Role of Type I Interferon Signaling and Microglia in the Abnormal Long-term Potentiation and Object Place Recognition Deficits of Male Mice With a Mutation of the Tuberous Sclerosis 2 Gene

Abstract

Background: Tuberous sclerosis complex is a genetic disorder associated with high rates of intellectual disability and autism. Mice with a heterozygous null mutation of the Tsc2 gene (Tsc2^+/-) show deficits in hippocampal-dependent tasks and abnormal long-term potentiation (LTP) in the hippocampal CA1 region. Although previous studies focused on the role of neuronal deficits in the memory phenotypes of rodent models of tuberous sclerosis complex, the results presented here demonstrate a role for microglia in these deficits.

Methods: To test the possible role of microglia and type I interferon in abnormal hippocampal-dependent memory and LTP of Tsc2^+/- mice, we used field recordings in CA1 and the object place recognition (OPR) task. We used the colony stimulating factor 1 receptor inhibitor PLX5622 to deplete microglia in Tsc2^+/- mice and interferon alpha/beta receptor alpha chain null mutation (Ifnar1^-/-) to manipulate a signaling pathway known to modulate microglia function.

Results: Unexpectedly, we demonstrate that male, but not female, Tsc2^+/- mice show OPR deficits. These deficits can be rescued by depletion of microglia and by the Ifnar1^-/- mutation. In addition to rescuing OPR deficits, depletion of microglia also reversed abnormal LTP of the Tsc2^+/- mice. Altogether, our results suggest that altered IFNAR1 signaling in microglia causes the abnormal LTP and OPR deficits of male Tsc2^+/- mice.

Conclusions: Microglia and IFNAR1 signaling have a key role in the hippocampal-dependent memory deficits and abnormal hippocampal LTP of Tsc2^+/- male mice.

Keywords: Hippocampal memory; LTP; Microglia; OPR; Tsc2; Type I IFN.

Figure 1

Male but not female Tsc2^+/− mice show OPR deficits. (A) OPR outline. (B) Graph shows the percentage of time that mice spent actively exploring the objects in the novel and familiar locations. WT male (n = 14; p < .0001, t = 7.81), WT female (n = 15; p < .0001, t = 6.35), and Tsc2^+/− female (n = 8; p < .0001, t = 7.59) mice show normal OPR (they explore significantly more the object in the novel location than the object in the familiar location), but Tsc2^+/− male (n = 16; p = .25, t = 1.16) mice show OPR deficits tested 24 hours after training (show no preference for the object in the novel location). Data represent mean ± SEM as well as individual data. ∗∗∗∗p < .0001. F, familiar; OPR, object place recognition; N, novel; n.s., not significant; WT, wild-type.

Figure 2

Effects of rapamycin on OPR deficits of male Tsc2^+/− mice. (A) Outline of treatment with rapamycin or DMSO and behavior approach. (B)Tsc2^+/− male mice before rapamycin treatment (n = 13; p = .11, t = 1.64) show OPR deficits tested 24 hours after training. Tsc2^+/− mice during rapamycin treatment (n = 12; p < .0001, t = 8.90) show normal OPR. In contrast, Tsc2^+/− mice 2 months after rapamycin treatment (n = 9; p = .58, t = 0.55) again show OPR deficits. Data represent mean ± SEM as well as individual data points. ∗∗∗∗p < .0001. n.s., not significant; OPR, object place recognition; Rap, rapamycin.

Figure 3

Evidence for a critical role of microglia in OPR memory deficits of male Tsc2^+/− mice. (A) Outline of treatment with PLX, which depletes microglia, or control chow and behavior approach. (B) IBA1 immunostaining of Tsc2^+/− control mice, PLX mice, and mice 2 months after PLX. Treatment with PLX led to elimination of microglia in the whole brain (hippocampus shown as example) compared with the control group (hippocampus shown as example). Two months after PLX, the microglia had repopulated the brain (hippocampus shown as example). (C)Tsc2^+/−/PLX mice (n = 8; p < .0001, t = 6.50), but not Tsc2^+/−/control mice (n = 10; p = .73, t = 0.33), show normal OPR memory tested 24 hours after training. (D)Tsc2^+/− mice 2 months after PLX (n = 8; p < .0001, t = 9.20), but not Tsc2^+/− mice 2 months after control (n = 6; p = .86, t = 0.17), show normal OPR memory. Data represent mean ± SEM as well as individual data. Scale bar = 200 μm. ∗∗∗∗p < .0001. n.s., not significant; OPR, object place recognition; PLX, PLX5622.

Figure 4

Tsc2 mutation restricted to microglia during development (but not in adult mice) leads to OPR memory deficits. (A) Outline of behavior approach for Cx3cr1^Cre–Tsc2^Flox. (B) Male WT (n = 7; p < .05, t = 2.69) mice show normal OPR. In contrast, Cx3cr1^Cre–Tsc2^Flox (Tsc2^+/−) (n = 11; p = .76, t = 0.3) mice show deficits in OPR tested 24 hours after training. Data represent mean ± SEM as well as values for individual mice. As indicated, in (B), Tsc2^+/− represents Cx3cr1^Cre–Tsc2^Flox. ∗p < .05. F, familiar; N, novel; n.s., not significant; OPR, object place recognition; WT, wild-type.

Figure 5

Effects of microglia depletion in the CA1 LTP of male Tsc2^+/− Ep mice. (A) Outline of poly(I:C) injections, treatment with PLX, and analysis approach. (B) Initial fEPSP slopes recorded in the CA1 region of hippocampal slices are shown before (baseline, 15 minutes) and following LTP induction (with a 1-second, 100-Hz tetanus delivered at time 0) for the tetanized pathway and for a separate, untetanized pathway (control). Data are plotted in 1-minute blocks and fEPSP slopes are normalized to the average baseline response. Sample traces show responses (5 responses were averaged) during baseline (gray) and during the last 5 minutes of recordings (color). Scale bars = 1 mV and 10 ms. (C) Averaged fEPSPs during the last 20 minutes of LTP recording. Regular two-way analysis of variance, pathway × genotype interaction: F_2,54 = 10.51, p < .001; Sidak’s multiple comparisons for tetanized vs. control pathway fEPSPs, ∗∗∗∗p < .0001 for Tsc2^+/−/VEH group, not significant for other two groups; Tukey’s multiple comparisons for tetanized pathway fEPSPs, ∗∗p < .01 for Tsc2^+/−/PLX group vs. Tsc2^+/−/VEH group. (D) Averaged PTP fEPSPs (5 minutes) recorded immediately after a 100-Hz tetanus. Regular two-way analysis of variance, pathway factor: F_1,54 = 105.1, p < .001; genotype factor: F_2,54 = 1.488, p > .05; Sidak’s multiple comparisons for tetanized vs. control pathway, ∗∗∗∗p < .0001 for all 3 groups. Tsc2^+/−/VEH slices, n = 10 from 6 mice. Tsc2^+/−/PLX slices, n = 11 from 7 mice. WT/PLX slices, n = 9 slices from 5 mice. All data are shown as mean ± SEM. fEPSP, field excitatory postsynaptic potential; L-LTP, late long-term potentiation; LTP, long-term potentiation; n.s., not significant; P, postnatal day; PLX, PLX5622; Poly I:C, polyinosinic:polycytidylic acid; PTP, post-tetanus potentiation; Tsc2^+/− Ep, Tsc2^+/− with early postnatal immune activation; VEH, vehicle; WT, wild-type.

Figure 6

Male Tsc2^+/− mice show normal OPR memory 60 minutes after training. (A) Outline of behavior approach. (B) Graph shows the percentage of time the mice spent actively exploring the objects in the novel and familiar locations. Tsc2^+/− mice tested 60 minutes after training (Tsc2^+/−/60 min; n = 10; p < .0001, t = 6.97) show normal OPR (they explore the object in the novel location significantly more than the object in the familiar location), but Tsc2^+/− mice tested 24 hours after training (Tsc2^+/−/24h; n = 10; p = .73, t = 0.33) show OPR deficits (show no preference for the object in the novel location). Data represent mean ± SEM as well as individual data. ∗∗∗∗p < .0001. Cont. control; F, familiar; N, novel; n.s., not significant; OPR, object place recognition.

Figure 7

Effects of the null homozygous mutation of type I IFN receptor (Ifnar1^−/−) in Tsc2^+/− OPR deficits. (A) Behavior approach. (B)Tsc2^+/− mice (n = 9; p = .24, t = 1.19) show OPR deficits tested 24 hours after training. WT (n = 6; p < .001, t = 5.71), Ifnar1^−/− (n = 6; p < .0001, t = 6.34), and Tsc2^+/−/Ifnar1^−/− (n = 8; p < .0001, t = 6.78) mice show normal OPR. Data represent mean ± SEM as well as values for individual mice. ∗∗∗p < .001, ∗∗∗∗p < .0001. IFN, interferon; n.s., not significant; OPR, object place recognition; WT, wild-type.