The paracrine factor Myeloid Derived Growth Factor regulates the inflammatory fate and motility of human induced pluripotent stem cell-derived neutrophils

Abstract

Neutrophils are necessary for host defense but also contribute to tissue damage. Understanding the molecular mechanisms that govern human neutrophil development and fate is critical for developing interventions for treating and preventing human disease. Human induced pluripotent stem cell (iPSC)-derived neutrophils (iNeutrophils) are genetically tractable and provide a robust source of neutrophils to dissect molecular regulation and for potential use in infusion therapies. Here, we show that deletion of the paracrine factor myeloid derived growth factor (MYDGF) in human iPSCs results in the development of iNeutrophils with improved zymosan-induced reactive oxygen species (ROS) generation and antimicrobial functions. MYDGF-deficient iNeutrophils showed improved transendothelial migration to microbes in organotypic models and motility in confined microchannels. We identify MYDGF as a new endogenous inhibitor of cytosolic calcium dynamics that suppresses neutrophil inflammatory activity. Our findings show that MYDGF mediates neutrophil-to-neutrophil communication during migration through a paracrine mechanism. Since MYDGF-deficient iNeutrophils show improved effector and antimicrobial function, these cells may improve future therapeutic use of iNeutrophils for infusion therapies. Finally, soluble MYDGF provides an attractive strategy to limit chronic neutrophil-mediated inflammation and tissue damage.

Introduction

As first responders to injury and infection, neutrophils employ effector functions to recognize and eliminate pathogens^[1-4]. In healthy individuals, neutrophil recruitment to sites of tissue damage and infection is efficiently resolved – a process required for normal wound healing. However, dysregulated neutrophil behavior, including excessive neutrophil accumulation, can contribute to pathology^{[5, 6]}. Therefore, neutrophils or the tissues they migrate through must have mechanisms to limit neutrophil recruitment and their inflammatory activity.

Previous work on neutrophil recruitment has focused on pro-migratory signaling. Injury or infection causes tissue-resident cells to release DAMPs, including heat shock and S100 family proteins^[7]. This is quickly followed by chemokine and cytokine production, promoting additional neutrophil recruitment. Further, neutrophil intrinsic signaling, such as release of swarming factor LTB4 boosts neutrophil accumulation at sites of infection through paracrine signaling^[8]. However, the pathways that either limit neutrophil inflammation or aid in resolution are less clear. Prior research has shown that neutrophil-intrinsic Semaphorin 3E acts as a chemorepellent that inhibits IL-8-induced migration through the regulation of Ras-associated GTPase activity^[9]. Similarly, NLRP12 has been shown to specifically inhibit neutrophil migration towards CXCL1 through ERK signaling[10]. While evidence of endogenous neutrophil suppressors exists, more work is needed to understand neutrophil intrinsic pathways that quell inflammatory behavior.

Myeloid derived growth factor (MYDGF) is a secreted paracrine factor ^[11] which is highly conserved across species and was originally identified in eosinophilic vesicles^[12]. Our lab previously reported that MYDGF was upregulated in neutrophils and macrophages in response to tail fin wounding in larval zebrafish^[13]. Further, depletion of MYDGF impaired inflammation resolution and wound healing, suggesting MYDGF acts as a damage signal that regulates neutrophil responses^[13]. Recently, therapeutic properties of MYDGF were highlighted in studies showing protective effects of MYDGF on cardiomyocytes following myocardial infarction^[14] and a role for MYDGF in suppressing NF-κB-driven inflammation in Non-Alcoholic Fatty Liver Disease^[15]. Additionally, MYDGF was shown to inhibit sarco/endoplasmic reticulum Ca²⁺-ATPase 2a, alleviating calcium-driven cardiac hypertrophy^[16], suggesting roles in calcium regulation. Calcium signaling is central to neutrophil effector functions including reactive oxygen species (ROS) production, phagocytosis, and migration^[17]. Recent work reported that Ca²⁺ release-activated Ca²⁺ channel regulator 2A promotes neutrophil motility through activation of αMβ2 integrin ligand binding^[18]. These observations led us to hypothesize that MYDGF may act to limit neutrophil inflammation in part through the regulation of calcium signaling.

Although primary human neutrophils from peripheral blood (PBNs) have long been used to investigate neutrophil function, there are limitations associated with their use in vitro^[19]. Specifically, PBNs have a short lifespan in vitro and are not genetically tractable. We previously reported that human induced pluripotent stem cell (iPSC)-derived neutrophils (iNeutrophils) retain neutrophil identity and effector functions including phagocytosis, ROS production, and response to inflammatory cytokines^[20-22]. Moreover, iNeutrophils represent a genetically tractable neutrophil model to uncover new mechanisms of human neutrophil function. We and others have used iNeutrophils to study cellular signaling^{[20, 23]} and to identify factors that improve antimicrobial function in human neutrophils^{[22, 24]}. This is particularly important because of the potential future use of iNeutrophils to treat human disease in infusion therapies^{[25, 26]}.

Here, we use iNeutrophils to investigate the neutrophil intrinsic role of the paracrine factor MYDGF. We show that MYDGF dampens cytosolic calcium and limits ROS production and anti-fungal activity. Additionally, depletion of MYDGF improves neutrophil transmigration and motility under confinement. These findings identify MYDGF as a new endogenous inhibitor of neutrophil inflammatory behavior and motility that mediates neutrophil-to-neutrophil communication.

Materials and Methods

Generation of MYDGF Knockout and Knockin BM9-iPSCs and Differentiation into iNeutrophils

Two single guide RNAs (sgRNAs) were designed (Synthego) targeting human MYDGF exon 4: sgRNA1 and sgRNA2 (Supp. Table 1). Before nucleofection, BM9 iPSCs were singularized using TrypLE Select (Life Technologies). 5μg of each sgRNA and 5μg of Cas9 protein (PNA Bio) were incubated in double-distilled water (ddH₂O) for 10 minutes at room temperature. Cells were nucleofected using Human Stem Cell Nucloefector Kit 2 (#VPH-5022; Lonza) using Nucleofector protocol number B-16 and plated at 25 cell/cm² in mTeSR-Plus media with CloneR supplement (1X; STEMCELL Technologies) on a Cultrex-coated tissue culture plate. Two days after nucleofection, media was changed to mTeSR-Plus only. Individual cell colonies were picked after seven days and expanded. To confirm the MYDGF mutation, genomic DNA was extracted from clones and screened using polymerase chain reaction (PCR) for a 42 base pair deletion using primer1, primer2, and primer3 (Supp. Table 1). iPSCs were differentiated from knockout clones using previously published protocols^{[20, 22]}. cDNA was generated using SuperScript IV First-Strand Synthesis System (Invitrogen) from human neutrophil total RNA. MYDGF was amplified using infusion primers for cloning first into Addgene vector (#34931) pcs2+8 using EcoR1 forward primer and Xba1 reverse primer (Supp. Table 1). Then MYDGF was cloned into Addgene vector (#68375) AAVS1_Puro_PGK1_3xFLAG_Twin_Strep using Kpn forward infusion primer and Not1 reverse primer (Supp. Table 1). MYDGF was knocked into iPSC line AAVS1 safe harbor using 3ug of MYDGF_AAVS1_Puro_PGK1_3xFLAG_Twin_Strep with 1μg Addgene vector (#80494) pXAT2-AAVS1 sgRNA expression vector using Nucleofector Kit 2 for Human Stem cells with protocol B-16^[27-29]. Clones were picked and selected in 0.3μg/ml puromycin/mTeSR Plus media and checked for protein expression using human MYDGF and anti-Flag antibodies (Supp. Table 3).

See Supplementary Methods for further description of differentiation.

Flow Cytometry

iNeutrophils were stained with antibody panels (Supp. Table 2) in DPBS + 1% human serum albumin (HSA) and Brilliant Buffer (Thermo Fisher Scientific) (1:1) and fixed with 4% paraformaldehyde (PFA) (16%; Thermo Fisher Scientific). Sample acquisition and data collection were performed on Aurora Cytometer (CytekBio). All data analyzed with FlowJo Software (v10.8.1; TreeStar).

Cytospin and iNeutrophil Staining

8x10⁴ iNeutrophils were spun onto glass slides (1200rpm; 5 minutes) using the Shandon Cytospin 3 centrifuge (Thermo Fisher Scientific). Slides were stained with Differential Quick III Stain Kit (Electron Microscopy Sciences) following manufacturer protocol. Samples were imaged using Olympus IX70 microscope with UplanFI 10X/0.30 NA objective.

Immunoblotting

iNeutrophils were pelleted and collected in lysis buffer. Lysis buffer consisted of HEPES (25mM, pH 7.5), NaCl₂ (150mM), Nonidet P-40 (1%), MgCl₂ (10mM), EDTA (1mM), glycerol (10%) supplemented with pepstatin A (1g/mL), aprotinin (2g/mL), leupeptin (1g/mL), phenylmethanesulfonyl fluoride (200nM), sodium orthovanadate (1mM), protease inhibitor mixture 2 (Sigma-Aldrich). Lysates were incubated on ice for 10 minutes and sonicated for 3 x 5 seconds at 20% amplitude. Cells were clarified using centrifugation (15,000g, 4° C for 10 minutes). Protein concentrations were evaluated using BCA Protein Assay (Thermo Fisher Scientific) and samples stored at −80° C. Lysates were boiled with 1X SDS running buffer (20X; Thermo Fisher Scientific) and immunoblotting performed. Blots were imaged using an infrared imaging system (Odyssey; LI-COR Biosciences). See Supplementary Table 3 for antibodies. See Supplementary Methods for immunoprecipitation protocols.

Single cell RNA sequencing

2x10⁵ iNeutrophils were collected at day 7 following transfer into Media C, spun down, and resuspended in 200μL of PBS + 0.04% Bovine Serum Albumin (BSA; Sigma-Aldrich). All cells were directly used for library construction in a Chromium controller. Libraries were constructed according to the Chromium Single Cell 3’ Reagent Kit v3.1 user guide (10x Genomics, Pleasanton, CA). Cell concentration and viability were quantified on the Luna-FX7 (Logos Biosystems, Anyang, South Korea) using acridine orange/propidium iodide stain. Cells were loaded onto a Chromium Single Cell Chip G chip for the single cell partitioning GEM reaction on the Chromium iX platform. Next, GEMs were transferred to emulsion safe strip tubes for reverse transcription and cell/transcript barcoding using an Eppendorf MasterCycler Pro thermocycler (Eppendorf, Hamburg, Germany). Following reverse transcription, GEMs were broken, and pooled single cell cDNA was amplified for 13 cycles, purified using SPRIselect beads (Beckman Coulter, Brea, CA), and quantified on a Bioanalyzer 2100 (Agilent, Santa Clara, CA) using the High Sensitivity DNA kit. A portion of the full-length cDNA was fragmented and used to generate cDNA libraries according to standard 10X Genomics workflow. See supplementary materials for description of data processing, quality control, data integration, and annotations.

Intracellular ROS Assay

A 96-well black, clear bottom plate was coated with 1X fibronectin bovine plasma (100X; Sigma-Aldrich) and incubated at 4° C overnight. On the day of experimentation, cells were pelleted and resuspended in RPMI 1640 (no phenol red; Thermo Fisher Scientific) + 2% fetal bovine serum (FBS) (R&D Systems). Dihydrorhodamine-123 (5mM; Thermo Fisher Scientific) was added to resuspended cells at a 1:2000 dilution. Cells were plated at 1x10⁵ cells/well and incubated for 30 minutes at 37°C, 5% CO₂. During incubation, phorbol myristate acetate (PMA) (Sigma-Aldrich) and Zymosan (Invivogen) stimuli were prepared at 2X concentration, so that final concentration was 100ng/mL and 100μg/mL, respectively. Stimuli or media (control) were added to appropriate wells following incubation. Fluorescent signals were read at 485/535 nm in a microplate reader (Synergy H1; Bio-Tek Instruments). Background fluorescence of the control wells was subtracted from each sample prior to data analysis.

Trans-endothelial migration

Endothelial lumens were prepared using LumeNEXT protocols^{[23, 30]}. See Supplementary Methods for endothelial cell differentiation and culture. Endothelial cells were cultured in lumen for two days. iNeutrophils were stained with Calcein AM (1:1000 in DPBS; Fisher Scientific) for 15 minutes at 37°C. Next, cells were resuspended in endothelial media at 7.5x10⁶ cells/mL and seeded into lumen channels. The top port of the lumen received Pseudomonas aeruginosa (provided by Samuel M. Moskowitz, University of Washington, Seattle, WA, USA) to induce migration. Migration was tracked over 12 hours on a Nikon Eclipse TE300 inverted fluorescent microscope with PlanFluor 20X/0.50 objective and movies were analyzed using ImageJ.

Confined migration assay in liquid-walled microchannels

Liquid-walled microchannels were constructed as previously described^{[31, 32]}. Channels were coated with fibrinogen (10μg/mL), fibronectin (1μg/mL) and collagen IV (1μg/mL) in RPMI + 2% fetal bovine serum (FBS) for 30 minutes at 37°C, 5% CO₂. For experiments requiring quantification of intracellular calcium dynamics, cells were stained with Fluo4-AM fluorescent dye (1μM) in RPMI media + 2% FBS for 20 minutes at 37°C, 5% CO₂. For BAPTA AM experiments, iNeutrophils were pretreated for 30 minutes with BAPTA-AM (100μM). After incubation of the channels, media was removed and 1.5 μL cells were seeded onto the inlet droplet at 60,000 cells/μL in RPMI media + 2% FBS. Immediately after, 1.5 μL of fMLP (10μM) was seeded onto the outlet droplet and the channels moved to an on-stage incubator on a Nikon Eclipse TE300 inverted fluorescent microscope with a PanFluor 20X/0.50 NA objective or on Nikon Eclipse Ti2 microscope using Aprochromat 60X/1.42 NA or S-Fluor 20X/0.75 NA objectives in combination with Tokai Hit stage incubation system. For experiments tracking speed and/or fluo4 fluorescent intensity, images were acquired every 2.5 seconds for 5 minutes in GFP and brightfield channels. See Supplementary Methods for analysis techniques.

Aspergillus fumigatus co-culture assay and imaging

See Supplementary Methods for Aspergillus fumigatus culture and collection. Following spore collection, 1mL of a 4x10³ suspension was added to wells of a 24-well plate coated with fibrinogen (10μg/mL). This plate was incubated at 37°C until germlings were visible. iNeutrophils were collected and diluted to a concentration of 8x10⁵ cells/mL in RPMI + 2% FBS. Once germlings developed, glucose minimal medium (GMM) was removed from all wells and 500μL of iNeutrophil suspensions were plated for a final neutrophil:germling ratio of 100:1. Images were taken every 3 minutes for 12 hours on a Nikon Eclipse TE300 inverted fluorescent microscope using PlanFluor 20X/0.50 NA objective with environmental chamber set to 37°C with 5% CO₂. Videos were compiled with NIS Element and ImageJ softwares. ImageJ was used to quantify hyphal growth. Fungal size was measured for individual and viable fungi at time 0 (start) and time 12 (end).

Human Peripheral Blood Neutrophil Isolation

Blood was collected from healthy donors and drawn in correspondence with the University of Wisconsin-Madison Minimal Risk Research Institutional Review Board – approved protocol (ID: 2017-0032) per the Declaration of Helsinki. Blood was collected from donors who provided informed written consent in correspondence with the University of Wisconsin-Madison Internal Review Board approved protocol. Human PBNs were isolated using MACSxpress negative antibody selection kit (Miltenyi Biotec) and purified using MACSxpress erythrocyte depletion kit (Miltenyi Biotec), following manufacturer protocols.

Results

Depletion of MYDGF does not alter iNeutrophil maturation or heterogeneity

Our previous work suggested that MYDGF limits neutrophil inflammatory responses in situ using zebrafish^[13]. To investigate a neutrophil-intrinsic role for MYDGF, we generated MYDGF-deficient (MYDGF KO) iNeutrophils using previously published protocols^{[20, 22]} (Fig. 1A-B, Fig. S1A). To confirm that MYDGF is expressed in human neutrophils we performed western immunoblotting and detected MYDGF in both primary human neutrophils and WT iNeutrophils, but not MYDGF KO iNeutrophils (Fig. S1B). A defining feature of neutrophils is the presence of a multi-lobed nucleus^[33]. Using hematological staining, we found that both wild-type (WT) and MYDGF KO iNeutrophil lines maintain the polymorphonuclear morphology (Fig. 1C). Additionally, human neutrophils express CD11b, a key granulocyte marker^[34], along with CD15 and CD16 which are surface markers for neutrophil maturity and pro-phagocytic capacity, respectively^[35]. Using flow cytometry, we found similar expression of these markers in WT and MYDGF KO lines (Fig. 1D-F, Fig. S1C-F).

Figure 1. Characterization of MYDGF KO iNeutrophils.

(A) Schematic demonstrating the CRISPR/Cas9-mediated knockout and differentiation of MYDGF KO iNeutrophils. (B) Western blot confirming the absence of hMYDGF in iPSC MYDGF KO. Mouse beta-tubulin used as a loading control. (C) Cytospin, hematoxyin and eosin (H&E) staining images of WT and MYDGF KO iNeutrophils. Scale bar = 20μm. (D-F) Flow cytometry-derived histograms of (D) CD11b, (E) CD15 and (F) CD16 expression on WT and MYDGF KO iNeutrophils measured using flow cytometry. Grey outline represents WT iNeutrophils and red outline denotes MYDGF KO iNeutrophils. Single cells identified using forward and side scatter parameters and live cells identified using Zombie NIR dye. Monocytes were identified using CD14 and excluded from results. WT, KO: N=3. (G) UMAP visualization of unsupervised clustering from single-cell transcriptomes of WT and MYDGF KO iNeutrophils. Color code for subpopulation is shared with (H). (H) Subpopulation proportion with WT and MYDGF KO iNeutrophils. Experiments were performed a minimum of three times.

Recent work has shown that human neutrophils exhibit distinct subpopulations and heterogeneity in homeostatic and pathological conditions[36]. We performed single-cell RNA sequencing (scRNA-seq) to determine the population composition and maturity profile of iNeutrophils (Fig. 1G-H, Fig. S2A-C). Using a cross-dataset label transfer approach, we first referred to a published human neutrophil scRNA profile to inform iNeutrophil classification^[36]. We found that both WT and MYDGF KO cells show molecular heterogeneity that match the full maturation trajectory of human primary neutrophils (Fig. S2D-E). Taken together, these data demonstrate that deletion of MYDGF does not significantly influence iNeutrophil identity as compared to WT iNeutrophils.

MYDGF depletion increases ROS production and anti-fungal activity

To determine how MYDGF regulates neutrophil effector functions, we performed unbiased Gene Ontology (GO) enrichment analysis on the scRNA-seq data comparing MYDGF-deficient to WT iNeutrophils within each subpopulation. The findings revealed that several terms associated with a pro-inflammatory response were increased in the MYDGF KO iNeutrophils (GO:0001819 positive regulation of cytokine production, GO:0002718 regulation of cytokine production involved in the immune response) (Fig. 2A). To determine if specific effector functions were altered, we first tested ROS production. When stimulated with zymosan, a bioparticle primarily composed of fungal β-glucan, MYDGF KO iNeutrophils produced higher amounts of ROS (Fig. 2B), suggesting that MYDGF dampens the neutrophil inflammatory response to microbial factors. Importantly, this was not due to differences in ROS production at rest (Fig. 2C). Re-expression of full-length MYDGF by genetic knock-in into the MYDGF KO iNeutrophils was sufficient to reduce ROS production to WT levels (Fig. 2B, Fig. S1G), demonstrating that MYDGF dampens ROS production in response to zymosan.

Figure 2. MYDGF impairs ROS production and antifungal activity.

(A) Gene Ontology (GO) terms enriched in genes highly expressed in MYDGF KO Neutrophils compared to WT by subpopulation from single-cell transcriptomes. GO terms associated with leukocyte inflammatory responses are highlighted in blue and terms associated with leukocyte migration are highlighted in magenta. (B-C) Reactive oxygen species (ROS) production of WT, MYDGF KO, and MYDGF knock-in iNeutrophils at 120 minutes (B) at rest or (C) following stimulation with zymosan (100μg/mL). WT, KO, knock-in: N=3. (D) Representative images of iNeutrophils co-cultured with red fluorescent protein (RFP)-labelled Aspergillus fumigatus at 0- and 12-hours following co-incubation. Scale bar = 50μm. (E) Quantification of individual spore fungal area at the terminal time point (12 hours) after co-incubation of iNeutrophils with A. fumigatus. Experiments were performed a minimum of three times, replicates are indicated by dots on each plot (B-C). Individual points represent single spores. Number of germlings quantified: WT, n=39; KO, n=42. (E). Means ± SEM are shown. P values were calculated by one-way Anova (B), unpaired Student’s t-test (C) or Mann-Whitney test (E).

We previously reported that GATA1 KO iNeutrophils had improved anti-fungal activity due, at least in part, to increased expression of human fungal pathogen recognition receptors^[22]. Here, we found that WT and MYDGF KO iNeutrophils express TLR2, TLR4 and Dectin-1, a key mediator of the neutrophil fungal response at similar levels (Fig. S3D-F). We compared the antifungal activity of WT and MYDGF KO iNeutrophils against the common human fungal pathogen A. fumigatus strain CEA10 expressing red fluorescent protein^[22]. Co-culture of iNeutrophils with A. fumigatus revealed that MYDGF KO iNeutrophils were significantly better at limiting fungal growth (Fig. 2D-E). This may be due to the increased ROS production or enhanced inflammatory cytokine production found in the MYDGF KO iNeutrophils in response to zymosan (Fig. S3A-B). There was no significant difference in NET release between iNeutrophil lines (Fig. S3C). Finally, we assessed iNeutrophil phagocytic capacity against the bacterium Pseudomonas aeruginosa and found that MYDGF KO iNeutrophils were still able to effectively engulf bacteria for phagocytosis, similar to WT iNeutrophils (Fig. S3G-H). Collectively, these results demonstrate that MYDGF limits iNeutrophil ROS generation and anti-microbial activity against fungi.

Deletion of MYDGF enhances iNeutrophil migration through an endothelial barrier and under confinement

GO term analysis further revealed an increase in pathways associated with motility and chemotaxis in MYDGF-deficient cells (GO:0002685 regulation of leukocyte migration, GO:0050918 positive chemotaxis) (Fig. 2A). Since neutrophil motility is essential for neutrophil inflammatory responses^[37], we tested the effects of MYDGF on iNeutrophil migration. First we used the LumeNext model which allows for in vitro assessment of neutrophil migration across an endothelial cell lined lumen to a pathogen source embedded in a 3-dimensional matrix^[30] (Fig. 3A). iNeutrophil extravasation showed that MYDGF KO iNeutrophils had a nearly five-fold increase in the number of cells that migrated across the endothelial barrier after 12 hours (Fig. 3B, Supp. Video 1).

Figure 3. MYDGF negatively regulates transendothelial migration and migration under confinement.

(A) Schematic of endothelial lumen – in vitro model of neutrophil migration. Image made using Biorender.com. Representative images of WT and MYDGF KO iNeutrophil migration through an endothelial barrier towards Pseudomonas aeruginosa at 0, 4, and 12 hours. (B) Quantification of cell number migrated across the endothelial barrier at 4, 8, and 12 hours. WT, KO: N=3. (C) Schematic of underoil confined migration system (UOMS). Image made using Biorender.com. (D) Quantification of the velocity (μm/min) of WT, MYDGF KO and MYDGF knock-in iNeutrophils in confinement towards fMLP (10μM). Channels quantified: WT, n=7; KO, n=7; Knock-in, n=6. Individual points represent at least 5 channels across three biological replicates. Means ± SEM are shown. P values were calculated by one-way simple linear regression (B) or one-way Anova (D).

Neutrophil extravasation requires expression of adhesion molecules like integrins and selectins^[38]. Therefore, we hypothesized that MYDGF may inhibit extravasation by downregulating expression of these surface proteins. Examination of common adhesion markers revealed no differences between WT and MYDGF KO iNeutrophils (Fig. S4A-D). These findings led us to consider the possibility that MYDGF may suppress extravasation through an alternative mechanism. Neutrophils commonly employ paracellular migration to squeeze between the tight intercellular junctions of neighboring endothelial cells to undergo TEM^[39]. We previously developed a model for investigating neutrophil migration under confinement using a system which mimics physiologically relevant pressures of interstitial migration^[31] (Fig. 3C). Using this microchannel system, we found that MYDGF-deficient iNeutrophils migrated two-times faster than their WT counterparts (Fig. 3D). We determined that this effect was MYDGF-specific, as re-expression of MYDGF was sufficient to dampen migration under confinement by over 50%, a level in line with the motility of WT iNeutrophils (Fig. 3D). Taken together, these findings indicate that MYDGF impairs neutrophil paracellular and interstitial confined migration.

Paracrine MYDGF signaling modulates iNeutrophil migration

MYDGF was previously identified as a paracrine factor that regulates inflammatory signaling through its release from bone marrow cells^[14-16]. While we demonstrated that neutrophil intrinsic MYDGF regulates motility and ROS production, we next sought to investigate its role in the collective neutrophil response and intercellular signaling.To determine if MYDGF is secreted by iNeutrophils, we probed for secretion using Western immunoblotting and immunoprecipitation from supernatants and found that MYDGF is secreted by WT but not MYDGF KO iNeutrophils (Fig. 4A). To test whether secreted MYDGF acts as a paracrine signal, we mixed equal numbers of WT and MYDGF KO iNeutrophils in confinement and found that MYDGF KO iNeutrophils in this mixed population migrated significantly slower than MYDGF KO cells alone and at a similar speed to WT iNeutrophils in the mixed population (Fig. 4B-C).

Figure 4. Secreted MYDGF acts as a paracrine mediator of neutrophil migration.

(A) Western blot of cell lysates or conditioned media (10mL at Day 6 of differentiation) removed from WT and MYDGF KO iNeutrophil differentiation preps that show presence or absence of hMYDGF protein. (B) Representative image of mixed population of WT (green) and MYDGF (red) iNeutrophils stained with Calcein AM (green; red-orange) in UOMS migrating towards fMLP (10μM). Scale bar = 20μm. (C) Quantification of migration speed (μm/min) of WT and MYDGF KO iNeutrophil migration in confinement following population mixing. Channels quantified: KO, n=5; mixed KO, n=5; mixed WT n=5. (D) Migration speed (μm/min) of MYDGF KO iNeutrophils following treatment with MYDGF KO or WT media in confinement towards fMLP (10μM). Channels quantified: KO in KO media, n=6; KO in WT media, n=6. (E) Migration speed (μm/min) of human peripheral blood neutrophils (PBNs) following treatment in MYDGF KO or WT media in confinement towards fMLP (10μM). Channels quantified: PBNs in KO media, n=9; PBNs in WT media, n=9. Individual points represent at least 5 channels across three biological replicates (C-E). Means ± SEM are shown. P values were calculated by one-way Anova (C) or unpaired Student’s t-test (D-E).

To rule out any effects of direct contact between cells, we cultured the MYDGF KO iNeutrophils in either WT or MYDGF KO conditioned media (Fig. 5SA). We found that MYDGF KO iNeutrophils migrated slower in the presence of media from WT iNeutrophils but not MYDGF KO conditioned media (Fig. 4D). Finally, we tested whether conditioned media from WT iNeutrophils regulates the population-level motility of human peripheral blood neutrophils (PBNs). We found that the velocity of PBNs was dampened 2.5-fold in the presence of WT conditioned media compared to MYDGF KO media (Fig. 4E). Taken together these data suggest that secreted MYDGF acts as an inhibitory signal that limits the migration of human neutrophils.

Deletion of MYDGF enhances calcium signaling in motile neutrophils

To identify a mechanism for the enhanced confined motility in MYDGF KO iNeutrophils, we investigated whether MYDGF regulates calcium handling. This is a particularly intriguing question since MYDGF is localized in the endoplasmic reticulum – a major store for intracellular calcium^[40]. First, we used intracellular calcium-sensitive dye Fluo-4 AM and found that calcium transients and migration speed of WT iNeutrophils were inhibited by treatment with BAPTA AM, an intracellular calcium chelator (Fig. 5A-B). Surprisingly, MYDGF KO iNeutrophils had a higher peak in calcium intensity compared to WT cells (Fig. 5C-D), suggesting that MYDGF acts as an endogenous inhibitor of neutrophil calcium flux. Additionally, we observed that WT cells appeared to distribute their calcium more evenly across the cells, while MYDGF KO iNeutrophils had more isolated pockets of high, intracellular calcium, suggesting the presence of calcium microdomains that may stimulate migratory signaling pathways^[41]. To confirm that deletion of MYDGF was not driving a chronic increase in cytosolic calcium, we measured resting calcium in the absence of stimulation and found no difference under resting conditions, suggesting that this is only observed after stimulation with fMLP (Fig. S6A). Along these lines, we found that during confined migration, MYDGF KO iNeutrophils maintained consistently higher Ca²⁺ intensity (Fig. 5E-F, Supp. Video 2) and that re-expression of MYDGF was sufficient to dampen this calcium increase (Fig. 5G). To confirm that the increased motility of MYDGF KO cells was mediated by calcium flux, we treated the cells with BAPTA AM and found that BAPTA-treated MYDGF KO iNeutrophils showed impaired migration under confinement (Fig. 5H-I). Together, these findings suggest that MYDGF dampens calcium transients, and this may provide a mechanism for its inhibitory effect on neutrophil motility.

Figure 5. MYDGF regulates calcium dynamics during confined migration.

(A) Representative images showing reduced calcium fluctuations in iNeutrophils in UOMS following treatment with BAPTA (100 μM) measured by Fluo-4 AM intensity. Scale bar = 10μm. (B) Migration speed (μm/min) of WT iNeutrophils in confinement with or without BAPTA (100μM) treatment for 30 mins. Channels quantified: WT, n=8; WT in BAPTA, n=7. (C) Representative single cell time lapse images of calcium waves over 2 minutes of migration to fMLP (10μM). Cells were stained with Fluo-4 AM. Scale bar = 10μm. (D) Quantification of single cell calcium transient intensity pictured in (C). (E) Representative images of WT and MYDGF KO iNeutrophils stained with Fluo-4 AM migrating to fMLP (10μM) in UOMS. Scale bar = 20μm. (F) Total calcium intensity (Fluo-4 AM signal) measured from migrating WT and MYDGF KO iNeutrophils in confinement towards fMLP (10μM). Channels quantified: WT, n=14; KO, n=14. (G) Total calcium intensity of MYDGF KO or MYDGF knock-in iNeutrophils in confinement towards fMLP (10μM). Channels quantified: KO, n=5; knock-in, n=5. (H) Representative images of MYDGF KO iNeutrophils in UOMS with or without BAPTA (100μM) treatment. Scale bar = 20μm. (I) Migration speed (μm/min) of MYDGF KO iNeutrophils in confinement towards fMLP (10 μM) with or without BAPTA (100μM) treatment for 30 mins. Channels quantified: KO, n=6; KO in BAPTA, n=5. Individual points represent 5 channels across three biological replicates (B, F-G, I). Means ± SEM are shown. P values were calculated by unpaired Student’s t-test (B, F, G, I).

Discussion

Innate immune inflammation is required for host defense against infection. However, excessive inflammation causes tissue damage and contributes to disease^[42]. Much of the work related to neutrophil biology focuses on pro-inflammatory signaling, including rapid migration to sites of tissue injury and infection. Drivers of directed migration include self-amplifying signals like LTB4 which enable rapid neutrophil chemotaxis and swarming^{[8, 43].} In homeostasis, inflammation is tightly controlled to prevent tissue damage. However, there is limited understanding of the neutrophil intrinsic mechanisms that dampen neutrophil inflammation^[44]. Here we identify MYDGF as a neutrophil intrinsic signal that puts the brakes on neutrophil motility and inflammatory responses.

We show that MYDGF dampens signaling mechanisms, including ROS production, essential for antimicrobial functions. It is intriguing that MYDGF KO iNeutrophils have improved antimicrobial activity even though they have a heterogeneous maturity profile compared to the more uniformly mature GATA1 KO iNeutrophils^[22]. These findings suggest that neutrophil maturity is not necessary for improved antimicrobial activity of iNeutrophils. It is also interesting that unlike GATA1 KO iNeutrophils MYDGF KO cells do not have increased NETosis. This is particularly helpful in that they are less likely to have tissue-damaging effects and may provide a useful option for infusion therapies for patients with resistant infections in the future.

Our findings suggest that MYDGF may function, at least in part, through effects on intracellular calcium signaling to regulate neutrophil motility. Previous studies have shown that transient increases in calcium are required for the migration of human neutrophils^[45]. In a recent study, it has been shown that STIM1/2 at endoplasmic reticulum-plasma membrane contact sites maintains cytosolic calcium levels necessary for adhesion and actin cytoskeletal rearrangements^[46]. Since MYDGF localizes to the ER, it is in proximity to these regulatory hubs which presents an intriguing potential mechanism for mediating intracellular calcium, but further work will be required to address this question. We found that MYDGF affected fMLP-induced motility and calcium transients, suggesting that MYDGF may work through a G-protein coupled receptor (GPCR) dependent pathway. Indeed, LTB4 signals through GPCR to induce neutrophil motility in swarming and further calcium flux in a self-amplifying pathway^[47]. It will be interesting to determine if MYDGF affects this positive feedback mechanism. However, further research will be required to fully understand the relationship between MYDGF, cell motility and calcium dynamics. A challenge for future studies will be to identify the MYDGF receptor that mediates MYDGF paracrine signaling to limit neutrophil motility, since the MYDGF receptor remains unknown.

Previous research from our group demonstrated a role for MYDGF as a regulator of neutrophil infiltration in zebrafish^[13], although it remained unclear whether this effect was environmental or neutrophil intrinsic. Our findings now show that MYDGF functions as a neutrophil intrinsic paracrine factor that provides cell-cell communication between neutrophils. It is possible that MYDGF has additional intracellular roles in the ER to modulate calcium transients and further affect cell substrate adhesions and actin dynamics.

Our findings further support the use of iNeutrophils to study fundamental properties of human neutrophil function and motility, in a genetically tractable system. iNeutrophils address a key gap in the neutrophil field. It is also possible that iNeutrophils will have therapeutic use to treat invasive infections or in cancer immunotherapy ^{[48, 49]}. MYDGF KO cells demonstrate improved function over WT cells with improved motility and antimicrobial effects. Moreover, the MYDGF KO cells show similar lack of maturity and heterogeneity as the WT cells, despite having improved function. While heterogeneity has been criticized as an issue associated with non-primary models of human neutrophils^[50], recent evidence indicates that primary human neutrophils exhibit significant heterogeneity, which may potentially improve the neutrophil response^[51]. Although it was believed that neutrophils must be mature to function effectively, our findings suggest that this may not be the case. MYDGF KO iNeutrophils have improved effector functions even though they have similar maturity to the WT cells, suggesting that maturity is not necessary for effective neutrophil function in this context.

iNeutrophils provide a promising opportunity for cell-based, neutrophil-targeted therapies to address conditions such as neutropenia which affects millions worldwide^[52]. Neutropenic patients are susceptible to infection and current treatment options are limited by donor need, short neutrophil half-life, and poor efficacy^[53]. iNeutrophils address these shortcomings in addition to being genetically tractable, presenting the possibility of harnessing key neutrophil pathways, such as MYDGF, to enhance their efficacy and future use to treat patients.

Human iPSC culture and iNeutrophil differentiation

Bone marrow-derived human iPSCs (hiPSCs) – IISH2i-BM9 (WiCell) – were differentiated into neutrophils as previously described^{[20, 22]}. hiPSCs were cultured in mTesR-Plus medium (STEMCELL Technologies) on tissue culture plates coated with Cultrex^™ (Biotechne).

hiPSCs were first differentiated into hemogenic endothelium in TeSR-E8 medium (STEMCELL Technologies) using ETV2 modified messenger RNA (mmRNA) (TriLink Biotechnologies), TasnIT reagent and mRNA boost (Mirus Bio). Cells were detached using TrypLE Select (Life Technologies) one hour prior to transfection. Lifted cells were then plated onto tissue culture plates coated with human collagen type IV (2.4μg/mL; Millipore Sigma) in TeSR-E8 medium supplemented with ROCK inhibitor Y-27632 (ROCKi; Tocris). One day after transfection, the medium was modified to StemLineII media (Sigma Aldrich) supplemented with FGF2 (20ng/mL; PeproTech) and VEGF-165 (20ng/mL;PeproTech) to induce hemogenic endothelial cell differentiation; this cocktail is known as Media A. After two days media was changed to induce granulocyte-monocyte progenitor differentiation using StemLineII media (Sigma Aldrich) supplemented with FGF2 (20ng/mL; PeproTech), GMCSF (25ng/mL; Peprotech), and UM171 (50nM; Xcess Biosciences); this cocktail is known as Media B. On days 8 to 10 following transfection, suspension cells were collected and plated for terminal neutrophil differentiation on tissue culture treated plates. These cells were cultured using StemSpan SFEM II medium (STEMCELL Technologies) supplemented with ExCyte 0.2% (Merck Millipore), GlutaMAX (1X; Thermo Fisher Scientific), and AM580 (2.5μM; STEMCELL Technologies) at a density of 1x10⁶ cells/mL; this cocktail is known as Media C. After four days, Media C was refreshed on the suspension cells. iNeutrophils were then collected from the supernatant 8 to 10 days after this media change.

Immunoprecipitation

10x10⁶ WT or MYDGF KO granulocyte-monocyte progenitors (GMPs) were seeded in 10mL Media C and fed with 10mL Media C 4 days later. Differentiation continued for 6-10 days from GMP collection. On day 6-10 media was collected, and cells were removed by centrifugation and kept for analysis. HALT protease/phosphatase inhibitor was added to cleared media and either from at −80°C or immediately used for immunoprecipitation of secreted MYDGF. 1μg of Rabbit anti-MYDGF or 1μg Rabbit IgG from pre-immunized rabbit was added to 10mL of media on ice and incubated end-over-end overnight. 10μL of Protein A/G Plus Agarose was added and incubated for 4 hours end-over-end to capture antibody-protein complexes. Beads were collected by centrifugation and washed 3 times in cold PBS, then eluted with 5X Protein Loading Buffer and resolved on 10-20% gradient gels.

Sequencing, Initial Data Processing and Quality Control

All libraries were pooled and sequenced at DNA Sequencing Facility (RRID:SCR_017759) on a NovaSeq 6000 sequencer S1 flow cell with paired end, 50bp sequencing. Data was processed with bcl2fastq for the generation of FASTQ files. Raw sequencing reads in the FASTQ files were mapped to GRCh38 human genome by Cell Ranger (v6.1.2, 10X genomics). Filtered cell-by-gene matrices were used for downstream analysis in R with Seurat^[54] (v4.3.0). Based on general distribution, only cells with more than 50 genes captured and a percentage of mitochondrial genes below 10% were kept. Doublets were labeled by DoubletFinder^[55] (v2.0.3) with principal components (PCs) 1:30, and parameters for generated artificial doublets set at 25%, and neighborhood size at 0.09. Post doublet removal, each sample was normalized individually by SCTransform, regressing out the impact of number of features.

Data integration, Unsupervised clustering and population annotations

To integrate normalized wildtype and MYDGF^−/− samples into one data object, we chose top 3000 highly variable genes in each sample as features for creating anchors. Samples were integrated by IntegrateData using the anchors as input, with normalization method set to SCT to match with SCTransform normalization process. Post integration, linear and nonlinear dimensional reduction and unsupervised clustering were performed. By examining the distribution of quality control measurements, we identified and removed one cluster that was low in number of features and did not have differentially highly expressed genes comparing to the rest. Post cluster removal, integration and dimensional reduction with principal component analysis (PCA) and UMAP were repeated. Specifically, PCs 1:30 were used in UMAP and neighbor-finding. Unsupervised clustering was done at a resolution of 0.1. For cluster annotations, differential expression analysis was done by Wilcoxon rank-sum test using FindAllMarkers with log2 fold change threshold set to 0.25. Functional annotation for differentially expressed gene signatures was done by gene ontology enrichment analysis through Metascape^[56] (v3.5.20250101) and individual gene curation through literature review.

Neutrophil maturation estimation

For maturation stage reference, we used the processed Seurat object provided by Montaldo et al..^[36] Briefly, SCTransform was performed for both the reference neutrophil atlas and our iNeutrophil data. Anchors between the two datasets were identified by FindTransferAnchors using PCA with PCs 1:30 as reference reduction space, and SCT as normalization approach. Label transfer from reference to iNeutrophils was performed by TransferData with the same dimensions used in anchor identification. Predicted maturation stage names were added to iNeutrophils as a new metadata.

iPSC-derived endothelial cell differentiation and culture

iPSC-derived endothelial cells were generated using ETV2 transfection and differentiating wild-type IISH2i-BM9 iPSCs (WiCell) through the hemogenic endothelium stage as previously described^[23]. The IISH2i-BM9 iPSCs were transfected with ETV2 mmRNA and cultured in StemLineII media (Sigma-Aldrich) supplemented with FGF2 (20ng/mL; Peprotech) and VEGF-165 (20ng/mL; PeproTech). Two days following transfection, medium was change to Vasculife Basal Medium (LifeLine Cell Technologies) supplemented with the iCell-Endothelial Cells Medium Supplement (CDI). iPSC-derived endothelial cells were plated in tissue culture flasks pre-incubated with 1X fibronectin bovine plasma (100X; Sigma-Aldrich).

Aspergillus fumigatus culture and collection

Red fluorescent protein (RFP)-expressing Aspergillus fumigatus (CEA10) was grown as previously described^[57]. Briefly, A. fumigatus was grown on glucose minimal medium (GMM) plates in the dark at 37°C to induce asexual conidiation. The A. fumigatus was plated at 1 x 10⁶ conidia/10cm plate for 3 days. The conidia were collected in water + 0.1% Tween by scraping with an L-spreader and subsequent passage through 0.4 μm Miracloth into a 50mL conical tube. A 5mL aliquot of this spore suspension was transferred to a new tube and then centrifuged for 5 min at 800 g and resuspended in 50 mL of PBS (1X). The collected conidia were the counted using a hemacytometer and the final concentration was adjusted to 4 x 10³ spores/mL.

NETosis assay

A plate reader, fluorescent signal-based assay was used to evaluate production of intracellular ROS by the iNeutrophil cell lines. One day prior to experimentation, a 96-well black, clear-bottom tissue culture treated plate was coated with 1X fibronectin bovine plasma (100X; Sigma-Aldrich) and incubated at 4° C overnight. Cells were resuspended in of Phenol Red Free RPMI supplemented with 2% FBS and 200,000 cells in 100μL were plated into each well. PMA (100ng/mL final concentration) was added to each well and the plate was incubated for 6 hours at 37°C. Following incubation, Sytox Green (Invitrogen) was added at a final concentration of 375 nM. An initial plate read was taken at 0 minutes and additional readings were taken at 10 and 20 minutes; the microplate was incubated between readings. Fluorescent readings were acquired using a Victor3V microplate reader (PerkinElmer) to quantify release of extracellular DNA using fluorescence (500/528 nm). The background signal from unstimulated cells with Sytox Green was subtracted from corresponding cell lines treated with PMA.

Inflammatory cytokines enzyme-linked immunosorbent assays (ELISAs)

ELISA was used to determine the concentrations of secreted inflammatory cytokines IL-8, IL-1β, IL-6 and TNFα by iNeutrophil cell lines. A tissue culture-treated 12-well plate was coated with fibronectin (10μg/mL). The cells were resuspended in Phenol Red Free RPMI + 2% FBS and 1.5 million iNeutrophils were plate per well of the 12-well plate. The cells were left to rest for 30 minutes at 37°C prior to stimulation with either E. coli LPS (200ng/mL) or zymosan (10μg/mL) for 4 hours at 37°C. iNeutrophils were collected and spun down at 350 xg. All supernatants were collected and frozen at −80° C. Cytokines concentrations were assessed using Human IL-8/CXCL8 DuoSet ELISA (Bio-Techne), Human IL-1 beta/IL1F2 DuoSet ELISA (Bio-Techne), Human IL-6 DuoSet ELISA (Bio-Techne), and Human TNFα DuoSet ELISA (Bio-Techne) following manufacturer instructions. All supernatant samples were appropriately diluted to fit within the standard curve and normalized to cell input number.

Analysis of confined migration assay in liquid-walled microchannels

Cell tracking was performed using the Manual Tracking PlugIn available within the FIJI image analysis software. For tracking migration speed in the liquid microchannels, between 3 and 10 cells were tracked within at least 5 channels pooled over 3 independent biological replicates (i.e. independent neutrophil isolation). With respect to calcium analysis, mean intensity of Fluo4-AM was calculated in manually drawn regions of interest (ROI’s) outlining individual cells (5 per channel). All fluorescent intensity values within each individual biological replicate (i.e. neutrophil isolation) were normalized to range between 0 and 1 based on the range of intensities for that given replicate

Phagocytosis Assay with Pseudomonas aeruginosa

Pseudomonas aeruginosa GFP (PAK:GFP) was grown overnight in LB+ 200ug/ml carbenicillin, then 2ml’s was transferred to 6ml’s of LB+ 200ug/ml carbenicillin and grown until OD600=0.7, about 2.5 hours. 1ml of PAK:GFP was pelleted at 2000g for 3 min. and washed 2x 1ml PBS. PAK:GFP was then opsonized for 1 hour at 37c in 50% normal human serum. PAK:GFP was washed 3x ddH2O then resuspended in 100ul and diluted 1:100 for OD600. An OD600=2.5 was determined to be 1x106cfu’s/ul. 5x10⁶ opsonized PAK:GFP was added to 2x106 iNeuts in 500ul StemSpanII+ 2% human serum (IgG depleted and heat inactivated) and incubated end-over-end for 1 hour at 37c. iNeuts were pelleted at 350xg for 3min and washed in phagocytosis media + 50ug/ml gentamicin. iNeuts were incubated for 1 hour in phagocytosis media + 50ug/ml gentamicin to kill extracellular bacteria. 4x105 iNeuts were plated per well of an 8 well ibidi camber with NucBlue live cell stain (Invitrogen) and allowed to settle for 30 min before imaging. Images were collected using a Nikon Eclipse Ti2 microscope equipped with a Crest X-light V3 spinning disc and Kinetix 22 monochrome camera. Imaging was performed using Aprochromat 60X/1.42 NA or S-Fluor 20X/0.75 NA objectives in combination with Tokai Hit stage incubation system used for imaging iNeutrophil and PAK co-cultures.

Supplementary Material

Fig. S1

Supplementary Figure 1. Generation of MYDGF KO and rescue iNeutrophils.

(A) Schematic of 42 base pair deletion in exon 4 of hMYDGF (human MYDGF) that generated functional knockout of MYDGF in iPSCs. (B) Western blot demonstrating presence of MYDGF in human primary neutrophils and WT iNeutrophils and absence in MYDGF KO iNeutrophils. (C-E) Mean fluorescence intensity (MFI) values for (C) CD11b, (D) CD15, (E) CD16, and (F) CD15/CD16 double positive expression on iNeutrophils measured by flow cytometry. WT, KO: N=3. (G) Western blot demonstrating successful knock-in of hMYDGF into both WT and MYDGF KO iPSCs to generate a MYDGF rescue cell line for differentiation into iNeutrophils. Experiments were performed a minimum of three times, replicates are indicated by dots on each plot (C-F). Means ± SEM are shown. P values were calculated by unpaired Student’s t-test (C-F).

Fig. S3

Figure S3. MYDGF suppresses inflammatory cytokine release and phagocytosis, but not NET production.

(A-B) ELISA-based measurement of (A) IL-8 and (B) TNFa release from WT and MYDGF KO iNeutrophils following stimulation with LPS or zymosan. WT, KO: N=6. (C) Fluorescent-based measurement of NET release over time following stimulation with PMA (100ng/mL) using SYTOX green. WT, KO: N=3. (D-F) MFI and associated histograms of (D) Dectin-1, (E) TLR2, and (F) TLR4 measured using flow cytometry. WT, KO: N=3. (G) Representative images of iNeutrophil phagocytosis of GFP-tagged Pseudomonas aeruginosa. Scale bar=10μm. (F) Quantification of iNeutrophil phagocytosis of P. aeruginosa in a co-culture assay. Quantifications calculated using the ratio of nuclear (blue) signal to bacterial (green) signal. WT, KO: N=3. Experiments were performed a minimum of three times, replicates are indicated by dots on each plot (A-F, H). Means ± SEM are shown. P values were calculated by unpaired Student’s t-test (A-F, H).

Fig. S2

Figure S2. iNeutrophil cluster identity confirmation.

(A) Relative expression level of top signature genes in each cluster as labeled in Figure 1G. Darkness of a dot represents level of expression, and its size represents percentage of cells with detectable expression. (B) Expression distributions of common neutrophil markers across iNeutrophil clusters. (C) Maturity level of iNeutrophils predicted from Montaldo et al.. UMAP axes are the same as Figure 1G. (D) Expression distributions of maturation stage markers across iNeutrophil clusters.

Fig. S4

Figure S4. WT and MYDGF KO iNeutrophils express similar levels of adhesion molecules.

(A-D) Flow-cytometry measurement of (A) CD49d, (B) CD11c, (C) CD18, and (D) ITGB7 surface markers of neutrophil adhesion. WT, KO: N=3 (A, B, D). WT, KO: N=5 (C). Experiments were performed a minimum of three times, replicates are indicated by dots on each plot (A-D). Means ± SEM are shown. P values were calculated by unpaired Student’s t-test (A-D).

Fig. S5

Figure S5. Collected media from iNeutrophil differentiation can be used to demonstrate the effects of secreted MYDGF on migration.

(A) Schematic depicting the experimental workflow for treating iNeutrophils and human peripheral blood neutrophils (PBNs) with ‘spent’ differentiation media collected from WT and MYDGF KO iNeutrophils. ‘Spent’ media refers to media which cells had been grown in for full differentiation. Image made using Biorender.com.

Fig. S6

Figure S6. Calcium intensity does not differ at homeostasis.

(A) Mean calcium intensity of unstimulated iNeutrophils, measured using Fluo-4 AM calcium stain. iNeutrophils were unconfined and not migrating. Single cell calcium intensity quantified: WT, n=12; KO, n=12. Individual points represent 12 cells across two biological replicates. Means ± SEM are shown. P values were calculated by unpaired Student’s t-test (A).

Video S1

Supplementary Video 1. iNeutrophil transendothelial migration.

SV1. WT and MYDGF KO iNeutrophils migrate out of an endothelial lined lumen model of extravasation. Timelapse video collected on Nikon TE300 Epifluorescent Micropscope at 10X magnification. Images were taken every 10 minutes for 16 hours and timestamp and scale bar were added using Fiji-ImageJ software. Scale bar = 250μm.

Video S2

Supplementary Video 2. iNeutrophil calcium dynamics during migration in confinement.

SV2. WT and MYDGF KO iNeutrophils in an underoil microchannel model of confined cell migration towards a source of fMLP (10μM). Cells were stained with Fluo4-AM intracellular calcium-labelling dye. Timelapse videos collected on Nikon Eclipse Ti2 Confocal Microscope at 20X magnification. Images were taken every 2.5 seconds for 5 minutes and timestamp and scale bar (20μm) were added using Fiji-ImageJ software.

Supplemental Text

Supplementary Table 1. Guide RNAs and Primers.

Supplementary Table 2. Flow Cytometry Antibodies.

Supplementary Table 3. Immunoblotting Antibodies.

Data Accessibility Statement – Data Accessibility Statement:

All raw and processed sequencing data generated in this study have been submitted to the NCBI Gene Expression Omnibus (GEO; http://ncbi.nlm.nih.gov/geo/) under accession number GSE308566. scRNA-seq analysis codes are available at: https://github.com/k326xh/iNeut_scRNA_analysis.