Human Sample Analysis
Post-mortem human brain and serum samples were procured in accordance with institutional guidelines, with full ethical approval and de-identification to protect donor privacy. The primary analysis focused on prefrontal cortex tissue from participants in the Religious Orders Study and Rush Memory and Aging Project (ROSMAP), a long-term clinical-pathological study on aging and Alzheimer’s disease (AD). For a subset of these individuals, cerebellar tissue and pre-mortem serum were also available. A second independent cohort of frontal cortex samples was obtained from brain banks at Massachusetts General Hospital, Duke University, and Washington University.
All human samples were categorized based on pathological diagnosis as either having AD, mild cognitive impairment (MCI), or no cognitive impairment (NCI). Within these groups, samples were matched for age and sex to ensure comparability. The study population consisted of approximately 40% males and 60% females.
Metal Level Quantification and Brain Fractionation
To measure metal concentrations, researchers employed inductively coupled plasma mass spectrometry (ICP-MS), a highly sensitive technique optimized for ultra-trace element detection. The protocol involved meticulous sample digestion and rigorous quality control, including the use of ultra-trace grade reagents, analysis of digestion blanks, and double-blinded measurements to prevent bias. The findings of reduced cortical lithium levels in AD cases were consistently replicated using independent methods, in different laboratories, and with alternative sample preparation protocols, confirming the robustness of the results.
To investigate the distribution of metals relative to AD pathology, brain tissue was fractionated to separate amyloid plaque-enriched material from non-plaque portions. In parallel, laser-ablation ICP-MS (LA-ICP-MS) was used to create high-resolution maps of metal distribution directly on brain tissue slices. These slices were cross-referenced with adjacent slices stained for amyloid-beta (Aβ) to precisely determine metal concentrations within and around amyloid plaques. This analysis confirmed that iron, copper, and zinc were enriched in plaques, consistent with previous studies.
In Vitro and Animal Model Studies
To explore the biological effects of lithium, the study utilized multiple mouse models of AD, including the 3xTg and J20 transgenic lines, which develop AD-like pathology. Animals were administered precisely controlled doses of lithium salts in their drinking water or were fed chemically defined diets that were either standard or lithium-deficient. To assess the impact of these interventions on cognition, mice underwent a comprehensive suite of behavioral tests, including the Morris water maze for spatial learning and memory, the novel-object recognition test, and the Y-maze for working memory.
Researchers also performed in vitro assays to determine if lithium binds directly to Aβ, the main component of amyloid plaques. Both oligomeric and fibrillar forms of Aβ were incubated with lithium solutions, and advanced dialysis and ICP-MS techniques were used to quantify the amount of bound lithium.
Neuropathology, Gene, and Protein Expression Analysis
Following the experimental period, mouse brains were processed for detailed neuropathological examination. Histological and immunofluorescence techniques were used to quantify Aβ plaque burden, tau pathology, synaptic density, myelination, and the activation states of brain immune cells like microglia and astrocytes. Advanced imaging, including confocal and transmission electron microscopy, provided high-resolution views of cellular and subcellular structures.
To understand the molecular mechanisms underlying lithium’s effects, the team performed extensive transcriptomic and proteomic analyses. Single-nucleus RNA sequencing (snRNA-seq) was conducted on hippocampal tissue to profile gene expression changes in individual cells. In parallel, bulk RNA sequencing was performed on microglia purified from mouse brains to specifically investigate how lithium levels affect these crucial immune cells. The vast datasets were analyzed using sophisticated bioinformatic tools, including Ingenuity Pathway Analysis (IPA), to identify key differentially expressed genes and signaling pathways. These findings were further validated by comparing the mouse gene expression data with transcriptomic data from human AD brains.
Finally, proteomic analysis using mass spectrometry was performed on hippocampal tissue to measure changes in protein abundance, providing a comprehensive view of the molecular alterations occurring in response to varying lithium levels. All animal experiments were conducted with institutional approval, and investigators were blinded to treatment conditions to ensure unbiased data collection and analysis.
