The zebrafish experiment was conducted in accordance with internationally accepted standards. The Animal Care and Use Protocol was approved by The Purdue Animal Care and Use Committee (PACUC), adhering to the Guidelines for the use of Zebrafish in the NIH Intramural Research Program (protocol number: 1401001018). Zebrafish roraa gene was cloned from zebrafish mRNA using SuperScript III RT-kit (Invitrogen #18080044) and amplified with the following primers. zroraa wt+: 5′- aaaaccccggtcctatgcatatgatgtattttgtgatttcagctatgaaagct-3′; zroraa wt-: 5′-ctgattatgatctagactacccgtcaacgggcat-3; zroraa LBD deletion DN+: 5′- aaaaccccggtcctatgcatatgatgtattttgtgatttcagctatgaaagctcaaatcg-3′; zroraa LBD deletion DN -: 5′- ctgattatgatctagagtccaggccggattgatcagg-3 and inserted into a Tol2-lyzC-mcherry-2A backbone by using infusion cloning kit (Takara #638920). The construction method for neutrophil-specific Cas9 expression and the guide RNA expression fish lines has been described in our previous study (26). The protospacer sequences for targeting dre-roraa in this research are 5′-taattccctgcaagatctg-3′; 5′-gcacgttatcacgccataa-3′, both guide RNAs target on dre roraa exon2. Microinjections of fish embryos were performed by injecting 1nl of a mixture containing 25 ng/µl plasmids and 35 ng/µl tol2 transposase mRNA in an isotonic solution into the cytoplasm of embryos at the one-cell stage. The stable lines were generated as previously described (27). At least two founders (F0) for each line were obtained. Experiments were performed using F2 larvae produced by F1 fish derived from multiple founders to minimize the artifacts associated with random insertion sites.

Kidney marrow was dissected from 3 adults from Tg(lyzC: miR-99-1-dendra2)^pu25 , Tg(lyzC: vector-dendra2)^pu7 , Tg(lyzC: mcherry-2a)^{pu 32} and Tg(mcherry-2a-RORα DN) ^pu33 and neutrophils were sorted using FACS. Total RNA was extracted using RNeasy Plus Mini Kit (Qiagen #74104). RNAseq was performed at The Center for Medical Genomics at Indiana University School of Medicine. Samples were polyA enriched and sequenced with Illumina HiSeq 4000 ultra-low with reads ranging from 37M to 44M. The RNA-seq aligner from the STAR (v2.5) (28) was employed to map RNA-seq reads to the reference genome, zebrafish (GRCz11), with previously the following parameter: “–outSAMmapqUnique 60”. Uniquely mapped sequencing reads were assigned to genes using featureCounts (from subread v1.5.1) (29) with the following parameters: “-p -Q 10”. The genes were filtered for further analysis if their count per million (CPM) of reads was less than 0.5 in more than three samples. The method of a trimmed mean of M values (TMM) was adopted for gene expression normalization across all samples, followed by differential expression analysis between different conditions using edgeR (v3.20.8). Differentially expressed gene was determined for the comparison if its false discovery rate (FDR) adjusted p-value was less than 0.05 and the amplitude of fold change (FC) was larger than linear 2-fold. The functional analysis was performed on DEGs of our interest with a cutoff of FDR < 0.05 to identify significantly over-represented Gene Ontology (GO) and/or KEGG pathways by using the DAVID (30). To compare the down-regulated DEG set in rorα DN neutrophil to the published THP-1/HUVEC RORα CHIP-seq data, the genes from CHIP-seq result were converted to their orthologues in zebrafish. The gene duplicates after the conversion from human cells were removed. The zebrafish genes not having the orthologues to human were also removed before comparison.

Total RNA was purified using MiRVANA miRNA purification kit (ThermoFisher) for miRNA or RNeasy Plus Mini Kit (Qiagen #74104) for mRNA. MicroRNAs were reverse transcribed with Universal cDNA Synthesis Kit II (Exiqon). MicroRNA RT-qPCR was performed with ExiLENT SYBR^® Green master mix (Exiqon), and mRNA with one-step RT-qPCR was performed with SuperScript^® III Platinum^® SYBR^® Green kit (Invitrogen) using LightCycler ^® 96 Real-Time PCR System (Roche Life Science). The specificity of the primers was verified with a single peak in the melt curve. The relative fold change with correction of the primer efficiencies was calculated following instructions provided by Real-time PCR Miner (http://ewindup.info/miner/data_submit.htm) and normalized to U6. miRNA Primers used in this study were purchased from Exiqon. The relative fold change with correction of the primer efficiencies was calculated following instructions provided by Real-time PCR Miner and normalized to rpl32. Primers: dre-grn1+: 5′-aacccagccagcaagatg-3′, dre-grn1−: 5′-ccaccgggatagacagatca-3′, dre-fabp7a+: 5′-cagcagacgatagacatgtgaagt-3′, dre-fabp7a−: 5′-tcccacctctgaacttggac-3′, dre-il1b+: 5′-atgctcatggcgaacgtc-3′, dre-il1b−: 5′-tggttttattgtaagacggcact-3′, dre-tgfbr1+: 5′-tgcaacaagaacccaaaagtt-3′, dre-tgfbr1−: 5′-gggactcatagactggggttc-3′, dre-roraa+: 5′-gaatgatcagatagtgcttctcaaag-3′, dre-roraa−: 5′-cacggcacattctgacga-3′, dre-rorc+: 5′-tcttttcctatccaacctctctaca-3′, dre-rorc−: 5′-gagtggtctctttatgtgagcgta-3′, dre-rpl32+: 5’- tcagtctgaccgctatgtcaa-3’; dre-rpl32-: 5’- tgcgcactctgttgtcaatac-3’; dre-ef1a+:5’- tgccttcgtcccaatttcag-3’; dre-ef1a-:5’- taccctccttgcgctcaatc-3’; dre-roraa-overexpression+:5’- tcaggcatccattatggcg-3’; dre-roraa-overexpression-:5’- cggtcaatcaggcagttctt-3’; dre-arhgap17b+:5’- ttggcggatgaagaggattc-3’; dre-arhgap17b-:5’- ccaatcgcacattttccagg-3’; dre-ptk2ba+:5’- attacatgcagcacaacgcc-3’; dre-ptk2ba-:5’-ccagacccacatctttctcca-3’; dre-mhc1zba+:5’- gaagattcccaaacagcactgg-3’; dre-mhc1zba-:5’-cgttccatcagaatgttgacgttc-3’; dre-tcerg1l+:5’-aagcagagaacagcggataag-3’; dre-tcerg1l-:5’-ttagcctctgggtttggattg-3’.

zebrafish roraa 3′UTR was amplified from zebrafish genomic DNA with SuperScript III RT-kit (Invitrogen #18080044) and cloned into psiCHECK2 (Promega) at XhoI and NotI cloning sites with the following primers: zroraa+:5’- taggcgatcgctcgacaaggcagctcactaggaacagaactg-3’; zroraa-: 5’- ttgcggccagcggccggaggacagggaaacgtagctgtac-3’. Mutated 3’UTR constructs were generated using Infusion HD cloning kit (Clontech) with the following primers: zroraa mut+: 5’- acaaaatgcccattctacagggtaaccactgctgtgg-3’; zroraa mut-: 5’- gaatgggcattttgtcgagcgacggacgtc-3’. DNA encoding miR-99 or vector was amplified from the construct used for expression in zebrafish and inserted into pcDNA3.1 at the HindIII/XbaI cloning sites using the following primers: pcDNA+: 5;- gtttaaacttaagcttgccaccatggatgaggaaatcgc-3’; pcDNA-: 5’-aaacgggccctctagagaccggtacccccgggctgc-3’.

Larvae at three dpf were settled on a glass-bottom dish, and imaging was performed at 28 °C. Time-lapse spinning disk confocal microscopy (SDCM) was performed with a Yokogawa scanning unit (CSU-X1-A1) mounted on an Olympus IX-83 microscope, equipped with a 20X 0.5–numerical aperture (NA) UPlanSApo oil objective (Olympus) and an Andor iXon Ultra 897BV EMCCD camera (Andor Technology). GFP and mCherry were excited with 488 nm and 561 nm and fluorescence emission collected through 525/30-nm and 607/36-nm filters, respectively, to determine the localization of stable actin in migrating neutrophils in vivo. Images were captured using MetaMorph version 7.8.8.0 software at 10 s intervals for a total of 5 min. Time-lapse fluorescence images in the head mesenchyme were acquired with a laser-scanning confocal microscope (LSM710, Zeiss) with a Plan-Apochromat 20×/0.8 M27 objective. The fluorescent stacks were flattened using the maximum intensity projection and overlaid with a single slice of the bright-field image. The velocity of neutrophils was quantified using ImageJ with the MTrackJ plugin.

Zebrafish wounding and infection were performed as described (31). Briefly, three dpf larvae were amputated posterior to the notochord or inoculated with P. aeruginosa (PAK) into the left otic vesicle or the vasculature at 1000 CFU/embryo. The larvae were fixed in 4% paraformaldehyde at one h post-wounding or -infection. Neutrophils were stained with Sudan Black, and the number at the indicated regions was quantified.

HL-60 cells were obtained from ATCC (CCL-240), and cultured using RPMI-1640 with HEPES supplemented with 10% FBS and sodium bicarbonate. The lentiviral backbone pLIX_403 was a gift from David Root (Addgene plasmid # 41395). The DNA sequence flanking mir-99 was cloned from HL-60 cell genomic DNA and cloned into a backbone containing dendra2. The miRNA and dendra2 reporter were then cloned into pLIX_403 vector using the NheI/AgeI sites with primer set: pLIX-mir+: 5’ tggagaattggctagcgccaccatggatgaggaaatcgc-3’, pLIX-mir-: 5’-catacggataaccggttaccacacctggctgggc-3’. Stable HL-60 cell lines were generated as described (32, 33). Briefly, HEK-293 cells were transfected with pLIX_403 or pLKO.1, together with VSV-G and CMV8.2 using lipofectamine 3000 (ThermoFisher). The viral supernatant was harvested on day three and concentrated with Lenti-X concentrator (Clontech 631232). HL-60 cells were infected with concentrated lentivirus in complete RPMI medium supplemented with four µg/ml polybrene (Sigma TR-1003-G) at 2500g for 1.5h at 32°C and then selected with one µg/ml puromycin (Gibco A1113803) to generate stable lines.

Cells were stained with Alexa Fluor 647 conjugated CD11b (neutrophil differentiation marker, Biolegend, 301319, clone IV-M047, 2ul to 1 x 10⁶ cells) or the isotype control antibody (Biolegend, 400130, clone MOPC-21, 2ul to 1 x 10⁶ cells), RUO AnnexinV (apoptosis marker, BD, 563973, 5ul to 1 x 10⁶ cells) and RUO propidium iodide (dead cell marker, BD, 556463, 5ul to 1 x 10⁶ cells). Cells were stained in staining buffer (1% BSA, 0.1% NaN3) and incubated on ice for 1 hour, washed three times with staining buffer, and resuspended in suitable volumes. Cell cycle profiling was obtained using Vybrant™ DyeCycle™ Ruby Stain (Thermo #V10309). Fluorescence intensity was collected using BD LSR Fortessa. Results were analyzed with Beckman Kaluza 2.1 software.

Transwell assays were performed as described (32). Briefly, 2x10⁶ cells/ml differentiated HL-60 cells were resuspended in HBSS with 0.5%FBS and 20 mM HEPES. 100 µl were placed into a 6.5 mm diameter 3 µm pore size transwell insert (Corning#3415) and allowed to migrate for two h at 37°C towards 100 nM fMLP in 500 µl of HBSS in a 24-well plate. Loading controls were done by directly adding 100 µl of cells to 400 µl of HBSS. Cells that migrated to the lower chamber were released with 0.5 M EDTA and counted using a BD LSRFortessa flow cytometer with an acquisition time of 30 s. The counts were normalized with the total numbers of cells added to each well the data was then gated for live cells and analyzed with Beckman Kaluza 2.1 software.

Differentiated HL-60 cells (dHL60) were resuspended in HBSS in 20 mM HEPES with 0.5% FBS (mHBSS) and allowed to attach to fibrinogen-coated slides for 30 min at 37°C. Neutrophil extracellular traps (NETs) were induced with 50 nM PMA (sigma #P1585) in HBSS for 4 h at 37°C. NETs were enumerated using cell-permeable Hoechst 33258 at 1 µg/ml (Thermo #62249) and cell impermeable Sytox Green at 1 µM (Thermo #S7020). Images were acquired using Lionheart FX Automated Microscope (Biotek) 10x phase objective, Plan Fluorite WD 10 NA 0.3 (1320516). The percentage of cells forming NETs was calculated by dividing the number of Sytox Green-positive cells by that of the Hoechst-positive cells.

Amplex™ Red Hydrogen Peroxide/Peroxidase Assay Kit (Thermo #A22188) was used to detect extracellular ROS. dHL60 cells (20,000 cells per well) were incubated with 0.1µg/mL of PMA (sigma #P1585) in 96 well black/clear bottom plate (Thermo #165305). ROS reading was taken 30 - 45 minutes after stimulation using Synergy Neo2 (BioTek) at Ex/Em of 571/585 nm. The amount of hydrogen peroxide release was calculated by comparing to standards prepared.

BioParticles from pHrodo™ Green BioParticles^® Conjugates (Thermo # P35366) were reconstituted and sonicated for 5 minutes at 10% amplitude with 10 seconds intervals. dHL60 cells (5x10⁵) were resuspended in 100uL of BioParticle suspension and incubated at 37°C for 1 hour. The reaction is stopped by placing on ice before running flow cytometry to quantify the mean FITC intensity in the cell population. Cells with beads placed on ice for 1 hour were used as the negative control.

ROR inhibitors SR3335 (Cayman # 12072), SR2211 (Cayman # 11972), SR1001 (Cayman # 10922), and VPR-66 (Novus # NBP2-29335) were dissolved in DMSO to make a 100 mM stock, then further diluted in E3 to working concentrations (10-100 µM for recruitment and motility assays, and indicated concentration for survival assays). For neutrophil recruitment and random motility assays, larval were pretreated with the inhibitor for one h before experimental procedures. For survival assays, larval were pretreated and kept in the inhibitors from 1-hour post-infection.

Primary human neutrophils were obtained from the peripheral blood of healthy adult donors, collected under approval by Purdue University’s Institutional Review Board (IRB). Primary human neutrophils were isolated with Milteny MACSxpress Neutrophil Isolation Kit (Macs # 130-104-434), and RBC lysed for 10 min RT (Thermo # 00-4300-54). 3x10⁶ cells/ml PMNs were incubated in HBSS with 0.1% FBS and 20mM HEPES with or without respected inhibitors for 1 hour. Cells were then mixed 3:1 with matrigel (corning # 356234) and loaded into collagen-coated IBIDI chemotaxis µ-slides (ibidi # 80326) and incubated at 37°C for 20 minutes to allow 3D matrix formation. 15 μl of 1000 nM fMLP was loaded into the right reservoir yielding a final fMLP concentration of 187 nM. Chemotaxis was recorded every 1 min for two h, with a Lionheart FX Automated Microscope (Biotek) 10x phase objective, Plan Fluorite WD 10 NA 0.3 (1320516). Chemotaxis tracking was done with the MTrackJ image J plugin and plotted in Prism 6.0 (GraphPad). Cell viability of primary human neutrophils treated with ROR inhibitor SR3335 (Cayman # 12072) was measured by staining with propidium iodide and analyzed using flow cytometry.

1.5x10⁶ cells/ml of primary human neutrophils were resuspended in DPBS. They were lysed with ice-cold lysis solution in a 1:1 ratio (20% TCA, 40 mM NaF, and 20 mM β-glycerophosphate). Cells were incubated on ice for 1-2 hours and then pelleted down. The protein pellet was washed twice with 0.75 mL 0.5% TCA and resuspended in 30 μl of 2x Laemmli sample buffer (Sigma #S3401). 15 μl of the suspension is loaded per well on a 10% SDS-PAGE gel. Samples were transferred to polyvinylidene difluoride (PVDF) membrane and blocked with 2.5% non-fat milk in PBST. Primary antibodies used were RORα Polyclonal Antibody (Thermo #PA1-812) and vinculin (Thermo #MA5-11690) diluted into 1% BSA at 1:1000 and 1:5000 ratios, respectively.

To determine the mutation efficiency in Tg(LyzC : Cas9, Cry : RFP, U6a/c: rora guides, LyzC : GFP)^pu34 , the roraa loci around the sgRNA binding site was amplified by PCR using the following primers: dre roraa+, 5′- cgactggtgacagtgaaaaatc-3′; dre roraa-, 5′- tttagcttcttaccttgcagcc-3′. PCR products were deep sequenced using WideSeq at the Purdue Genomics Core Facility. Mutation efficiencies were calculated using the CRISPResso2 package (www.github.com/pinellolab/CRISPResso2).

Larvae at three dpf were injected with one nl of 25 ng/nl LPS or 1000 CFU P. aeruginosa (PAK) into the tail vein and incubated individually in 96-well plates. Survival was tracked for four days or when one group reached 100% mortality. Representative results of at least three independent experiments (n≥20 larvae in each experiment) were shown.

Statistical analysis was carried out using PRISM 6 (GraphPad). Mann–Whitney test (comparing two groups), Kruskal–Wallis test (when compared to a single group), and Gehan–Breslow–Wilcoxon test (for survival curve comparison) were used in this study as indicated in the figure legends. Individual p values are indicated in the figure, with no data points excluded from statistical analysis. One representative experiment of at least three independent repeats is shown. For qPCR, each gene was normalized to a reference gene and compared with the Holm-Sidak test (individual comparison with paired value), where each p-value and DF were adjusted for respective multiple comparisons.

We use the lyzc promoter to drive gene expression, specifically in zebrafish neutrophils (34, 35). The microRNA gene was inserted into the intron of a fluorescent reporter gene ( Figure 1A ) to allow visualization of neutrophils expressing the transgene. To further characterize the zebrafish line Tg(lyzC: miR-99-1/dendra2) ^pu25 with this referred to as miR-99, we first isolated neutrophils and performed miRNA RT-qPCR to confirm miR-99 overexpression. Compared with the control line that only expresses the vector, the miR-99 level is significantly higher in the miR-99 line ( Figure 1B ). The two abundant neutrophil miRNAs, miR-223 and let-7e, were used as controls to demonstrate that this overexpression did not disrupt physiological microRNA levels. miR-99 overexpression did not affect the total neutrophil numbers in embryos ( Figures 1C, D ) but reduced neutrophil recruitment to either the tail injury or ear infection sites ( Figures 1E–H ). In addition, the spontaneous neutrophil motility in the head mesenchyme was significantly slower upon miR-99 overexpression ( Figures 1I, J and Movie 1 ). We then sought to determine whether miR-99 inhibits human neutrophil migration. Here we used HL-60 cells, a myeloid leukemia cell line that can be differentiated into neutrophil-like cells as our model (33, 36). We have developed a platform to express microRNAs in HL-60 cells after cell differentiation ( Figure 2A ) (33). As expected, MIR-99 overexpression did not affect cell differentiation, cell death, or the expression of two other miRNAs, MIR-223 and LET-7 ( Figures 2B–F ). The differentiated (dHL-60) cells were then subjected to a transwell assay using N-Formylmethionyl-leucyl-phenylalanine (fMLP) as the chemoattractant. Overexpressing MIR-99, but not the vector control, significantly reduced chemotaxis ( Figure 2G ), indicating that overexpression of MIR-99 inhibits neutrophil migration in both zebrafish and human cell models. In addition, MIR-99 suppressed other functions in HL-60 cells, including phagocytosis, production of reactive oxygen species, and neutrophil extracellular traps ( Figures 2H–J ).

To understand the underlying molecular mechanism of how miR-99 regulates neutrophil migration, we sequenced the transcriptome of neutrophils expressing miR-99. Since microRNAs generally suppress the expression of target transcripts (37), we focused on the down-regulated differentially expressed genes (DEG), which are associated with pathways in the cell cycle, microtubule-based movement, and DNA damage stimulus ( Figures 3A, B and SI Appendix , Dataset S1 , S2 ). Only two genes among the down-regulated DEGs, roraa, and dec1, contain predicted miR-99 binding sites in the 3’UTR ( Figure 3C ). Roraa is one of the retinoic acid receptor-related orphan receptor (ROR) gene family members in zebrafish. Identical to humans, there are three ROR genes in zebrafish: rora, rorb, and rorc. The protein product of rora, RORα, functions in metabolic regulation, energy homeostasis, thymopoiesis, and integration of the circadian clock (38, 39). The functions of RORβ are mainly unknown. It perhaps acts as a novel regulator in osteogenesis (26, 40). RORγt, the protein product of rorc, guides Th17 cell differentiation and promotes IL-17 production (19). Compared to dec1, the roraa gene orthologues are more conserved between zebrafish and humans (zebrafish roraa sequence is 89.96% identical to its human ortholog rora while dec1 in zebrafish is 59.55% identical to its human orthologue). In addition, RORα polymorphisms correlate with different diseases, such as multiple sclerosis, breast cancer, asthma in human patients (41–43). However, whether RORα plays a regulatory role in neutrophil migration is unknown. Therefore, we chose roraa to follow up. To validate the RNAseq results, we selected four other genes, grn1, fabp7, tgfb1, il1b, and confirmed their reduced transcript levels in the miR-99-overexpressing neutrophils ( Figure 3D ). To be noted, the expression of roraa, but not its homolog, rorc, was suppressed by miR-99. We then performed dual-luciferase reporter assays to determine whether miR-99 can directly bind to roraa 3’UTR and suppress reporter expression. miR-99 significantly reduced the relative luciferase activity, dependent on the roraa 3’UTRs ( Figures 3E, F ), indicating that miR-99 can directly target roraa in zebrafish.

RORα binds DNA as a monomer and constitutively activates transcription (44, 45). Although the endogenous ligands of this receptor are not entirely clear, synthetic ligands can be used to suppress or enhance RORα activities (46). To test whether Rorα regulates neutrophil migration in homeostasis and inflammation conditions, we used SR3335, an inverse agonist of RORα, SR2211, a RORγt inverse agonist, and VPR66, an inverse agonist of both RORα and RORγt. None of the inhibitors affected zebrafish survival or neutrophil numbers ( Figure S1A ). Neutrophil motility was explicitly reduced by the RORα, but not the RORγ inhibition ( Figures 4A, B and Movie 2 ). Consistently, neutrophil recruitment to the infected ear ( Figures 4C, D ) or tailfin amputation sites ( Figures 4E, F ) was significantly and specifically inhibited by the RORα inhibitor. Human neutrophils possess a limited transcriptional activity and do not require transcription for chemotaxis (47, 48). To ensure that our findings are not restricted to zebrafish, we isolated primary human neutrophils and treated them with the ROR inverse agonists. Similar to our observation made in zebrafish, the inverse agonist of RORα, but not RORγ, reduced chemotaxis of primary human neutrophils 1 hr after treatment, without causing apoptosis ( Figures 4G, H and Figures S1B-G and Movie 3 ). RORα protein is detected in primary human neutrophils ( Figure S1H ). In addition, the ROR inverse agonists SR3335 suppressed neutrophil chemotaxis in fish in a dose-dependent manner ( Figures 4I, J ). In summary, we report a previously unrecognized role for RORα in regulating neutrophil chemotaxis.

Due to the potential off-target effects of the inhibitors, we next sought genetic evidence for RORα function in neutrophil migration. RORα protein consists of one DNA-binding domain and one ligand-binding domain, working as a monomer and activates transcription when binding to the DNA response element. A truncated version of RORα lacking the ligand-binding domain acts as a dominant-negative by occupying the DNA response elements without activating transcription. We constructed zebrafish lines overexpressing Rorα (1-180aa) in neutrophils, Tg(lyzC:mcherry-2a-rora-dn)^{pu 32}, together with a control line Tg(lyzC:mcherry-2a)^pu33 ( Figure 5A ). Despite multiple attempts, we could not generate the line overexpressing the full-length rora as a control. Expressing the rora DN did not affect total neutrophil numbers but reduced neutrophil motility in the head mesenchyme ( Figures 5B–E and Movie 4 ). Consistently, Rorα DN significantly reduced neutrophil recruitment to infection or tail amputation sites ( Figures 5F–I ). The control or the Rorα DN line was crossed with Tg(mpx: GFP-UtrCH) to label stable actin in neutrophils (27). In the control neutrophils, stable actin was enriched in the retracing tail. In comparison, in Rorα DN expressing neutrophils, stable actin mislocalized to cell front or the middle of the cell body, indicating disrupted cell polarity ( Figure 5J and Movie 5 ). In addition, we knocked down rora specifically in neutrophils using the CRISPR-Cas9 system recently optimized in our lab where Cas9 expression is restricted in neutrophils and two sgRNAs targeting rora are expressed ubiquitously (49). Consistently, neutrophil migration is significantly inhibited upon rora deletion ( Figures 5K–N and Movie 6 ). Furthermore, SR3335 does not further reduce neutrophil speed when the rora gene is disrupted ( Figure S2 ). Taken together, our results indicate that the RORα mediated transcription is required for neutrophil polarization and efficient chemotaxis.

Our current finding suggests that miR-99 overexpression or RORα inhibition suppresses neutrophil chemotaxis. We then investigated whether RORα is required for defending against bacterial infection using P. aeruginosa (50). Survival of the zebrafish embryos was significantly reduced upon expression of miR-99 or Rorα DN in neutrophils ( Figures 6A, B ). In line with this observation, inhibiting RORα, but not RORγ, using small molecules also reduced survival ( Figures 6C, D ). Taken together, RORα plays an essential role in regulating neutrophil activation and the defense against bacterial infection.

RORα is expressed in primary human and zebrafish neutrophils, but not in the human neutrophil model cell line, HL-60, after different differentiation methods (51). Due to the possible off-target effects of inhibitors and the difficulty of genetically altering primary human neutrophils, we used zebrafish neutrophils overexpressing the Rorα DN to understand the RORα regulated gene network., 216 DEGs are down-regulated in zebrafish neutrophils overexpressing Rorα DN compared with the control with a cut off of false discovery rate of 0.05 ( Datasets S3, S4 ). However, the consensus ROR binding motif (AxxTxGGTCA) is not enriched in the downregulated DEGs, suggesting direct and indirect transcriptional regulation by RORα. The downregulated DEGs are enriched in cell death, transmembrane signaling receptor activity, and protein phosphorylation pathways ( Figures 7A, B ). Receptor signaling and protein phosphorylation pathways are involved in cell migration and activation, supporting the biological outcome of RORα inhibition in neutrophils. We further compared our gene list to a previously published list of RORα target genes in monocytic and endothelial cells identified using chromatin immunoprecipitation (ChIP) (52). We converted downregulated DEGs to their human homologs. Ten are putative RORα target genes in both THP-1 and HUVEC cells, and 13 and 23 are putative RORα target genes in either one of the cells ( Figure 7C ) ( Supplementary Table 1 ). Finally, we selected five transcripts that are putative RORα target genes to validate this RNA seq result ( Figure 7D ). As expected, the level of roraa is significantly higher due to overexpression of the DN construct. Rho GTPase Activating Protein 17 (Arhgap17b) is a signaling molecule regulating Cdc42 and Rac1 (53). Protein Tyrosine Kinase 2 Beta (Ptk2ba) regulates protein tyrosine kinase activity and is implicated in human neutrophil migration (54) and host defense response to bacterial infection (55). MHC Class I Polypeptide-Related Sequence B (Mhc1zba) is essential in self-antigen presentation under stressed conditions recognized by innate-like lymphocytes (56). Calcium Voltage-Gated Channel Auxiliary Subunit Gamma 3 (Cacng3b) is among the most significantly altered DEGs. It is a voltage-dependent calcium channel subunit on the plasma membrane, regulating the receptor activity (57). Another most significantly changed DEG we validated is the transcription elongation regulator one like (Tcerg1l). Together, the validated downregulation of these molecules provides a molecular understanding of how RORα-dependent gene expression is required for proper neutrophil migration and activation.