RESUMEN
BACKGROUND: Acute and chronic brain damage in type 2 diabetes mellitus (DM) determines the need to investigate the neuroprotective potential of glucose-lowering drugs. The purpose was to directly compare the neuroprotective effects of glucagon-like peptide-1 receptor agonists (GLP-1RAs) with different duration of action and sodium-glucose cotransporter-2 inhibitors (SGLT-2i) in type 2 diabetic rats with and without stroke. METHODS: DM was modelled using high-fat diet and nicotinamide+streptozotocin protocol. The following groups (n = 15 each) were formed: DM without treatment, treatment with liraglutide, dulaglutide, canagliflozin as well as control group without DM and treatment. After 8 weeks, 10 rats from each group underwent middle cerebral artery occlusion. In the reperfusion period neurological deficit, neuroglial damage markers and brain necrosis were evaluated. Brain slices from the remaining 5 animals in each group were histologically examined for microglial activation and neuronal damage. RESULTS: Brain damage was similar in "DM" and "Control" (17.53 [14.23; 26.58] and 15.87 [13.40; 22.68] % of total brain volume, respectively). All study drugs diminished damage volume comparing with "DM" and "Control" whereas the necrosis volume in "DM+Liraglutide" was smaller than in "DM+Canagliflozin" and did not significantly differ from "DM+Dulaglutide" (2.9 [1.83; 4.71], 6.17 [3.88; 8.88] and 4.57 [3.27; 7.90] %). The neurological deficit was more prominent in "DM" than in "Control", while all the drugs demonstrated similar positive effect. Neurofilament light chains (NLC) did not differ between "DM" and "Control". Dulaglutide and canagliflozin caused a marked decrease in NLC. Protein S100BB level was similar in "DM" and "Control". Liraglutide caused the largest S100BB decrease, while canagliflozin did not influence it. In chronic brain ischaemia, all drugs increased the number of normal neurons, but GLP-1RAs had a more pronounced effect. DM was accompanied by increased number of activated microglial cells in Cornu Ammonis (CA)1 hippocampal region. Both GLP-1RAs reduced the number of Iba-1-positive cells, with dulaglutide being more effective than liraglutide, whereas canagliflozin did not affect this parameter. CONCLUSIONS: GLP-1RAs and SGLT-2i have neuroprotective properties against acute and chronic brain damage in diabetic rats, although the infarct-limiting effect of GLP-1RAs may be more pronounced. GLP-1RAs and SGLT-2i exert their protective effects by directly influencing neuronal survival, whereas GLP-1RAs also affect microglia.