To investigate whether catalpol protects against diabetic retinal vascular endothelial injury by targeting the methyltransferase-like 3 (METTL3)-m⁶A-thioredoxin-interacting protein (TXNIP) axis and inhibiting nucleotide-binding oligomerization domain (NOD)-like receptor family pyrin domain containing 3 (NLRP3) inflammasome activation.

A streptozotocin-induced diabetic mouse model (n=20 per group) was used to assess retinal function via electroretinogram (ERG) and vascular integrity via Evans Blue leakage. Human retinal vascular endothelial cells (HRVECs) were exposed to high glucose (HG, 30 mmol/L) with or without catalpol or the METTL3 inhibitor STM2457. NLRP3 inflammasome components (Western blot), oxidative stress (DCFH-DA probe), global m⁶A levels (Dot blot), and TXNIP expression were measured. The binding of catalpol to METTL3, NLRP3, TXNIP, and interleukin-1β (IL-1β) was analyzed via molecular docking and dynamics simulations.

Catalpol treatment improved ERG amplitudes [a-wave, b-wave, oscillatory potentials (OPs)] and reduced vascular leakage in diabetic mice (P<0.05), while downregulating retinal vascular endothelial growth factor (VEGF), NLRP3, IL-1β, and IL-18 protein levels. In HG-stimulated HRVECs, catalpol inhibited the NLRP3-apoptosis-associated speck-like protein containing a CARD (ASC)-caspase-1 inflammasome, reduced reactive oxygen species, and suppressed METTL3 expression and global m⁶A methylation (P<0.05). It also attenuated HG-induced TXNIP upregulation. METTL3 inhibition by STM2457 mimicked all protective effects of catalpol. Molecular simulations confirmed stable binding of catalpol to METTL3, NLRP3, TXNIP, and IL-1β.

Catalpol alleviates diabetic retinal vascular endothelial injury by inhibiting the NLRP3 inflammasome. This effect is mediated, at least in part, through downregulating METTL3-dependent m⁶A RNA methylation of TXNIP.

To investigate systemic immune-inflammation index (SII), neutrophil-to-lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR) levels in patients with type 2 diabetes at different stages of diabetic retinopathy (DR).

This retrospective study included 141 patients with type 2 diabetes mellitus (DM): 45 without diabetic retinopathy (NDR), 47 with non-proliferative diabetic retinopathy (NPDR), and 49 with proliferative diabetic retinopathy (PDR). Complete blood counts were obtained, and NLR, PLR, and SII were calculated. The study analysed the ability of inflammatory markers to predict DR using receiver operating characteristic (ROC) curves. The relationships between DR stages and SII, PLR, and NLP were assessed using multivariate logistic regression.

The average NLR, PLR, and SII were higher in the PDR group than in the NPDR group (P=0.011, 0.043, 0.009, respectively); higher in the NPDR group than in the NDR group (P<0.001 for all); and higher in the PDR group than in the NDR group (P<0.001 for all). In the ROC curve analysis, the NLR, PLR, and SII were significant predictors of DR (P<0.001 for all). The highest area under the curve (AUC) was for the PLR (0.929 for PLR, 0.925 for SII, and 0.821 for NLR). Multivariate regression analysis indicated that NLR, PLR, and SII were statistically significantly positive and independent predictors for the DR stages in patients with DM [odds ratio (OR)=1.122, 95% confidence interval (CI): 0.200–2.043, P<0.05; OR=0.038, 95%CI: 0.018–0.058, P<0.05; OR=0.007, 95%CI: 0.001–0.01, P<0.05, respectively).

The NLR, PLR, and SII may be used as predictors of DR.