Automated quantification of axons within the nerves samples was performed using StrataQuest version 5.1.249 and TissueQuest version 4.0.1.0128 (TissueGnostics, Vienna, Austria) as described previously and validated by Gesslbauer et al. (2017) (link). Per sample three cross-sections were selected for quantification analysis. The results were calculated using a custom-made script made specifically for this staining protocol (“Fibers_v3_16bit”). Axons were identified and quantified according to the following criteria. NF signals were used as the focus channel as this identifies all axons. The ChAT-positive axons were counted when overlapping with NF signals. All single positive as well as double positive axons were counted and visualized in the nerve cross section. Manual post-analysis correction of the falsely identified axons was applied to every single sample in alle three cross-sections. The variance between different cross-sections from the same sample remained under 3%.
For axonal quantification of the cross-sections stained using anti-NF and anti-MBP, QuPath version 0.3.0 was used (Bankhead et al., 2017 (link)). NF-positive axons were detected using the cell detection module. Subsequently, object classification via a single measurement classifier was used to classify MBP-positive axons by thresholding for mean MBP intensity in a 1 μm encircling each single axon.
For axonal quantification of the cross-sections stained using anti-NF and anti-MBP, QuPath version 0.3.0 was used (Bankhead et al., 2017 (link)). NF-positive axons were detected using the cell detection module. Subsequently, object classification via a single measurement classifier was used to classify MBP-positive axons by thresholding for mean MBP intensity in a 1 μm encircling each single axon.
Free full text: Click here
