Fig 2: Hypoxia-induced changes of IL-17A and ER stress markers in retinas and macrophages. a Western blotting analysis of GRP78, ATF-4 and IL-17A protein levels at P12, P15, P17, P21 in retinas of OIR mice compared with their age-matched controls. b Immunofluorescent staining of F4/80 (red, specific expressed by macrophage) and ATF-4 (green), GRP78 (green), IL-17A (green) on representative sections of retinas from P17 OIR mice and normal controls. Boxed areas are magnified in the images. Yellow represents co-localization of GRP78, ATF-4 or IL-17A and macrophages infltration. Scale bars, 100 μm. c, d Analysis of mRNA (c) and protein (d) levels of GRP78, ATF-4 and IL-17A in macrophages exposed to hypoxia for 0, 4, 6, 8, 12 or 24 h. e Immunofluorescent staining of F4/80 (red) and ATF-4 (green), GRP78 (green), IL-17A (green) in macrophages cultured under hypoxic conditions for 0 and 24 h. Yellow represents cellular localization of GRP78, ATF-4 or IL-17A. Scale bars, 50 μm. β-actin served as an endogenous reference for normalization. Data are shown as mean ± SEM, n = 6–8 per group for Real-time RT-PCR, n = 3 per group for western blotting. Each experiment repeated three times. ns, no significance. * P < 0.05 and **P < 0.01 compared with control groups. ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer

Fig 3: Effects of MCC950 on the production of IL-17A under hypoxia conditions both in vivo and in vitro. a Representative immunoblots and densitometric analyses of TXNIP, NLRP3, IL-1β and IL-17A protein expression levels in OIR retinas treated with different concentrations of MCC950 (0, 0.01, 0.1, 1, 10 mM). b Representative immunoblots and densitometric analyses of TXNIP, NLRP3, IL-1β and IL-17A protein expression levels in macrophages cultured under hypoxic conditions treated with different concentrations of MCC950 (0, 0.01, 0.1, 1, 10 μM). c Representative immunoblots and densitometric analyses showed alterations of TXNIP, NLRP3, IL-1β and IL-17A protein expression levels in macrophages cultured under hypoxic conditions following TXNIP gene silenced with TXNIP shRNA. β-actin served as an endogenous reference for normalization. Data are shown as mean ± SEM, n = 6–8 per group for Real-time RT-PCR, n = 3 per group for western blotting. Each experiment repeated three times. ns, no significance. * P < 0.05 and **P < 0.01 compared with control groups or each other

Fig 4: Neutralization or knockout of IL-17A dampened ER stress in macrophages cultured under hypoxic conditions. a, d Real-time RT-PCR and western blotting analysis of GRP78, ATF-4 and CHOP mRNA (a) and protein (d) levels in macrophages pretreated with rmIL-17A for 24 h at different concentrations (0, 5, 10, 25, 50, 100 ng/ml). b, e Real-time RT-PCR and western blotting analysis of GRP78, ATF-4, CHOP and IL17A mRNA (b) and protein (e) levels in macrophages exposed to hypoxia for 12 h pretreated with IL-17ANab at different concentrations (0, 0.5, 1.0 μg/ml). c Changes of GRP78, ATF-4 and CHOP mRNA levels in macrophages from WT mice and IL-17A KO mice cultured under hypoxic conditions for 12 h. β-actin served as an endogenous reference for normalization. Data are shown as mean ± SEM, n = 6–8 per group for Real-time RT-PCR, n = 3 per group for western blotting. Each experiment repeated three times. ns, no significance. * P < 0.05 and **P < 0.01 compared with control groups

Fig 5: Inhibition of ER stress or IL-17A alleviated RNV. a Immunofluorescence staining of fluorescein-conjugated lectin in retinal flat-mounts of normal and OIR WT mice at P17 treated with 4-PBA or IL-17ANab or vehicle, as well as in retinal flat-mounts of normal and OIR IL-17A KO mice. Areas of vaso-obliteration (within yellow lines) and NV (red) were quantified. b, c Quantification of the avascular area (b) and NV area (c) of retinal flat mounts. N = 6 per group. ns, no significance. ***P < 0.001 compared with OIR (WT)-vehicle group. ###P < 0.001 compared with OIR (WT)-control group. Scale bars, 500 μm. Data are shown as mean ± SEM. ns, no significance