Sp8 scanning confocal microscope

Agrobacterium tumefaciens (strain GV3101) carrying the different gene constructs in the pEG100 plant expression vector were cultured in LB media and resuspended in 10-mM MgCl₂ 10-mM MES pH 5.6 to an optical density (OD)₆₀₀ of 0.2. The culture was incubated for several hours with 3′,5′-Dimethoxy-4′-hydroxyacetophenone (Acetosyringone) to induce virulence and then injected into N. benthamiana leaves. After 48 h of incubation, leaves were excised and mounted for imaging the fluorescence of the expressed gene products on a Leica SP8 scanning confocal microscope using imaging parameters as described below.

Images of single MLIs were taken on a Leica SP8 scanning confocal microscope, using a ×40 oil objective (NA = 1.30). Z-stacks were collected with a 0.5 μm (mature dataset) or 1 μm (developmental dataset) step size throughout the depth of the cells including the entire dendritic and axonal arbor. Only neurons where arbors were not cut off by the vibratome were imaged and analyzed. To optimize the consistency of the morphological comparisons, we selected neurons in the cerebellar vermis to capture cells in the optimal sagittal orientation and with complete dendritic and axonal arborizations within the 100 μm slice. The Z-compensation feature was used to avoid signal saturation throughout the depth of the tissue. Images were acquired with slightly different XY pixel sizes to accommodate varying arbor sizes, and to be able to capture single MLIs within one field of view. XY pixel sizes used were maintained at ~135 nm.

smFISH images were acquired on a Leica SP8 scanning confocal microscope with integrated Lightening deconvolution (Leica), a white light laser capable of super-continuous excitation at any user-defined wavelength between 470–670 nm, and using a 40X oil objective (NA = 1.30). Z-stacks were collected with a 0.33 μm step size throughout the depth of the tissue, and following nyquist requirements for lateral sampling. Five-color smFISH images were acquired using DAPI, Alexa 488 (excitation: 496 nm; emission: 502–546 nm), Alexa 546 (excitation: 550 nm; emission: 560–580 nm), Alexa 594 (excitation: 600 nm; emission: 606–663 nm), and Alexa 647 (excitation: 653 nm; emission: 658–775 nm).

3D image stacks were collected on a Leica SP8 scanning confocal microscope, using a pinhole size of 1 Airy unit and a 63X oil immersion (NA 1.4) objective. Approximately, 35 optical sections were collected per sample, with each section 700 nm thick. Sections were spaced 345 nm apart. DAPI, ATTO 565, and ATTO 633 were excited by 405, 555, and 630 nm lasers, respectively. ATTO dye fluorescence was collected using a HyD detector on photon counting mode and a scanning speed of 200 Hz, with 16X line accumulation. DAPI fluorescence was collected using a PMT detector using 8X line averaging. Pixel intensity values are 12-bit, and x-y pixel sizes are 76 nm. We modeled each z-section like a plane of width 345 nm for analysis, but in reality the edges of the z point spread function (PSF) overlap between sections. Since the PSF resembles a Gaussian distribution, most of the light is coming from the center of that distribution. Therefore, overlap is needed between sections to ensure equivalent sampling of the entire specimen.

Samples were dissected as described for TEM, except that dissections were performed in PBS. Septa were immediately transferred to 4% paraformaldehyde (PFA) in PBS and fixed for 3 to 24 hours at 4°C. Septa were washed and stored in PBS at 4°C. For imaging, septa were mounted in a chamber of double-sided tape on glass slides with SlowFade Gold mountant (Invitrogen) and high-precision 1.5 weight coverslips (Deckglässer) sealed with nail polish. Samples were imaged on a Leica SP8 scanning confocal microscope. For each sample, 5 fields of view were spaced approximately evenly along the anterior–posterior axis of the olfactory epithelium. Within each field of view, the cell at the center of each quadrant of the field was imaged such that the z-stack included all centrioles within the dendritic knob. The lowest-quality image from each field of view was excluded from the analysis, giving 15 cells per animal (N = 6 animals, n = 90 cells total). Individual dendrites were identified by their tightly clustered centrioles. Image stacks were processed by semiautomated detection in the program Imaris x64 9.2.1 (Oxford Instruments) using the Surfaces function and separating touching objects by seed points of 0.3-μm diameter. Dot plots were generated using Statistika [38 (link)]. Data are available in S3 Data.

For analysis of the effect of LPS on phagocytosis, microglia (1 × 10⁵cells/well were plated onto coverslips coated with poly-D-lysine (5 μg/ml; Merck Millipore Ltd., UK). After 48 h, cells were preincubated with LPS (100 ng/ml; Ezo Life Sciences, UK) for 4 h then stimulated for 24 h with Aβ_1–40 and Aβ_1–42 peptides (5 μM; Invitrogen, UK), and pulse-centrifuged for 15 s in a benchtop microcentrifuge. This procedure removed large aggregates and microscopic analysis indicated that that the preparation comprised mainly oligomers. The ratio of Iba⁺ cells that engulfed Aβ was calculated by enumerating the total number of Iba⁺ cells and the number of Iba⁺ Aβ⁺ cells in 3D projection images generated from Z stacks using a Leica SP8 scanning confocal microscope and processed using LAS AF Lite software. A total of 10 fields per experiment in triplicate were analysed.