The Long Isoform of Intersectin-1 Has a Role in Learning and Memory

Abstract

Down syndrome is caused by partial or total trisomy of chromosome 21 and is characterized by intellectual disability and other disorders. Although it is difficult to determine which of the genes over-expressed on the supernumerary chromosome contribute to a specific abnormality, one approach is to study each gene in isolation. This can be accomplished either by using an over-expression model to study increased gene dosage or a gene-deficiency model to study the biological function of the gene. Here, we extend our examination of the function of the chromosome 21 gene, ITSN1. We used mice in which the long isoform of intersectin-1 was knocked out (ITSN1-LKO) to understand how a lack of the long isoform of ITSN1 affects brain function. We examined cognitive and locomotor behavior as well as long term potentiation (LTP) and the mitogen-activated protein kinase (MAPK) and 3'-kinase-C2β-AKT (AKT) cell signaling pathways. We also examined the density of dendritic spines on hippocampal pyramidal neurons. We observed that ITSN1-LKO mice had deficits in learning and long term spatial memory. They also exhibited impaired LTP, and no changes in the levels of the phosphorylated extracellular signal-regulated kinase (ERK) 1/2. The amount of phosphorylated AKT was reduced in the ITSN1-LKO hippocampus and there was a decrease in the number of apical dendritic spines in hippocampal neurons. Our data suggest that the long isoform of ITSN1 plays a part in normal learning and memory.

Keywords: cell signaling; cognition; dendritic spines; down syndrome; intersectin-1; learning; memory.

Figure 1

No locomotor function deficits were observed in ITSN1-LKO mice. An open field test showed that there were no differences between ITSN1-LKO and WT mice at 6 months of age. (A) Time spent moving (move/s). (B) Distance traveled (cm). (C) Velocity (cm/s). (D) Time spent in the margins (s). (E) Time spent in the center (s). (F) The number of rears over 30 min. Data are expressed as mean ± SEM. Unpaired two-tailed t-test, n = 6 per genotype.

Figure 2

Deletion of the ITSN1 long isoform impaired learning and long term spatial memory in an age-dependent manner. A Morris Water Maze (MWM) was used to assess spatial and long term memory in WT and ITSN1-LKO mice. (A) On day 0, there was a significant increase in the swimming velocity of 9–12 weeks old ITSN1-LKO mice compared with age-matched WT controls (p = 0.0186, n = 12 per group). (B) No differences between ITSN1-LKO and WT mice at 6 months of age were observed (n = 11 per group, unpaired two-tailed t-test). Over 6 days, both young (C) and older (D) ITSN1-LKO mice took longer to find the hidden platform compared with their WT counterparts, with the difference between WT and KO mice more evident in the older mice. In the probe trial test on day 7, there was no difference in the time spent in the NW quadrant between the genotypes at 9–12 weeks of age (E) but at 6 months of age, (F) ITSN1-LKO mice did poorly compared with the WT (p = 0.0056), implying that loss of long-term memory is age-related in mice deficient for the long isoform of ITSN1. (G) Comparison of average escape time as a % of the respective WT at 9–12 weeks and 6 months of age. Two-way ANOVA showed significant main effects of age (p < 0.0001) and genotype (p < 0.0001) and there was an interaction between age and genotype (p = 0.0156). A Sidak’s post hoc analysis indicated there was a significant difference between 9–12 weeks and 6-months-old WT mice (p < 0.0001). The older WT mice were faster at finding the platform than their younger counterparts, suggesting that learning and spatial memory had improved with age. There was no difference between young and old KO mice. (H) Comparison of average time spent in the NW quadrant as a % of the respective WT at 9–12 weeks and 6 months of age. Two-way ANOVA showed significant main effects of age (p = 0.0091) and genotype (p = 0.0178). A Sidak’s post hoc analysis indicated there was a significant difference between 9–12 weeks and 6-months-old LKO mice (p = 0.0069). The older KOs spent significantly less time in the quadrant that had previously housed the platform, suggesting that long term memory declined with age in ITSN1-LKO mice. Error bars show means ± SEM, n = 11–12 per genotype. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. ns, not significant.

Figure 3

Short term working memory was intact in ITSN1-LKO mice. For both young and old mice, the amount of time spent in the novel arm was equivalent across the genotypes (A,B), suggesting that a lack of the long isoform of ITSN1 had no adverse effect on short term memory. Error bars show means ± SEM, n = 11–12 per genotype.

Figure 4

Deletion of ITSN1-L impaired hippocampal long term potentiation (LTP). Field excitatory synaptic potentials (fEPSPs) in the hippocampal CA1 region were recorded to measure LTP and are expressed as mean percentage ± SEM change over baseline. (A) LTP was diminished in 9–12 weeks old ITSN1-LKO mice compared with WT controls (*p < 0.05, n = 5 per genotype). (B) LTP was also reduced in ITSN1-LKO mice at 6 months of age (****p < 0.0001, 2–3 slices per mouse, n = 5 per genotype). Although the initial amplitude of the potentiation resulting from tetanic stimulation was similar between genotypes, after three trains of high-frequency stimulation to induce LTP, WT mice sustained this potentiation for a greater period of time than ITSN1-LKO mice at both ages, with a more pronounced decline apparent in knockout mice at 6 months of age. A two-tailed t-test for each time point was used for statistical analysis. (C) Comparison of the average change in LTP as a % of the respective WT at 9–12 weeks and 6 months of age. Two-way ANOVA showed significant main effects of age (p < 0.0001) and genotype (p < 0.0001) and there was an interaction between age and genotype (p < 0.0001). A Sidak’s post hoc analysis indicated a significant difference between WT and ITSN1-LKO mice at both ages and the WTs at 6 months of age showed enhanced LTP compared with their 9–12 weeks old counterparts. There was no difference between 9–12 weeks and 6-months-old LKO mice. Error bars show means ± SEM, n = 5 per genotype. ns, not significant.

Figure 5

Active mitogen-activated protein kinase (MAPK) was unchanged in the ITSN1-LKO hippocampus but AKT activation was reduced. MAPK and AKT activities were measured by western blotting in mouse hippocampal homogenates. Levels of phosphorylated ERK1 and 2 were measured to determine basal MAPK cell signaling activity. Phosphorylated AKT was measured to determine basal AKT cell signaling activity. Representative Western blots are shown. In 6-months-old mice, no differences were observed for the activated forms of ERK1/2 compared with the WT (A,B). The only difference was a modest decrease in the activity of AKT in LKO mice compared with the controls (C; p = 0.0306, n = 10 per genotype, unpaired two-tailed t-test). The data are presented as mean ± SEM. *p < 0.05.

Figure 6

ITSN1-LKO hippocampal pyramidal neurons exhibited decreased dendritic spine density. As shown in (A) spine density was significantly decreased on apical dendrites of ITSN1-LKO pyramidal neurons compared with controls (n = 3, p = 0.0024, unpaired two-tail t-test). Spines on basal dendrites were unaffected. (B) Representative image of ITSN1-LKO and WT Golgi-stained apical dendrites showing spines. Images were captured with a 100× oil immersion lens. **p < 0.01.