Vis software

The fixed pancreas was weighed, processed for paraffin embedding, and randomly cut into four slabs for stereological assessment of alpha and beta cell mass. Alpha cells were detected using a mouse anti-glucagon antibody (Novo Nordisk), and beta cells were visualised using a guinea pig anti-insulin antibody (Dako, Denmark). The stained slides were scanned using a VS120 slide scanner (Olympus, Germany). Total tissue area, glucagon area (alpha cells) and insulin area (beta cells) were quantified in the digital images using VIS software (Visiopharm, Denmark). Alpha and beta cell mass were estimated by multiplying the fractional areas by the total pancreas mass as described previously [32 (link)].

After fixation in formalin for 48–96 h and subsequent embedding in paraffin, 3 µm sections of liver tissue from the left lateral, the right medial and the caudate lobe were cut and stained with Haematoxylin and Eosin (H&E, Sigma-Aldrich, Soeborg, Denmark) and Picro Sirius Red (PSR, Sigma-Aldrich) for evaluation of hepatic steatosis and collagen deposition, respectively. Additional immunohistochemistry (IHC) was performed for CD68 to characterize development of inflammation and for α-smooth muscle actin (α-SMA) to characterize activation of hepatic stellate cells. IHC for CD68 and α-SMA was performed according to protocols shown in Supplemental Table S2. After staining, all sections were scanned using a NanoZoomer 2.0 HT slide scanner (Hamamatsu, Hamamatsu City, Japan) and subsequently evaluated using NanoZoomer Digital Pathology Image Software (Hamamatsu). Threshold-based image analysis was performed using VIS software (Visiopharm, Hoersholm, Denmark), which quantified the fractional area for each animal with positive staining for each marker of interest (i.e., PSR, CD68, and α-SMA). These area fractions were expressed as the percentage of the total area of liver tissue per animal.

The numbers of TH- and 5HT-immunoreactive cell numbers in the striatum were estimated using an unbiased stereological quantification method by employing the optical fractionator principle [27] (link). All quantifications were done after blinding the identity of the sections by a coding system. The borders for the region of interest was defined by using a 4x objective, whereas the actual counting was performed using a 60x Plan-Apo oil objective (Numerical aperture = 1.4) on a Nikon 80i microscope equipped with an X-Y motorized stage, a Z-axis motor and a high-precision linear encoder (Heidenhein). All three axes and the input from the digital camera that were controlled by a Computer Assisted Toolbox Software (New CAST) module in VIS software (Visiopharm A/S, Denmark), which carries out the procedure with a random start and systematic sampling routine. The sampling interval in the X-Y axis was adjusted so that at least 100 cells were counted in each grafted striatum. Upon completion of the quantification of batches, samples were moved to a database for further analysis using appropriate statistical and graphical tools. Coefficient of error attributable to the sampling was calculated according to Gundersen and Jensen [28] (link) and values ≤0.10 were accepted.

Hematoxylin and eosin (H&E), Ki67 and IBA1 immunohistochemistry (IHC) stainings were performed by the molecular pathology core platform of the CRCHUM using the Shandon multiprogram robotic slide stainer on 4 μm formalin-fixed paraffin-embedded (FFPE) hepatic whole tissue sections. Images were acquired with the whole slide scanner Olympus BX61VS. Quantification of the necrotic area was performed with the VIS software (Version 2018.4, Visiopharm, Hørsholm, Denmark) using protocols that automatically detected the tissue and manual delineation of the necrotic area by visual inspection. References for antibodies and H&E reagents are provided in Tables S1, S2 in the Supplementary Material.