bioRxiv2020
Study questionAre there differences in operant learning and memory between mice born through intracytoplasmic sperm injection (ICSI) and naturally-conceived control (CTL) mice? Summary answerICSI females exhibited deficits in acquisition learning relative to CTL females, whereas ICSI males exhibited deficiency in discrimination learning and memory relative to CTL males during initial assessments. ICSI and CTL groups exhibited equally poor long-term retention of learned discrimination and memory performances at old age. What is known alreadySome human outcome studies have suggested that ICSI might be associated with an increased risk of certain cognitive disorders, but only one of two behavioral studies with ICSI mouse models have reported differences between ICSI and CTL females. No studies to date have investigated associative learning in ICSI mice. Study design, size, duration36 ICSI mice (18 male, 18 female) and 37 CTL mice (19 male, 18 female) aged 3-6 months were compared in a series of operant learning procedures that assessed acquisition of a new behavior, discrimination learning, and memory. 16 ICSI mice (9 male, 7 female) and 17 CTL mice (10 males, 7 females) received follow-up discrimination learning and memory assessments at 12 months of age (six months after the end of initial training) to evaluate retention and reacquisition of learned performances. Participants/materials, setting, methodsMice received daily operant learning sessions in experimental chambers in which all stimulus events and the recording of responses were automated. Food rewards were delivered for responding under different conditions of reinforcement, which varied by procedure. Subjects received a successive series of sessions of nose poke acquisition training, discrimination training, and the delayed non-matching-to-position (DNTMP) memory procedure. Mixed repeated measures ANOVAs in which the between-subjects factor was group (ICSI vs. CTL) and the within-subjects factor was repeated exposures to learning procedures (i.e., sessions) were used to analyze data. Main results and the role of chanceIn comparisons between all mice (i.e., males and females combined), CTL mice exhibited superior performance relative to ICSI in response acquisition (p = 0.03), discrimination (p = 0.001), and memory (p = 0.007). Sex-specific comparisons between the groups yielded evidence of sexual dimorphism. ICSI females exhibited a deficit in acquisition learning relative to CTL females (p < 0.001) but there was not a significant difference between CTL and ICSI males. In the discrimination and memory tasks, ICSI males exhibited deficits relative to CTL males (p = 0.002 and p = 0.02, respectively) but the differences between females in these tasks were not significant. There was no difference in discrimination or memory retention/re-acquisition assessments conducted with mice at 12 months of age. ICSI males and females weighed significantly more than CTL counterparts at all points during the experiment. Limitations, reasons for cautionThe study was not blinded. All learning assessments utilized food reward; other assessments of operant, Pavlovian, and nonassociative learning are needed to fully characterize learning in ICSI mice and speculate regarding the implications for cognitive function in humans conceived via ICSI. Wider implications of the findingsStudying learning and memory processes in mouse models has the potential to shed light on ICSI outcomes at the level of cognitive function. Future research should use multiple learning paradigms, assess both males and females, and investigate the effects of variables related to the ICSI procedure. Studying cognitive function in ICSI is an interdisciplinary endeavor and requires coordination between researchers at the genetic and psychological levels of analysis. Study funding/competing interest(s)This work was supported, in part, by grants from NIH (P30GM110767, HD071736 and HD085506 to WY), the Templeton Foundation (Grant ID: 61174 to WY), and a New Scholarly Endeavor Grant from the University of Nevada, Reno Office of Research and Innovation (to ML, YW, HZ, LH, and WY). The authors declare no competing interests.