Pushing the boundaries of high content imaging

摘要

Microscopy performance is routinely assessed based on three criteria (i) Speed, (ii) Resolution, and (iii) Sensitivity. Whilst we, the research community, have access to a spectacular array of advanced equipment, alas the perfect instrument does not exist—we cannot simultaneously have optimal performance in each criterion. For this reason, the ubiquitous triangle of compromise (Fig. 1) is drawn to demonstrate these limitations. Thankfully, the triangle is continuously shrinking with new technological improvements. The triangle of compromise of microscopy performance. There are additional important considerations, like the need keep cells healthy in live-cell imaging experiments; ensuring the phenomenon being observed is not simply a consequence of phototoxicity; or how effective a given technique is in terms of depth-penetration into a 3D sample. Great technological strides have been made in improving instrument performance, but when one criteria is improved it often comes at a cost to another. For example, super resolution microscopy 1 allows us to surpass the diffraction barrier that otherwise limits the resolution (to ∼250 nm in the lateral, xy, direction), however these techniques are generally slow and less cell-friendly relative to more conventional approaches. In the first part of this special issue 2 we highlight the continuous advancements improving performance for high-content (HC) and high-throughput (HT) imaging for a variety of applications from basic to applied and clinical research 3, in terms of [i] Reagents and assay tools 4-6, [ii] Microscopy advancements 6, 7, [iii] Automation and robotics 7 and [iv] Computational advancements 8, 9. Whilst the performance in terms of speed is key for HT imaging, the image quality in terms of both resolution and signal:noise ratio (A.K.A. sensitivity) is critical for HC imaging. The ideal HC instrument requires all three criteria to be enhanced. We want it all and we want it now! As will be clear from Part 2 of the special issue on High Throughput and High Content Imaging and Cellular Informatics, newer demands on HC imaging for live-cells, 3D samples, and whole organisms are becoming increasingly prevalent. This will provide the impetus for continued technological development to enable faster, higher-resolution, and more sensitive HC imaging of these ever more challenging samples. This second part features five articles; three reviews and two original research articles each of which describe assay, technological or computational developments that work toward improving HC imaging for varied applications. Bougen-Zhukov and colleagues (this issue, page 115) review the current experimental and computational methods that enable phenotypic profiling from the enormous datasets derived from 2D HC image based screens in which many thousands of features are extracted from images of millions of cells. The article by Zhang et al. (this issue, page 126) describes how an Ultrahigh-speed Simultaneous Framing Optical Electronic Camera (USFOEC) is able to achieve exposure times as short as 20 μs. Together with novel microfluidic design, they assess the morphological and mechanical properties of red blood cells to probe the effects of different approaches to blood banking/storage. Agrawal et al. (this issue, page 133) describe a novel DNA methylation detection system to enable screening potential epigenetic drugs. Aberrant DNA methylation is common in cancers, and this study utilized their system not only in 2D cell culture but also in living 3D cultures and xenografts, to provide better tumor models and monitor the dynamics changes in DNA methylation within them. As 3D cell culture, spheroids and organotypic models have increased in popularity amongst biologists 10, novel imaging techniques have been developed with these applications in mind. Imaging of whole animal models through their development, e.g. the worm Caenorhabditis elegans 11 and the zebrafish Danio rerio is now possible 12. HC