Deep learning image recognition enables efficient genome editing in zebrafish by automated injections

Bringing Artificial Intelligence (AI) into Environmental Toxicology Studies: A Perspective of AI-Enabled Zebrafish High-Throughput Screening

The booming development of artificial intelligence (AI) has brought excitement to many research fields that could benefit from its big data analysis capability for causative relationship establishment and knowledge generation. In toxicology studies using zebrafish, the microscopic images and videos that illustrate the developmental stages, phenotypic morphologies, and animal behaviors possess great potential to facilitate rapid hazard assessment and dissection of the toxicity mechanism of environmental pollutants. However, the traditional manual observation approach is both labor-intensive and time-consuming. In this Perspective, we aim to summarize the current AI-enabled image and video analysis tools to realize the full potential of AI. For image analysis, AI-based tools allow fast and objective determination of morphological features and extraction of quantitative information from images of various sorts. The advantages of providing accurate and reproducible results while avoiding human intervention play a critical role in speeding up the screening process. For video analysis, AI-based tools enable the tracking of dynamic changes in both microscopic cellular events and macroscopic animal behaviors. The subtle changes revealed by video analysis could serve as sensitive indicators of adverse outcomes. With AI-based toxicity analysis in its infancy, exciting developments and applications are expected to appear in the years to come.

Comparing robotic and manual injection methods in zebrafish embryos for high-throughput RNA silencing using CRISPR-RfxCas13d

In this study, the authors compared the efficiency of automated robotic and manual injection methods for the CRISPR-RfxCas13d (CasRx) system for mRNA knockdown and Cas9-mediated DNA targeting in zebrafish embryos. They targeted the no tail (TBXTA) gene as a proof-of-principle, evaluating the induced embryonic phenotypes. Both Cas9 and CasRx systems caused loss of function phenotypes for TBXTA. Cas9 protein exhibited a higher percentage of severe phenotypes compared with mRNA, while CasRx protein and mRNA showed similar efficiency. Both robotic and manual injections demonstrated comparable phenotype percentages and mortality rates. The findings highlight the potential of RNA-targeting CRISPR effectors for precise gene knockdown and endorse automated microinjection at a speed of 1.0 s per embryo as a high-throughput alternative to manual methods.

Deep learning in CRISPR-Cas systems: a review of recent studies

In genetic engineering, the revolutionary CRISPR-Cas system has proven to be a vital tool for precise genome editing. Simultaneously, the emergence and rapid evolution of deep learning methodologies has provided an impetus to the scientific exploration of genomic data. These concurrent advancements mandate regular investigation of the state-of-the-art, particularly given the pace of recent developments. This review focuses on the significant progress achieved during 2019-2023 in the utilization of deep learning for predicting guide RNA (gRNA) activity in the CRISPR-Cas system, a key element determining the effectiveness and specificity of genome editing procedures. In this paper, an analytical overview of contemporary research is provided, with emphasis placed on the amalgamation of artificial intelligence and genetic engineering. The importance of our review is underscored by the necessity to comprehend the rapidly evolving deep learning methodologies and their potential impact on the effectiveness of the CRISPR-Cas system. By analyzing recent literature, this review highlights the achievements and emerging trends in the integration of deep learning with the CRISPR-Cas systems, thus contributing to the future direction of this essential interdisciplinary research area.

Technologies bringing young Zebrafish from a niche field to the limelight

Fundamental life science and pharmaceutical research are continually striving to provide physiologically relevant context for their biological studies. Zebrafish present an opportunity for high-content screening (HCS) to bring a true in vivo model system to screening studies. Zebrafish embryos and young larvae are an economical, human-relevant model organism that are amenable to both genetic engineering and modification, and direct inspection via microscopy. The use of these organisms entails unique challenges that new technologies are overcoming, including artificial intelligence (AI). In this perspective article, we describe the state-of-the-art in terms of automated sample handling, imaging, and data analysis with zebrafish during early developmental stages. We highlight advances in orienting the embryos, including the use of robots, microfluidics, and creative multi-well plate solutions. Analyzing the micrographs in a fast, reliable fashion that maintains the anatomical context of the fluorescently labeled cells is a crucial step. Existing software solutions range from AI-driven commercial solutions to bespoke analysis algorithms. Deep learning appears to be a critical tool that researchers are only beginning to apply, but already facilitates many automated steps in the experimental workflow. Currently, such work has permitted the cellular quantification of multiple cell types in vivo, including stem cell responses to stress and drugs, neuronal myelination and macrophage behavior during inflammation and infection. We evaluate pro and cons of proprietary versus open-source methodologies for combining technologies into fully automated workflows of zebrafish studies. Zebrafish are poised to charge into HCS with ever-greater presence, bringing a new level of physiological context.

Knock-in of a Large Reporter Gene via the High-Throughput Microinjection of the CRISPR/Cas9 System

The non-viral delivery of the prokaryotic clustered regularly interspaced short palindromic repeats (CRISPR)-associated protein 9 (Cas9) nuclease system provides promising solutions for gene therapy. However, traditional chemical and physical delivery approaches for gene knock-in are confronted by significant challenges to overcome the draw-backs of low efficiency and high toxicity. An alternative method for directly delivering CRISPR components into single cells is microinjection. Here, we present the high-throughput robotic microinjection of CRISPR machinery plasmids to produce gene insertions. We demonstrate that the microinjection of CRISPR/Cas9 with an enhanced green fluorescent protein (eGFP) donor template into single HepG2 cells can achieve re-porter gene knock-in targeting the adeno-associated virus site 1 locus. Homology-directed repair-mediated knock-in can be ob-served with an efficiency of 41%. Assessment via T7E1 assay indicates that the eGFP knock-in cells exhibit no detectable changes at potential off-target sites. A case study of injecting the eGFP knock-in cells into zebrafish (Danio rerio) embryos to form an in vivo tumor model is conducted. Results demonstrate the efficiency of combining microinjection with the CRISPR/Cas9 system in achieving gene knock-in.

Zebrafish Embryos and Larvae as Alternative Animal Models for Toxicity Testing

Prerequisite to any biological laboratory assay employing living animals is consideration about its necessity, feasibility, ethics and the potential harm caused during an experiment. The imperative of these thoughts has led to the formulation of the 3R-principle, which today is a pivotal scientific standard of animal experimentation worldwide. The rising amount of laboratory investigations utilizing living animals throughout the last decades, either for regulatory concerns or for basic science, demands the development of alternative methods in accordance with 3R to help reduce experiments in mammals. This demand has resulted in investigation of additional vertebrate species displaying favourable biological properties. One prominent species among these is the zebrafish (Danio rerio), as these small laboratory ray-finned fish are well established in science today and feature outstanding biological characteristics. In this review, we highlight the advantages and general prerequisites of zebrafish embryos and larvae before free-feeding stages for toxicological testing, with a particular focus on cardio-, neuro, hepato- and nephrotoxicity. Furthermore, we discuss toxicokinetics, current advances in utilizing zebrafish for organ toxicity testing and highlight how advanced laboratory methods (such as automation, advanced imaging and genetic techniques) can refine future toxicological studies in this species.

Pharmacokinetics in Zebrafish Embryos (ZFE) Following Immersion and Intrayolk Administration: A Fluorescence-Based Analysis

Zebrafish embryos (ZFE) have increasingly gained in popularity as a model to perform safety screenings of compounds. Although immersion of ZFE is the main route of exposure used, evidence shows that not all small molecules are equally absorbed, possibly resulting in false-negative readouts and incorrect conclusions. In this study, we compared the pharmacokinetics of seven fluorescent compounds with known physicochemical properties that were administered to two-cell stage embryos by immersion or by IY microinjection. Absorption and distribution of the dyes were followed at various timepoints up to 120 hpf by spatiotemporal fluorescence imaging. The concentration (10 µM) and dose (2 mg/kg) used were selected as quantities typically applied in preclinical experiments and zebrafish studies. The data show that in the case of a lipophilic compound (log D: 1.73) the immersion procedure resulted in an intrabody exposure which is similar or higher than that seen after the IY microinjection. In contrast, zero to low intrabody exposure was reached after immersion of the embryos with less lipophilic compounds. In the latter case IY microinjection, a technical procedure that can be easily automated, is highly recommended.

Deploying Big Data to Crack the Genotype to Phenotype Code

Mechanistically connecting genotypes to phenotypes is a longstanding and central mission of biology. Deciphering these connections will unite questions and datasets across all scales from molecules to ecosystems. Although high-throughput sequencing has provided a rich platform on which to launch this effort, tools for deciphering mechanisms further along the genome to phenome pipeline remain limited. Machine learning approaches and other emerging computational tools hold the promise of augmenting human efforts to overcome these obstacles. This vision paper is the result of a Reintegrating Biology Workshop, bringing together the perspectives of integrative and comparative biologists to survey challenges and opportunities in cracking the genotype to phenotype code and thereby generating predictive frameworks across biological scales. Key recommendations include promoting the development of minimum "best practices" for the experimental design and collection of data; fostering sustained and long-term data repositories; promoting programs that recruit, train, and retain a diversity of talent; and providing funding to effectively support these highly cross-disciplinary efforts. We follow this discussion by highlighting a few specific transformative research opportunities that will be advanced by these efforts.

Spatiotemporal imaging and pharmacokinetics of fluorescent compounds in zebrafish eleuthero-embryos after different routes of administration

Zebrafish (Danio rerio) is increasingly used to assess the pharmacological activity and toxicity of compounds. The spatiotemporal distribution of seven fluorescent alkyne compounds was examined during 48 h after immersion (10 µM) or microinjection (2 mg/kg) in the pericardial cavity (PC), intraperitoneally (IP) and yolk sac (IY) of 3 dpf zebrafish eleuthero-embryos. By modelling the fluorescence of whole-body contours present in fluorescence images, the main pharmacokinetic (PK) parameter values of the compounds were determined. It was demonstrated that especially in case of short incubations (1-3 h) immersion can result in limited intrabody exposure to compounds. In this case, PC and IP microinjections represent excellent alternatives. Significantly, IY microinjections did not result in a suitable intrabody distribution of the compounds. Performing a QSPkR (quantitative structure-pharmacokinetic relationship) analysis, LogD was identified as the only molecular descriptor that explains the final uptake of the selected compounds. It was also shown that combined administration of compounds (immersion and microinjection) provides a more stable intrabody exposure, at least in case of a prolonged immersion and compounds with LogD value > 1. These results will help reduce the risk of false negative results and can offer an invaluable input for future translational research and safety assessment applications.

Modeling Neuronal Diseases in Zebrafish in the Era of CRISPR

BACKGROUND

Danio rerio is a powerful experimental model for studies in genetics and development. Recently, CRISPR technology has been applied in this species to mimic various human diseases, including those affecting the nervous system. Zebrafish offer multiple experimental advantages: external embryogenesis, rapid development, transparent embryos, short life cycle, and basic neurobiological processes shared with humans. This animal model, together with the CRISPR system, emerging imaging technologies, and novel behavioral approaches, lay the basis for a prominent future in neuropathology and will undoubtedly accelerate our understanding of brain function and its disorders.

OBJECTIVE

Gather relevant findings from studies that have used CRISPR technologies in zebrafish to explore basic neuronal function and model human diseases.

METHODS

We systematically reviewed the most recent literature about CRISPR technology applications for understanding brain function and neurological disorders in D. rerio. We highlighted the key role of CRISPR in driving forward our understanding of particular topics in neuroscience.

RESULTS

We show specific advances in neurobiology when the CRISPR system has been applied in zebrafish and describe how CRISPR is accelerating our understanding of brain organization.

CONCLUSION

Today, CRISPR is the preferred method to modify genomes of practically any living organism. Despite the rapid development of CRISPR technologies to generate disease models in zebrafish, more efforts are needed to efficiently combine different disciplines to find the etiology and treatments for many brain diseases.

ZF-AutoML: An Easy Machine-Learning-Based Method to Detect Anomalies in Fluorescent-Labelled Zebrafish

Background: Zebrafish are efficient animal models for conducting whole organism drug testing and toxicological evaluation of chemicals. They are frequently used for high-throughput screening owing to their high fecundity. Peripheral experimental equipment and analytical software are required for zebrafish screening, which need to be further developed. Machine learning has emerged as a powerful tool for large-scale image analysis and has been applied in zebrafish research as well. However, its use by individual researchers is restricted due to the cost and the procedure of machine learning for specific research purposes. Methods: We developed a simple and easy method for zebrafish image analysis, particularly fluorescent labelled ones, using the free machine learning program Google AutoML. We performed machine learning using vascular- and macrophage-Enhanced Green Fluorescent Protein (EGFP) fishes under normal and abnormal conditions (treated with anti-angiogenesis drugs or by wounding the caudal fin). Then, we tested the system using a new set of zebrafish images. Results: While machine learning can detect abnormalities in the fish in both strains with more than 95% accuracy, the learning procedure needs image pre-processing for the images of the macrophage-EGFP fishes. In addition, we developed a batch uploading software, ZF-ImageR, for Windows (.exe) and MacOS (.app) to enable high-throughput analysis using AutoML. Conclusions: We established a protocol to utilize conventional machine learning platforms for analyzing zebrafish phenotypes, which enables fluorescence-based, phenotype-driven zebrafish screening.

Mechanistic and Quantitative Understanding of Pharmacokinetics in Zebrafish Larvae through Nanoscale Blood Sampling and Metabolite Modeling of Paracetamol

Zebrafish larvae are increasingly used for pharmacological research, but internal drug exposure is often not measured. Understanding pharmacokinetics is necessary for reliable translation of pharmacological results to higher vertebrates, including humans. Quantification of drug clearance and distribution requires measurements of blood concentrations. Additionally, measuring drug metabolites is of importance to understand clearance in this model organism mechanistically. We therefore mechanistically studied and quantified pharmacokinetics in zebrafish larvae, and compared this to higher vertebrates, using paracetamol (acetaminophen) as a paradigm compound. A method was developed to sample blood from zebrafish larvae 5 days post fertilization. Blood concentrations of paracetamol and its major metabolites, paracetamol-glucuronide and paracetamol-sulfate, were measured. Blood concentration data were combined with measured amounts in larval homogenates and excreted amounts and simultaneously analyzed through nonlinear mixed-effects modeling, quantifying absolute clearance and distribution volume. Blood sampling from zebrafish larvae was most successful from the posterior cardinal vein, with a median volume (interquartile range) of 1.12 nl (0.676-1.66 nl) per blood sample. Samples were pooled (n = 15-35) to reach measurable levels. Paracetamol blood concentrations at steady state were only 10% of the external paracetamol concentration. Paracetamol-sulfate was the major metabolite, and its formation was quantified using a time-dependent metabolic formation rate. Absolute clearance and distribution volume correlated well with reported values in higher vertebrates, including humans. Based on blood concentrations and advanced data analysis, the mechanistic and quantitative understanding of paracetamol pharmacokinetics in zebrafish larvae has been established. This will improve the translational value of this vertebrate model organism in drug discovery and development. SIGNIFICANCE STATEMENT: In early phases of drug development, new compounds are increasingly screened in zebrafish larvae, but the internal drug exposure is often not taken into consideration. We developed innovative experimental and computational methods, including a blood-sampling technique, to measure the paradigm drug paracetamol (acetaminophen) and its major metabolites and quantify pharmacokinetics (absorption, distribution, elimination) in zebrafish larvae of 5 days post fertilization with a total volume of only 300 nl. These parameter values were scaled to higher vertebrates, including humans.

A Smart Imaging Workflow for Organ-Specific Screening in a Cystic Kidney Zebrafish Disease Model

The zebrafish is being increasingly used in biomedical research and drug discovery to conduct large-scale compound screening. However, there is a lack of accessible methodologies to enable automated imaging and scoring of tissue-specific phenotypes at enhanced resolution. Here, we present the development of an automated imaging pipeline to identify chemical modifiers of glomerular cyst formation in a zebrafish model for human cystic kidney disease. Morpholino-mediated knockdown of intraflagellar transport protein Ift172 in Tg(wt1b:EGFP) embryos was used to induce large glomerular cysts representing a robustly scorable phenotypic readout. Compound-treated embryos were consistently aligned within the cavities of agarose-filled microplates. By interfacing feature detection algorithms with automated microscopy, a smart imaging workflow for detection, centring and zooming in on regions of interests was established, which enabled the automated capturing of standardised higher resolution datasets of pronephric areas. High-content screening datasets were processed and analysed using custom-developed heuristic algorithms implemented in common open-source image analysis software. The workflow enables highly efficient profiling of entire compound libraries and scoring of kidney-specific morphological phenotypes in thousands of zebrafish embryos. The demonstrated toolset covers all the aspects of a complex whole organism screening assay and can be adapted to other organs, specimens or applications.