Abstract

   Neutrophils have been classically viewed as a homogenous population.
   Recently, neutrophils were phenotypically classified into
   pro-inflammatory N1 and anti-inflammatory N2 sub-populations, but the
   functional differences between the two subtypes are not completely
   understood. We aimed to investigate the phenotypic and functional
   differences between N1 and N2 neutrophils, and to identify the
   potential contribution of the S100A9 alarmin in neutrophil
   polarization. We describe distinct transcriptomic profiles and
   functional differences between N1 and N2 neutrophils. Compared to N2,
   the N1 neutrophils exhibited: i) higher levels of ROS and oxidative
   burst, ii) increased activity of MPO and MMP-9, and iii) enhanced
   chemotactic response. N1 neutrophils were also characterized by
   elevated expression of NADPH oxidase subunits, as well as activation of
   the signaling molecules ERK and the p65 subunit of NF-kB. Moreover, we
   found that the S100A9 alarmin promotes the chemotactic and enzymatic
   activity of N1 neutrophils. S100A9 inhibition with a specific
   small-molecule blocker, reduced CCL2, CCL3 and CCL5 chemokine
   expression and decreased MPO and MMP-9 activity, by interfering with
   the NF-kB signaling pathway. Together, these findings reveal that N1
   neutrophils are pro-inflammatory effectors of the innate immune
   response. Pharmacological blockade of S100A9 dampens the function of
   the pro-inflammatory N1 phenotype, promoting the alarmin as a novel
   target for therapeutic intervention in inflammatory diseases.

   Keywords: neutrophil polarization, N1 neutrophils, N2 neutrophils,
   S100A8/A9, ABR-238901, RNA-Seq, neutrophil chemotaxis

Introduction

   Neutrophils are the first responders in host defense, with an important
   role in promoting the innate immune response. They originate from the
   bone marrow and are released in the circulation when they mature and
   are stimulated by invasive pathogens and inflammatory signals that
   facilitate their migration to sites of infection or tissue injury. At
   the site of infection, neutrophils eliminate the invading pathogens
   utilizing a combination of NADPH oxidase-derived reactive oxygen
   species (ROS), cytotoxic granule components, and neutrophil
   extracellular traps (NETs) ([51]1).

   Although regarded for a long time as a homogenous population with
   conserved phenotype and function, recent evidence has suggested the
   existence of neutrophil heterogeneity with different functional
   phenotypes, both in healthy individuals and in pathological conditions
   including cancer, infections, and autoimmune and inflammatory disorders
   ([52]2, [53]3). The heterogeneity of neutrophil populations is
   characterized by differences in life span, cytokine release, surface
   proteins, antibacterial responses, as well as pro-inflammatory,
   proangiogenic, or immunosuppressive functions ([54]2–[55]5). It has
   been reported that a unique neutrophil population emerging during acute
   inflammation suppresses T cell function, a process dependent of
   neutrophil Mac-1 and ROS ([56]6). Infection with Staphylococcus aureus
   leads to two subsets of murine polymorphonuclear neutrophils with
   important differences in their expression of surface markers, cytokine
   production and macrophage activation potential ([57]7). In systemic
   lupus erythematosus and other autoimmune diseases, a subpopulation of
   low-density neutrophils (LDN) with an unclear physiological role has
   been detected ([58]8). The LDN population with immunosuppressive
   properties has also been found to accumulate in tumor-bearing mice and
   cancer patients. In contrast, the high-density neutrophils (HDN) have
   been shown to have anti-tumorigenic functions ([59]9). Moreover,
   circulating neutrophil subsets in advanced lung cancer patients have
   unique immune signatures and are associated with the disease prognosis
   ([60]10).

   Recently, the consecutive myocardial infiltration of two neutrophil
   subpopulations has been described in a mouse model of myocardial
   infarction (MI). Cardiac N1 neutrophils isolated on day one post-MI,
   during the inflammatory phase, showed high levels of pro-inflammatory
   markers (CCL3, IL-1β, IL-12a, and TNF-α). In contrast, cardiac N2
   neutrophils isolated at days 5 and 7, during the reparatory phase,
   exhibited increased expression of anti-inflammatory markers CD206 and
   IL-10. Moreover, neutrophils polarized in vitro with a combination of
   lipopolysaccharide (LPS) and interferon-γ (IFN-γ) for N1 or
   interleukin-4 (IL-4) for N2, exhibited similar markers as the
   sub-populations found in vivo ([61]11). Uncovering the potentially
   important role of the different neutrophil subtypes in driving
   inflammation or the resolution of inflammation could have significant
   therapeutic relevance, as targeting a specific subpopulation may
   modulate the course of the disease.

   S100A8/A9 is an immune mediator abundantly secreted by neutrophils that
   plays a complex role in various pathologies with an immune and
   inflammatory component. S100A9 and its dimerization partner S100A8 are
   rapidly released as the S100A8/A9 heterodimer upon cell activation
   ([62]12) and functions as a damage-associated molecular pattern (DAMPs)
   that binds to toll-like receptor 4 (TLR4) ([63]13), and to the receptor
   for advanced glycation end products (RAGE) ([64]14). Activation of TLR4
   by S100A8/A9 has been shown to have an important pro-inflammatory role
   in the pathogenesis of endotoxin-induced shock ([65]15), autoimmune
   disease and cancer ([66]16).

   After MI, S100A8/A9 is abundantly secreted by activated neutrophils and
   promotes cardiac inflammation by stimulating myeloid cell production
   and trafficking to the ischemic myocardium ([67]17). We have recently
   found that short-term S100A9 blockade with the specific blocker
   ABR-238901 during the inflammatory phase of MI reduces myocardial and
   systemic inflammation, and improves cardiac function ([68]17). The
   precise mechanisms behind these beneficial therapeutic effects remain
   to be investigated. Interestingly, binding of S100A8/A9 to TLR4 on
   neutrophils has subsequently been shown to drive IL-1β production,
   leading to increased myelopoesis in MI ([69]18). As IL-1β secretion is
   characteristic for the N1 neutrophil phenotype, we hypothesize that
   S100A8/A9 might play an important role in the development of this
   particular subpopulation.

   In this work, our main aims were: i) to perform a comparative study of
   N1 and N2 neutrophil genotype, phenotype and function, and ii) to
   investigate the effects of S100A9 blockade with ABR-238901 on the
   functions of the two neutrophil subpopulations. Elucidating the
   immunomodulatory properties of S100A9 inhibition is highly relevant for
   further development of the compound toward a potential
   anti-inflammatory treatment in MI and other immune and inflammatory
   diseases.

Materials and Methods

Mice

   Male and female C57BL/6J mice, between 12-16 weeks old, were bred and
   housed in pathogen-free conditions at the Institute of Cellular Biology
   and Pathology (ICBP) “Nicolae Simionescu”. The mice were euthanized
   through cervical dislocation, and the femurs and tibias were collected
   in a Petri dish containing ice-cold RPMI 1640 supplemented with 10% FBS
   and 1% Penicillin/streptomycin, for further isolation of bone marrow.

   All animal experiments were performed in strict accordance with the
   European Guidelines for animal welfare (Directive 2010/63/EU) and
   approved by The National Sanitary Veterinary and Food Safety Authority
   (nr. 425/22.10.2018). All procedures were approved by the Institutional
   Ethics Committee of ICBP “N. Simionescu” (Bucharest, Romania).

Isolation and Polarization of Neutrophils

Isolation of Neutrophils

   Cells were isolated from mouse bone marrow by Percoll gradient
   centrifugation, using a simplified and improved version of a previous
   protocol ([70]19). Briefly, the bones were placed in HBSS-Prep to
   prevent drying, the ends were cut and the bone marrow (BM) was flushed
   into a 50 ml conical tube with HBSS-Prep and centrifuged at 400 × g for
   5 min. For erythrocyte lysis, the pellet was resuspended in 10 ml NaCl
   0.2% for 30-40 s and the osmolarity was then restored with 10 ml 1.6%
   NaCl. The resulting suspension was centrifuged in 62.5% Percoll in
   HBSS-Prep for 30 min at 1000 × g, without brake. At the end of
   centrifugation, the neutrophils-containing pellet was transferred to
   another 15 ml tube, washed twice with HBSS and cells were resuspended
   in RPMI. The purity of isolated neutrophils was confirmed by flow
   cytometry using the neutrophil marker Ly-6G and by fluorescence
   microscopy using Hoechst/PI staining ([71]Figure S1).

Polarization of Neutrophils

   Freshly isolated neutrophils were pooled from 6–10 mice and cultured
   for 2h/18h in RPMI medium, in the presence of 100 ng/ml
   lipopolysaccharide (LPS) and 20 ng/ml interferon gamma (IFNγ) or 20
   ng/ml interleukin 4 (IL-4) - in order to obtain polarized N1
   (inflammatory) and N2 (anti-inflammatory) neutrophil subsets
   respectively. This polarization protocol has previously been shown by
   Ma et al. to generate neutrophil sub-populations with a similar
   phenotype as the N1/N2 neutrophils isolated from infarcted hearts in
   vivo ([72]11). Unstimulated neutrophils were used as controls (N). In
   the experiments when S100Ab was blocked, N1 and N2 neutrophils were
   polarized in the presence of the S100A9 inhibitor ABR-238901 (100µM,
   Active Biotech AB, Sweden).

mRNA-Sequencing

   To profile the gene expression of N1 and N2 neutrophils after 2h
   polarization, we used 3 samples per condition and 20x10^6 neutrophils
   per sample, pooled from several mice. Total RNA was isolated using
   TRIzol reagent and Phasemaker Tubes (Thermo Fischer, Waltham,
   Massachusetts, US) and was sent to Novogene (Cambridge, UK) for
   mRNA-seq analysis. RNA degradation and contamination were monitored on
   1% agarose gels. RNA purity was checked using the NanoPhotometer^®
   spectrophotometer (IMPLEN, CA, USA). RNA integrity and quantitation
   were assessed using the RNA Nano 6000 Assay Kit with the Bioanalyzer
   2100 system (Agilent Technologies, CA, USA). A sample from the LPS+IFNγ
   polarization did not pass the quality control test and was excluded
   from the downstream analysis.

   A total amount of 1 µg RNA per sample was used as input material for
   RNA analysis. Sequencing libraries were generated using the NEBNext
   UltraTM RNA Library Prep Kit for Illumina (NEB, USA) following the
   manufacturer’s recommendations and index codes were added to attribute
   sequences to each sample. Library quality was assessed on the Agilent
   Bioanalyzer 2100 system (Agilent Technologies, CA, USA). The clustering
   of index-coded samples was performed on a cBot Cluster Generation
   System using the PE Cluster Kit cBot-HS (Illumina). After cluster
   generation, the library preparations were sequenced using Illumina
   NovaSeq 6000 (Illumina) and paired-end reads were generated.

   Raw data (raw reads) of FASTQ format were first processed through fastp
   ([73]20). Clean data were obtained by removing reads containing adapter
   and poly-N sequences and reads with low quality from raw data.
   Simultaneously, Q20, Q30 and GC content of the clean data were
   calculated ([74]Supplementary Table 1). Paired-end clean reads were
   aligned to the Ensembl mouse reference genome (GRCm38.p6) ([75]21)
   using the Spliced Transcripts Alignment to a Reference (STAR) software
   ([76]22). A summary of the mapping result is presented in
   ([77]Supplementary Table 2). Gene expression values FPKM (expected
   number of Fragments Per Kilobase of transcript sequence per Millions
   base pairs sequenced) were calculated and used for the PCA and Pearson
   correlation coefficient matrix, using R software ([78]23).

Differential Expression Analysis

   Differential expression analysis was performed using the DESeq2R
   package (2_1.6.3) ([79]24). The resulting P values were adjusted using
   Benjamini and Hochberg’s approach for controlling the false discovery
   rate (FDR). Genes with an adjusted P-value <0.05 found by DESeq2 were
   assigned as differentially expressed (DEGs). Using a built-in R
   package, pheatmap, a hierarchical clustering heatmap was generated
   presenting the log2(FPKM+1) of DEG union within all comparison groups.
   Volcano plots were realized using EnhancedVolcano R package ([80]25).

Functional Analysis of DEGs

   Functional enrichment analysis of the up-regulated N1 gene cluster was
   performed using g:GOSt function in gProfiler version
   e102_eg49_p15_7a9b4d6, database updated on 15/12/2020 ([81]26). The
   selected organism was Mus musculus, the significance threshold was
   g:SCS, with a user threshold of 0.01. Gene Ontology, pathways from
   KEGG, Reactome and regulatory motif matches from TRANSFAC databases
   were inquired.

   GO enrichment and KEGG database enrichment analysis was performed using
   the clusterProfiler R package ([82]27) on all the DEGs, either down or
   up-regulated and the terms with a corrected P value less than 0.05 were
   considered significantly enriched.

Quantitative RT-PCR

   Validation of key molecules found to be highly increased by RNA-seq was
   performed by qPCR using RNA obtained from pooled neutrophils isolated
   from subsequent experiments. Total cellular RNA was extracted from N,
   N1 and N2 neutrophils using TRIzol or Qiagen PureLink RNA Kit (Ambion™,
   Carlsbad, CA). First-strand cDNA synthesis was performed employing 1 μg
   of total RNA and MMLV reverse transcriptase, according to the
   manufacturer’s protocol (Invitrogen). Assessment of mRNA expression was
   done by amplification of cDNA using a LightCycler 480 Real-Time PCR
   System (Roche) and SYBR Green I. The primer sequences for the mRNAs of
   interest are shown in [83]Supplementary Table 3. The relative
   quantification was done by the comparative CT method and expressed as
   arbitrary units. Beta-actin was used as reporter gene.

Cytokine Array

   The presence of soluble pro-inflammatory cytokines and chemokines in
   the neutrophil condition media was analyzed using the Proteome Profiler
   Mouse Cytokine Array Kit (ARY006, R&D Systems) in conditioned media
   from the N, N1 and N2 subpopulations. Detection of the chemiluminescent
   signal was performed using the Luminescent image analyzer LAS 4000
   (Fujifilm). The mean pixel density of each point was calculated using
   ImageJ (Bethesda, MD).

Enzyme Linked Immunosorbent Assay (ELISA)

   The supernatant was harvested from control (N) or polarized N1 and N2
   neutrophils cultured in the presence or absence of ABR-238901 (100 µM).
   We measured the amount of the proteins of interest released in the
   condition media by using specific kits (R&D Systems & Mabtech),
   following the manufacturer’s instructions.

Measurement of Reactive Oxygen Species

   Control (N) or activated neutrophils (N1 and N2) were assayed for
   intracellular ROS using 2′,7′-dichlorofluorescein diacetate (DCFH-DA)
   as previously described ([84]28). Briefly, the cells were incubated
   with 5 μM DCFH-DA (30 min at 37°C) and the DCF fluorescence emission
   was detected at 535 nm with an excitation wavelength of 485 nm in a
   96-well microplate reader (GENios, Tecan). Immediately after DCF
   measurements, cells were further incubated for 20 min with Hoechst
   33342 (0.2 µg/ml) and the fluorescence was measured at 460 nm (with an
   excitation wavelength of 345 nm). ROS was expressed as DCF/Hoechst
   fluorescence units.

The Cellular Energetics of N1 and N2 Neutrophils

   An XFp Extracellular Flux Analyzer (Seahorse, Agilent Technologies) was
   used to measure the oxygen consumption rate (OCR) and the proton efflux
   rate (PER) as a measure of extracellular acidification in control (N)
   or polarized neutrophils (N1 and N2). Immediately following isolation,
   neutrophils were added at 5 x 10^5 cells/well onto a poly-L-lysine
   coated XFp plate and stimulated for 2 h to obtain polarized N1 and N2
   neutrophil subsets. The OCR and PER were measured in XF media
   (non-buffered DMEM containing 10 mM glucose, 4 mM L-glutamine, and 2 mM
   sodium pyruvate) under basal conditions and in response to phorbol
   12-myristate 13-acetate (PMA) activation. After 3 h, 500 nM Rotenone
   was injected to measure the amount of OCR due to mitochondrial
   activity. After the measurements, data were normalized to the cell
   number by staining the cells for 10 min with Hoechst 33342 (5 µg/ml)
   followed by measurement on a microplate reader (GENios, Tecan). The
   assay was performed twice in duplicates for each condition.

Determination of Cell Migration by Chemotaxis Assay

   Real-time migration was monitored using CIM-plate-16 and the
   xCELLigence System RTCA DP Instrument (Roche). We used a 16-well
   modified Boyden chamber composed of an upper chamber (UC) and a lower
   chamber (LC) that snapped together to form a tight seal. The bottom of
   the UC consists of a microporous polyethylene terephthalate membrane
   that permits the translocation of cells from the upper to the bottom
   side. Cell migration was monitored by interdigitated gold
   microelectrode sensors that generate an impedance signal by contact
   with the migrated cells. IL-8 (300ng/ml) was added as a chemoattractant
   in the LC. We seeded 4x10^5 neutrophils in the UC of the CIM-plate-16
   in RPMI medium without FCS. Cell migration was monitored for up to
   20 h.

Gelatin Zymography

   The gelatinolytic activity of the MMP‐9 released by N, N1 and N2
   neutrophils in the culture medium was evaluated by gelatin zymography,
   as previously described ([85]29). Briefly, the cell culture medium was
   collected and the nonreducing Laemmli’s buffer was added to the
   cell‐free neutrophil supernatants and subjected to electrophoresis
   under non‐reducing conditions on 10% polyacrylamide gels containing 1
   mg/mL gelatin as substrate. After electrophoresis, the gels were
   re‐natured in 2.5% Triton X‐100 (2 × 30 minutes) and incubated with
   50 mmol/L Tris‐HCl pH 7.4, containing 10 mmol/L CaCl[2] and 0.2 mmol/L
   PMSF (18 hours, 37°C). The gels were subsequently stained with 0.2%
   Coomassie brilliant blue R‐250 and de‐stained with 10% acetic acid and
   25% methanol. The white bands against the blue background were
   indicative of the gelatinolytic activity of MMP-2/-9. Image acquisition
   was done with a transillumination imaging system LAS 4000 (Fujifilm).
   Data are presented as fold increase over the unstimulated control.

Western Blot

   Following polarization, neutrophils were rapidly chilled by the
   addition of ice‐cold HBSS. Neutrophils were pelleted, supernatants were
   collected and the cell pellets were lysed using RIPA lysis buffer
   supplemented with a protease inhibitor cocktail. After centrifugation
   (12000 × g), the proteins were quantified by bicinchoninic acid (BCA)
   Protein Assay Kit. Samples (30 μg protein) were separated on 10%
   SDS-PAGE (sodium dodecyl sulfate-polyacrylamide) gel electrophoresis
   and transferred to nitrocellulose membranes, which were subsequently
   probed with specific antibodies. The signals were visualized using
   SuperSignal West Pico chemiluminescent substrate (Pierce) and
   quantified by densitometry employing the gel analyzer system
   Luminescent image analyzer LAS 4000 (Fujifilm) and the Image reader LAS
   4000 software.

Detection of Myeloperoxidase (MPO) and Nitric Oxide (NO)

   Quantification of MPO activity was performed in the cell lysate, and of
   NO levels in the conditioned media, using specific kits from
   Elabscience (K074 and K035, respectively). Following polarization,
   conditioned media from N, N1, N2 neutrophils were harvested and used
   for NO detection, while neutrophils were rapidly chilled with ice‐cold
   PBS, centrifugated and the resulted cell lysate used for MPO
   quantification according to the manufacturer’s instructions.

Statistical Analyses

   GraphPad Prism 7.0 with data points expressed as mean ± standard
   deviation (SD) was used for all statistical analyses. We used a
   two-tailed Student’s t-test when comparing two experimental groups and
   a one-way ANOVA and Tukey’s multiple comparison test when comparing
   more than two groups. A p-value of p<0.05 was considered statistically
   significant.

Data Access

   All sequencing data have been deposited in the ArrayExpress database,
   [86]https://www.ebi.ac.uk/arrayexpress/E-MTAB-10508. All other data are
   available from the authors on request.

Results

N1 and N2 Neutrophils Exhibit a Distinct Transcriptomic Profile

   RNA obtained from freshly isolated neutrophils polarized for 2h with
   LPS+IFNγ (N1) or IL-4 (N2), was analyzed by RNA-seq. The sample
   replicates showed a high degree of correlation, as determined by
   Pearson correlation matrix ([87]Figure S1).

   The hierarchical clustering analysis showed a distinct transcriptomic
   profile of the N1 and N2 neutrophil populations compared with control
   neutrophils (N) ([88]Figure 1A). Additionally, we performed a
   differential expression analysis to generate modules of genes that are
   significantly modulated in each neutrophil population. With a cutoff
   criterion of absolute fold change ≥ 1.0 and adjP < 0.05, 966 genes were
   found to be differentially expressed in N1 neutrophils compared to
   control (771 genes were increased and 195 genes were decreased), and
   532 genes were found to be differentially expressed in N2 neutrophils
   (408 genes were increased and 124 genes were decreased)
   ([89]Figures 1B, C). In N1 neutrophils, a substantial number of
   up-regulated genes code for inflammatory cytokines and chemokines such
   as TNF-α, IL-10, IL-12, IL-1β, IL-1α, CCL3, CCL4, CCL5, CCL7, CCL9,
   CXCL1, CXCL2, CXCL3, CXCL10, CXCL16 ([90]Figure 1D). These molecules
   were either unmodified in N2 neutrophils or were down-regulated
   compared to controls (TNF-α, IL-1β, CXCL16, CXCL2, CXCL10)
   ([91]Figure 1D).

Figure 1.

   [92]Figure 1
   [93]Open in a new tab

   Gene expression profiling of the different neutrophil subsets. (A)
   Hierarchical Clustering Heatmap analysis of N, N1, and N2 neutrophils.
   Hierarchical clustering analysis was conducted of log2(FPKM+1) of
   differential expression genes union within all comparison groups. The
   color coding indicates different levels of expression: red indicates
   genes with high expression, and blue indicates genes with low
   expression levels. A major cluster of DEGs up-regulated in N1 is
   highlighted in a black square. (B, C) Volcano plot of differential gene
   expression between N1/N2 and N cells. The red dots represent
   significantly up-regulated and down-regulated genes with – adjP < 0.05
   and Log2FC > 1: 771 genes were up-regulated and 195 genes were
   down-regulated in N1 vs N neutrophils; 532 genes were up-regulated and
   124 genes were down-regulated in N2 vs N cells. (D) Heatmap showing
   log2 Fold change and adjusted p-value for selected inflammatory
   cytokines and chemokines differentially expressed either in N1 (upper
   panel) or N2 neutrophils (lower panel). Red color indicates the
   up-regulated, and blue the down-regulated genes. (E) Manhattan plot
   illustrating the results of the enrichment analysis of the gene cluster
   of highly up-regulated genes in N1 compared with N and N2 neutrophils.
   The functional terms are grouped and color-coded by data sources, i.e.,
   molecular function (MF) in red, biological processes (BP) in orange,
   cellular components (CC) in green, KEGG in pink and TRANSFAC in dark
   blue. Numbered terms are detailed below the plot with their respective
   adjP values.

   The gene ontology (GO) enrichment analysis highlighted biological
   processes and functions that are significantly associated with the
   modified genes in N1 and N2 neutrophils. The differentially modulated
   genes in N1 neutrophils are associated with cytokine production, cell
   response to LPS, cell chemotaxis and cytokine mediated-signaling
   pathways ([94]Figure S2A). In contrast, genes found to be modified in
   N2 cells are related to T cell activation and differentiation,
   cell-cell adhesion and immune response ([95]Figure S2B).

   Interestingly, a central cluster of 391 highly up-regulated genes is
   well represented in N1 neutrophils compared with N2 and control cells
   ([96]Figures 1A and [97]S3). Functional enrichment analysis for the
   genes in this cluster revealed as significantly enriched the following
   terms: i) biological process - “defense response”; ii) molecular
   function - “cytokine activity” and “cytokine receptor binding”; iii)
   KEGG pathway analysis - “TNF signaling pathway”, “NF-kappa B signaling
   pathway”; iv) TRANSFAC - Factor: RelA-p65 as the most enriched
   transcription factor ([98]Figure 1E). The results emphasize that the
   highly up-regulated genes in the N1 subset are associated with an
   inflammatory response. A list of the 10 most enriched terms for each
   database searched is presented in [99]Figure S4.

   Following the RNA-seq analysis, we validated a selection of
   differentially expressed inflammatory/anti-inflammatory genes by qPCR
   in neutrophils polarized for 2h or 18h. Compared to control
   neutrophils, the N1 neutrophils exhibited higher gene expression of the
   pro-inflammatory mediators TNF-α, IL-12, IL-1β, CCL2 (MCP-1), CCL3
   (MIP-1α), and CCL5 (RANTES) both at 2h and 18h ([100]Figures 2A–G). The
   gene expression of inflammatory cytokines was time-dependent, reaching
   higher levels at 18h compared with 2h of activation. IL-6 was highly
   induced in N1 neutrophils only after 18h of activation
   ([101]Figure 2D). The anti-inflammatory markers CD206, Ym1 and Arg1
   were increased in N2 neutrophils, but not in N1 cells
   ([102]Figures 2I–K). However, the IL-10 gene expression was unchanged
   in N2 neutrophils and was overexpressed in N1 cells ([103]Figure 2H).
   The data are in agreement with the results of RNA-seq expression where
   the gene encoding for IL-10 was found to be ~10-fold upregulated in N1
   neutrophils.

Figure 2.

   [104]Figure 2
   [105]Open in a new tab

   Gene expression of inflammatory and anti-inflammatory markers in
   neutrophils polarized for 2 or 18h with LPS+IFNγ (N1) or IL-4 (N2).
   (A–G) qPCR for inflammatory markers: TNFα, IL-12, IL-1β, IL-6, CCL3,
   CCL5, MCP-1, in N1 and N2 compared to control neutrophils (N). (H–K)
   qPCR for anti-inflammatory markers: IL-10, CD206, Ym1 and Arg1, in N1,
   N2 and N. n = 5, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001
   (N1 or N2 vs. N).

   Expression of inflammatory genes in N2 neutrophils was similar to
   unstimulated controls, except for the IL-1β that was decreased
   ([106]Figure 2C). These results are in agreement with the results
   obtained by RNA-seq, where IL-1β and TNF-α were significantly decreased
   in N2 neutrophils compared with controls.

   To validate these findings in an inflammatory state in humans, we
   analyzed the gene expression of CCL3, IL-6, IL-1β, and CD206 in human
   blood neutrophils isolated from MI patients during the first 24h after
   infarction. We found that the gene expression of CCL3 and IL-6 was
   significantly increased in neutrophils from these patients compared
   with healthy controls ([107]Figure S6). These data demonstrate the
   involvement in human pathology of N1-like inflammatory neutrophils with
   a gene expression profile similar to the N1 neutrophils isolated from
   the infarcted myocardium ([108]11) and to the N1 neutrophils derived in
   vitro. The gene expression of the N2 marker CD206 was not significantly
   affected in this early stage of the disease.

   Additionally, we investigated the expression of N1/N2 surface markers
   by flow cytometry in a mouse model of endotoxemia (LPS-induced acute
   systemic inflammation). Circulating neutrophils were analyzed at 24h
   after the LPS treatment. Neutrophils from these mice were characterized
   by a significantly higher surface expression of CD11b (Mac-1) and
   ICAM-1 ([109]Figures S7D, E), molecules involved in neutrophil
   adhesion, rolling and recruitment into the tissue. These results offer
   in vivo support for our in vitro data: RNA-seq data where ICAM-1
   expression was 11 times increased in N1 compared with control
   neutrophils (log2FC:3,56; [110]Figure 1D), cytokine array showing the
   increased shedding of sICAM-1 in N1 neutrophils ([111]Figures 4A, B),
   and with the increased chemotactic activity of N1 neutrophils
   ([112]Figure 5G). Similar to MI patients, LPS treatment did not
   modulate the surface CD206 expression of mice circulating neutrophils
   ([113]Figures S7D, E), suggesting that N2-like neutrophils are not
   present in blood during the inflammatory stage.

Figure 4.

   [114]Figure 4
   [115]Open in a new tab

   Effects of S100A9 blockade on neutrophil mediators released in the
   conditioned media after 18h of culture. (A) Quantification of mediators
   present in the culture medium from N1, N1 treated with ABR-23901, and
   N2 neutrophils compared to control neutrophils (N), as detected by the
   Proteome Profiler mouse cytokine array. The culture medium was pooled
   from three experiments. Protein levels were normalized to references on