Save the brain! Studies on GLP-1-mediated neuroprotection

Abstract

Background: Patients with type 2 diabetes (T2D) suffer stroke more often and have worse recovery. Moreover, hyperglycemic stroke patients treated with thrombolytic therapy have more intracranial bleedings. Insulin is the gold standard for treatment of hyperglycemia, but has a significant risk of hypoglycemia, especially if rapid normalization of glucose is the goal. Glucagon like peptide-1 based drugs, such as receptor agonists (GLP-1RA) and dipeptidylpeptidase-4 inhibitors (DPP-4i) that raise the endogenous GLP-1 levels are used to treat T2D. Interestingly these drugs have been shown to reduce stroke incidence; and, importantly for this thesis, to exert acute neuroprotective effects in animal models if given before a stroke.

Aim: To determine if GLP-1RA and DPP-4i are neuroprotective in animal models when given after stroke and to identify some of the underlying mechanisms. Furthermore, the aim was to determine the GLP-1 levels in humans after stroke and their relationship to outcome. Finally, to test treating hyperglycemic stroke patients with GLP-1RA early after stroke: in the ambulance.

Study 1. An experimental animal study determining the effect of giving the GLP-1RA exendin-4 (Ex-4) after the onset of stroke. We did this in both young and healthy animals as well as aged and obese/diabetic animals. Experimental stroke was induced by middle cerebral artery occlusion (MCAO). The main finding was that exendin-4 is neuroprotective if given after MCAO, both in young and old animals. The effect was time-sensitive with effect at 1.5 and 3 h after MCAO, but lost at the 4.5 h time point. Study 2. An experimental animal study where we determined if DPP-4i treatment had neuroprotective properties when started after the onset of stroke. Additionally, we studied if the effect was dependent of the GLP-1 receptor by using a GLP-1R -/- knockout model mouse. The main finding was that DPP-4 inhibitors require chronic pre-treatment to be effective. Furthermore, we showed that the effect is not dependent on the GLP-1 receptor. Study 3. In rats with diabetes (GK-rats) or without (Wistars) we studied the impact of aging in diabetes on regulatory GABA-ergic interneurons (key cells involved in stroke recovery) and whether treatment with the GLP-1RA exendin-4 could revert the observed changes. Main findings: 1: The number of GABAeric neurons decreased in aged diabetic animals. 2: Treatment with Ex-4 had effect on the subpopulation of GABAeric neurons positive for calbindin. Study 4. We determined the endogenous GLP-1 levels in patients treated with thrombolytic therapy for ischemic stroke. A group of 59 patients underwent a OGTT at day 2-4 during the hospital stay. A repeat OGTT was performed 3 months later. At the three-month follow-up functional stroke outcome was measured with mRS. 27 healthy controls underwent one OGTT. The main findings were that the GLP-1 level was higher in stroke patients and remained unchanged 3 months later. The GLP-1 level was not associated with functional outcome. Study 5. A randomized clinical trial examining the feasibility of prehospital treatment with exenatide in hyperglycemic patients (8-15 mmol/L) with suspected stroke. Patients were followed for 24 h. 19 patients were randomized, 8 received exenatide. There was no evidence of a difference in the main outcome of glucose at 4 hours. No adverse events were reported.

Conclusion: Both GLP-1RA and DPP-4i are neuroprotective against stroke. The effect is, however, time sensitive with the need for early initiation after stroke (GLP-1RA) or chronic prestroke treatment (DPP-4i). Additionally, the effect of DPP-4i is GLP-1 receptor independent. The findings encourage the use of these drugs for the treatment of diabetes.