Hamza Anwer, Dominic Mason, Susanne Zajitschek, Daniel Hesselson, Daniel W.A. Noble, Margaret Morris, Malgorzata Lagisz, Shinichi Nakagawa

The obesity epidemic is among the most serious and rapidly growing public health challenges of the 21st century. This rapid increase is concerning as obesity appears to negatively impact cognition and behaviour. Furthermore, some studies suggest that this negative effect could be carried across generations from both mothers and fathers although evidence is not consistent. Here, we attempt to address how obesogenic diets in the parental generation (F0) can impact offspring’s cognition and anxiety intergenerationally (F1) in a zebrafish model. We compare both mean trait values and their variances. Using a multifactorial design, we created a total of four groups: F1T (treatment mothers x treatment fathers); F1M (treatment mothers x control fathers); F1P (treatment fathers x control mothers); and F1C (control mothers x control fathers, F1C); and subjected them to anxiety tank tests and aversive learning assays. When both parents were exposed, offspring (F1T) displayed the poorest aversive learning, while offspring that only had one parent exposed (F1P and F1M) learnt the aversive learning task the best. Zebrafish in all groups displayed no statistically significant differences in anxiety-associated behaviours. Males and females also performed similarly in both anxiety and aversive learning assays. While all F1 groups had similar levels of fasting blood glucose, variance in glucose levels were reduced in F1P and F1T indicating the importance of investigating heteroskedasticity between groups. Furthermore, anxiety behaviours of these two groups appeared to be less repeatable. To our knowledge, this is the first study to test the intergenerational effects of an obesogenic diet on zebrafish cognition. Our multifactorial design as well as repeated tests also allowed us to disentangle maternal and paternal effects (as well as combined effects) and accurately detect subtle information such as between-individual variation. However, our work is restricted to aversive learning. It would be interesting to see whether appetitive and other types of learning produce similar results in zebrafish offspring.