Applications of High Throughput Zebrafish Screening

Zebrafish screening offers an extraordinary potential for in vivo small-molecule phenotypic drug screening and is an outstanding model for human disease. Each breeding couple can produce hundreds of embryos. Because of their rapid development and transparency, they are suitable for microscopy-based screening methods that are often limited to cell culture. Thousands of animals can be photographed in a matter of days, unlike other in vivo screening technologies, enabling a relatively high throughput screen with the benefits of screening in intact living animals.

The swift repurposing of discovered drugs into clinical trials has shown the value of such in vivo screens. This is demonstrated by dmPGE2, which was found to boost hematopoietic stem cells in a zebrafish screen using in situ hybridization and entered clinical trials as a therapeutic for patients undergoing umbilical cord transplantation. Additionally, ORC-13661, found in hair cells in zebrafish embryos during zebrafish screenings, is undergoing clinical trials as a treatment for hearing loss brought on by aminoglycoside antibiotic-induced hair loss.

In-Vivo Zebrafish Screening

Zebrafish have also emerged as a valuable patient-derived xenograft model for drug screening for novel chemotherapeutic compounds or to identify patient-specific responses, having several advantages over mouse models, including far more rapid time scales, cost-effectiveness, and the ability to use large numbers. Zebrafish xenografts have been successfully utilized to uncover or validate chemotherapeutic drugs, such as the identification of regorafenib against adenoid cystic carcinoma and the validation of BPIQ against lung cancer cells.

Although there have been several cases where zebrafish screenings have been used as an effective high-throughput screening platform for xenograft models, there have only been a few instances of these being used on a large scale. Given the variety and complexity of different models and the type of read-outs for outcomes, such as tumor size or cell count, a platform with greater flexibility and customization would substantially enhance the use of xenograft models for drug screening.

Using Microscopy as a Tool for Studying Zebrafish is Challenging.

Because of their optical transparency and genetic tractability, zebrafish (Danio rerio) provides an appealing model organism for investigating human disease pathology. Due to their lower cost, higher throughput, and fewer ethical concerns than mammalian screening, they make a fantastic option. Fluorescently labeled tissues can be directly visualized in transgenic zebrafish embryos using a range of imaging modalities. Moreover, valuable information can be extracted using simply transmitted light microscopy for label-free studies to measure fish morphology, like curvature, and internal organs can directly be visualized. The size of the fish benefits microscopy as they can fit within multi-well plates to potentially facilitate high experimental throughput.

However, image acquisition and analysis workflows are often challenging to automate when using microscopy to study Zebrafish. While obtaining images of zebrafish is frequently straightforward, image analysis presents itself as a significant bottleneck when in large volume zebrafish screening or when looking to automate analysis.

The first difficulty is that these fish are living, three-dimensional objects. So, even when anesthetized, they can roll and drift about, making it unlikely they will maintain a perfect positioning within the wells. In a large-scale screen, it is impractical to manually go through the images to exclude those improperly positioned fish, so any automated analysis must be robust to deviations in fish positioning.

Another critical image analysis challenge is the ability to examine one specific anatomical region within the zebrafish embryo while ignoring others. For example, when looking at cells in the tail, standard fluorescence analysis using thresholding picks up bright spots throughout the entire fish, obscuring the biology of interest by detecting both cells located outside the anatomy of interest and undesired autofluorescence. The yolk sac is particularly problematic for autofluorescence, especially in the commonly used GFP channel. It is commonplace to either ignore the visible anatomy and use simple fluorescence analysis methods, such as intensity thresholding, or manually select the region of interest. The former reduces the information extracted from each embryo, disregarding the anatomical context of the signal. At the same time, the latter makes automated workflows impossible and limits achievable throughput and the size of drug libraries that can be studied.

Automated and Unbiased Zebrafish Screening–Now Possible

How can microscopic analysis of zebrafish embryos and larvae be easier and more efficient? There are many steps and bottlenecks in this experimental workflow, and we at IDEA Bio-Medical focus on two key areas: image analysis and automated image acquisition.

Microscopy permits direct visualization of the biology of interest. Zebrafish anatomy is often clearly visible to human observers but not readily detected by computers, given the contrast of brightfield images. To increase throughput and robustness and permit unbiased image analysis for zebrafish screening, we have developed a novel AI-based algorithm to detect the fish contour and internal anatomy in brightfield images. By identifying these hard-to-detect structures, our software maintains the anatomical context of associated fluorescence signals to enable TRUE high-content imaging in zebrafish. This way, region-specific spot/cell counting, fluorescence measurement, and zebrafish morphology are readily quantified.

It is an affordable, user-friendly platform for reliable, automated zebrafish image-based analysis. The software is suited for researchers who only occasionally image a handful of fish and for researchers imaging hundreds and thousands of fish in multi-well plates for large-scale screens. The software is also available as a stand-alone product and accepts microscopy images in multiple image formats, including proprietary ones (e.g. images from Zeiss, Leica, and Nikon microscopes). So, all researchers using manual microscopes or automated systems from other vendors can readily use our software to extract quantitative, meaningful information from their images.

Idea Bio will soon release this software as a web-based platform, allowing zebrafish researchers worldwide to improve their image analysis workflows and enable TRUE high-content zebrafish screening.

Consistent image acquisition is equally crucial for reproducible results, which we achieve with our Hermes high-content imaging platform. It holds a key advantage for imaging zebrafish, as it was the first microscope in the market to offer a mobile objective, leaving the sample stationary during imaging. This gentle sample scanning approach does not disturb zebrafish orientation, while the microscope obtains multi-colour images, including in Z-stack.

References and further reading:

Applications of High Throughput Zebrafish Screening