Vs120 slide scanner

PLNs from WT males (n=2 mice, 4 PLNs) were harvested, embedded in optimal cutting temperature (OCT) compound (Cat# 4583; Sakura, Osaka, Japan), and frozen in liquid nitrogen. Six tissue sections of 10μm were placed within individual capture areas of a 10X Genomics Optimization slide and stored at -80°C. The tissue sections were then fixed in cold methanol and stained with H&E according to the Visium Methanol Fixation and H&E Staining guide (10X Genomics). An Olympus VS120 slide scanner was used for brightfield imaging of the H&E-stained sections. To evaluate the optimal permeabilization time, each capture area was permeabilized for a different time period (3, 6, 12, 18, 24, or 30 minutes) followed by reverse transcription with fluorescent nucleotides to label cDNA. Tissue sections were then removed, and the slide was imaged again using the Olympus VS120 slide scanner to evaluate fluorescence with the TRITC filter. The optimal permeabilization time was determined to be 12-minutes by balancing the brightest fluorescent signal intensity with the least amount of diffusion (Supplementary Figure 1), and the 12-minute permeabilization time was then applied to the sections of the gene expression slide.

The left lobes of the lungs were fixed in 10% neutral-buffered formalin for 24 h before storing in 70% ethanol. Tissues were processed, paraffin-embedded and sectioned vertically at a thickness of 4 µm. Coronal sections that uniformly contain bronchi and >8 airways were then stained with hematoxylin and eosin (H&E) and scanned using a VS120 Slidescanner (Olympus, Japan). Lung injury was assessed in a blinded manner on a scale of 0–3 (none to severe) in peribronchial, perivascular and interstitial/alveolar regions individually based on the degree of inflammatory cell infiltration, epithelial/endothelial destruction, and alveolar septal thickening as previously described [24 (link)] where total scores were presented. NET staining on lung sections was performed as previously described [25 (link)] using primary goat anti-mouse MPO antibody (1:40 dilution, AF3667, R&D Systems, US) and rabbit anti-mouse citrullinated histone antibody (1:100, ab5103 Abcam, UK). Sections were then labelled with Alexa Fluor 568 donkey anti-rabbit IgG (1:200) and Alexa Fluor 488 donkey anti-goat IgG (1:200) secondary antibodies and counterstained with DAPI (1:1000) (all from Thermofisher Scientific, US). Whole slides were scanned using a VS120 Slidescanner (Olympus, Japan). NETs were recognised by the yellow staining after fluorescence channel overlay.

E18.5 wild-type and MafB^−/− pancreata (n=4 in each group) were paraffin embedded and sectioned (see above). For histological analysis, sections with a distance of 70 µm were stained for Sox10, β III tubulin or mucin with insulin, glucagon and DAPI. Stained sections were scanned on an Olympus slide scanner (VS120) with a 20× objective. Pictures were analyzed using Pathology AI (Visiopharm). The application was first trained to recognize islet, pancreatic, gut and ganglionic structures from wild-type and MafB^−/− sections by manual outlining of the respective structures, the application was then allowed to train for 143,000 iterations using Deep Labv3⁺ classification. Accuracy of tissue recognition was assessed afterwards. Insulin-, glucagon-, β III tubulin-, mucin- and Sox10-stained area was then determined using a fluorescence threshold, and respective areas were then normalized against pancreatic, islet or ganglionic areas. Mucin structures were categorized into stained areas larger or smaller than 10 µm² and the size composition of mucin lumina per islet and pancreatic area was assessed. Branching points were automatically recognized and the average number of branching points per mucin structure in islet versus pancreatic area determined.