2.3. Manganese‐enhanced MRI

For MEMRI, the slices were removed from the cell culture insert. The membrane insert was cut around the slice and rinsed in artificial cerebrospinal fluid (aCSF; 124 mM NaCl₂, 5 mM KCl, 1.23 mM NaH₂PO₄, 26 mM NaHCO₃, 10 mM dextrose, 1.5 mM MgCl₂ and 2.5 mM CaCl₂ in water), placed in a perfusion chamber then fixed with a plastic anchor (Figure S1). To keep the cultures alive during imaging, slices were continuously perfused with the aCSF at 0.9 μL/min, and bubbled with 95% O₂ and 5% CO₂ at 37.5 ± 0.5°C. On those rare occasions when microbubbles were observed, they were not included in the ROI measurements.

Imaging was performed on an 11.7 T/30 cm horizontal bore magnet (Magnex Scientific, Oxford, UK) with an Avance III console (Bruker Biospin, Billerica, MA, USA), using a volume transmit coil and a surface coil (diameter = 1 cm) for detection. 3D imaging was performed to determine the slice position in the MRI. 3D T_1w images were then acquired with a fast low angle shot (FLASH) sequence, with 50 μm isotropic resolution, 25^o pulse, FOV = 19.2 × 19.2 × 3.2 mm³, matrix size = 384 × 384 × 64, TR/TE = 25/4.7 ms, and a total scan time of approximately 11 minutes. One to two MRI slices from the center of the hippocampal slices were selected for ROI analysis. T₁ mapping is more quantitative, as T_1w imaging is not linearly proportional to Mn over all the concentrations used in our experiments. However, T_1w images and Mn concentrations were roughly linearly within the range in which most experiments were performed. Further, the goal of the experiments was to order the routes of Mn uptake in slices instead of precisely calculating the Mn levels, thus T_1w imaging was used instead of T1 or R1 mapping. Further, T_1w imaging reduces scan time to maintain slice viability and can increase resolution to detect Mn²⁺ signal in hippocampal subregions.