M7248

Carotid cross sections were stained for SR (Direct Red 80, Sigma-Aldrich Chemie GmbH, Taufkirchen, Germany; 365548), ki67 (with the antibody M7248 from Dako, Agilent Technologies, Santa Clara, CA, USA; 1:10 diluted) and SMA (M0851; Dako; 1:100 diluted). These sections were scanned. Subsequent analysis was conducted with QuPath software (version 0.2.3, Queen’s University, Belfast, UK) [61 (link)]: once more, the areas of media and neointima were first designated manually before the subsequent parameters were automatically quantified for these two areas. The positive cell detection command in QuPath was used for cell detection, counting and classification. In the ki67-stained sections, the overall cell number, as well as the cell density per mm² and the number of ki67+ cells per cross-section were determined. The threshold for positive cell detection was set using the pixel intensity histogram in order to ensure a standardized classification while still taking staining variation between slides into account. The formula used was:
The relative proportion of SMA+ and SR+ areas was measured. For quantification of SMA-positive cells, it is common to analyze the stained areas [62 (link),63 (link)], as a reliable assignment to individual cells is not possible.

Paraffin-embedded tumor issues were sliced into 5-mm thick section and tissue slides were deparaffinized in Tissue-Clear (1466; Sakura) and rehydrated with sequential immerse in 99–95–70% ethanol. Frozen tissue samples embedded in the OCT Embedding Compound (SAKURA) were sectioned using a cryostat, followed by fixation with 4% PFA for 20 min. Tissue slides were stained with a rat anti-mouse endomucin (1:400; 14-5851-85; eBioscience) antibody, a mouse anti-human αSMA (1:200; M0851; clone 1A4; DAKO) antibody, a rabbit anti-mouse NG2 (1:400; AB5320; Millipore) antibody, and a rat anti-mouse Ki67 (1:100; M7248; DAKO) antibody, followed by staining with species-matched secondary antibodies as follows: an Alexa Fluor 555-labeled goat anti-rat (1:400; A21434; Invitrogen), an Alexa Fluor 488-labeled donkey anti-mouse (1:400; A21202; Invitrogen), and an Alexa Fluor 488-labeled donkey anti-rabbit (1:400; A21206, Invitrogen). Positive signals were detected using a fluorescence microscope equipped with a camera (Nikon, DS-QilMC). Images were analyzed using an Adobe Photoshop software (CS5; Adobe) program. Vascular coverage was quantified as a percentage of vessels covered by αSMA⁺ cells by calculating overlapping area of endomucin and αSMA⁺ positive signals.

Immunohistochemistry was performed with the following antibodies: BRCA1 (ab16780, Abcam) or CTSS (sc-271619, Santa Cruz Biotechnology) and mouse anti-Ki-67 (M7248, Dako). Human breast cancer tissue slides were purchased from US Biomax. For antigen retrieval, slides were placed in citric acid buffer (pH 6.0) and heated at 100 °C for 10 min. Slides were incubated overnight at 4 °C with antibodies. Quantification of images was measured with image analyzer (Image J, NIH, Bethesda, MD, USA). All statistical analyses of images were performed using GraphPad Prism software 5.0 (GraphPad Software, San Diego, CA, USA).