Comparing CRISPR-Cas9-mediated knock-in efficiency between Atlantic Salmon (Salmo salar) and Zebrafish (Danio rerio)

Abstract

Summary CRISPR gene editing technology has revolutionized the field of molecular biology enabling precise editing in the genome of all living organisms. In any gene sequence CRISPR-Cas9 can induce precise double stranded breaks (DSB), allowing us to insert new sequences or impair existing ones in the gene. Salmon aquaculture is among the areas CRISPR can be utilized by targeting some of the environmental issues facing this industry. However, Atlantic salmon’s long generation time and limited egg availability is slowing some of this progress. Zebrafish on the other hand is known for its short generation time, year-round egg laying ability and is an established genetic model with a wide range of experimental tools. It might thus serve as a model for CRISPR technology development in salmon. To assess CRISPR-Cas9 potential to improve animal welfare in salmon and use zebrafish as a model for testing new associated technologies, three separate experiments where conducted. Initially, we attempted knockout (KO) of the thyrotropin gene (tshβb), which is associated with smoltification in Atlantic salmon. This KO was successful, albeit with fish displaying a low average rate of 7.8% (std = 3.56) mutation in the tshβb gene (std = 3.56). The low rate of KO is suspected to be caused by a non-optimal sgRNA, highlighting the need for better sgRNA predictive tools when designing them. Secondly, we attempted to create albino zebrafish and salmon through BASE editing, a new approach enabling precise single base pair (bp) substitution. However, BASE editing was not achieved in zebrafish or salmon. In addition, the lack of albino fish in the positive control groups makes it challenging to determine the cause of failure. Lastly, we performed CRISPR-Cas9-mediated homology directed repair (HDR) in zebrafish targeting the slc45a2 gene, comparing the rates and indel locations with results in Atlantic salmon. HDR rates in zebrafish (2.4-4.0%) were 2-3 times lower than what was observed in salmon. The locations of indels were however similarly dependent on template polarity, suggesting similar mechanisms in both species of repair by synthesis-dependent strand annealing (SDSA) as proposed by Straume et al., 2020. We speculate that the different temperatures are a potential factor explaining the lower HDR witnessed in zebrafish compared to salmon. We show zebrafish to be a viable model for various aspects of gene editing that have transferable value for salmon. Also, improving software prediction tools could help to overcome challenges of optimal sgRNA design. Lastly, we propose future studies on the effect of temperature on HDR in both zebrafish and salmon by rearing microinjected eggs at varying temperatures. This might shed light on the observed differences in HDR rates.