Hif-1α stabilisation polarises macrophages via cyclooxygenase/prostaglandin E2 in vivo

Macrophage subtypes are poorly characterised in disease systems in vivo. The initial innate immune response to injury and infectious stimuli through M1 polarisation is important for the outcome of disease. Appropriate macrophage polarisation requires complex coordination of local microenvironmental cues and cytokine signalling to influence immune cell phenotypes. If the molecular mechanisms behind macrophage polarisation were better understood then macrophages could be pharmacologically tuned to better deal with bacterial infections, for example tuberculosis. Here, using zebrafish tnfa:GFP transgenic lines as in vivo readouts of M1 macrophages, we show that hypoxia and stabilisation of Hif-1α polarises macrophages to a tnfa expressing phenotype. We demonstrate a novel mechanism of Hif-1α mediated macrophage tnfa upregulation via a cyclooxygenase/prostaglandin E2 axis, a mechanism that is conserved in human primary macrophages. These findings uncover a novel macrophage HIF/COX/TNF axis that links microenvironmental cues to macrophage phenotype that may have implications in inflammation, infection and cancer, where hypoxia is a common microenvironmental feature and where cyclooxygenase and Tnfa are major mechanistic players.

4 neutrophils, are exquisitely sensitive to tissue hypoxia, a signature of many disease 8 3 microenvironments (eg infections and cancer) due to lack of local blood supply and a high 8 4 turnover of local oxygen by pathogens (5,18 demonstrated to be upregulated in macrophages after tailfin injury in a zebrafish model (31).

3 9
Here, we demonstrate that this injury induced upregulation of macrophage tnfa:GFP is also other tnfa promoter driven line, (Tg(tnfa:eGFP-F)ump5Tg, crossed to Tg(mpeg1:mCherry-2 5 1 F)ump2Tg) ( Figure S1). Our data demonstrate that injury and Mm induced tnfa expression 2 5 2 occurs in macrophages as part of an early M1 response. Hif-1α can be stabilised in zebrafish both genetically and pharmacologically using  pharmacological Hif-1α stabilisation ( Figure 1J). Together, these data indicate that tnfa 2 7 4 expression is part of a pro-inflammatory M1 macrophage response to hypoxia and stabilised 2 7 5 Hif-1α, a response that is targetable by pharmacological agents and has the potential to aid 2 7 6 the host response to bacterial challenge. Eicosanoids are lipid signalling molecules produced by macrophages during early was significantly abrogated to basal levels by both SC560 and NS398 (Figure 2A-B). We NS398 caused a small decrease that did not reach basal levels ( Figure S3B-C). These data indicate that Hif-1α-induced tnfa expression is caused by a product of the cyclooxygenase regulation of M1 pro-inflammatory cytokines by Hif-1α stabilisation.
2 9 3 We next tested whether injury-and infection-induced tnfa are cyclooxygenase 2 9 4 dependent processes. Macrophage tnfa:GFP expression induced after injury was not cyclooxygenase inhibition using either SC560 or NS398 did not alter the expression of Mm- induced tnfa:GFP ( Figure 2F-G). Our data indicate that macrophage tnfa expression can be cyclooxygenase independent mechanisms ( Figure 2H). We have previously shown that Hif-1α stabilisation is host-protective in the zebrafish 3 0 1 Mm model, an effect that is dependent on macrophage expression of proinflammatory il-1β 3 0 2 (33). We therefore hypothesised that priming macrophages with increased tnfa via Hif-1α demonstrated that Hif-1α-induced tnfa requires cyclooxygenase activity in Figure 2A-C).

0 9
Neither SC560 nor NS398 mediated inhibition of Hif-1α-induced tnfa expression diminished F). Taken together these data indicate that injury-and Mm-induced tnfa is not further macrophages is not required for the Hif-1α mediated reduction of bacterial burden. To investigate whether the effect of the cyclooxygenase inhibitors on tnfa was 3 1 8 specific to the prostaglandin path of the arachidonic acid pathway, we targeted the lipoxin 3 1 9 and leukotriene producing arms using the 15-Lipoxygenase inhibitor PD146176 and as DA Hif-α ( Figure 3A-B). PD146176 also did not affect tnfa:GFP expression after Mm PD146176 was sufficient to decrease Mm burden, but not to the same extent as DA Hif-1α 3 2 6 ( Figure 3C). Leukotriene B4 inhibition, using the BLTR1/2 antagonists U75302 and prostaglandins. The best characterised of these as a regulator of macrophage function is 3 3 6 prostaglandin E2 (PGE2). We tested whether PGE2 was a mediator in the HIF/COX/TNF  The HIF-COX-TNF axis is conserved in human macrophages To translate our findings from zebrafish to humans we tested whether HIF-1α 3 4 9 stabilisation in human macrophages induces TNF expression. We found that human 3 5 0 monocyte derived macrophages (hMDMs) produced higher levels of TNF protein in hypoxia 3 5 1 (0.8% oxygen) than those in normoxia, measured by an anti-human TNF ELISA ( Figure 5A).

5 2
Furthermore, treatment with the COX-2 inhibitor, NS398, reduced this hypoxia-induced TNF 3 5 3 back to the equivalent levels found in normoxia ( Figure 5A). This was replicated when HIF- 1α was stabilised in hHDMs using the hypoxia mimetic FG4592 ( Figure 5B). These data 3 5 5 indicate that the HIF/COX/TNF axis is a conserved mechanism in human macrophages and 3 5 6 could be important in human disease. signalling. Understanding of signalling pathways that link the microenvironment with 3 6 2 macrophage phenotypic outcomes will identify potential novel therapeutic avenues for leading to different macrophage phenotypes are not well-defined in vivo.

6 5
Here we show that a disease relevant microenvironmental cue, hypoxia signalling via Hif-1α, upregulates macrophage tnfa expression in a cyclooxygenase dependent manner in 3 6 7 vivo. TNF regulation by hypoxia has been shown in a range of mammalian cells, and its 3 6 8 promoter region contains HIF responsive elements (HREs), resulting in some level of direct concerning il-1β activation (33). Not only could tnfa activation be achieved by genetic 3 7 3 stabilisation of Hif-1α, but also using hypoxia mimetics and physiological hypoxia. These findings indicate that the Hif-1α pro-inflammatory switch is targetable by pharmaceuticals cyclooxygenase and further shows that the mechanism is likely to be via the production of PGE2. The degradation metabolite 15-keto-PGE2 did not rescue the loss of tnfa expression 3 8 0 after cyclooxygenase inhibition, consistent with previous reports that 15-keto-PGE2 is widely studied in human, mice and zebrafish and is rich in pro-inflammatory cytokine 3 8 7 production (32, 33, 65-69). Here, we observed that an M1 pro-inflammatory tnfa response 3 8 8 was induced at pre-granuloma stages as well as during granuloma formation in vivo. A key 3 8 9 advantage of using fluorescent transgenic zebrafish lines, such as tnfa:GFP, is that they enable identification of the cell type producing the transgene in an intact organism. This is it eliminates the risk of activation during a sorting process. TNF is required for the control of therapies for immune diseases such as Crohn's disease and rheumatoid arthritis, which 3 9 5 have proved effective treatments for these debilitating illnesses, have an associated it has been widely demonstrated that Tnfa is required for control of early infection, with 3 9 8 perturbation of Tnfa signalling leading to high infectious burdens (75). When Hif-1α driven 3 9 9 tnfa was downregulated by cyclooxygenase inhibition, there was no effect on the decrease  Our data reveal a novel mechanism of TNF activation in macrophages via a Hif-1α/ 4 1 9 cyclooxygenase/PGE2 axis. Importantly, this pathway is conserved in human macrophages, infections such as TB that might be influenced by a HIF-1α mediated M1/TNF switch.

2 8
Hypoxia is a key hallmark of cancers with high levels of HIF-1α widely found in those that 4 2 9 produce large tumours where the centre is hypoxic (78). This is also true of HIF/COX/TNF mechanism. The zebrafish larvae is small enough to be fully oxygenated at where hypoxia is a key hallmark of disease pathology is required to uncover the full 4 3 8 therapeutic relevance of this potentially important novel macrophage pathway.