Limits...
Parvalbumin-expressing interneurons coordinate hippocampal network dynamics required for memory consolidation

View Article: PubMed Central - PubMed

ABSTRACT

Activity in hippocampal area CA1 is essential for consolidating episodic memories, but it is unclear how CA1 activity patterns drive memory formation. We find that in the hours following single-trial contextual fear conditioning (CFC), fast-spiking interneurons (which typically express parvalbumin (PV)) show greater firing coherence with CA1 network oscillations. Post-CFC inhibition of PV+ interneurons blocks fear memory consolidation. This effect is associated with loss of two network changes associated with normal consolidation: (1) augmented sleep-associated delta (0.5–4 Hz), theta (4–12 Hz) and ripple (150–250 Hz) oscillations; and (2) stabilization of CA1 neurons' functional connectivity patterns. Rhythmic activation of PV+ interneurons increases CA1 network coherence and leads to a sustained increase in the strength and stability of functional connections between neurons. Our results suggest that immediately following learning, PV+ interneurons drive CA1 oscillations and reactivation of CA1 ensembles, which directly promotes network plasticity and long-term memory formation.

No MeSH data available.


Related in: MedlinePlus

PV+ interneurons promote consistent reactivation of CA1 neural ensembles following learning.(a) Generation of a FSM. The FSM displays the similarity of functional connectivity patterns (that is, matrices from Fig. 6b) across all time intervals. (b) NREM CA1 network FSMs from a representative hM4Di-expressing mouse at baseline, and over the first 6 h post CFC (vehicle and CNO conditions). Colour in the body denotes the degree of similarity between NREM functional connectivity patterns at any given time point in the recording, and NREM patterns at all other time points. Scale bars, 20 min of recording time. (c) Distributions of minute-to-minute similarity values (at baseline, and following CFC) for the data shown in b. (d) Distributions of NREM sleep similarity values from all stably recorded mice at baseline, and following CFC (n=3/condition). For both (c,d), similarity distributions were significantly shifted to higher values following CFC in vehicle-treated mice, but not CNO-treated mice (P<0.001, Mann–Whitney rank sum test).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5384212&req=5

f7: PV+ interneurons promote consistent reactivation of CA1 neural ensembles following learning.(a) Generation of a FSM. The FSM displays the similarity of functional connectivity patterns (that is, matrices from Fig. 6b) across all time intervals. (b) NREM CA1 network FSMs from a representative hM4Di-expressing mouse at baseline, and over the first 6 h post CFC (vehicle and CNO conditions). Colour in the body denotes the degree of similarity between NREM functional connectivity patterns at any given time point in the recording, and NREM patterns at all other time points. Scale bars, 20 min of recording time. (c) Distributions of minute-to-minute similarity values (at baseline, and following CFC) for the data shown in b. (d) Distributions of NREM sleep similarity values from all stably recorded mice at baseline, and following CFC (n=3/condition). For both (c,d), similarity distributions were significantly shifted to higher values following CFC in vehicle-treated mice, but not CNO-treated mice (P<0.001, Mann–Whitney rank sum test).

Mentions: To visualize how CA1 network communication patterns change in the hours following CFC, we generated a functional similarity matrix (FSM; schematized in Fig. 7a) for each recording. The FSM illustrates the degree of similarity between a network's functional connectivity pattern at a given time point and at all other time points across a recording. FSMs are shown for NREM epochs a representative hM4Di-expressing mouse at baseline and for the first 6 h after CFC following either vehicle or CNO administration, respectively (Fig. 7b). Distributions of similarity values were quantified and compared across baseline and post-CFC recording periods (Fig. 7c). In vehicle-treated mice, there was more frequent and consistent repetition of specific network functional connectivity patterns throughout the first 6 h of post-CFC NREM, shown as a rightward shift in the similarity distribution after CFC (Fig. 7c,d; see also Supplementary Fig. 10). As was true for delta and theta LFP power changes, stable repetition of network patterns waxed and waned (and in some cases, appeared to do so in concert with LFP power changes; see Supplementary Fig. 11). Such changes in connectivity patterns across time were not seen when PV+ interneurons were inhibited. These findings show that in addition to promoting network oscillations after learning, PV+ interneurons promote consistent reactivation of CA1 neural ensembles over time.


Parvalbumin-expressing interneurons coordinate hippocampal network dynamics required for memory consolidation
PV+ interneurons promote consistent reactivation of CA1 neural ensembles following learning.(a) Generation of a FSM. The FSM displays the similarity of functional connectivity patterns (that is, matrices from Fig. 6b) across all time intervals. (b) NREM CA1 network FSMs from a representative hM4Di-expressing mouse at baseline, and over the first 6 h post CFC (vehicle and CNO conditions). Colour in the body denotes the degree of similarity between NREM functional connectivity patterns at any given time point in the recording, and NREM patterns at all other time points. Scale bars, 20 min of recording time. (c) Distributions of minute-to-minute similarity values (at baseline, and following CFC) for the data shown in b. (d) Distributions of NREM sleep similarity values from all stably recorded mice at baseline, and following CFC (n=3/condition). For both (c,d), similarity distributions were significantly shifted to higher values following CFC in vehicle-treated mice, but not CNO-treated mice (P<0.001, Mann–Whitney rank sum test).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5384212&req=5

f7: PV+ interneurons promote consistent reactivation of CA1 neural ensembles following learning.(a) Generation of a FSM. The FSM displays the similarity of functional connectivity patterns (that is, matrices from Fig. 6b) across all time intervals. (b) NREM CA1 network FSMs from a representative hM4Di-expressing mouse at baseline, and over the first 6 h post CFC (vehicle and CNO conditions). Colour in the body denotes the degree of similarity between NREM functional connectivity patterns at any given time point in the recording, and NREM patterns at all other time points. Scale bars, 20 min of recording time. (c) Distributions of minute-to-minute similarity values (at baseline, and following CFC) for the data shown in b. (d) Distributions of NREM sleep similarity values from all stably recorded mice at baseline, and following CFC (n=3/condition). For both (c,d), similarity distributions were significantly shifted to higher values following CFC in vehicle-treated mice, but not CNO-treated mice (P<0.001, Mann–Whitney rank sum test).
Mentions: To visualize how CA1 network communication patterns change in the hours following CFC, we generated a functional similarity matrix (FSM; schematized in Fig. 7a) for each recording. The FSM illustrates the degree of similarity between a network's functional connectivity pattern at a given time point and at all other time points across a recording. FSMs are shown for NREM epochs a representative hM4Di-expressing mouse at baseline and for the first 6 h after CFC following either vehicle or CNO administration, respectively (Fig. 7b). Distributions of similarity values were quantified and compared across baseline and post-CFC recording periods (Fig. 7c). In vehicle-treated mice, there was more frequent and consistent repetition of specific network functional connectivity patterns throughout the first 6 h of post-CFC NREM, shown as a rightward shift in the similarity distribution after CFC (Fig. 7c,d; see also Supplementary Fig. 10). As was true for delta and theta LFP power changes, stable repetition of network patterns waxed and waned (and in some cases, appeared to do so in concert with LFP power changes; see Supplementary Fig. 11). Such changes in connectivity patterns across time were not seen when PV+ interneurons were inhibited. These findings show that in addition to promoting network oscillations after learning, PV+ interneurons promote consistent reactivation of CA1 neural ensembles over time.

View Article: PubMed Central - PubMed

ABSTRACT

Activity in hippocampal area CA1 is essential for consolidating episodic memories, but it is unclear how CA1 activity patterns drive memory formation. We find that in the hours following single-trial contextual fear conditioning (CFC), fast-spiking interneurons (which typically express parvalbumin (PV)) show greater firing coherence with CA1 network oscillations. Post-CFC inhibition of PV+ interneurons blocks fear memory consolidation. This effect is associated with loss of two network changes associated with normal consolidation: (1) augmented sleep-associated delta (0.5&ndash;4&thinsp;Hz), theta (4&ndash;12&thinsp;Hz) and ripple (150&ndash;250&thinsp;Hz) oscillations; and (2) stabilization of CA1 neurons' functional connectivity patterns. Rhythmic activation of PV+ interneurons increases CA1 network coherence and leads to a sustained increase in the strength and stability of functional connections between neurons. Our results suggest that immediately following learning, PV+ interneurons drive CA1 oscillations and reactivation of CA1 ensembles, which directly promotes network plasticity and long-term memory formation.

No MeSH data available.


Related in: MedlinePlus