Unraveling neurotoxicity discrepancies: comparative in vitro and In vivo analysis of colistin and polymyxin B and the underlying mechanisms

Abstract

Abstract Background: Polymyxins, including colistin and polymyxin B, are a final resort against Gram-negative bacterial infections. However, its clinical application is restricted due to concerns related to neurotoxicity. Despite the similar antibacterial spectrum and mode of action shared between colistin and polymyxin B, there is still a lack of definitive evidence to support the idea that their neurotoxicity profiles are identical. Purpose: To comprehensively compare the neurotoxicity between colistin and polymyxin B both in vivo and in vitro and establish a theoretical foundation to guide the rational use of polymyxins within clinical settings. Methods: In vitro experiments simulated nerve damage by exposing N2a and RSC96 cells to colistin and polymyxin B. The evaluation of nerve injury included assessments of cell viability and apoptosis. To discern the variance in the mechanisms of nerve injury between colistin and polymyxin B, oxidative stress levels were examined, such as SOD, CAT, GSH, and MDA. In in vivo experiments, a rat nerve injury model was created through intraventricular injections of colistin and polymyxin B, respectively. The impact of these drugs on brain injury in rats, particularly within the hippocampus and medulla oblongata, was measured using HE and Nissl staining. The potential influence of polymyxins on the ferroptosis pathway was evaluated by assessing the levels of LPO and Fe2+ and the degree of mitochondrial impairment. Results: At equivalent doses, colistin demonstrated a reduced level of neurotoxicity compared to polymyxin B, both in vitro and in vivo. In vitro experiments revealed greater cell viability and a lower apoptosis rate after colistin treatment than after polymyxin B treatment. This variance in outcomes could be attributed to the comparatively lower levels of oxidative stress associated with colistin administration. In a rat model, nerve injury resulted in observable damage to both the hippocampus and the medulla oblongata. A comprehensive assessment of the extent of damage in the CA1 to CA4 regions of the hippocampus, the nucleus of the solitary tract, and the hypoglossal nucleus of the medulla oblongata underscored that the neurotoxic effects of colistin remained milder compared to those elicited by polymyxin B. Even when evaluated at equivalent multiples of clinically recommended doses, colistin exhibited lower neurotoxicity in vivo than polymyxin B. For the first time, this study demonstrated the role of ferroptosis in polymyxin B-induced nerve damage. The activation levels observed within the ferroptosis pathway due to polymyxin B exceeded those triggered by colistin. Conclusions: Colistin exhibited a marked reduction in neurotoxicity compared to polymyxin B, evident in both the equivalent and clinically recommended doses. These findings suggest that, from the perspective of neurotoxicity, colistin presents a more favorable option for clinical use.