb8 allowing conditional immortalization of granulocyte-macrophage progenitors (34). In the presence of β-estradiol, granulocyte-macrophage progenitors remain in an undifferentiated and self-renewing state. Upon removal of β-estradiol, the ER-Hoxb8 is no longer active in the nucleus and the cells undergo synchronous differentiation from granulocyte-macrophage progenitors into mature neutrophils (35). Assays of mature neutrophils were completed in cRPMI unless otherwise noted.
C. albicans killing assay
Mature neutrophils were plated in a clear flat-bottom 96-well plate (Corning) at a concentration of 1 × 105 cells per well. Ibrutinib, R406, or DMSO control were added to each well and incubated at 37°C, 5% CO2 for 30 min. Then, LPS or dH2O control was added to each well and incubated at 37°C, 5% CO2 for an additional 30 min. Live C. albicans was then cocultured with neutrophils at a multiplicity of infection (MOI) of 2. A serial dilution series of C. albicans only was plated to establish a standard curve. Following this incubation, 1% Nonidet P-40 containing 10 mM Tris HCl, 150 mM sodium chloride, and 5 mM magnesium chloride at pH 7.5 was added for complete lysis of the neutrophils. Proceeding lysis, MOPS-RPMI and 10% PrestoBlue (Thermo Fisher Scientific) was added to each well and read continuously on an SpectraMax i3x reader (Molecular Devices, San Jose, CA) at wavelengths of 560 and 590 nm every hour for 18 h at 30°C. The inflection point of each well reading was determined, and the amount of live yeast in each well was determined using the yeast standard curve. Percent killing was determined using the equation [1 − (no. of C. albicans in well with neutrophils/no. of C. albicans alone)] × 100, based on control wells containing C. albicans alone as maximal growth and compared with WT plus vehicle.
Neutrophil gelatinase-associated lipocalin
Neutrophils were plated in a clear flat-bottom 96-well plate (Corning) at 1 × 105 cells per well with LPS or vehicle control and incubated at 37°C, 5% CO2 for 10 min. Live C. albicans was then added at an MOI of 10 and coincubated for 2 h at 37°C, 5% CO2. Supernatant was collected and diluted to 1:2048 in reagent diluent (R&D Systems, Minneapolis, MN). Diluted supernatants were applied to a mouse lipocalin-2/neutrophil gelatinase-associated lipocalin (NGAL) DuoSet ELISA and run according to the manufacturer’s protocol to determine the concentration of NGAL. The ELISA was analyzed on a SpectraMax i3x reader at 650 nm. NGAL concentrations were determined by comparison with a NGAL standard curve and expressed as pg/ml.
In a white-walled flat-bottom 96-well plate (Corning), 1 × 105 neutrophils were added to each well followed by a given TLR agonist or its respective vehicle control. The plate was then incubated at 37°C, 5% CO2 for 30 min. After, heat killing C. albicans, hyphae or vehicle control was added to each well at a ratio of 20:1 (C. albicans to neutrophils), followed by lucigenin (Thermo Fisher Scientific) at a final concentration of 15 μM in DMSO. The plate was read in a SpectraMax i3x with an analysis of luminescence at wavelength 578 nm every 5 min for 4 h at 37°C. Data are expressed as a sum of all readings over time in arbitrary units and compared with WT plus LPS plus HK C. albicans.
Neutrophils were added to microcentrifuge tubes at 5 × 106 per ml in cRPMI along with LPS, cytochalasin D, or vehicle control and incubated for 30 min at 37°C, 5% CO2 followed by the addition of HK C. albicans hyphae labeled with AF647 at a ratio of 10:1 (pathogen to neutrophils) and coincubated for 4.5 h at 37°C, 5% CO2. Following coincubation, the medium was washed out and the samples were fixed in 10% formalin and stained for CD11b-PE in FACS buffer at 1:50 for 30 min in the dark on ice. Flow cytometry was performed on a BD FACSCelesta (BD Biosciences). Percent phagocytosis was determined by gating on CD11b+ neutrophils and determining the percentage of those events that are double positive CD11b+/AF647+, which represent Candida-bound neutrophils. Percent double-positive events were calculated using FlowJo 10 software (BD Biosciences).
Mouse neutrophils were incubated in a microcentrifuge tube at 2.5 × 106 per ml in cRPMI along with LPS (400 ng/ml) and/or HK C. albicans hyphae at a ratio of 10:1 (pathogen to neutrophils) for 40 min at 37°C, 5% CO2. Cells were then collected and lysed with Laemmli sample buffer (Bio-Rad, Hercules, CA) containing protease inhibitors (cOmplete mini; Roche Diagnostics, Indianapolis, IN), reducing agent (NuPAGE sample reducing agent; Thermo Fisher Scientific), and phosphatase inhibitors (sodium orthovanadate; New England Biolabs, Ipswich, MA). Proteins were resolved by SDS-PAGE, transferred onto a polyvinylidene difluoride membrane, and blocked with PBS with 0.1% Tween 20/5% BSA (Sigma-Aldrich). Phosphorylated c-Fos levels were probed using rabbit mAb D82C12 anti–phospho-c-Fos (Cell Signaling Technology, Danvers, MA) (1:1000) and total c-Fos levels were probed using rabbit mAb 9F6 anti–c-Fos (Cell Signaling Technology). For β-actin, mouse mAb AC-15 anti–β-actin (Sigma-Aldrich) (1:200,000) was used to confirm equal loading.
Neutrophil swarming assay
Neutrophils were added to microcentrifuge tubes at 5 × 105 cells per ml followed by Hoechst staining at 1:500 for 10 min at 37°C, 5% CO2. The cells were washed with PBS and resuspended in swarming media. R406, ibrutinib, or vehicle control were added, and the cells were preincubated at 37°C, 5% CO2 for 30 min, followed by the addition of LPS or vehicle control and again incubated at 37°C, 5% CO2 for an additional 30 min. Sytox Green was then added to a final concentration of 0.5 μM in swarming media. Swarming experiments and analysis were performed as previously described (36). In brief, 16-well ProPlate wells (Grace Bio-Labs, Bend, OR) were bound to glass slides with printed poly-l-lysine/ZETAG arrays. A solution of C. albicans yeast was added for 10 min to bind to the slides and then subsequently washed to remove unbound pathogen. The neutrophils from the microcentrifuge tubes were then transferred to the appropriate well and time-lapse imaging of selected swarming targets was performed using an automated Nikon Ti-E microscope (Minato City, Tokyo, Japan) with a ×10 Plan Fluor Ph1 DLL (numerical aperture = 0.3) lens. The area of fungal growth was determined by manually outlining the perimeter of fungal growth for each fungal cluster after 16 h using Fiji (ImageJ v2.0.0-rc 59/1.52p, National Institutes of Health) software.
In vivo homing to C. albicans
B6.SJL-Ptprca Pepcb/BoyJ (Pep Boy) mice (The Jackson Laboratory, Bar Harbor, ME) were cared for and housed in accordance with the Institutional Animal Care and Use Committee guidelines, and prior approval for experiments was obtained. Mice were lethally irradiated with a total of 900 cGy through x-ray irradiation in two equally fractionated doses 4–6 h apart. Mice were rested for 1 d followed by retro-orbital injection of either WT or Syk KO granulocyte-macrophage progenitors in PBS (total of 40 × 106 cells over two transfusions). Mice were then housed for a subsequent 4 d to allow the granulocyte-macrophage progenitors to fully mature into neutrophils. On the fourth day after cell transfusions the mice were challenged with either PBS, LPS, HK C. albicans plus PBS, or HK C. albicans plus LPS i.p. C. albicans was prior labeled with succinimidyl ester-AF405. After 5 h, mice were euthanized and peritoneal lavage was collected along with bone marrow. Samples were immediately fixed in 10% formalin, then washed and stored in FACS buffer. Cells were stained with CD11b-PE and CD45.2-AF647 and subjected to flow analysis. Adoptively transferred WT and Syk KO cells were identified by CD45.2 labeling in the CD45.1 recipient mice. The percentages of neutrophils in lavage were determined by gating on CD45.2 and CD11b double-positive events, and the percentages of transfused cells in the bone marrow were determined by gating on CD45.2+ cells.
All statistical analyses for normality and significance were performed using GraphPad Prism 9 software (San Diego, CA). A p value <0.05 was considered significant.
LPS improves the fungicidal capacity of Syk-deficient or -inhibited murine neutrophils
We investigated the capacity to prime mouse neutrophils toward the killing of C. albicans (Fig. 1A). Without LPS priming, WT neutrophils kill significantly more C. albicans than do Syk KO neutrophils. When WT and Syk KO neutrophils are primed with LPS, the fungicidal capacity of neutrophils is significantly improved, with Syk KO neutrophil levels reaching those of WT neutrophils. We then investigated the ability for LPS to rescue WT neutrophils from drug inhibition of Syk using R406 (Fig. 1B). R406 alone results in complete loss of neutrophil killing capacity of C. albicans. However, LPS was able to partially rescue WT neutrophil fungicidal activity in the presence of Syk.
LPS priming improves phagocytosis, degranulation, and swarming but not ROS production of Syk-deficient neutrophils
Having observed the clear improvement in killing of C. albicans, we sought to determine which of the neutrophil effector mechanisms were improved following LPS priming. Using a flow cytometry–based phagocytosis assay, we gated on live CD11b+ cells to select for mature neutrophils (Fig. 2A) and determined the percentage that were physically bound to AF647-labeled HK C. albicans. As a control, we used the same gating strategy with C. albicans alone to demonstrate that there are no events in the “neutrophil-bound C. albicans” gate, confirming that these events represent double-positive neutrophil/C. albicans cells. We next determined the percentage phagocytosis of HK C. albicans by WT and Syk KO neutrophils. We found Syk KO neutrophils are severely impaired in their ability to bind HK C. albicans (Fig. 2B). Priming WT and Syk KO neutrophils with LPS significantly improved phagocytosis in both cell types, while cytochalasin D, an actin polymerization inhibitor, acted as a potent suppressor of phagocytosis.
We then quantified the ability of WT and Syk KO neutrophils to degranulate via measurement of NGAL release into supernatant following live C. albicans stimulation. Compared to their resting state, Syk KO neutrophils had impaired NGAL degranulation compared with the WT, although NGAL release significantly augmented when stimulated with LPS, C. albicans, or both stimuli (Fig. 2C).
Next, we investigated the ability for these cells to engage in a coordinated multistep response to C. albicans as determined by swarming function of WT and Syk KO neutrophils primed with LPS (Fig. 2D). WT neutrophils significantly suppress fungal growth compared with C. albicans alone or Syk KO neutrophils plus C. albicans. LPS priming did not improve the ability of WT neutrophils to suppress fungal growth, and although LPS priming of Syk KO neutrophils did enhance control of fungal growth, the effect was not statistically significant.
Lastly, we determined ROS production of WT and Syk KO neutrophils in response to HK C. albicans hyphae with and without LPS (Fig. 2E). LPS did not boost ROS production in WT or Syk KO neutrophils, although the response was more robust following stimulation of WT neutrophils with the combination of HK C. albicans hyphae with and without LPS. Syk KO neutrophil ROS production was not significantly improved with either HK C. albicans hyphae or hyphae in combination with LPS.
Specific TLR receptor pathway activation improves candidacidal killing by Syk-deficient neutrophils
Given the results of TLR4 signaling by LPS, we examined whether other common TLR receptor agonists could improve the response of Syk KO neutrophils against C. albicans. Percent killing by Syk KO neutrophils was improved following priming with ligands to TLRs 1/2, 2, 4, 5, and 6/2 (Fig. 3A); conversely, ligation of the intracellular receptors, that is, TLRs 3, 7, and 9, decreased Syk KO neutrophil killing.
We also investigated the ability of TLR ligation to augment ROS production from Syk KO neutrophils (Fig. 3B, 3C). Values were based on the percentage of ROS production of WT plus LPS plus C. albicans. WT neutrophils did not produce significantly more ROS when primed with LPS; however, HK C. albicans boosted production compared with LPS alone, and both stimulants together improved production above HK C. albicans alone. Across all TLR ligands, there was not a significant improvement in ROS production by Syk KO neutrophils in response to LPS, HK C. albicans, or LPS plus HK C. albicans compared with vehicle.
LPS modulates killing of human neutrophils against C. albicans in the setting of kinase inhibitors
We also sought to determine whether human neutrophil fungicidal activity is augmented in the presence of LPS. Primary human neutrophils treated with inhibitors of Syk (R406) or Btk (ibrutinib) were primed with LPS and then challenged with C. albicans (Fig. 4A). LPS augmented human neutrophil killing against C. albicans whereas ibrutinib significantly decreased the killing capacity of human neutrophils. LPS, however, was able to rescue the inhibitory effect of ibrutinib on fungicidal activity of neutrophils to those similar to control neutrophils. R406 also significantly decreased human neutrophil killing and, in a similar process, LPS priming was able to partially restore fungicidal activity against C. albicans.
Additionally, we analyzed the ability of human neutrophils to restrict the C. albicans area of fungal growth through swarming in the setting of Syk inhibition. Bright-field and fluorescence microscopy images were taken at designated time points of wells containing either C. albicans alone, C. albicans plus neutrophils, C. albicans plus neutrophils and R406, or C. albicans plus neutrophils plus R406 and LPS. We visualized iRFP C. albicans and neutrophil nuclei through Hoechst staining (Fig. 4B). Analyzing the area of fungal growth when normalized to C. albicans alone, the presence of R406 causes severe limitation of the ability of neutrophils to restrict the fungal growth of C. albicans compared with vehicle control. However, when the neutrophils are primed with LPS prior to Syk inhibition, there is a partial restoration of the ability to suppress fungal growth compared with control. As a control, we tested whether LPS priming alone had influence on the suppression of fungal growth versus vehicle control, which showed that there was no LPS effect on neutrophil function.
WT but not Syk-deficient transfused neutrophil progenitors persist in the bone marrow and mature into neutrophils
The ER-Hoxb8 granulocyte-macrophage progenitors were transfused into recipient mice following conditioning with radiation (37). The transfused progenitors home to the bone marrow of the recipient where they undergo normal maturation to terminally differentiated neutrophils. This provides a model system in which to study the in vivo activity of WT versus Syk-deficient neutrophil function.
To investigate the capacity for engraftment, differentiation, and homing to the site of a HK C. albicans challenge, we transfused WT and Syk-deficient granulocyte-macrophage progenitors (CD45.2) into ablated CD45.1 mice. After 4 d, mice were challenged i.p. with HK C. albicans, and after 5 h the mice were euthanized for analysis of their bone marrow and i.p. lavage.
We gated on live transfused cells as marked by CD45.2 expression, which were present in all mice (Fig. 5A). There was a significantly larger percentage of transfused WT cells in the bone marrow compared with Syk KO. Furthermore, the percentage of mature neutrophils in the bone marrow, as defined by being CD45.2 and CD11b double-positive, was significantly higher in recipients of WT cells compared with Syk KO (Fig. 5C), suggesting that Syk may play a role in granulocyte-macrophage progenitor persistence in bone marrow.
Neutrophils lacking Syk are incapable of homing to an i.p. C. albicans challenge
We also examined whether the transfused cells could home to the site of an infection following an i.p. challenge with HK C. albicans. Following peritoneal lavage, the number of CD45.2 and CD11b double-positive cells in the lavage was determined, and we found a statistically significant increase in recruited neutrophils in mice that received WT granulocyte-macrophage progenitors that were challenged with HK C. albicans plus LPS compared with unchallenged mice (Fig. 6). There was no significant change in WT granulocyte-macrophage progenitor recipient mice that received vehicle control, LPS, or HK C. albicans plus vehicle. Among the groups that were transfused with Syk KO granulocyte-macrophage progenitors, there was not a significant difference in mice that received vehicle control, LPS, or HK C. albicans with or without LPS compared with no challenge.
Syk-independent c-Fos activation in WT and Syk KO neutrophils following Candida and LPS stimulation
To investigate potential alternate pathways of neutrophil activation capable of circumventing loss of Syk, we analyzed c-Fos activation in response to LPS and HK C. albicans hyphae through Western blot analysis. WT neutrophils showed minimal levels of phosphorylated c-Fos that became more pronounced following stimulation with LPS or LPS plus HK C. albicans hyphae. Syk KO neutrophils likewise did not show robust levels of c-Fos phosphorylation at rest or with stimulated LPS or HK C. albicans hyphae; however, prominent activation was detected in the presence of both LPS and C. albicans, suggesting activation through a Syk-independent alternate pathway (Fig. 7). As a control, LPS plus HK C. albicans hyphae alone revealed no detectable levels of phosphorylated c-Fos. Despite even protein loading as shown by β-actin, fluctuations in total c-Fos were seen across the conditions. WT neutrophils at rest, Syk KO neutrophils at rest, and Syk KO neutrophils stimulated with HK C. albicans hyphae appeared to have minimal levels of c-Fos, whereas the other conditions yielded robust levels of total c-Fos. These data suggest that the total available pool of c-Fos, as well as the fraction that results in a phosphorylated and activated state, is dynamic depending on stimulation condition.
Neutrophils rely on a critical set of receptors for initial recognition of invasive fungal pathogens and subsequent activation of a fungicidal response. Many of these receptors belong to the lectin family, which rely on Syk for intracellular signaling and cellular activation. We have previously demonstrated an essential reliance on Syk for neutrophil response against Candida spp. (22). In this study, we sought to determine whether alternate pathways exist in Syk-deficient neutrophils capable of restoring function against C. albicans. Given TLR expression on neutrophils, we examined TLR agonists as potential alternative activation signaling pathways. Previous data support the augmentation of neutrophil fungicidal activity through TLR4 recognition of LPS (38). Interestingly, TLR4 also has the capacity to recognize C. albicans through the recognition of cell wall O-mannan (39) and to trigger downstream Syk activation, further complicating potential mechanisms of Syk-independent activation through TLRs (40).
In this study, we present data that support our hypothesis of TLR-mediated rescue of neutrophil function in the setting of Syk signaling deficits, both in mouse and human neutrophils. This improvement was seen across many, but not all, effector functions, including phagocytosis, degranulation, ROS production, and neutrophil swarming (Fig. 8). For example, ROS production was not significantly improved in the Syk KO neutrophils when responding to C. albicans in the setting of LPS priming, which corroborates other published work that LPS is not a potent activator of ROS in mouse or human neutrophils (41, 42). This finding was confirmed to likely be a common feature of TLR signaling, as no members of a TLR agonist panel were able to significantly boost ROS production in Syk KO neutrophils. In contrast, depending on the TLR receptor activated, there appear to be differences in neutrophil augmentation. Syk KO neutrophil killing of C. albicans was improved by priming with Pam3CSK4, HK Listeria monocytogenes (HKLM), and FSL-1, ligands to the TLRs 1/2, 2, and 6/2, respectively. Pam3CSK4 has been previously reported to induce a primed human neutrophil state, including shedding of L-selectin and induction of NADPH oxidase activity, which may account for the increased fungicidal activity seen in our mouse neutrophils (43). Interestingly, HKLM has previously been shown to boost candidacidal activity of peritoneal cells as well as evade destruction through activation of macrophage mitophagy (44, 45). In part, low mitochondria density in neutrophils may play a role in suppressing HKLM evasion. Additionally, work has been done to show that agonists to TLRs 1/2, 2/6, and 4 improve macrophage phagocytosis of zymosan particles, whereas agonist to TLRs 3 and 5 do not (46). Taken together, our data suggest that specific TLR activation can influence downstream neutrophil effector functions with improved fungicidal activity.
LPS could restore specific neutrophil functions, including phagocytosis and swarming. Syk KO neutrophils without LPS demonstrate nearly complete absence of a response; however, following LPS priming, the ability to phagocytose and swarm against C. albicans was significantly improved. One possible explanation is the differential expression of pattern recognition receptors following LPS priming such as upregulation of CD11b expression (47, 48), which is known to have the capacity to bind fungal cell wall components including β-glucan (49), although CD11b relies, in part, on Syk for subsequent signaling and cellular activation (20). The exact molecular mechanism that defines rescue of Syk-deficient neutrophils through a LPS-dependent pathway will require additional investigation.
As with mouse neutrophils, human neutrophil function is also linked to Syk activity. R406, a small molecule inhibitor of Syk, severely impairs human neutrophil capacity to kill C. albicans and limits neutrophil swarming. Using R406 and ibrutinib, a Btk inhibitor, which also plays a critical role in the neutrophil response to invasive fungal and bacterial infections (26, 50), we examined whether LPS was able to rescue human neutrophil dysfunction. Our data show a restoration of human neutrophil function in the setting of kinase inhibition, suggesting that these alternate activation pathways appear to be conserved in both mouse and human neutrophils.
We next asked whether restoration of function might be replicated in vivo. Following bone marrow ablation, WT and Syk-deficient granulocyte-macrophage progenitors were transfused in irradiated mouse recipients. Interestingly, there was a stark difference in the proportion of WT versus Syk-deficient transfused cells that could home and persist in the bone marrow. WT granulocyte-macrophage progenitors composed 85–90% of the bone marrow cells analyzed, whereas <15% of Syk KO granulocyte-macrophage progenitors were found, raising the possibility of a critical role of Syk in granulocyte-macrophage progenitor engraftment, persistence, or survival in the bone marrow. Syk has been implicated in general hematopoiesis (51), and our previous work showed that Syk protein levels increase from the granulocyte-macrophage progenitor to the neutrophil developmental pathway (22). Despite not being an abundant kinase in early granulocyte-macrophage progenitor stages, it does raise the possibility of Syk playing a role in the bone marrow microenvironment as granulocyte-macrophage progenitors mature into neutrophils. The influence of Syk on granulocyte-macrophage progenitors appears to be an in vivo phenomenon specific to the bone marrow microenvironment, as previous data indicate that Syk-deficient neutrophils differentiate and mature in vitro without loss of viability, granule protein production, or expression of surface markers (22). The impact of Syk loss may also include the possible impairment of neutrophil homing to sites of inflammation given the contribution of Syk to molecular machinery responsible for adhesion, rolling, and extravasation from circulation. Interestingly, LPS dramatically improved the recruitment of WT neutrophils to peritoneal fungal inflammation, consistent with other studies using LPS priming of tissue-resident macrophages in antifungal responses (52). These Syk-independent pathways highlight possible therapeutic avenues in patients for more precision-based immune augmentation.
The mechanisms of improvement seen following LPS priming in the setting of Syk deficiency remain unclear. One possibility is crosstalk or feedthrough between TLR4 and Dectin-1 signaling pathways. We investigated possible downstream targets that could act as a bridge. In macrophages, activation downstream of Syk and TLR4 involves c-Fos–dependent cytokine release (53). Further investigation identified c-Fos activation in response to C. albicans infections in human epithelial cells (54). Our Western blot analysis showed there to be differential levels of c-Fos phosphorylation when WT and Syk-deficient neutrophils were stimulated with LPS, C. albicans, or both. Although the total levels of c-Fos fluctuated as well, there was increased c-Fos phosphorylation with LPS and C. albicans compared with either stimulus alone. This seemingly synergistic effect may shed insight on downstream activating proteins of TLR4 and Dectin-1 signaling, although further investigation is required to further define the molecular mechanisms of this crosstalk pathway.
These data show that loss or inhibition of essential kinases that significantly impair the neutrophil response to C. albicans can be partially rescued through priming with LPS. Syk-deficient mouse neutrophil effector functions are significantly diminished; however, LPS priming can provide a robust, augmented response to some, but not all, of these, suggesting specificity to the restoration. Specifically, phagocytosis, degranulation, swarming, and overall fungicidal activity were all improved. In contrast, ROS production was not improved, and further studies are needed to define where this signaling potentiation may be occurring and whether it is dependent on activator chaperones downstream of Syk. Despite differences between human and mouse neutrophils, including diminished myeloperoxidase levels and absent defensins (55), LPS rescue of neutrophil function also restored killing and swarming of impaired human neutrophils in the setting of small molecular inhibitors to Syk and Btk. As more kinase inhibitors are approved and brought to market to treat a broad spectrum of human diseases, these results provide insight into how we may address neutrophil dysfunction in the face of these drug interactions and improve outcomes for patients with invasive fungal infections.
This work was supported in whole or in part by a Shriners Fellowship (to A.H.), National Cancer Institute Grant K08 CA201640 (to D.B.S.), National Institutes of Health/National Institute of General Medical Sciences Grant R01 GM092804 (to D.I.), and National Institutes of Health/National Institute of Allergy and Infectious Diseases Grant R01 AI132638 (to M.K.M.).
The author list was determined by contribution. A.L.V., N.J. Alexander, A.H., A.K.S., D.B.S., and M.K.M. contributed to experimental design. A.L.V., K.D.T., N.J. Alexander, R.B., J.M., A.H., A.K.S., D.B.S., and M.K.M. executed the experiments. A.L.V., N.J. Alexander, A.H., A.K.S., D.B.S., and M.K.M. performed analysis and interpretation of the experiments. A.L.V., K.D.T., N.J. Alexander, R.B., J.M., A.H., N.J. Atallah, A.K.S., D.B.S., D.I., and M.K.M. contributed to writing the manuscript. Fig. 8 was created with recommendations by A.L.V. and M.K.M.
Abbreviations used in this article:
Alexa Fluor 405
Alexa Fluor 647
Bruton’s tyrosine kinase
HK Listeria monocytogenes
multiplicity of infection
neutrophil gelatinase-associated lipocalin
reactive oxygen species
spleen tyrosine kinase
M.K.M. reports consultation fees from Vericel, Pulsethera, NED Biosystems, GenMark Diagnostics, Clear Creek Bio, and Day Zero Diagnostics; grant support from Thermo Fisher Scientific and Genentech; and medical editing/writing fees from UpToDate, outside the submitted work. M.K.M. also reports patents 14/110,443 and 15/999,463 pending. D.B.S. is a co-founder and holds equity in Clear Creek Bio, is a consultant in SAFI BioSolutions, and is a consultant for Keros Therapeutics. The other authors have no financial conflicts of interest.