Gadolinium-based Contrast Agents Trigger Gadolinium Accumulation Within the Hippocampus In Vivo and Mitochondrial Alterations in Organotypic Hippocampal Tissue Cultures Ex Vivo.

Abstract

Background: In multiple sclerosis (MS), mitochondrial dysfunction contributes to disease progression and may impair cognitive function, particularly in vulnerable regions such as the hippocampus. Gadolinium-based contrast agents (GBCAs) are widely used for imaging-based diagnosis of MS; however, evidence indicates that gadolinium (Gd) can be released from low-stability agents and retained in the brain. Whether retained Gd affects neuronal mitochondria, particularly under inflammatory conditions, remains unclear. We investigated hippocampal Gd retention following in vivo GBCA exposure and assessed associated mitochondrial effects in hippocampal slice cultures.

Methods: Mice with experimental autoimmune encephalomyelitis (EAE) and healthy controls received repeated injections of linear gadopentetate-dimeglumine (gadopentetate) or macrocyclic gadobutrol (cumulative dose: 20 mmol/kg body weight). Hippocampal Gd distribution and retention kinetics were analyzed 1, 10, and 40 days post-final injection (p.f.i.) using laser ablation inductively coupled plasma-time-of-flight mass spectrometry (LA-ICP-ToF-MS). Ex vivo, organotypic hippocampal slice cultures from B6.Cg-Tg(Thy1-CFP/COX8A)S2Lich/J mice expressing fluorescently labeled neuronal mitochondria were exposed to TNF-alpha (paradigm of inflammation)±gadopentetate or gadobutrol for 48 hours (0.1, 1, 10, 50 mM). Mitochondrial dynamics were quantified using semi-automated and manual analyses of confocal microscopy images, and Gd slice content was determined by ICP-MS.

Results: In vivo, neuroinflammation increased hippocampal Gd accumulation following administration of both GBCAs in EAE mice. However, long-term retention up to 40 days p.f.i. was observed exclusively for gadopentetate, particularly in EAE mice. Ex vivo, prolonged exposure to both GBCAs at low-to-intermediate concentrations (0.1 to 10 mM) reduced mitochondrial length and area, particularly under inflammatory conditions, and altered mitochondrial motility relative to untreated controls. At these concentrations, intratissue Gd levels, quantified by ICP-MS under representative conditions, were within a biologically relevant, sub-cytotoxic range. Marked motility impairment and mitochondrial rounding occurred only at the highest gadopentetate concentration (50 mM). Interestingly, although inflammation did not alter Gd concentrations within the ex vivo slices, mitochondrial alterations were more pronounced under inflammatory conditions.

Conclusion: In the hippocampus, long-term Gd retention in vivo appears to be restricted to linear GBCAs, whereas ex vivo, exposure to both linear and macrocyclic GBCAs at biologically relevant concentrations alters neuronal mitochondrial morphology and dynamics. Inflammation further amplifies these effects, highlighting the importance of considering GBCA-associated mitochondrial damage when evaluating their potential neurotoxic risks, particularly in neuroinflammatory diseases.