Semiautomatic High-Content Analysis of Complex Images from Cocultures of Vascular Smooth Muscle Cells and Macrophages: A CellProfiler Showcase

Abstract

Automatization in microscopy, cell culture, and the ease of digital imagery allow obtainment of more information from single samples and upscaling of image-based analysis to high-content approaches. Simple segmentation algorithms of biological imagery are nowadays widely spread in biomedical research, but processing of complex sample structures, for example, variable sample compositions, cell shapes, and sizes, and rare events remains a difficult task. As there is no perfect method for image segmentation and fully automatic image analysis of complex content, we aimed to succeed by identification of unique and reliable features within the sample. Through exemplary use of a coculture of vascular smooth muscle cells (VSMCs) and macrophages (MPs), we demonstrate how rare interactions within this highly variable sample type can be analyzed. Because of limitations in immunocytochemistry in our specific setup, we developed a semiautomatic approach to examine the interaction of lipid-laden MPs with VSMCs under hypoxic conditions based on nuclei morphology by high-content analysis using the open-source software CellProfiler (www.cellprofiler.org). We provide evidence that, in comparison with fully automatic analysis, a low threshold within the analysis workflow and subsequent manual control save time, while providing more objective and reliable results.