Macrophage tnfa:GFP is upregulated by injury, Mm infection and Hif-1α stabilization. (A) Confocal micrographs of 3 days post-fertilization larvae with or without tailfin wound (16 hpw) at the tailfin site. tnfa expression was detected using the TgBAC(tnfa:GFP)pd1028 transgenic line. Macrophages are shown in red using a Tg(mpeg1:mCherryCAAX)sh378 line. The white arrowhead indicates the presence of a tnfa:GFP negative, mpeg:mCherry positive macrophage in the tailfin of a non-wounded control. Asterisk indicates a neuromast that is fluorescent in all channels as a marker of position. Representative images of three independent experiments with 20–30 larvae per group per experiment. (B) Confocal micrographs of 1 dpi Mm mCrimson infected larvae, prior to granuloma formation (left panels), and 4 dpi, early granuloma (right panels) stages of infection using the TgBAC(tnfa:GFP)pd1028 line. Representative images of three independent experiments with 20–30 larvae per group per experiment. (C) Confocal micrographs of 1 day post-infection (dpi) with Mm mCrimson at the caudal vein region of infection in the TgBAC(tnfa:GFP)pd1028 line. Macrophages are shown in red using a Tg(mpeg1:mCherryCAAX)sh378 line. Mm in right panels with PVP control in left panels. a Dotted lines indicate the yolk extension of the larvae where there is non-specific fluorescence. Representative images of three independent experiments with 20–30 larvae per group per experiment. (D) Confocal micrographs of 2 dpf larvae in the caudal vein region in the TgBAC(tnfa:GFP)pd1028 line. Macrophages are shown in red using a Tg(mpeg1:mCherryCAAX)sh378 line. Larvae were injected at the 1 cell stage with dominant active (DA) Hif-1α or phenol red (PR) control. Representative images of three independent experiments with 20–30 larvae per group per experiment. (E) Confocal micrographs of 2 dpf zebrafish imaged around the caudal vein region in the TgBAC(tnfa:GFP)pd1028 line. Larvae were injected at the 1 cell stage with dominant negative (DN) or dominant active (DA) Hif-1α or phenol red (PR) control. (F) Corrected fluorescence intensity levels of tnfa:GFP in larvae in (E). Mean ± SEM, n = 66 cells from 12 embryos accumulated from three independent experiments and corresponds to images in (E). P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, one way ANOVA with Bonferonni post-test adjustment. (G) Confocal micrographs of 2 dpf TgBAC(tnfa:GFP)pd1028 larvae treated with the PHD inhibitor FG4592 or a DMSO solvent control. Graph shows corrected fluorescence intensity levels of tnfa:GFP. Mean ± SEM, n = 54 cells from nine embryos accumulated from three independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, unpaired, two-tailed t-test. (H) Confocal micrographs of 2 dpf TgBAC(tnfa:GFP)pd1028 larvae treated with the PHD inhibitor FG4592 or a DMSO solvent control. Larvae were injected at the 1 cell stage with dominant negative (DN) Hif-1α or phenol red (PR) control. (I) Corrected fluorescence intensity levels of tnfa:GFP in larvae in (H). Mean ± SEM, n = 60 cells from 10 embryos accumulated from three independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment. (J) Confocal micrographs of 2 dpf TgBAC(tnfa:GFP)pd1028 larvae incubated in room normoxia or 5% oxygen (hypoxia) for 6 h between 32 and 38 hpf. Graph shows corrected fluorescence intensity levels of tnfa:GFP. Mean ± SEM, n = 72 cells from 12 embryos accumulated from three independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, unpaired, two-tailed t-test.

Hif-1α-activated tnfa is cyclooxygenase dependent. (A) Confocal micrographs of 2 dpf caudal vein region of larvae in the TgBAC(tnfa:GFP)pd1028 line. Phenol red (PR) and dominant active Hif-1α (DA1) injected larvae were treated with DMSO, SC560 (Cox-1 inhibitor), and NS398 (Cox-2 inhibitor). Dotted lines indicate the yolk extension of the larvae where there is non-specific fluorescence. (B) Corrected fluorescence intensity levels of tnfa:GFP in larvae in (A). Mean ± SEM, n = 36 cells from six embryos representative of three independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment. (C) Confocal micrographs of 2 dpf caudal vein region of larvae in the TgBAC(tnfa:GFP)pd1028 line. DMSO and FG4592 treated larvae were co-treated with DMSO, SC560 (Cox-1 inhibitor), and NS398 (Cox-2 inhibitor). Dotted lines indicate the yolk extension of the larvae where there is non-specific fluorescence. (D) Corrected fluorescence intensity levels of tnfa:GFP in larvae in (C). Mean ± SEM, n = 48 cells from eight embryos representative of three independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment. (E) TNF ELISA of human monocyte derived macrophages treated with LPS and incubated in normoxia or 0.8% hypoxia with or without treatment with NS398. LPS negative controls were performed but TNF produced in these groups was below detectable levels. Mean ± SEM, n = 5–8 biological repeats from 3 to 4 donors. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment. (F) Schematic of the arachidonic pathway indicating that stabilizing Hif-1α upregulates tnfa, an effect that is blocked using the Cox1/2 inhibitors SC560/NS398.

Injury and infection induced tnfa:GFP is cyclooxygenase independent. (A) Confocal micrographs of the TgBAC(tnfα:GFP)pd1028 line at 3 days post-fertilization larvae with or without tailfin wound (induced 16 h previously) in phenol red (PR) or dominant negative (DN) Hif-1α embryos. The dotted line indicates the border of the tailfin wound. (B) Corrected fluorescence intensity levels of tnfa:GFP in larvae in (A). Mean ± SEM, n = 60 cells from 10 embryos accumulated from three independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment. (C) Confocal micrographs of 1 dpi caudal vein region of larvae in the TgBAC(tnfa:GFP)pd1028 line. Phenol red (PR) and dominant negative Hif-1α (DN1) injected larvae were infected with Mm Crimson or PVP controls. (D) Corrected fluorescence intensity levels of tnfa:GFP in larvae in (C). Mean ± SEM, n = 30 cells from five embryos accumulated from two independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment. (E) Confocal micrographs of the TgBAC(tnfα:GFP)pd1028 line at 3 days post fertilization larvae with or without tailfin wound (induced 16 h previously) in phenol red (PR) or dominant active Hif-1α embryos. Embryos were treated with DMSO or NS398 (Cox-2 inhibitor). (F) Corrected fluorescence intensity levels of tnfa:GFP in larvae in (E). Mean ± SEM, n = 24 (in not wounded) or n = 36 (in wounded) cells from six embryos accumulated from two independent experiments. n.b. Not-wounded fish had fewer macrophages at the tailfin site. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment. (G) Confocal micrographs of 1 dpi caudal vein region of Mm Crimson infected larvae in the TgBAC(tnfa:GFP)pd1028 line. Phenol red (PR) and dominant active Hif-1α (DA1) injected larvae were treated with DMSO, SC560 (Cox-1 inhibitor), and NS398 (Cox-2 inhibitor) and infected with Mm Crimson. (H) Corrected fluorescence intensity levels of tnfa:GFP in larvae in (G). Mean ± SEM, n = 36 cells from six embryos representative of three independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment.

Blocking 15-lipoxygenase or leukotriene B4 receptors does not abrogate DA-Hif-1α-upregulation of tnfa:GFP. (A) Schematic of the arachidonic pathway indicating where the inhibitors act. (B) Confocal micrographs of 1 dpi caudal vein region of Mm mCrimson and PVP infected larvae in the TgBAC(tnfa:GFP)pd1028 line. Phenol red (PR) and dominant active Hif-1α (DA1) injected larvae were treated with DMSO or PD146176 (15-Lipoxygenase inhibitor). Dotted lines indicate the yolk extension of the larvae where there is non-specific fluorescence. (C) Corrected fluorescence intensity levels of tnfa:GFP in larvae in (B). Mean ± SEM, n = 42 cells from seven embryos accumulated from three independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment. (D) Bacterial burden at 4 dpi after injection of DA Hif-1α (DA1) and 24 h of the 15-lipoxygenase inhibitor PD146176, using DMSO as a negative solvent control. Data shown are mean ± SEM, n = 48–50 as accumulated from three independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment. (E) Confocal micrographs of 1 dpi caudal vein region of Mm mCrimson and PVP infected larvae in the TgBAC(tnfa:GFP)pd1028 line. Phenol red (PR) and dominant active Hif-1α (DA1) injected larvae were treated with DMSO or U75302/LY255283 (BLTR1/2 inhibitors). (F) Corrected fluorescence intensity levels of tnfa:GFP confocal z-stacks in uninfected larvae (PVP, empty bars) and infected larvae (Mm, filled bars) at 1 dpi of data shown in (E) after treatment with DMSO or U75302/LY255283 (BLTR1/2 inhibitors). Data shown are mean ± SEM, n = 30 cells from five embryos accumulated from two independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment.

Hif-1α-induced tnfa:GFP requires active prostaglandin E2. (A) Confocal micrographs of 1 dpi caudal vein region in the TgBAC(tnfa:GFP)pd1028 line. Phenol red (PR) and dominant active Hif-1α (DA1) injected larvae were treated with DMSO or NS398 (Cox-2 inhibitor) in the presence or absence of endogenous prostaglandin E2 (PGE2). All larvae are PVP injected. (B) Corrected fluorescence intensity levels of tnfa:GFP in larvae in (A). Mean ± SEM, n = 54 cells from nine embryos accumulated from three independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment. (C) Confocal micrographs of 1 dpi caudal vein region in the TgBAC(tnfa:GFP)pd1028 transgenic line. Phenol red (PR) and dominant active Hif-1α (DA1) injected larvae were treated with DMSO or NS398 (Cox-2 inhibitor) in the presence or absence of endogenous 15-keto prostaglandin E2 (15K). All larvae are PVP injected. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment. (D) Corrected fluorescence intensity levels of tnfa:GFP in larvae in (C). Mean ± SEM, n = 24 cells from four embryos accumulated from two independent experiments. (E) Confocal micrographs of 1 dpi caudal vein region in the TgBAC(tnfa:GFP)pd1028 transgenic line. Phenol red (PR) and dominant active Hif-1α (DA1) injected larvae were treated with DMSO or AH2348 (EP4 inhibitor). All larvae are PVP injected. (F) Corrected fluorescence intensity levels of tnfa:GFP in larvae in (E). Mean ± SEM, n = 72 cells from 12 embryos accumulated from three independent experiments. P-values shown are: *P < 0.05, **P < 0.01, and ***P < 0.001, two way ANOVA with Bonferonni post-test adjustment.