Investigating the effects of inflammation on resident peritoneal macrophage autonomy
Louwe, Pieter Adriaan
Inflammation is known to lead to long-term changes in the responsiveness to subsequent inflammatory challenges but the mechanism driving this alteration is poorly understood. In the peritoneal cavity a similar phenomenon has been described and historic peritoneal inflammation is thought to lead to dampened neutrophil recruitment in response to a subsequent insult. Acute peritoneal inflammation is characterized by the transient loss of embryonically-seeded resident macrophages and infiltration by large numbers of monocyte-derived inflammatory macrophages. The long-term fate of inflammatory macrophages post-inflammation has remained controversial. Hence, I hypothesized that inflammatory macrophages persist long-term but could remain functionally distinct. In addition, I aimed to explore what regulates the survival and identity of these cells by examining how severity of the initial inflammatory insult changes the post inflammation micro-environment. Furthermore, a combination of cell-intrinsic regulation and environmental signals limits inflammatory macrophages from converting into resident macrophages. To investigate these questions, I developed a flow-cytometric method based on dye-labelling of resident macrophages in conjunction with conventional antibody staining to unequivocally delineate resident from inflammatory macrophages during resolution of a well characterised model of sterile peritoneal inflammation induced by injection of zymosan A. Following a low dose of zymosan, both resident and inflammatory macrophages persisted in the peritoneal cavity by day 3 whereas severe inflammation resulting from injection of a high dose of zymosan led to complete loss of resident macrophages. I then used adoptive transfer of purified resident or inflammatory macrophages present during resolution of mild peritonitis into congenic recipient mice that harboured an equivalent inflammatory response to investigate their long-term survival and identity. To investigate if the post resolution environment dictates these features, cells were transferred into naïve recipient mice. To ascertain if competition with incumbent resident macrophages for niche signals alters this fate and phenotype, donor cells were transferred into recipient mice transiently-depleted of endogenous resident macrophages. Combined these data allowed me to ascertain to what degree survival and identity of inflammatory macrophages is dictated by environmental niche cues, competition with incumbent resident macrophages and to what degree these features are pre-determined. These adoptive transfer experiments revealed that following mild peritonitis both resident and inflammatory macrophages persist through inflammation resolution, when the macrophage compartment contracts to pre-inflammation size. Long-term fate mapping indicated that transferred cells survived to an equivalent degree up to 8 weeks. This suggests that cells that persisted shortly after resolution become long-lived. However, Nanostring mRNA analysis at 8 weeks demonstrated that inflammatory macrophages acquired some resident features such as CD102 expression but remained transcriptionally distinct. Importantly, inflammatory macrophages failed to acquire equivalent levels of the peritoneal identity transcription factor GATA6 whilst maintaining high expression of inflammatory macrophage markers, most notably MHCII and Sema4a. Critically, inflammatory macrophages transferred into macrophage-deplete recipients adopted a more complete resident macrophage identity, including equivalent expression of GATA6 and MHCII to their resident counterparts, whilst retaining differential expression of few genes, including the embryonic macrophage marker Tim4. Combined these data indicate that the failure of inflammatory macrophages to adopt a resident identity is largely environment and competition dictated. However, over time more complete conversion occurs, as fate mapping until 5 months after mild peritonitis indicated that inflammatory macrophages gradually reprogram both environment dependent and independent features. In contrast, after severe peritoneal inflammation resident macrophages were completely and irrevocably lost, which was confirmed using tissue-protected bone marrow chimeric mice. Inflammatory macrophages recruited during severe inflammation adopted a somewhat ‘resident like’ identity including high levels of GATA6, equivalent to resident macrophages. However, after severe peritonitis, and complete loss of resident macrophages, inflammatory macrophages retained high levels of MHCII. As MHCII was rapidly downregulated by low dose inflammatory macrophages after transfer into macrophage-deplete recipients but not after transfer into naïve or equivalently inflamed recipients these data indicate that signals unique to the severely inflamed cavity drive expression of MHCII. Consistently, naïve resident macrophages expressing low levels of MHCII transferred into high dose zymosan treated recipients rapidly upregulated MHCII. Finally, I showed that inflammatory macrophages that persisted 8 weeks after mild peritonitis were more proliferative, less phagocytic and produced less TNFa in response to LPS stimulation both in vitro and in vivo. One feature of inflammatory macrophages, irrespective of the severity of peritonitis was an inability to produce the B1 cell chemokine CXCL13 and as a likely consequence peritoneal B1 cells failed to accumulate post inflammation. Hence, these data support a model where inflammation drives the integration of functionally divergent monocyte-derived inflammatory macrophages into the peritoneal cavity. Moreover, during mild inflammation competitive pressure from incumbent resident macrophages for environmental signals largely dictates the extent to which inflammatory macrophages adopt a more resident-like identity. Conversely, after severe peritonitis no competing resident macrophages persist but environmental signals are altered and consequently inflammatory macrophages adopt a unique identity.