Transcellular biosynthesis of leukotriene B4 orchestrates neutrophil swarming to fungi

Summary Neutrophil swarming is an emergent host defense mechanism triggered by targets larger than a single neutrophil’s capacity to phagocytose. Swarming synergizes neutrophil functions, including chemotaxis, phagocytosis, and reactive oxygen species (ROS) production, and coordinates their deployment by many interacting neutrophils. The potent inflammatory lipid mediator leukotriene B4 (LTB4) has been established as central to orchestrating neutrophil activities during swarming. However, the details regarding how this eicosanoid choreographs the neutrophils involved in swarming are not well explained. Here we leverage microfluidics, genetically deficient mouse cells, and targeted metabolipidomic analysis to demonstrate that transcellular biosynthesis occurs among neutrophils to generate LTB4. Furthermore, transcellular biosynthesis is an entirely sufficient means of generating LTB4 for the purposes of orchestrating neutrophil swarming. These results further our understanding of how neutrophils coordinate their activities during swarming, which will be critical in the design of eventual therapies that can harness the power of swarming behavior.


INTRODUCTION
Neutrophils have long been known for their critical role in the protection against fungal infections, featuring an armament of antimicrobial defenses (Desai and Lionakis, 2018). Among the most recently described is the behavior of swarming, during which neutrophils coordinate their own exponential recruitment to concentrate antimicrobial action against large targets (Kienle and Lä mmermann, 2016). The role of LTB 4 as a critical mediator of neutrophil swarming has been well established in mice and humans by lipidomic analysis, antagonizing LTB 4 receptors, inhibiting LTB 4 biosynthesis, genetically manipulating intracellular LTB 4 signaling, and disrupting LTB 4 biosynthesis pathways Lammermann et al., 2013;Malawista et al., 2008;Reategui et al., 2017). Despite these advances, details of LTB 4 biosynthesis and transport by neutrophils while swarming remain largely unexplored. A hint to potential complexity is provided by earlier studies showing that during inflammation, neutrophils exchange significant quantities of eicosanoid intermediates with other immune and non-immune cells (Serhan et al., 1984a(Serhan et al., , 2020. Transcellular eicosanoid biosynthesis adds flexibility and robustness to coordinate responses resulting from interactions between neutrophils and platelets Kienle et al., 2021;Lammermann et al., 2013;Reategui et al., 2017), neutrophils and red blood cells (Stern and Serhan, 1989), neutrophils and endothelial cells (Claesson and Haeggstrom, 1988), neutrophils and airway epithelial cells, (Bigby et al., 1989) neutrophils and epidermal cells (Sola et al., 1992), neutrophils and lymphocytes (Odlander et al., 1988), etc. However, in this rich context, it is unknown if transcellular LTB 4 biosynthesis plays a role in coordinating neutrophil-neutrophil interactions during swarming. Here, we employ microfluidics, genetically deficient mouse cells, and targeted metabolipidomic analysis to probe the role of transcellular biosynthesis of LTB 4 during neutrophil swarming and restriction of pathogen growth.

RESULTS
We tested the ability of LTB 4 to restore the functions of neutrophils from mice with knockout genotypes at two critical steps in the LTB 4 synthesis pathway: 5-lipoxygenase (5-LOX) and leukotriene A 4 hydrolase (LTA 4 H), encoded by alox5 and lta4h, respectively (Wan et al., 2017). We verified that there were no significant differences in expression of the primary LTB 4 receptor BLT1 ( Figure S1A) and no differences in chemotaxis toward LTB 4 ( Figure 1A) between knockout and wild-type cell counterparts (C57/BL6 vs. Figure 1. Neutrophil swarming is absent and cannot be restored by LTB 4 in neutrophils from alox5 À/À and lta 4 h À/À mice, whereas common neutrophil functions are comparable to wild-type or could be restored by LTB 4 (A) Transmigration toward LTB 4 (0.2 ng/mL) through a membrane with 3 mm pores is comparable for bone marrow cells from B6 (C57/BL6 mice, the wild-type control for the alox5 À/À mice) and alox5 À/À or S129 (129S1/SvImJ mice, the wildtype control for the lta4hÀ/À mice) and lta 4 h À/À mice. There are no significant differences within the HBSS or the LTB 4 groups.
(B and C) The ability to phagocytose (B)C. albicans and produce ROS (C) by enriched neutrophils from alox5 À/À mice is restored by LTB 4 (0.6 nM) to levels comparable to C57 mice. N = 6 mice per genotype across 2 independent experiments. Phagocytosis events were differentiated from cell surface adherence events with the cytoskeletal inhibitor cytochalasin D (CytoD) at a concentration of 30 mM. (D and E) Neutrophils enriched from the bone marrow of B6 and alox5 À/À mice, 500,000 neutrophils per genotype. Concentrations of 10 nM or 1 nM refer to LTB 4 . (D) The amount of fungal growth of C. albicans was quantified at 16 h after the start of the assay. CA refers to Candida alone, a condition in which only media is added to live Candida albicans targets. NR 282 swarms across three independent experiments. (E) The area covered by the neutrophil swarm was quantified at the indicated timepoint. N = 48 swarms across 3 independent experiments. Mean and SD are shown, except for A, which is SEM ****p%0.0001 by Kruskal-Wallis or one way ANOVA. See also Figure S1. iScience Article alox5 À/À and S129 vs. lta4h À/À ). Consistent with the essential roles of LTB 4 in stimulating neutrophil phagocytosis and ROS production, we found significant differences in phagocytosis and ROS production between knockout and wild-type cells. These differences were corrected by the addition of exogenous LTB 4 ( Figures 1B, 1C and S1B).
Next, we tested the swarming of mouse neutrophils triggered by 100 mm diameter clusters of live Candida albicans, a common example of an opportunistic fungal pathogen. We found swarming against these clusters as well as restriction of their growth to be completely defective in alox5 À/À cells, and these functions were not restored by the addition of LTB 4 ( Figures 1D and 1E). In addition, the application of exogenous LTB 4 appeared to disrupt the ability of wild-type cells to swarm effectively and restrict fungal growth ( Figures 1D and 1E). This result was surprising as it was established earlier that the process of swarming is LTB 4 -signaling dependent Kienle et al., 2021;Lammermann et al., 2013;Malawista et al., 2008;Reategui et al., 2017). Our results confirm a critical role for LTB 4 signaling, as blocking of the primary LTB4 receptor BLT1 disrupts swarming and restriction of C. albicans growth (Figures S2A and S2B). Of interest, LTB 4 levels appeared higher in the anti-BLT1 treated condition ( Figure S2C). This may be due to an inability of the BLT1 receptor to bind and remove LTB 4 from the media. Despite this increased LTB 4 , swarming is completely compromised, demonstrating the importance of sensing LTB 4 to an effective swarm response. Together, these results highlight the unique requirements for LTB 4 during swarming.
These requirements depend not only on the presence of LTB 4 as observed with chemotaxis, phagocytosis, and ROS production, but also on context.
Further investigation of the relationship between LTB 4 and neutrophil swarming revealed that mixing the bone marrow cells derived from alox5 À/À and lta4h À/À mice in a 1:1 ratio restores their capacity to swarm (Figures 2A and 2B). This finding stands in stark contrast to their failure to swarm or restrict fungal growth when in genetically homogeneous populations (Figures 2 and S2D-S2F). The restoration is significant, with the ability of the mixed population of knockout neutrophils to restrict fungal growth comparable to that of their wild-type counterparts ( Figure 2C). Full restoration in swarming is also observed when cells from knockout mice were mixed 1:1 with their appropriate wild-type counterparts ( Figures S2D-S2F). These results suggest that, when mixed, cells with defects at different steps along the LTB 4 biosynthesis pathway can collaborate and compensate for their defects to restore their capacity to swarm and restrict fungal growth. We confirmed this finding using an enriched population of neutrophils ( Figure S3), which matched those results obtained with bone marrow cells ( Figure 2).
We hypothesized that the restoration of swarming in mixed knockout conditions might be due to transcellular biosynthesis of LTB 4 ( Figure 3A). According to this hypothesis, the lta4h À/À neutrophils synthesize LTA 4 and share this precursor with neighboring cells, of which the alox5 À/À could complete the synthesis and the release of the LTB 4 , which helps coordinate the activities of all mutant neutrophils possessing the BLT1 receptor. To directly test this hypothesis, we blocked LTB 4 signaling using an antagonist of BLT1. In agreement with our previous results ( Figure S2), disruption of the LTB 4 signaling blocked swarming and reduced fungal restriction for the mixed wild-type cells ( Figures 3B and 3C). Critically, blocking BLT1 signaling also disrupted swarming and fungal restriction for mixed knockout cells ( Figures 3B and 3C). This result demonstrates that the restored ability of the mixed knockout cells to swarm is dependent on LTB 4 signaling, suggesting that transcellular LTB4 biosynthesis is likely to occur when the knockout cells are mixed.
We measured LTB 4 release during swarming by ELISA and found that the mixed combination of alox5 À/À and lta4h À/À cells did release LTB 4 , consistent with an occurrence of transcellular synthesis ( Figure S4A). The amount of LTB 4 recovered from the mixed knockout population was less than that of LTB 4 recovered from the mixed combination of their respective wild-type cells. However, the magnitude of the swarming responses and the ability to restrict fungal growth is effectively the same as wild-type levels ( Figures 3B and  3C). These results suggest that the generation of LTB 4 exclusively through transcellular biosynthesis of LTB 4 is sufficient to drive robust swarming responses even though the amount of LTB 4 generated appears to be less (Figures 2, 3 and S4A).
Paradoxically, a homogeneous population of lta4h À/À cells appears to be producing low levels of LTB 4 as measured by ELISA, despite lacking a critical biosynthetic enzyme for this process ( Figure S4A). This observation was confirmed by ELISA with LTB 4 assayed from supernatant following stimulation with calcium ll OPEN ACCESS iScience 25, 105226, October 21, 2022 3 iScience Article ionophore in the absence of C. albicans, suggesting that the presence of fungi and the possibility of a fungal source of LTA 4 H is not an explanation for the ELISA signal associated with lta4h À/À neutrophils ( Figure S4B). To better resolve whether this ELISA signal from lta4hÀ/À neutrophils is real, given its low level, the supernatant was also collected using an enriched population of neutrophils responding to live C. albicans, which once more confirmed the presence of this ELISA signal associated with lta4h À/À neutrophils ( Figure S4C). One potential explanation for why lta4h À/À neutrophils generated a positive LTB 4 ELISA signal is because, unlike alox5 À/À cells, lta4h À/À cells remain capable of producing leukotriene A 4 (LTA 4 ). In the absence of LTA 4 H, LTA 4 is rapidly converted non-enzymatically to inactive breakdown metabolites, including 6-trans-LTB 4 (Haeggstrom, 2018). It is possible that LTA 4 and breakdown metabolites are indistinguishable from LTB 4 in this ELISA, and supernatant from lta4h À/À neutrophils that produce LTA 4 breakdown products yield a positive signal despite not containing any actual LTB 4 . In support of this Fluorescence imaging (Hoechst) of mouse bone marrow cells swarming against live C. albicans target. 500,000 bone marrow cells from wild-type, alox5 À/À , lta 4 h À/À , and equal numbers of alox5 À/À + lta 4 h À/À mice were added to swarming arrays. Representative images of swarming from wild-type cells, alox5 À/À or lta 4 h À/À cells alone and alox5 À/À + lta 4 h À/À mixed together 1:1 are shown. T is in minutes. iScience Article interpretation, a bioactivity assay that takes advantage of the chemotactic potential of LTB 4 (not shared by LTA 4 breakdown metabolites) was developed and revealed that the supernatant from swarming chambers containing either alox5 À/À or lta4h À/À neutrophils alone failed to elicit a response, consistent with a lack of true LTB 4 production for lta4h À/À neutrophils ( Figure S4D). A significant increase in directed migration of neutrophils across a permeable Transwell was observed when supernatant was derived from a 1:1 mix of alox5 À/À and lta4h À/À , thereby revealing the presence of chemotactic bioactivity that was exclusively associated with the mixture of alox5 À/À and lta4h À/À cells and this bioactivity likely represents LTB 4 generated by transcellular means ( Figure S4D).
To further confirm that transcellular biosynthesis of LTB 4 is occurring when mixing alox5 À/À and lta4h À/À cells and to unambiguously clarify the nature of molecules detected by the ELISA, we conducted mass spectrometry on collected supernatants ( Figure 4A and Table 1). Bone marrow cells derived from alox5 À/À mice failed to generate LTB 4 and produced trace amounts of LTA 4 non-enzymatic breakdown metabolites (Figures 4B, 4C and Table 1). Cells from lta4h À/À mice also failed to generate LTB 4 but did produce significant LTA 4 and breakdown metabolites, 6-trans-LTB 4 , 6-trans-12-epi-LTB 4 , 5S,6S-diHETE, 5S,6R-diHETE as anticipated ( Figures 4B, 4C and Table 1). The mixture of the alox5 À/À and lta4h À/À cells resulted in the biosynthesis of LTB 4 through transcellular processes as individual knockout neutrophils in isolation are incapable of generating LTB 4 (Figures 4B, 4C and Table 1). In agreement with our ELISA results ( Figure S4A), the amount of LTB 4 produced in this condition of exclusively transcellular biosynthesis

DISCUSSION
We measured mouse neutrophil swarming against C. albicans cluster targets and found that transcellular biosynthesis of LTB 4 drives swarming responses that restrict the growth of fungi. Interfering with the LTB 4 biosynthesis through deletion of key synthetic enzymes in alox5 À/À and lta4h À/À mouse neutrophils and antagonizing LTB 4 receptors disrupts swarming. Notably, the swarming of alox5 À/À mouse neutrophils cannot be restored by the addition of LTB 4 . These results reveal an essential role for the coordinated LTB 4 release from neutrophils in accomplishing the swarming choreography. The dependence on coordinated LTB 4 release distinguishes swarming from other 'traditional' neutrophil functions. For example, phagocytosis and ROS production are also altered when LTB 4 biosynthesis is prevented in alox5 À/À mouse neutrophils, but, unlike swarming, phagocytosis and ROS production in these cells are restored by exposing the neutrophils to extrinsic LTB 4 , consistent with the previous reports (Miralda et al., 2017).
Our study shows that swarming can only be restored when mixing alox5 À/À and lta4h À/À neutrophils, where transcellular biosynthesis of LTB 4 becomes possible. Furthermore, antagonizing LTB 4 receptors disrupts swarming in these mixed cell experiments. These results highlight swarming as a unique and higher-order function of neutrophil coordination, which is more than simply the sum of activities manifested by individually functioning neutrophils.
Swarming, as an emergent neutrophil behavior, has been recently visualized in the context of mechanical (Alexander et al., 2020;Barros et al., 2021;Hopke et al., 2020;Knooihuizen et al., 2021;Yonker et al., 2021), thermal (Lammermann et al., 2013), or infected wounds (Chtanova et al., 2008) in mice and zebrafish. For human neutrophils, ex vivo testing revealed disrupted swarming in patient populations at risk for fungal infections, e.g., transplant recipients, cirrhosis, trauma, chronic granulomatous disease, cystic fibrosis, etc. Hopke et al., 2020;Knooihuizen et al., 2021;Yonker et al., 2021). Furthermore, in a single patient case study, we found that restoring neutrophil swarming correlated with reduced numbers of infections experienced by that patient (Alexander et al., 2020). In parallel efforts, lipid mediators are increasingly understood to be consequential in various pathological processes, and targeting biosynthesis may have therapeutic benefits in these circumstances (Haeggstrom, 2018). The range of conditions that could be corrected by the manipulation of lipid mediator levels spans from common infections (Jordan and Werz, 2021) to complex conditions like Alzheimer's disease (Emre et al., 2022).
Our understanding of transcellular biosynthesis of lipid mediators in homogeneous cell populations benefits from earlier studies in heterogeneous mixtures of neutrophils with other cell types (Fabre et al., 2002). Transcellular biosynthesis helps coordinate the activity of immune and non-immune cells sharing the same space, e.g., neutrophils, lymphocytes, platelets, and endothelial cells (Claesson and Haeggstrom, 1988;Fiore and Serhan, 1990;Marcus et al., 1982;Odlander et al., 1988;Serhan et al., 1984aSerhan et al., , 1984bSerhan et al., , 2020. Transcellular biosynthesis is facilitated by the proximity of two distinct cell types that individually lack but collectively express all necessary enzymes to synthesize a particular mediator (Corey et al., 1980). One of the eicosanoid intermediates that is most shared among immune and non-immune cells is LTA 4 , produced  iScience Article and released in large amounts by neutrophils (Afonso et al., 2012;Fiore and Serhan, 1989). When LTA 4 is taken up by endothelial cells, keratinocytes, erythrocytes, or alveolar macrophages, which express LTA 4 hydrolase, these cells can biosynthesize LTB 4 (Dieterle et al., 2020). An indication that transcellular biosynthesis is likely to be quite common is the observation that close to half of the LTA 4 produced by neutrophils is released extracellularly rather than converted to LTB 4 (Claesson and Haeggstrom, 1988;Fiore and Serhan, 1990;Marcus et al., 1982;Odlander et al., 1988;Serhan et al., 1984aSerhan et al., , 1984bSerhan et al., , 2020. Our study raises several important questions that will be addressed in future studies. It is not fully understood how transcellular biosynthesis intermediates are transported between neutrophils. For transcellular biosynthesis of LTB 4 to occur, LTA 4 must be passed from lta4h À/À neutrophils, which have functional 5-LOX, to the alox5 À/À neutrophils, which have functional LTA 4 hydrolase. By collaborating in this fashion, the alox5 À/À and lta4h À/À cells can produce functional LTB 4 , which can then be released to drive the recruitment and swarming of both cell types. LTA 4 has a short half-life (Fiore and Serhan, 1989;Haeggstrom, 2018;Stsiapanava et al., 2017) and is likely hydrolyzed immediately after release (Fiore and Serhan, 1989). Transportation modes that increase the biological half-life of LTA 4 should be considered. For example, associations of LTA 4 to lipid membranes (Fiore and Serhan, 1989) and to chaperone molecules, like albumin (Fitzpatrick et al., 1982), have been proposed to protect LTA 4 in the extracellular space between various cell pairs. A transport mediated by exosomes has also been suggested for shuttling LTB 4 from neutrophils to other neutrophils (Dieterle et al., 2020) and may also be applicable to LTA 4 . This mechanism may also be consistent with the relay model of neutrophil signaling during swarming (Dieterle et al., 2020).

Limitations of the study
Much of this work was conducted with bone marrow cells that feature 40% or less mature neutrophils within the total cell population. It is, therefore, possible that other cells within the bone marrow may influence the swarming observed herein. Our observations were confirmed using enriched neutrophil populations (65-75% mature neutrophils) from bone marrow. Nevertheless, further work is needed to exclude potential influences of the non-neutrophil cellular component within the bone marrow.
The relevance of our findings in mice to human neutrophils remains to be examined. Human neutrophils display multiple levels of redundancy in swarming, with additional factors besides LTB 4 , like IL-8 and complement factors, partially compensating for the loss of LTB 4 . Unlike human neutrophils, LTB 4 appears to be the only driving factor of neutrophil swarming in mice Kienle et al., 2021;Lammermann et al., 2013;Reategui et al., 2017). Our study demonstrates that transcellular LTB 4 biosynthesis is necessary and sufficient to orchestrate swarming and restriction of fungal growth by a mixture of genetically deficient mouse neutrophils that are individually incapable of completing LTB 4 synthesis. We suggest that transcellular LTB 4 biosynthesis is likely to be important in orchestrating wild-type mouse neutrophil swarming as well. Transcellular LTB 4 biosynthesis is facilitated by the large proportion of LTA 4 released from wildtype neutrophils as revealed by detection of LTA 4 non-enzymatic breakdown metabolites (Afonso et al., 2012;Fiore and Serhan, 1989). Future investigations are necessary to characterize the role of LTB 4 transcellular biosynthesis in human and mouse swarming and its contribution to neutrophil-mediated host defense following sterile injury and infection.

STAR+METHODS
Detailed methods are provided in the online version of this paper and include the following:

ACKNOWLEDGMENTS
We acknowledge funding from the National Institutes of Health (GM092804 to DI, AI095338 to BPH, AI132638 to MKM, GM095467 & GM139430 to CNS), and from the Shriners Hospitals for Children (71010-BOS-22 to DI). Graphical Abstract illustrated by Nicole Wolf, MS, ª2022 (nicolecwolf@gmail.com).

Lead contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Bryan Hurley (bphurley@mgh.harvard.edu).

Materials availability
This study did not generate any new unique reagents. Slides and plates for the swarming assays are available through the BioMEMS Core at the Massachusetts General Hospital https://researchcores.partners. org/biomem/about.

Data and code availability
d All data reported in this article will be shared by the lead contact on request.
d This article does not report any original code.
d Any additional information required to reanalyze the data reported in this article is available from the lead contact on request.

EXPERIMENTAL MODEL AND SUBJECT DETAILS Animals
The following strains of mice were obtained from Jackson Laboratories: wild-type C57BL/6J and S129 (129S1/SvImJ), knockout mice alox5 À/À (B6.129S2-Alox5 tm1Fun /J) (Chen et al., 1994) and lta 4 h À/À (129-Lta 4 h tm1Bhk /J) (Byrum et al., 1999). Eight to twenty weeks old male and female mice of different genotypes were used to isolate bone marrow cells. The Institutional Animal Care and Use Committee at Massachusetts General Hospital (MGH) approved the animal protocols used in this study. The mice were housed and bred in the the animal facility of MGH. The laboratory animal care and use program at MGH is accredited by AAALAC International, has an assurance with the Office of Laboratory Animal Welfare (OLAW) and is registered with the United States Department of Agriculture (USDA).

Microbial strains
Candida albicans SC5314 far-red fluorescence expressing strain (SC5314 iRFP) was a kind gift of Robert Wheeler at the University of Maine. (Hopke et al., 2016) C. albicans was inoculated to fresh liquid YPD and grown overnight with shaking at 30 C.

Isolation of bone marrow cells and purification of bone marrow neutrophils
Bone marrow cells were isolated from C57BL/6, 129S1, alox5 À/À, and lta 4 h À/À mice (Jackson Laboratories) as described previously with mild modifications (Boxio et al., 2004 iScience Article and femurs and tibia were flushed with HBSS without calcium and magnesium (Thermo Fisher Scientific). Spicules or bone matrix were removed by 40 mm cell strainer (Fisher). Red blood cells were lysed in cold NH 4 Cl lysis buffer as described previously. (Hurley et al., 2004) About 1.5 -2.5 3 10 7 bone marrow cells were isolated per mouse and 26-40% of the bone marrow cells were CD11b+Ly6G+ neutrophils as confirmed by flow cytometry by using fluorescently-labeled antibodies for CD45, CD11b and Ly6G (Thermo Fisher Scientific). The majority of neutrophils (94%) were morphologically mature and functionally competent, as reported previously. (Boxio et al., 2004) This technique allows for rapid isolation of 17-fold more neutrophils than those isolated from peripheral blood per mouse. Additional neutrophil purification was performed using the EasySep mouse neutrophil enrichment kit (STEMCELL), following the manufacturer's recommended protocol. The purity of CD11b+Ly6G+ neutrophils was 65-75% as evaluated by flow cytometry analysis.
Phagocytosis and ROS production during C. albicans challenge Murine neutrophils were harvested from the tibias and femurs of B6 and alox5 À/À mice as previously described (Wang et al., 2006). Briefly, bones were crushed in FACs buffer (2% heat-inactivated fetal bovine serum in PBS), strained through a 40 mM filter, and red blood cells were lysed using 0.2% and 1.6% NaCl solutions. Neutrophils were then harvested by a Ficoll gradient (Histopaque, Sigma Aldrich). To assess phagocytosis and ROS production, neutrophils were co-incubated with a far-red fluorescent protein-expressing C. albicans strain (Hopke et al., 2016) at a ratio of 5 yeast cells per neutrophil for 1 h at 37 C in a 1.5-mL tube. Samples were incubated with dihydrorhodamine-1,2,3 (DHR-123 at 1 mM, Life Technologies, Eugene, OR) to assess ROS production for each condition. Where appropriate, neutrophils were treated with 30 mM of cytochalasin-D (Sigma) to inhibit phagocytosis or with LTB 4 (0.6 nM) for phenotype rescue. Following co-incubation, samples were placed on ice and labeled with Ly6G-PE (BioLegend) for 15 min, washed in FACs buffer, and plated in a 96-well U-bottom plate. A BD FACSCeleta Cell Analyzer (BD Biosciences) with a high-throughput plate adaptor running BD FACSDiva Software (v9.0). Percent ROS was measured by selecting doubly positive Ly6G-PE neutrophils and DHR-123 fluorescent cells, whereas percent phagocytosis was measured by fluorescent C. albicans in neutrophils. Flow data were analyzed using FlowJo 10 software (FlowJo, Ashland, OR).

Swarming array printing
Utilizing a microarray printing platform (Picospotter PolyPico, Galway, Ireland), we printed a solution of 0.1% poly-l-lysine (Sigma-Aldrich) and ZETAG targets with 100 mm diameter. For microscopy and ELISA experiments, we printed eight by eight arrays in a sixteen-well format on ultra-clean glass slides (Fisher Scientific). For LC-MS/MS experiments, we printed over 4500 targets covering the glass slide. Slides were screened for accuracy and then dried at 40 C for 2 h on a heated block. After 2 h, slides were removed from the heat block and left at room temperature until required.

Patterning of Candida albicans cluster targets
Swarming arrays were created as described . Briefly, 16-well ProPlate wells (Grace Biolabs) or single-well ProPlate wells were attached to glass slides with printed arrays of poly-l-lysine/ZETAG. A suspension of the desired target, in this case, live C. albicans(SC5314 iRFP) yeast in water, was added to each well (50 uL per well for the 16-well format, 1.5 mL for the single well) and incubated with rocking for 5 min. Following incubation, wells were thoroughly washed out with PBS to remove unbound targets from the glass surface. Wells were screened to ensure appropriate patterning of targets onto the spots with minimal non-specific binding before use.

Swarming experiments
All imaging experiments were conducted using a fully automated Nikon TiE microscope. Time-lapse imaging was conducted using a 10x Plan Fluor Ph1 DLL (NA = 0.3) lens, and endpoint images were taken with a 2x Plan Apo (NA = 0.10) lens. Swarming targets (C. albicans clusters) to be observed during time-lapse were selected and saved using the multipoint function in NIS elements prior to loading of cells. Bone marrow cells or enriched bone marrow neutrophils were stained with Hoechst (Thermo Fisher Scientific) and resuspended in IMDM with 20% FBS (Thermo Fisher Scientific). 500,000 cells were added to each well for individual genotype conditions. 250,000 cells each were added in mixed genotype conditions. All selected points were optimized using the Nikon Perfect Focus System before launching the experiment. In experiments using chemical inhibitors, neutrophils were pre-incubated with the chemical or ll OPEN ACCESS iScience 25, 105226, October 21, 2022 13 iScience Article appropriately matched vehicle control for 30 min before use. The supernatants were collected 2 h after the cells were added and saved at À80 C after removing the cells by centrifugation.

Image analysis
Area analysis was performed manually by outlining the swarms or areas of fungal growth in the NIS-elements (v4.00.12; Nikon Inc) or FIJI (FIJI is just ImageJ v2.0.0-rc-59/1.52p, NIH) software. For the area of the swarm, only the swarm itself (just the immune cells) was measured. This was done using the DAPI fluorescent channel image, using Hoechst staining to identify neutrophils. For areas of fungal growth, a combination of brightfield and fluorescent channels was used. Fungi used in experiments were always far-red fluorescent (Hopke et al., 2016). We combined the appropriate fluorescent channel with the brightfield image to be sure we included any escaped fungal elements, like lone hyphae, that may not show up well in the fluorescent channel.

Bone marrow cell culture
Bone marrow cells from single or mixed cell types were seeded in 96-well round-bottom tissue culture plates at 200 mL/well with 5310 6 cells/mL. Cells were incubated with calcium ionophore A23187 (Sigma-Aldrich) at 20 mg/mL at 37 C with 5% CO 2 for 1 h. Cells were removed by centrifugation at 500 3g for 5 min. The supernatants were saved at À80 C for LTB 4 ELISA assays.

LTB 4 quantification by ELISA
Supernatants from the swarming assay for each condition were collected at the indicated time points and subjected to a competitive LTB 4 ELISA (Cayman chemical) according to the manufacturer's protocol. Briefly, 50 mL LTB 4 standards diluted in 1:2 series and supernatants from the swarming assay were added to the 96-well plate precoated with mouse anti-rabbit IgG and incubated with LTB 4 antiserum and AChE linked to LTB 4 (tracer) at 4 C overnight. The plate was then washed five times with wash buffer, followed by incubation with Ellman's reagent for 90-120 min. The absorbance at 405 nm was measured by SpectraMax iD5 microplate reader (Molecular Devices). The readings of diluted standards were plotted as logit B/B 0 versus log LTB 4 concentration using a linear fit and were used to determine sample LTB 4 concentrations according to the manufacturer's instructions.

BLT1 receptor quantification on mouse neutrophils
Bone marrow cells isolated from C57BL/6, alox5 À/À , S129, and lta 4 h À/À mice were applied to LIVE/DEAD fixable Dead Cell staining by incubating with Near-IR fluorescent reactive dye (Thermo Fisher Scientific) in HBSS at room temperature for 15min in the dark, followed by two washes with HBSS. The cells were then resuspended in eBioscience TM Flow Cytometry staining buffer (Thermo Fisher Scientific) and incubated with rat anti-mouse CD16/CD32 monoclonal antibody on ice for 10 min to block the Fc receptor.

Chemotaxis assay
The chemoattractive activity of the supernatants obtained from swarming experiments was measured by performing a bone marrow cell transmigration assay using 96-well Transwell with a pore size of 3mm (Corning). Bone marrow cells were isolated from the femoral and tibial bones of C57BL/6J alox5 À/À or lta 4 h À/ mice, as described above. One hundred microliters of supernatant were added to the bottom well, and 10 6 bone marrow cells in 75 mL HBSS were added to the inside of the Transwell insert. After incubation at 37 C for 2 h, the inserts were removed. Bioactivity was determined by the number of neutrophils that migrate through the Transwell towards the conditioned swarming supernatant or LTB 4 (0.2 ng/mL). Moreover, MPO assay was performed with the cells migrated to the bottom wells as described previously. (McCormick et al., 1995) ll OPEN ACCESS