Macrophages are innate immune cells that adopt diverse activation states in response to their microenvironment. Editing macrophage activation to dampen inflammatory diseases by promoting the repolarization of inflammatory (M1) macrophages to anti-inflammatory (M2) macrophages is of high interest. Here, we find that mouse and human M1 macrophages fail to convert into M2 cells upon IL-4 exposure in vitro and in vivo. In sharp contrast, M2 macrophages are more plastic and readily repolarized into an inflammatory M1 state. We identify M1-associated inhibition of mitochondrial oxidative phosphorylation as the factor responsible for preventing M1→M2 repolarization. Inhibiting nitric oxide production, a key effector molecule in M1 cells, dampens the decline in mitochondrial function to improve metabolic and phenotypic reprogramming to M2 macrophages. Thus, inflammatory macrophage activation blunts oxidative phosphorylation, thereby preventing repolarization. Therapeutically restoring mitochondrial function might be useful to improve the reprogramming of inflammatory macrophages into anti-inflammatory cells to control disease.

Cxcl18b is a chemokine found in zebrafish and in other piscine and amphibian species. Cxcl18b is a reliable inflammatory marker; however, its function is yet to be elucidated. Here, we found that Cxcl18b is chemotactic towards neutrophils, similarly to Cxcl8a/Interleukin-8, the best characterised neutrophil chemoattractant in humans and teleosts. Like Cxcl8a, Cxcl18b-dependent recruitment required the chemokine receptor Cxcr2, while it was unaffected by depletion of the other two neutrophil receptors cxcr1 and cxcr4b. To visualise cxcl18b induction, we generated a Tg(cxcl18b:eGFP) reporter line. The transgene is induced locally upon bacterial infection with the fish pathogen Mycobacterium marinum, but strikingly is not directly expressed by infected cells. Instead, cxcl18b is induced by non-phagocytic uninfected cells that compose the stroma of the granulomas, typical inflammatory lesions formed upon mycobacterial infections. Together, these results suggest that Cxcl18b might be an important contributor to neutrophil chemotaxis in the inflammatory microenvironment and indicate that the zebrafish model could be explored to further investigate in vivo the biological relevance of different Cxcl8-like chemokine lineages.

Most macrophages generate energy to mount an inflammatory cytokine response by increased glucose metabolism through intracellular glycolysis. Previous studies have suggested that alveolar macrophages (AMs), which reside in a glucose-poor natural environment, are less capable to utilize glycolysis and instead rely on other substrates to fuel oxidative phosphorylation (OXPHOS) for energy supply. At present, it is not known whether AMs are capable to use glucose metabolism to produce cytokines when other metabolic options are blocked. Here, we studied human AMs retrieved by bronchoalveolar lavage from healthy subjects, and examined their glucose metabolism in response to activation by the gram-negative bacterial component lipopolysaccharide (LPS) ex vivo. The immunological and metabolic responses of AMs were compared to those of cultured blood monocyte-derived macrophages (MDMs) from the same subjects. LPS stimulation enhanced cytokine release by both AMs and MDMs, which was associated with increased lactate release by MDMs (reflecting glycolysis), but not by AMs. In agreement, LPS induced higher mRNA expression of multiple glycolytic regulators in MDMs, but not in AMs. Flux analyses of [¹³C]-glucose revealed no differences in [¹³C]-incorporation in glucose metabolism intermediates in AMs. Inhibition of OXPHOS by oligomycin strongly reduced LPS-induced cytokine production by AMs, but not by MDMs. Collectively, these results indicate that human AMs, in contrast to MDMs, do not use glucose metabolism during LPS-induced activation and fully rely on OXPHOS for cytokine production.

Circulating nonadherent monocytes can migrate to extravascular sites by a process that involves adherence. Alterations in intracellular metabolism shape the immunological phenotype of phagocytes upon activation. To determine the effect of adherence on their metabolic and functional response human monocytes were stimulated with LPS under nonadherent and adherent conditions. Adherent monocytes (relative to nonadherent monocytes) produced less TNF and IL-1β (proinflammatory) and more IL-10 (anti-inflammatory) upon LPS stimulation and had an increased capacity to phagocytose and produce reactive oxygen species. RNA sequencing analysis confirmed that adherence modified the LPS-induced response of monocytes, reducing expression of proinflammatory genes involved in TLR signaling and increasing induction of genes involved in pathogen elimination. Adherence resulted in an increased glycolytic response as indicated by lactate release, gene set enrichment, and [¹³C]-glucose flux analysis. To determine the role of glycolysis in LPS-induced immune responses, this pathway was inhibited by glucose deprivation or the glucose analogue 2-deoxy-d-glucose (2DG). Although both interventions equally inhibited glycolysis, only 2DG influenced monocyte functions, inhibiting expression of genes involved in TLR signaling and pathogen elimination, as well as cytokine release. 2DG, but not glucose deprivation, reduced expression of genes involved in oxidative phosphorylation. Inhibition of oxidative phosphorylation affected TNF and IL-10 release in a similar way as 2DG. Collectively, these data suggest that adherence may modify the metabolic and immunological profile of monocytes and that inhibition of glycolysis and oxidative phosphorylation, but not inhibition of glycolysis alone, has a profound effect on immune functions of monocytes exposed to LPS. Key Points Nonadherent monocytes produce more TNF and IL-1β upon LPS stimulation. Adherent, but not nonadherent, monocytes induce glycolysis upon stimulation. Treatment with 2DG, but not glucose deprivation, influences monocyte functions.

Hypoxia‐inducible factor‐ (HIF‐) 1α has been implicated in the ability of cells to adapt to alterations in oxygen levels. Bacterial stimuli can induce HIF1α in immune cells, including those of myeloid origin. We here determined the role of myeloid cell HIF1α in the host response during pneumonia and sepsis caused by the common human pathogen Klebsiella pneumoniae. To this end, we generated mice deficient for HIF1α in myeloid cells (LysM‐cre × Hif1α^fl/fl) or neutrophils (Mrp8‐cre × Hif1α^fl/fl) and infected these with Klebsiella pneumoniae via the airways. Myeloid, but not neutrophil, HIF1α‐deficient mice had increased bacterial loads in the lungs and distant organs after infection as compared to control mice, pointing at a role for HIF1α in macrophages. Myeloid HIF1α‐deficient mice did not show increased bacterial growth after intravenous infection, suggesting that their phenotype during pneumonia was mediated by lung macrophages. Alveolar and lung interstitial macrophages from LysM‐cre × Hif1α^fl/fl mice produced lower amounts of the immune enhancing cytokine tumor necrosis factor upon stimulation with Klebsiella, while their capacity to phagocytose or to produce reactive oxygen species was unaltered. Alveolar macrophages did not upregulate glycolysis in response to lipopolysaccharide, irrespective of HIF1α presence. These data suggest a role for HIF1α expressed in lung macrophages in protective innate immunity during pneumonia caused by a common bacterial pathogen.

The lipidome of immune cells during infection has remained unexplored, although evidence of the importance of lipids in the context of immunity is mounting. In this study, we performed untargeted lipidomic analysis of blood monocytes and neutrophils from patients hospitalized for pneumonia and age- and sex-matched noninfectious control volunteers. We annotated 521 and 706 lipids in monocytes and neutrophils, respectively, which were normalized to an extensive set of internal standards per lipid class. The cellular lipidomes were profoundly altered in patients, with both common and distinct changes between the cell types. Changes involved every level of the cellular lipidome: differential lipid species, class-wide shifts, and altered saturation patterns. Overall, differential lipids were mainly less abundant in monocytes and more abundant in neutrophils from patients. One month after hospital admission, lipidomic changes were fully resolved in monocytes and partially in neutrophils. Integration of lipidomic and concurrently collected transcriptomic data highlighted altered sphingolipid metabolism in both cell types. Inhibition of ceramide and sphingosine-1-phosphate synthesis in healthy monocytes and neutrophils resulted in blunted cytokine responses upon stimulation with lipopolysaccharide. These data reveal major lipidomic remodeling in immune cells during infection, and link the cellular lipidome to immune functionality.