A sensitive period for the development of episodic-like memory in mice

Abstract

Episodic-like memory is a later-developing cognitive function supported by the hippocampus. In mice, the formation of extracellular perineuronal nets in subfield CA1 of the dorsal hippocampus controls the emergence of episodic-like memory during the fourth postnatal week (Ramsaran et al., 2023). Whether the timing of episodic-like memory onset is hard-wired, or flexibly set by early-life experiences during a critical or sensitive period for hippocampal maturation, is unknown. Here, we show that the trajectories for episodic-like memory development vary for mice given different sets of experiences spanning the second and third postnatal weeks. Specifically, episodic-like memory precision developed later in mice that experienced early-life adversity, while it developed earlier in mice that experienced early-life enrichment. Moreover, we demonstrate that early-life experiences set the timing of episodic-like memory development by modulating the pace of perineuronal net formation in dorsal CA1. These results indicate that the hippocampus undergoes a sensitive period during which early-life experiences determine the timing for episodic-like memory development.

Figure 1.. Experience-dependent critical period mechanisms may control the maturation of episodic-like memory.

(A) In primary visual cortex (V1), the formation of WFA⁺ PNNs around PV⁺ interneurons begins after eye opening and is largely completed by the fifth postnatal week, consistent with the timing of critical period closure for ocular dominance plasticity (Gordon & Stryker, 1996). (B) In the barrel fields of primary somatosensory cortex (S1BF), the formation of WFA⁺ PNNs around PV⁺ interneurons has already occurred by the second postnatal week, consistent with the timing of critical period closure for barrel field plasticity (Fox, 1992). (C) In the dorsal CA1 of the hippocampus, the formation of WFA⁺ PNNs around PV⁺ interneurons occurs rapidly during the fourth postnatal week. (D) PNN-dependent critical period closure requires specific postnatal experiences during the critical period. In mice, vision loss or dark rearing from birth impairs PNN formation in the binocular zone of primary visual cortex (V1), resulting in impaired visual acuity (Hensch, 2005; Huang et al., 1999; Pizzorusso et al., 2002; Ye & Miao, 2013). Likewise, whisker removal from birth precludes PNN formation in layer 4 of barrel cortex (Fox, 1992; McRae et al., 2007). (E) If PNN formation in the dorsal CA1 marks the end of a hippocampal critical period, deprivation or enrichment should shift the developmental trajectories for PNN formation and episodic-like memory. Scale bar, 50 μm.

Figure 2.. Early-life adversity impairs the maturation of perineuronal nets in CA1 and memory precision in juvenile mice.

(A) Mice experienced adversity in the forms of daily maternal separation from P6 to P16 followed by early weaning on P17 (ELA group), or no separation and conventional weaning (Control group). (B) Labeling of WFA⁺ PNNs (red) and PV⁺ interneurons (cyan) in dorsal CA1 of Control and ELA mice on P24. (C) P24 mice with a history of adversity had fewer WFA⁺ PNNs (unpaired t-test, t₈ = 7.17, P < 0.0001), PV⁺ interneurons (unpaired t-test, t₈ = 2.47), and PNN-enwrapped PV⁺ interneurons (unpaired t-test, t₈ = 2.99, P < 0.05), but not PV⁻ PNNs (unpaired t-test, t₈ = 0.11, P = 0.91) in dorsal CA1, compared with Control mice. (D) Labeling of WFA⁺ PNNs (red) and PV⁺ interneurons (cyan) in dorsal CA3 of Control and ELA mice on P24. (E) There was no difference in WFA⁺ PNNs (unpaired t-test, t₈ = 2.00, P = 0.081), PV⁺ interneurons (unpaired t-test, t₈ = 1.65, P = 0.13), PNN-enwrapped PV⁺ interneurons (unpaired t-test, t₈ = 1.50, P = 0.17), or PV⁻ PNNs (unpaired t-test, t₈ = 0.67, P = 0.52) in dorsal CA3 of P24 Control and ELA mice. (F) Experimental schedule for assessing memory precision. Mice were trained in a contextual fear conditioning task on P24, and 1 d later, they were tested in the same context (Context A) or a similar, novel context (Context B). (G) P24 mice with a history of adversity showed imprecise contextual fear memories, compared with Control mice that showed age-appropriate memory precision (ANOVA, experience × test context interaction: F_1,27 = 14.27, P < 0.001). Data points represent individual mice with mean ± SEM. Scale bar, 50 μm. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 3.. Early-life adversity does not block the maturation of perineuronal nets in CA1 and memory precision in adult mice.

(A) Mice experienced adversity in the forms of daily maternal separation from P6 to P16 followed by early weaning on P17 (ELA group), or no separation and conventional weaning (Control group). (B) Labeling of WFA⁺ PNNs (red) and PV⁺ interneurons (cyan) in dorsal CA1 of Control and ELA mice on P60. (C) There was no difference in WFA⁺ PNNs (unpaired t-test, t₁₁ = 0.25, P = 0.81), PV⁺ interneurons (unpaired t-test, t₁₁ = 1.88, P = 0.088), PNN-enwrapped PV⁺ interneurons (unpaired t-test, t₁₁ = 1.58, P = 0.14), or PV⁻ PNNs (unpaired t-test, t₁₁ = 1.38, P = 0.19) in dorsal CA1 of P60 Control and ELA mice. (D) Experimental schedule for assessing memory precision. Mice were trained in a contextual fear conditioning task on P60, and 1 d later, they were tested in the same context (Context A) or a similar, novel context (Context B). (E) P60 Control and ELA mice showed adult-like memory precision (ANOVA, main effect of test context: F_1,29 = 171.78, P < 0.0001). (F) Compared to Control mice, ELA mice on P60 showed increased anxiety-like behavior (unpaired t-test, t₁₃ = 2.53, P < 0.05) but similar levels of locomotion (unpaired t-test, t₁₃ = 0.72, P = 0.48) in the open field. Data points represent individual mice with mean ± SEM. Scale bar, 50 μm. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 4.. BDNF treatment restores perineuronal nets in CA1 and memory precision in juvenile mice raised in adverse conditions.

(A) Mice experienced adversity followed by infusions of Vehicle or BDNF into CA1 on P21. (B) Labeling of WFA⁺ PNNs (red) and PV⁺ interneurons (cyan) in dorsal CA1 of ELA mice treated with Vehicle or BDNF. (C) P24 mice treated with BDNF following early-life adversity had more WFA⁺ PNNs (unpaired t-test, t₉ = 8.46, P < 0.0001), PV⁺ interneurons (unpaired t-test, t₉ = 3.01, P < 0.05), and PNN-enwrapped PV⁺ interneurons (unpaired t-test, t₉ = 5.25, P < 0.001), but not more PV⁻ PNNs (unpaired t-test, t₉ = 0.97, P = 0.36), than mice treated with Vehicle. (D) Experimental schedule for assessing memory precision. Mice were trained in a contextual fear conditioning task on P24, and 1 d later, they were tested in the same context (Context A) or a similar, novel context (Context B). (E) P24 mice treated with BDNF following early-life adversity showed age-appropriate contextual fear memory precision, compared with P24 mice treated with Vehicle, that showed imprecise memories (ANOVA, treatment × test context interaction: F_1,29 = 6.55, P < 0.05). Data points represent individual mice with mean ± SEM. Scale bar, 50 μm. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 5.. Early-life enrichment induces precocial maturation of perineuronal nets in CA1 and memory precision in preweaning mice.

(A) Mice were housed in large cages containing enrichment from P6 to P19 (ELE group) or were housed conventionally (Control group). (B) Labeling of WFA⁺ PNNs (red) and PV⁺ interneurons (cyan) in dorsal CA1 of Control and ELE mice on P20. (C) P20 mice previously raised in enriched environments had more WFA⁺ PNNs (unpaired t-test, t₇ = 4.69, P < 0.01) and PNN-enwrapped PV⁺ interneurons (unpaired t-test, t₇ = 7.84, P < 0.001) and equivalent PV⁺ interneurons (unpaired t-test, t₇ = 1.38, P = 0.21 and PV⁻ PNNs (unpaired t-test, t₇ = 1.21, P = 0.26) in dorsal CA1, compared with Control mice. (D) Labeling of WFA⁺ PNNs (red) and PV⁺ interneurons (cyan) in dorsal CA3 of Control and ELE mice on P20. (E) There was no difference in WFA⁺ PNNs (unpaired t-test, t₈ = 2.00, P = 0.081), PV⁺ interneurons (unpaired t-test, t₈ = 1.65, P = 0.13), PNN-enwrapped PV⁺ interneurons (unpaired t-test, t₈ = 1.50, P = 0.17), or PV⁻ PNNs (unpaired t-test, t₈ = 0.67, P = 0.52) in dorsal CA3 of P20 Control and ELA mice. (F) Experimental schedule for assessing memory precision. Mice were trained in a contextual fear conditioning task on P20, and 1 d later, they were tested in the same context (Context A) or a similar, novel context (Context B). (G) P20 mice previously raised in enriched environments showed precise contextual fear memories, compared with Control mice that showed age-appropriate memory imprecision (ANOVA, experience × test context interaction: F_1,28 = 11.64, P < 0.01). Data points represent individual mice with mean ± SEM. Scale bar, 50 μm. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 6.. ChABC treatment disrupts perineuronal nets in CA1 and memory precision in preweaning mice raised in enriched conditions.

(A) Mice raised in enriched environments from P6 to P18 were infused with Penicillinase or ChABC into CA1 on P18. (B) Labeling of WFA⁺ PNNs (red) and PV⁺ interneurons (cyan) in dorsal CA1 of ELE mice treated with Penicillinase or ChABC. (C) P20 mice treated with ChABC following early-life enrichment had fewer WFA⁺ PNNs (unpaired t-test, t₁₀ = 13.17, P < 0.0001), PV⁺ interneurons (unpaired t-test, t₁₀ = 4.49, P < 0.01), and PNN-enwrapped PV⁺ interneurons (unpaired t-test, t₁₀ = 12.84, P < 0.0001), but not fewer PV⁻ PNNs (unpaired t-test, t₁₀ = 2.08, P = 0.06), than mice treated with Penicillinase. (D) Experimental schedule for assessing memory precision. Mice were trained in a contextual fear conditioning task on P20, and 1 d later, they were tested in the same context (Context A) or a similar, novel context (Context B). (E) P20 mice treated with ChABC following early-life enrichment showed age-appropriate contextual fear memory imprecision, compared with P20 mice treated with Penicillinase, that showed precise memories (ANOVA, infusion × test context interaction: F_1,36 = 4.51, P < 0.05). Data points represent individual mice with mean ± SEM. Scale bar, 50 μm. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 7.. Early-life experience shapes trajectory for neuronal allocation to sparse engrams via PNNs.

(A) Mice were trained in a contextual fear conditioning task on P20 or P24 and their brains were removed for c-Fos and PV immunohistochemistry. (B) Labeling of c-Fos⁺ nuclei (white) in the dorsal CA1 pyramidal layer. (C) Sparse expression of c-Fos in dorsal CA1 was observed in mice with mature PNNs (P20 ELE+Penicillinase, P24 Control, P24 ELA+BDNF) but not in mice with immature PNNs (P20 Control, P20 ELE+ChABC, and P24 ELA+Vehicle) (ANOVA, age × group interaction: F_2,46 = 69.25, P < 0.00001). (D) Labeling of PV⁺ neurites (cyan) in the perisomatic region of putative engram (c-Fos⁺) and non-engram (c-Fos⁻) nuclei. (E) Following conditioning, PV⁺ neurites were more localized to the perisomatic region of non-engram cells in mice with mature PNNs (P20 ELE+Penicillinase, P24 Control, P24 ELA+BDNF) but not in mice with immature PNNs (P20 Control, P20 ELE+ChABC, and P24 ELA+Vehicle) (top, ANOVA, age × group × c-Fos interaction: F_2,46 = 25.12, P < 0.00001; bottom, ANOVA, age × group interaction: F_2,46 = 16.79, P < 0.00001). Data points represent individual mice with mean ± SEM. Scale bar, white: 50 μm, magenta: 20 μm. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 8.. Early-life experiences determine the timing of episodic-like memory development in mice.

The precision of event memories increases by P24, signaling the emergence of episodic-like memory function in mice. The development of episodic-like memory is supported by the maturation of extracellular PNNs around PV⁺ interneurons in CA1, which enable encoding of events by sparse neuronal engram ensembles. This developmental trajectory is decelerated in mice experiencing early-life adversity and is accelerated in mice experiencing early-life enrichment.