Phenothiazines reduced autophagy in ischemic stroke through endoplasmic reticulum (ER) stress-associated PERK-eIF2α pathway.

Background: Neuroprotective effects have been the main focus of new treatment modalities for ischemic stroke. Phenothiazines, or chlorpromazine plus promethazine (C + P), are known to prevent the generation of free radicals and uptake of Ca by plasma membrane; they have a potential as a treatment for acute ischemic stroke (AIS). This study aims to investigate the role of endoplasmic reticulum (ER) stress-associated PERK-eIF2α pathway underlying the phenothiazine-induced neuroprotective effects after cerebral ischemia/reperfusion (I/R) injury.

Methods: A total of 49 male Sprague Dawley rats (280-320 g) were randomly divided into 4 groups (n = 7 per group): (1) sham, (2) I/R that received 2 h of middle cerebral artery occlusion (MCAO), followed by 6 or 24 h of reperfusion, (3) MCAO treated by C + P without temperature control and (4) MCAO treated by C + P with temperature control. Human neuroblastoma (SH-SY5Y) cells were used in 5 groups: (1) control, (2) oxygen-glucose deprivation (OGD) for 2 h followed by reoxygenation (OGD/R), (3) OGD/R with C + P; (4) OGD/R with PERK inhibitor, GSK2656157, and (5) OGD/R with C + P and GSK2656157. The molecules of ER stress, unfolded protein response (UPR) (Bip, PERK, p-PERK, p-PERK/PERK, eIF2α, p-eIF2α, p-eIF2α/eIF2α), autophagy (ATG12, LC3II/I), and apoptosis (BAX, Bcl-XL) were measured at mRNA levels by real time PCR and protein levels by Western blotting.

Results: In ischemic rats followed by reperfusion, expression of Bip, p-PERK/PERK, p-eIF2α/eIF2α, ATG12, and LC3II/I, as well as BAX were all significantly increased. These markers were significantly reduced by C + P at both 6 and 24 h of reperfusion. Anti-apoptotic Bcl-XL expression was increased, while pro-apoptotic BAX expression was decreased by C + P. In SH-SY5Y cell lines, both C + P and GSK2656157 significantly reduced the level of autophagy and apoptosis after I/R, respectively. The combination of GSK2656157 and C + P did not promote the same effect, suggesting that C + P did not induce any neuroprotective effect by inhibiting autophagy and apoptosis through the PERK-eIF2α pathway when this pathway was already blocked by GSK2656157. In general, the reduction in body temperature by phenothiazines was associated with better neuroprotection but it did not reach significant levels.

Conclusion: The combined treatment of C + P plays a crucial role in stroke therapy by inhibiting ER stress-mediated autophagy, thereby leading to reduced apoptosis and increased neuroprotection. Our findings highlight the PERK-eIF2α pathway as a central mechanism through which C + P exerts its beneficial effects. The results from this study may pave the way for the development of more targeted and effective treatments for stroke patients.