Retinopathy of prematurity (ROP) is a leading cause of childhood blindness associated with retinal vaso‐obliteration in phase 1 and pathological neovascularization (NV) in phase 2; however, effective and safe treatments for ROP definitive treatment are yet to be determined. Anti‐vascular endothelial growth factor (VEGF) therapy mainly focuses on reducing abnormal NV in phase 2 but with high risks of late recurrence and systemic side effects. Previous studies have established that the severity of vaso‐obliteration in phase 1 largely influences subsequent stages, suggesting that prevention of vessels loss may be a potential therapeutic target for ROP. Herein, the therapeutic potential and safety of early Elabela intervention treatment in treating phase 1 ROP and the possible underlying mechanisms were investigated using an oxygen‐induced retinopathy (OIR) mouse model. It was observed that intraperitoneal injection of Elabela remarkably reduced the avascular retinal area and increased the vascular density in phase 1 of OIR mice. Further investigation revealed that mitochondrion‐dependent ferroptosis was involved in oxidative stress‐mediated vascular protection loss in phase 1 OIR. Furthermore, we demonstrated that Elabela could rescue mitochondria‐dependent ferroptosis via mediating the xCT/GPX4 axis. Collectively, our study revealed that ferroptosis may play a significant role in early ROP, while Elabela may be a safe and promising strategy for the early intervention of ROP.

No abstract

No abstract

Configural face processing precedes featural face processing under the face-attended condition, but their temporal sequence in the absence of attention is unclear. The present study investigated this issue by recording visual mismatch negativity (vMMN), which indicates the automatic processing of visual information under unattended conditions. Participants performed a central cross size change detection task, in which random sequences of faces were presented peripherally, in an oddball paradigm. In Experiment 1, configural and featural faces (deviant stimuli) were presented infrequently among original faces (standard stimuli). In Experiment 2, configural faces were presented infrequently among featural faces, or vice versa. The occipital-temporal vMMN emerged in the 200–360 ms latency range for configural, but not featural, face information. More specifically, configural face information elicited a substantial vMMN component in the 200–360 ms range in Experiment 1. This result was replicated in the 320–360 ms range in Experiment 2, especially in the right hemisphere. These results suggest that configural, but not featural, face information is associated with automatic processing and provides new electrophysiological evidence for the different mechanisms underlying configural and featural face processing under unattended conditions.