Immunofluorescence procedures, confocal image acquisition and analysis of CA1 neurons
Slices were incubated as described previously (Petrache et al. 2019), using GABAAR α5 primary antibody (abcam, raised in mouse, 1:100) incubated concomitantly with the primary antibody targeting one of the following: calretinin (Swant, raised in goat, 1:1000), somatostatin (Santa Cruz Biotechnology, raised in rabbit, 1:500), cholecystokinin (Frontier Institute, raised in rabbit, 1:1000) or CaMKII-α (Invitrogen, raised in goat, 1:100). The secondary antibodies used were as follows: FITC (Sigma-Aldrich, anti-mouse, 1:200), Texas Red (Invitrogen, anti-rabbit/anti-mouse, 1:500) or Alexa Fluor 488 (Abcam, anti-goat, 1:500). The sections were counterstained with the nuclear stain, DAPI (Sigma-Aldrich, 1:1000).
Images were acquired at 63× magnification using a ZEISS LSM 880 confocal microscope and processed using Zen Black 2009. Collapsed Z-stacks were imported into Fiji (Image J) as .tif files and split into individual channels. If needed, the background was removed using theBackground subtraction function in Image J. In the channel corresponding to the cell staining, the outline of the cells of interest was drawn manually to obtain regions of interest (ROIs). TheColoc2 plugin was then used to obtain Pearson’s R coefficient as a measure of colocalisation between the channels corresponding to the ROIs and to the α5 subunit, and Fisher’s transformation was applied to convert the coefficients to a normal distribution. The results so obtained were then averaged separately for wild-type andAppNL-F/NL-F animals, respectively, for each of the cells of interest. There were no age differences observed during the analysis, so the data were grouped without any age-dependent segregation, with ages from 2.5 months to 15 months.