Fig. 2: Behavioral and electrophysiological alterations in response to the absence of 6’SL from maternal milk and/or due to the deletion of the St6Gal1 gene.

Reduced access to 6′SL during lactation affects recognition memory (a), spatial reference memory (b), attentional capabilities (c), perseverative behavior (d), sensorimotor gating (e), and long-term potentiation (f). a Percent preference with 95% confidence intervals (CI, represented by whiskers) for the novel object during a 10-min test session performed 60min (open circles) and 24h (filled circles) after the habituation phase. The dashed line represents chance level; a CI intersecting the dashed line indicates that preference index was not statistically different from chance. In both test sessions, only CTRL subjects showed preference for the novel object compared to the familiar one. b Time spent in the target zone of the Barnes maze during the probe trial conducted 7 days after the last training session. Spatial memory retention varied between experimental groups (offspring genotype×maternal genotype: F_1,36 = 6.120, p = 0.020, N = 10 per group). Specifically, CTRL mice spent more time in the target zone compared to MILK group (p = 0.023 in post hoc tests). c Number of trials to attain the criterion (eight correct choices out of ten consecutive trials) in different stages of the task: simple discrimination (SD); compound discrimination (CD); compound discrimination reversal (CDR); intra-dimensional shift (IDS); and extra-dimensional shift (EDS). While there is no significant difference between groups in the SD phase (offspring genotype: F_1,28 = 0.529, p = 0.473; maternal genotype: F_1,28 = 3.871, p = 0.059; offspring genotype×maternal genotype: F_1,28 = 0.529, p = 0.473; CTRL: N = 8, MILK: N = 9, GENE: N = 8, GENE + MILK: N = 7) during CD, the number of trials needed to reach the criterion varied depending on the maternal genotype (offspring genotype: F_1,28 = 0.759, p = 0.391; maternal genotype: F_1,28 = 11.627, p = 0.002; offspring × maternal genotype: F_1,28 = 0.007, p = 0.934; CTRL: N = 8, MILK: N = 9, GENE: N = 8, GENE + MILK: N = 7). Specifically, compared to CTRL and GENE, MILK mice required more trials to attain the criterion (p = 0.0218 and p = 0.0041, respectively, in post hoc tests). Additionally, compared to GENE group, GENE + MILK required higher number of trials to complete the phase (p = 0.023 in post hoc tests). During CDR phase, groups did not show significant difference (offspring genotype: F_1,28 = 0.171, p = 0.682; maternal genotype: F_1,28 = 3.212, p = 0.084; offspring × maternal genotype: F_1,28 = 0.872, p = 0.358; CTRL: N = 8, MILK: N = 9, GENE: N = 8, GENE + MILK: N = 7). In IDs phase, the numbers of trial varied depending on maternal genotype (offspring genotype: F_1,28 = 0.425, p = 0.520; maternal genotype: F_1,28 = 7.892, p = 0.009; offspring × maternal genotype: F_1,28 = 2.546, p = 0.122; CTRL: N = 8, MILK: N = 9, GENE: N = 8, GENE + MILK: N = 7). Specifically, compared to CTRL and GENE group, MILK group required more trials to attain the criterion (p = 0.003 and p = 0.017, respectively, in post hoc tests). Finally, experimental groups varied depending on both offspring and maternal genotype in number of trial needed to reach the criterion during the EDs phase (offspring genotype: F_1,28 = 8.236, p = 0.007; maternal genotype: F_1,28 = 6.461, p = 0.017; offspring × maternal genotype: F_1,28 = 1.795, p = 0.191). Specifically, MILK group required more trials to attain the criterion compared to CTRL, GENE, GENE + MILK groups (p = 0.008, p = 0.0005, p = 0.006, respectively, in post hoc tests). Data are expressed as mean ± SEM; data on errors to criterion are reported in the Supplementary Information (see Supplementary Fig. 5). d Percentage of alternations (open circles) with 95% CI (whiskers) between the two arms of the T-maze. The dashed line represents chance level. CTRL subjects exhibited an intact natural tendency to alternate between the two arms of the maze (average 77.78%; 95% CI 68.19–87.36). e Percent inhibition of the startle reflex measured as PPI = [(A − B)/A × 100]: A is the Galvanic reflex in response to the startle stimulus alone, and B is the response to prepulse plus pulse stimuli (open circles) with 95% CI (whiskers). The dashed line represents chance level; a CI intersecting the dashed line indicates that the percent inhibition was not statistically different from chance and that mice failed to inhibit the startle reflex in response to the presentation of the prepulse. Mice of CTRL and GENE groups exhibited prepulse inhibition, while mice of MILK and GENE + MILK groups failed to exhibit it. f fEPSPs recorded in the CA1 area of hippocampal slices; long-term potentiation (LTP) was induced by theta-burst stimulation (TBS) of Schaffer collaterals and varied depending on the rearing dam (maternal genotype: F_1,14 = 8.421, p = 0.012). MILK mice (N = 6 slices from three animals) showed an increased LTP compared to CTRL subjects (p = 0.04 in post hoc tests). The number of slices for each condition were: CTRL: N = 10 slices from five animals; MILK: N = 6 slices from three animals; GENE: N = 11 slices from five animals; GENE + MILK: N = 10 slices from five animals; data concerning the GENE vs. GENE + MILK comparison are reported in Supplementary Fig. 6. * indicates p value < 0.05 compared to CTRL group; + indicates p value < 0.05 compared to GENE group; $ indicates p value < 0.05 compared to GENE + MILK group; p values were calculated using analysis of variance (ANOVA). All bars represent mean ± SEM. € indicates that experimental group did not differ from chance.