Fig. 1: Long-term behavioural consequences of the PA insult in males.

A Experimental design and PA insult. Seven-day-old male Wistar rat pups were subjected to 15 min of exposure to an asphyxia-inducing gas mixture (4% O₂, 20% CO₂) under normothermic conditions. In Experiment 1, different cohorts of animals underwent comprehensive behavioural testing from their juvenile period through adulthood, followed by immunohistochemical or differential gene expression analysis of the prefrontal cortex, likely to be involved in the observed deficits. In Experiment 2, animals were extensively characterised in the acute post-PA period (see Fig. 3). B Anxiety testing in the elevated plus-maze performed in young adulthood revealed a significant decrease in time spent in the open arms in the PA group (U = 9, p = 0.022). C PA animals showed slightly decreased social sniffing time in the social interaction test (U = 11, p = 0.031). D, E In the delay discounting task, PA animals exhibited marked motor impulsivity, indicated by the progressive increase in inadequate responses with the lengthening delay (the slope of the number of inadequate responses: t = 2.078, p = 0.055; insult × delay interaction F(7, 105) = 1.870, p = 0.081). F–H PA animals presented learning and attention deficits in the 5-choice serial reaction time task (acquired stage: test day F(2.015, 22.17) = 72.73, p < 0.001; treatment F(1,11) = 9.390, p = 0.010; test day × treatment interaction F(29,319) = 2.358, p < 0.001. PA animals were significantly slower in acquiring the task and were less able to proceed to more difficult stages (stage F (1.801,19.82) = 14,34, p < 0.001; treatment F(1,11) = 7.421, p = 0.019; stage × treatment interaction F(5,55) = 2.716, p = 0.029). This was the most prominent in stages 1 and 6. I–K In the Go/no-Go task, although PA animals were able to learn the Go cue (J), they showed marked inhibitory deficits when presented with the no-Go stimulus (K; trial day F(2.914,37.88) = 20.67, p < 0.001; treatment F(1,13) = 6.535, p = 0.023; trial day × treatment interaction F(12,156) = 2.028, p = 0.025). No significant differences in gross neuro-motor functioning were apparent between treatment groups during juvenile (L) and long-term assessment (M, N). Animals are depicted as individual data points. Error bars represent mean ± SEM; *p < 0.05; **p < 0.01; *** p < 0.001.