Aperio slide scanner

All IVDs were fixed with 10% neutralized formalin buffer, decalcified in 5% nitric acid for 1 to 2 months, and transferred to 75% ethanol prior to paraffin embedding. Samples were sectioned at 4 μm thickness through a mid‐coronal plane onto glass slides. After sectioning onto glass slides, samples were stained with alcian blue and imaged with brightfield microscopy with an Aperio slide scanner (Leica Biosystems, Buffalo Grove, Illinois).

CD31-positive blood vessels were stained following the manufacturer’s instructions. In brief, sections were deparaffinised and heat-mediated antigen retrieval in 10 mM citrate buffer (pH6) was performed for 10 minutes. Endogenous peroxidases were blocked with 0.3% v/v hydrogen peroxide followed by blocking of endogenous avidin/biotin (Vectorlabs, Burlingame, USA). Sections were then incubated for 1-hour with the primary antibody (1:20 rat anti-mouse CD31, clone SZ31, Dianova GmbH, Germany). After washing, a biotinylated anti-rat secondary antibody was applied for 30 minutes (1:200 Dako, UK). ABC/HRP solutions (ABC/HRP Complex Kit, Vectorlabs, Burlingame, USA) were then applied and visualised with 3,3′-Diaminobenzidine (DAB, Dako, UK). Slides were scanned using an Aperio slide scanner (Leica Biosystems, Wetzlar, Germany) at × 20 magnification. Ten 0.25 mm² boxes were placed randomly throughout the tumour section using RandomSpot software version 6.02^{38 (link)} and the number of CD31 positive vessel were counted manually. Microvessel density was calculated as the number of CD31 positive vessels per mm².

Gastric tissues were formalin-fixed and embedded in paraffin. Tissues were stained with Mayer’s Haematoxylin & Eosin or Periodic acid-Schiff (PAS)–Alcian Blue (AB) stains and imaged at ×20 magnification using a bright-field microscope (Nikon Instruments Inc., NY, USA). Inflammatory scores for neutrophils and lymphocytes were graded in a blinded fashion (J.S.P.) according to a previously described grading scheme^{27 (link)}. The mucosal thickness of the gastric corpus was determined in a blinded fashion (J.E.) on sections that had been digitised using an Aperio slide scanner (Leica Biosystems, Mount Waverley, VIC, Australia) and ImageScope software (Leica Biosystems). Three measurements were taken for each tissue section, with 3–4 sections available per stomach sample.

Tissues were fixed overnight to 24 h in 10% formalin (VWR, 48218–700) and stained with hematoxylin (Leica Biosystems, Wetzlar, Germany, 3801575) and eosin (Leica Biosystems, 3801606) using standard techniques. Slides were scanned using an Aperio slide scanner (Leica Biosystems), and images were analyzed using Aperio ImageScope.

Spleens were harvested and passed through a 70 µM strainer and lysed in ACK lysis buffer (Quality Biological, Gaithersburg, MD, USA) for 2 min prior to analysis. Heart tissue for flow cytometry analysis was cut into approximately 2 mm³ pieces in HBSS (Gibco, Gaithersburg, MD, USA), incubated for 20 min at 37°C with collagenase II (Sigma-Aldrich, St. Louis, MO, USA), DnaseI (Roche, Branford, CT, USA), and HEPES (Gibco, Gaithersburg, MD, USA) prior to passing through a 70 µM strainer. Heart tissue for histology was fixed in 10% formalin for approximately 48 h prior to transfer to 70% ethanol for storage. Tissue was embedded in paraffin and then sequential cuts were made for H&E staining, and immunohistochemistry stains for CD4 and FoxP3.
Slides were scanned at 20× magnification using Aperio Slide Scanner (Leica Biosystems, Buffalo Grove, IL, USA), and representative or whole sections were sampled and the total number or percentage CD4⁺ and FoxP3⁺ cells were manually quantified in a single blind manner.