GPR35 promotes neutrophil recruitment in response to serotonin metabolite 5-HIAA

Abstract

Rapid neutrophil recruitment to sites of inflammation is crucial for innate immune responses. Here, we reveal that the G-protein-coupled receptor GPR35 is upregulated in activated neutrophils, and it promotes their migration. GPR35-deficient neutrophils are less recruited from blood vessels into inflamed tissue, and the mice are less efficient in clearing peritoneal bacteria. Using a bioassay, we find that serum and activated platelet supernatant stimulate GPR35, and we identify the platelet-derived serotonin metabolite 5-hydroxyindoleacetic acid (5-HIAA) as a GPR35 ligand. GPR35 function in neutrophil recruitment is strongly dependent on platelets, with the receptor promoting transmigration across platelet-coated endothelium. Mast cells also attract GPR35⁺ cells via 5-HIAA. Mice deficient in 5-HIAA show a loss of GPR35-mediated neutrophil recruitment to inflamed tissue. These findings identify 5-HIAA as a GPR35 ligand and neutrophil chemoattractant and establish a role for platelet- and mast cell-produced 5-HIAA in cell recruitment to the sites of inflammation and bacterial clearance.

Keywords: 5-HIAA; GPCRs; GPR35; SSRI; inflammation; mast cells; migration; neutrophil; platelets; serotonin metabolite.

Figure 1.. GPR35 is a neutrophil chemotactic receptor.

a, qPCR for expression of Gpr35 (relative to Hprt) in Ly6G⁺ Ly6C⁺ CD11b⁺ CXCR2⁺ neutrophils sorted from BM (0hr, n=4; 2hr n=5; 18hr, n=3), blood (0hr, n=4; 2hr, n=4; 18hr, n=3) or peritoneum (2hr, n=5; 18hr, n=6) at indicated time points after TG injection. b, Intracellular flow cytometry for GPR35 in Ly6G⁺ CD11b⁺ Ly6C⁺ neutrophils from peritoneum (left) or blood (right) at indicated time points after TG. c, Representative flow cytometry plots of GPR35-GFP and control WEHI-231 cells migrating to 50 μM lodoxamide compared to input. d, Quantification of transwell migration assays of GPR35-GFP (black) or control (white) WEHI-231 cells to lodoxamide, KynA or 18:1 LPA at the indicated concentrations (n=3–6). e-f, Representative flow cytometry plots (e) and quantification (f) of transmigration assay with GPR35 WT (CD45.1) or KO (CD45.2) peritoneal neutrophils 2hr after TG, migrating to KynA at indicated concentrations (n=3–6). g, Quantification of adhesion assays performed with GPR35 WT or KO peritoneal neutrophils 2hr after TG. Data are pooled from at least 2 independent experiments. Statistics in a, d, f and g show comparison of all samples to BM 0hr (a) or Nil (d,f and g). P* < 0.05, **P < 0.005, ***P < 0.001, ****P < 0.0001. See also Figures S1.

Figure 2.. GPR35 sustains neutrophil recruitment to inflamed tissues.

a, Representative flow cytometry plots of spleen (top) and peritoneum (bottom) from mice injected with a 50:50 mix of GPR35 WT (Cell trace violet) and GPR35 KO (CMTMR) BM neutrophils, 18hr after i.p. TG treatment. b, Graphs showing neutrophil recruitment index (% KO / % WT) of mice exemplified in (a) (n=15–16). c-e, Quantification of transferred neutrophil recruitment index in mice injected with either i.p. TNF (2hr, n=8) (c), s.c. Listeria (20hr, n=10) (d), or pricked with Listeria-contaminated needle (2hr; ears, n=11; spleen, blood, n=7) (e). f, Intracellular flow cytometry for GPR35 in Ly6G⁺ CD11b⁺ Ly6C⁺ neutrophils from peritoneum of WT or GPR35 KO mice 20hr after Listeria injection. g, Quantification of endogenous neutrophil numbers 20hr after peritoneal Listeria infection (WT, n=9; GPR35 KO, n=6). h, Quantification of Listeria peritoneal CFUs (WT, n=8; KO, n=7) 20hr after i.p. infection in mice treated or not with Ly6G depleting antibody. i, Quantification of endogenous neutrophil numbers 20hr after peritoneal E. coli infection (n=4). j, k Quantification of E.coli peritoneal CFUs 6hr (j, n=4) and 20hr (k, n=4) after i.p. infection in mice treated or not with Ly6G depleting antibody. l, Quantification of Listeria peritoneal CFUs (n=5) in Mrp8-Cre x iDTR/ GPR35 WT or KO mixed chimeras treated with DT, 20hr after Listeria infection Data are pooled from at least 2 independent experiments. *P < 0.05, **P < 0.005, ***P < 0.001, **** P<0,0001. See also Figure S2 and S3.

Figure 3.. GPR35 supports neutrophil in vivo transendothelial migration.

a, Representative flow cytometry plots showing % of CD45-PE⁺ intravascular WT (left) or GPR35 KO (right) transferred neutrophils in the omentum 2hr after i.p. TG. b, Quantification of data shown in (a) (WT, n=11; HET, n=5; KO, n=9). Data are pooled from 3 independent experiments. c, Multiphoton intravital micrographs of Mrp8-Cre⁺ mTmG (blue) and GPR35 KO (green) neutrophils in the omentum of mixed BM chimeric mice injected with labeled dextran to identify vessels (red) and with Ly6G-PE to label neutrophils, 2hr post TG. Time shown in min:sec, See also Movie S1. d, Intravital micrograph showing transmigrating (WT blue, KO green) tracks from Movie S1. Image is representative of at least 3 independent intravital movies. e-f, Quantification of transmigrating (e) or intravascular (f) track % of WT or KO neutrophils in the omentum of mice of the type in c, 2hr after TG (n=6). g, Intravital micrographs of CTV-WT (blue) or CFSE-KO (green) transferred neutrophils in the omentum of mice injected with Dextran-Rhodamine and anti-CD31 PE to identify vessels (red), 2hr post TG (Movie S3). Time shown in min:sec. Arrow highlights transmigrating cell. Data are representative of 4 cell-transfer movies. **P < 0.005, ***P < 0.001. See also Figure S4.

Figure 4.. Platelet-derived 5-HIAA is a potent GPR35 ligand

a, Quantification of transwell migration assay of GPR35-GFP or control WEHI-231 cells to 10% serum or plasma (GPR35-GFP: n=6; Nil, n=4; Control: n=5; Nil, n=2). Data are pooled from 2 independent experiments. b-c, Representative flow cytometry plots (b) and quantification (c) of GPR35-GFP (black) or control (white) WEHI-231 cells migrating to activated or resting (control) platelet culture supernatants (n=4–7). Data are pooled from 3 independent experiments. d-e, Representative flow cytometry plots (d) and quantification (e) of GPR35-GFP and control WEHI-231 transwell migration to 5-HIAA at indicated concentrations (n=3–10). f, Quantification of WT or GPR35 KO peritoneal neutrophil (2hr after TG) transwell migration to 100nM 5-HIAA or serotonin (Nil, n=3; all others, n=4). Data are pooled from 4 (e) or 2 (f) independent experiments. g, Representative flow cytometry histograms of OX56-GPR35 levels in transduced WEHI-231 cells after incubation with 1mM KynA (yellow), 50μM lodoxamide (red), 100nM 5-HIAA (blue) or Nil (black). Image is representative of at least 4 independent experiments. h, Diagrams of KynA, serotonin and 5-HIAA molecular structures. i, Quantification of OX56 levels in murine OX56-GPR35 and OX56-LPAR3 WEHI-231 cells incubated with 5-HIAA at indicated concentrations (n=3–11). Data are pooled from 4 independent experiments. *P<0.05, **P < 0.005, ***P < 0.001, ****P < 0.0001. See also Figure S5.

Figure 5.. Activated platelets promote GPR35-dependent neutrophil transmigration.

a, Multiphoton intravital micrographs of CTV-WT (blue) or CMTMR-GPR35 KO (red) neutrophils in the omentum of Pf4-Cre x mTmG (platelets GFP⁺, green) recipients 2hr after TG (see also Movie S5). Vessels identified based on tdTom distribution (red). Images are representative of 3 independent intravital movies. Left: Composite image at time 0. Right: Time series showing example WT (upper) and KO (lower) cells transmigrating (arrow). Endothelium shown as an iso-surface based on tdTom fluorescence. Time stamp shows min:sec. b-c, Quantification of mean contact time with platelets (b) (n=6) and transmigrating track shortest distances from platelets (c) (n=6) of images of the type in (a) and Movie S5. Each dot represents individual movies. d, Representative flow cytometry plots showing % of CD41⁺ WT (left) or KO (right) transferred neutrophils in blood 2hr after TG. e-f, Quantification of (e) neutrophil recruitment index in WT (n=7) or platelet-deficient (n=8) and (f) intravascular neutrophil % in WT (left; WT, n=10; KO, n=9) or platelet-deficient (right; n=15) mice 2hr after Listeria skin-pricking. g, Quantification of neutrophil recruitment index in LNs and spleen of WT or platelet-deficient mice (n=8–13). Data are pooled from at least 2 independent experiments. *P < 0.05, **P< 0.005, ***P< 0.001. See also Figure S6.

Figure 6.. Activated mast cells are sources of 5-HIAA and collaborate with platelets to sustain GPR35-dependent neutrophil recruitment.

a, Graph showing ELISA quantification of 5-HIAA in control (n=4) or LPS-treated (n=3) peritoneal mast cell supernatants. Single dots represent biological replicates. Unpaired t test was applied: ****P < 0.0001. b, Quantification of GPR35-GFP WEHI-231 cell migration to control or LPS-treated peritoneal mast cell supernatants (n=5). c, Graph showing ELISA quantification of 5-HIAA in control (n=3) or LPS-treated (n=4) P815 mast cell line supernatants. d, e, Quantification of GPR35-GFP WEHI-231 cell migration to (d) 5-HIAA (5nM), mast cell supernatants (50%),(e) CXCL12 (100ng/ml) or serum (5%) with the addition of anti-5-HIAA (2ug/ml) as indicated (n=2–5). f, g, Transferred neutrophil recruitment index (n=5–8) (f) and i.v. labeling (n=6–7) (g) in WT or mast cell-deficient mice 2hr after Listeria pricking. h, Quantification of CD63⁺ cells out of total CD117⁺ mast cells in 2hr Listeria pricked skin of control (n=5), platelet-depleted (n=8), platelet-depleted + compound 48/80 s.c. (n=7) or untreated controls (n=3).. i Transferred neutrophil recruitment index (n=5–7) in control, platelet-depleted or platelet-depleted and 48/80 treated mice 2hr after Listeria pricking. Data pooled from 2 independent experiments. j, k, Transferred neutrophil recruitment index (j) and numbers (k) in mast cell-deficient mice 2hr after Listeria skin-pricking and s.c. injection with DMSO (control) or 5-HIAA (n=4). *P < 0.05, **P < 0.005, ***P < 0.001, ****P < 0.0001.

Figure 7.. Platelet and mast cell-derived 5-HIAA promote neutrophil recruitment in vivo

a-b, Representative flow cytometry plots (a) and quantification (b) of % transferred WT or KO neutrophils in ear skin of controls (green) or mice pretreated with Fluoxetine (purple) or Phenelzine (red), 2hr after Listeria skin-pricking (n=6–12). c, Graph showing quantification of transferred neutrophil recruitment index in controls or mice pretreated with Fluoxetine (purple) or Phenelzine (red), 2hr after Listeria skin-pricking (n=6–7). Graphs depict bars with mean +/− SEM. d, Quantification of CD45-PE⁺ intravascular transferred neutrophils in controls (d) and mice treated with Phenelzine (e), 2hr after Listeria skin-pricking (b, n=8, c, n=10). f,g, Graphs showing quantification of endogenous neutrophil % (f) (WT, n=9; Phenelzine, n=6; Fluoxetine, n=6) or CD45-PE⁺ intravascular endogenous neutrophils (g) in mice untreated (n=14) treated with Phenelzine (red) (n=9) or Fluoxetine (purple) (n=6). h, Quantification of GPR35-GFP WEHI-231 cell migration to activated platelet culture supernatants from mice treated with Phenelzine or Fluoxetine (Control platelet sup., n=6; all others, n=5). i, ELISA quantification of 5-HIAA in supernatants of activated platelets from control, phenelzine-treated or fluoxetine-treated mice. Data are representative of 2 independent experiments. j, Quantification of Ly6G⁺ CD11b⁺ endogenous neutrophils in the skin of WT (n=16), platelet-deficient (n=14) or platelet-deficient mice treated with phenelzine (n=7) 2hr after Listeria skin-pricking. Data are pooled from at least 3 independent experiments. k, l, Transferred neutrophil recruitment index in SERT Het or KO mice (k, n=4) or Tph1 Het or KO chimeras (l, n=8) 2hr after Listeria skin-pricking. Data are pooled from 2 independent experiments. *P < 0.05, **P < 0.005, ***P<0.001 ****P < 0.0001. See also Figures S7 and S8.