Search

Background Microglia cells represent the resident innate immune cells of the retina and are important for retinal development and tissue homeostasis. However, dysfunctional microglia can have a negative impact on the structural and functional integrity of the retina under native and pathological conditions. Methods In this study, we examined interferon-regulatory factor 8 (Irf8)–deficient mice to determine the transcriptional profile, morphology, and temporospatial distribution of microglia lacking Irf8 and to explore the effects on retinal development, tissue homeostasis, and formation of choroidal neovascularisation (CNV). Results Our study shows that Irf8-deficient MG exhibit a considerable loss of microglial signature genes accompanied by a severely altered MG morphology. An in-depth characterisation by fundus photography, fluorescein angiography, optical coherence tomography and electroretinography revealed no major retinal abnormalities during steady state. However, in the laser-induced CNV model, Irf8-deficient microglia showed an increased activity of biological processes critical for inflammation and cell adhesion and a reduced MG cell density near the lesions, which was associated with significantly increased CNV lesion size. Conclusions Our results suggest that loss of Irf8 in microglia has negligible effects on retinal homeostasis in the steady state. However, under pathological conditions, Irf8 is crucial for the transformation of resident microglia into a reactive phenotype and thus for the suppression of retinal inflammation and CNV formation. Supplementary Information The online version contains supplementary material available at 10.1186/s12974-021-02230-y.

INTRODUCTION: Hyalocytes have been recognized as resident tissue macrophages of the vitreous body since the mid-19(th) century. Despite this, knowledge about their origin, turnover, and dynamics is limited. AREAS COVERED: Historically, initial studies on the origin of hyalocytes used light and electron microscopy. Modern investigations across species including rodents and humans will be described. Novel imaging is now available to study human hyalocytes in vivo. The shared ontogeny with retinal microglia and their eventual interdependence as well as differences will be discussed. EXPERT OPINION: Owing to a common origin as myeloid cells, hyalocytes and retinal microglia have similarities, but hyalocytes appear to be distinct as resident macrophages of the vitreous body.

A chronic low-grade inflammation within adipose tissue (AT) seems to be the link between obesity and some of its associated diseases. One hallmark of this AT inflammation is the accumulation of AT macrophages (ATMs) around dead or dying adipocytes, forming so-called crown-like structures (CLS). To investigate the dynamics of CLS and their direct impact on the activation state of ATMs, we established a laser injury model to deplete individual adipocytes in living AT from double reporter mice (GFP-labeled ATMs and tdTomato-labeled adipocytes). Hence, we were able to detect early ATM-adipocyte interactions by live imaging and to determine a precise timeline for CLS formation after adipocyte death. Further, our data indicate metabolic activation and increased lipid metabolism in ATMs upon forming CLS. Most importantly, adipocyte death, even in lean animals under homeostatic conditions, leads to a locally confined inflammation, which is in sharp contrast to other tissues. We identified cell size as cause for the described pro-inflammatory response, as the size of adipocytes is above a critical threshold size for efferocytosis, a process for anti-inflammatory removal of dead cells during tissue homeostasis. Finally, experiments on parabiotic mice verified that adipocyte death leads to a pro-inflammatory response of resident ATMs in vivo, without significant recruitment of blood monocytes. Our data indicate that adipocyte death triggers a unique degradation process and locally induces a metabolically activated ATM phenotype that is globally observed with obesity.

Additional file 1. Supplemental figure 1 Irf8 is expressed predominantly in retinal MG and in some bipolar or Müller cells. No Irf8-VENUS expression could be detected co-localised with GFAP (A), βIII-Tubulin (B) and Collagen IV (E), indicating that IRF8 is not expressed in retinal astrocytes, ganglion cells or vessels. All IBA1+ cells exhibited a strong Irf8-VENUS signal (C) suggesting that all retinal MG express IRF8. Some CHX10+ cells could be co-localised with Irf8-VENUS expression (D) demonstrating that some bipolar cells or Müller cells express VENUS.

Additional file 3. Supplemental figure 3 Irf8 deficiency does not affect the retinal structure, function and vasculature. A-C Representative color fundus images (A), fluorescein angiography (B) and optical coherence tomography (OCT) images of Irf8 WT and Irf8 KO mice. C) Both Irf8 WT (blue, n=12) and Irf8 KO (red, n=12) mice displayed a regular retinal structure, a similar thickness of the inner nuclear layer and the outer retina at 100 and 200 μm from the optic nerve head in the optical coherence tomographs. Data are shown as mean ± SEM. ONH = Optic nerve head. D Electroretinography (ERG). No significant difference was found between the Irf8 WT (blue, n=9) and Irf8 KO (red, n=9) mice concerning the dark-adapted scotopic and light-adapted photopic ERG measurements at different flash intensities. Data are shown as mean ± SEM. E Staining against smooth muscle actin (SMA, red) reveals a comparable number of arteries (Irf8 WT (n=5); Irf8 KO (n=7)) and major vessels (Irf8 WT (n=11); Irf8 KO (n=13)) between both groups. Data are shown as mean ± SEM. F No significant differences in vessel branch formation in the superficial (upper panel) or deep plexus (lower panel) in the central or peripheral area of the retina could be observed, compared between Irf8 WT (n=6) and Irf8 KO (n=7). Data are presented as mean ± SEM.

Additional file 4. Supplemental figure 4 Spatial distribution of retinal microglia in CNV lesions. A Spatial distribution of microglia in the inner plexiform layer (IPL) in lasered and unlasered Irf8 WT and Irf8 KO mice. B Quantification of the numbers of microglia per field of view in the inner plexiform layer in lasered and unlasered Irf8 WT and Irf8 KO mice. Data are presented as mean ± SEM.

Additional file 5. Supplemental figure 5: Polarisation markers expressed by CNV-associated microglia. The M1 and M2 polarisation markers Cd86, H2-Ab1, Tlr2, Cd163 and Mrc1 (CD206) are shown as transcripts per million in comparison between Irf8 WT and KO under CNV conditions.

Additional file 7. Supplemental table 2: List of antibodies used for immunohistochemistry and flow cytometry.

Additional file 2. Supplemental figure 2 Temporal and spatial distribution of retinal microglia during development. A Representative pictures showing the numbers of microglia per field of view in Irf8 WT and Irf8 KO mice in comparison between the ganglion cell and inner plexiform layer (GCL/IPL) and the developing neuroblast layer (NBL) or outer plexiform layer (OPL), respectively, at postnatal day (P) 1, P7 and in adult mice. B Quantification thereof. Data are presented as mean ± SEM.