Müller cells and macrophages/microglia are likely important for the development of diabetic retinopathy; however, the interplay between these cells in this disease is not well understood. An inflammatory process is linked to the onset of experimental diabetic retinopathy. CD40 deficiency impairs this process and prevents diabetic retinopathy. Using mice with CD40 expression restricted to Müller cells, we identified a mechanism by which Müller cells trigger proinflammatory cytokine expression in myeloid cells. During diabetes, mice with CD40 expressed in Müller cells upregulated retinal tumor necrosis factor-α (TNF-α), interleukin 1β (IL-1β), intracellular adhesion molecule 1 (ICAM-1), and nitric oxide synthase (NOS2), developed leukostasis and capillary degeneration. However, CD40 did not cause TNF-α or IL-1β secretion in Müller cells. TNF-α was not detected in Müller cells from diabetic mice with CD40+ Müller cells. Rather, TNF-α was upregulated in macrophages/microglia. CD40 ligation in Müller cells triggered phospholipase C–dependent ATP release that caused P2X7-dependent production of TNF-α and IL-1β by macrophages. P2X7−/− mice and mice treated with a P2X7 inhibitor were protected from diabetes-induced TNF-α, IL-1β, ICAM-1, and NOS2 upregulation. Our studies indicate that CD40 in Müller cells is sufficient to upregulate retinal inflammatory markers and appears to promote experimental diabetic retinopathy and that Müller cells orchestrate inflammatory responses in myeloid cells through a CD40-ATP-P2X7 pathway.

Toxoplasma gondii resides in an intracellular compartment (parasitophorous vacuole) that excludes transmembrane molecules required for endosome - lysosome recruitment. Thus, the parasite survives by avoiding lysosomal degradation. However, autophagy can re-route the parasitophorous vacuole to the lysosomes and cause parasite killing. This raises the possibility that T. gondii may deploy a strategy to prevent autophagic targeting to maintain the non-fusogenic nature of the vacuole. We report that T. gondii activated EGFR in endothelial cells, retinal pigment epithelial cells and microglia. Blockade of EGFR or its downstream molecule, Akt, caused targeting of the parasite by LC3+ structures, vacuole-lysosomal fusion, lysosomal degradation and killing of the parasite that were dependent on the autophagy proteins Atg7 and Beclin 1. Disassembly of GPCR or inhibition of metalloproteinases did not prevent EGFR-Akt activation. T. gondii micronemal proteins (MICs) containing EGF domains (EGF-MICs; MIC3 and MIC6) appeared to promote EGFR activation. Parasites defective in EGF-MICs (MIC1 ko, deficient in MIC1 and secretion of MIC6; MIC3 ko, deficient in MIC3; and MIC1-3 ko, deficient in MIC1, MIC3 and secretion of MIC6) caused impaired EGFR-Akt activation and recombinant EGF-MICs (MIC3 and MIC6) caused EGFR-Akt activation. In cells treated with autophagy stimulators (CD154, rapamycin) EGFR signaling inhibited LC3 accumulation around the parasite. Moreover, increased LC3 accumulation and parasite killing were noted in CD154-activated cells infected with MIC1-3 ko parasites. Finally, recombinant MIC3 and MIC6 inhibited parasite killing triggered by CD154 particularly against MIC1-3 ko parasites. Thus, our findings identified EGFR activation as a strategy used by T. gondii to maintain the non-fusogenic nature of the parasitophorous vacuole and suggest that EGF-MICs have a novel role in affecting signaling in host cells to promote parasite survival.

Autophagy degrades pathogens in vitro. The autophagy gene Atg5 has been reported to be required for IFN-γ-dependent host protection in vivo. However, these protective effects occur independently of autophagosome formation. Thus, the in vivo role of classic autophagy in protection conferred by adaptive immunity and how adaptive immunity triggers autophagy are incompletely understood. Employing biochemical, genetic and morphological studies, we found that CD40 upregulates the autophagy molecule Beclin 1 in microglia and triggers killing of Toxoplasma gondii dependent on the autophagy machinery. Infected CD40−/− mice failed to upregulate Beclin 1 in microglia/macrophages in vivo. Autophagy-deficient Beclin 1+/− mice, mice with deficiency of the autophagy protein Atg7 targeted to microglia/macrophages as well as CD40−/− mice exhibited impaired killing of T. gondii and were susceptible to cerebral and ocular toxoplasmosis. Susceptibility to toxoplasmosis occurred despite upregulation of IFN-γ, TNF-α and NOS2, preservation of IFN-γ-induced microglia/macrophage anti-T. gondii activity and the generation of anti-T. gondii T cell immunity. CD40 upregulated Beclin 1 and triggered killing of T. gondii by decreasing protein levels of p21, a molecule that degrades Beclin 1. These studies identified CD40-p21-Beclin 1 as a pathway by which adaptive immunity stimulates autophagy. In addition, they support that autophagy is a mechanism through which CD40-dependent immunity mediates in vivo protection and that the CD40-autophagic machinery is needed for host resistance despite IFN-γ.