Processing of brain tissue and calculation of brain region volumes
We quantified the volumes of brain regions based on 3D reconstructions of serial-sectioned tissue as published in Lázaro et al. (2018b). Briefly, brains were cut sagittally in half and the two hemispheres were weighed to the nearest 0.001 g. Then we fixated them for two weeks in PBS/4% paraformaldehyde and then transferred them to PBS/0.1% sodium azide at 4 °C for long- term storage. We used the left hemispheres for all volume reconstructions. Before sectioning, we immersed the hemispheres in a series of PBS/10, 20 and 30% sucrose for cryoprotection. We cut the tissue in the coronal plane on a freezing sliding microtone (Reichert- Jung Hn-40) to obtain series of 30 µm-thick sections, of which we mounted every fifth section on microscope slides and stained them with 0.5% cresyl violet. We measured the following brain regions: olfactory bulb, neocortex, rhinal and piriform cortices, caudoputamen, amygdala, nucleus accumbens, thalamus, hypothalamus, hippocampus, dentate gyrus, CA1, CA2, CA3, subiculum and cerebellum and the total hemi-sphere (see Lázaro et al. (2018b) for details). To outline each region on the sections we use the software Neurolucida (MBF Bioscience, Williston, VT, USA) and we applied the Cavalieri principle to calculate the volume of each region based on the sum of the outlined areas multiplied by the section thickness and inter-section distance. This calculation was made automatically in Neurolucida Explorer. All data from Radolfzell were previously published in Lázaro et al. (2018b).
We accounted for shrinkage of tissue during the histological process with a correcting factor. This correcting factor was calculated for each brain as the quotient between the original hemisphere volume – determined by dividing the fresh hemisphere mass by the specific gravity of brain tissue (Stephan, 1960) – and the final volume of the hemisphere as determined by our reconstruction. The correction factor for each brain was then applied to the brain regions of that specimen. We also size corrected brain region volumes by the upper tooth row (which does not change in length over the year) obtained from X-ray images (Lázaro et al. , 2017).