Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jun 8;21(6):e3002154.
doi: 10.1371/journal.pbio.3002154. eCollection 2023 Jun.

Fear memory recall involves hippocampal somatostatin interneurons

Krisztián Zichó  1   2 , Katalin E Sos  1 , Péter Papp  1 , Albert M Barth  1 , Erik Misák  1 , Áron Orosz  1 , Márton I Mayer  1   2 , Réka Z Sebestény  1 , Gábor Nyiri  1
Affiliations

Fear memory recall involves hippocampal somatostatin interneurons

Krisztián Zichó et al. PLoS Biol. .

Abstract

Fear-related memory traces are encoded by sparse populations of hippocampal principal neurons that are recruited based on their inhibitory-excitatory balance during memory formation. Later, the reactivation of the same principal neurons can recall the memory. The details of this mechanism are still unclear. Here, we investigated whether disinhibition could play a major role in this process. Using optogenetic behavioral experiments, we found that when fear was associated with the inhibition of mouse hippocampal somatostatin positive interneurons, the re-inhibition of the same interneurons could recall fear memory. Pontine nucleus incertus neurons selectively inhibit hippocampal somatostatin cells. We also found that when fear was associated with the activity of these incertus neurons or fibers, the reactivation of the same incertus neurons or fibers could also recall fear memory. These incertus neurons showed correlated activity with hippocampal principal neurons during memory recall and were strongly innervated by memory-related neocortical centers, from which the inputs could also control hippocampal disinhibition in vivo. Nonselective inhibition of these mouse hippocampal somatostatin or incertus neurons impaired memory recall. Our data suggest a novel disinhibition-based memory mechanism in the hippocampus that is supported by local somatostatin interneurons and their pontine brainstem inputs.

PubMed Disclaimer

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. Re-inhibition of DG SOM cells inhibited during memory acquisition recalls fear memory.
(A) After infecting DG SOM cells with inhibitory opsin-containing (“eOPN3 mice”) or control AAVs (“CTRL mice”) bilaterally, we implanted optic fibers over DGs. Representative image: injection site and optic fiber position (blue). Scale bar: 200 μm. Day 6: baseline freezing behavior of mice without and with light illumination. Day 7: CTRL and “Associated eOPN3-mice” received foot shocks aligned with light illumination, “Not associated eOPN3-mice” received foot shocks but no light. Day 8, in environment C: 2 min OFF 3min ON 2 min OFF light cycle. Only “Associated eOPN3 mice” (red) could recall fear memory. (B) Individual percentages of time spent with freezing behavior during the light OFF-ON-OFF cycles on day 8. (C) Changes in freezing behavior between the first light OFF and ON periods (medians and interquartile ranges). (D) Freezing time differences during the light ON periods on day 8 (medians and interquartile ranges). (E) Bilateral infection of DG SOM cells with inhibitory opsin and bilateral optic fiber implantation. After handling, the mice spent 10 min in a novel environment, where only 1 side of the DG was light illuminated. (F) Representative images: c-Fos (green) labeling of non-illuminated (non) and illuminated side (inh, where SOM cells are inhibited) of the DG, 1 h after exploration (optic fiber is blue). Scale bar: 200 μm. (G) Differences in the density of c-Fos positive cells (cells/mm2) in the DG GC layer in non-illuminated (non) and light disinhibited (inh.) side of the DG in 6 eOPN3 mice (medians and interquartile ranges). (H) We infected DG SOM cells with chemogenetic inhibitory (“hM4Di mice”) or control AAVs (“CTRL mice”) bilaterally. Image shows a representative injection site. Scale bar: 200 μm. After handling, the mice received 4 foot shocks on day 6. Day 7: all mice received CNO 1 h before contextual memory readout. (I) Contextual fear of hM4Di mice almost completely diminished compared to CTRL mice on day 7 (medians and interquartile ranges). (J) hM4Di-staining (red) is present only in a subpopulation (yellow) of DG SOM cells (green). Scale bar: 50 μm (for details, see S1 Extended Data). The data underlying this figure can be found in S1 Data. AAV, adeno-associated virus; DG, dentate gyrus; eOPN3, encephalopsin 3; GC, granule cell; hM4Di, modified human M4 muscarinic receptor; SOM, somatostatin.
Fig 2
Fig 2. Unlike DG PV cells, the re-inhibition of CA1 SOM cells inhibited during memory acquisition recalls fear memory.
(A) After infecting DG PV cells with inhibitory opsin-containing (“eOPN3 mice”) or control AAVs (“CTRL mice”) bilaterally, we implanted optic fibers over DGs. A representative image: injection site and optic fiber position (blue). Scale bar: 200 μm. Day 6: baseline freezing behavior of mice without and with light illumination. Day 7: All mice received foot shocks aligned with light illumination. Day 8 in environment C: 2 min OFF 3 min ON 2 min OFF light cycle demonstrated that (unlike DG SOM cells) optical re-inhibition of DG PV cells could not recall fear memory. (B) No differences in freezing between PV-Cre mouse groups during the light ON period on day 8 (medians and interquartile ranges). (C) Individual percentages of time spent with freezing behavior during the light OFF-ON-OFF cycles on day 8. (D) Changes in freezing behavior between the first light OFF and ON periods (medians and interquartile ranges). (E) After infecting CA1 SOM cells with inhibitory opsin-containing (“ArchT mice”) or control AAVs (“CTRL mice”) bilaterally, we implanted optic fibers over the CA1 areas. Representative image: injection site and optic fiber position (orange). Scale bar: 200 μm. Day 6: baseline freezing behavior of mice without and with light illumination. Day 7: All mice received foot shocks precisely aligned with light illumination. Day 8 in environment C: 2 min OFF 3 min ON 2 min OFF light cycle demonstrated that optical re-inhibition of CA1 SOM cells in itself could recall fear memory. (F) Individual percentages of time spent with freezing behavior during the light OFF-ON-OFF cycles for each SOM-Cre mouse on day 8. (G) Significant difference in the changes in freezing behavior between the first light OFF and ON periods (medians and interquartile ranges). (H) Freezing time difference between SOM-Cre mouse groups during the light ON period on day 8 (medians and interquartile ranges; for details, see S2 Extended Data). The data underlying this figure can be found in S1 Data. AAV, adeno-associated virus; ArchT, archaerhodopsin T-3; DG, dentate gyrus; eOPN3, encephalopsin 3; PV, parvalbumin; SOM, somatostatin.
Fig 3
Fig 3. NI GABAergic cells can disinhibit DG GCs via inhibiting DG SOM interneurons.
(A) We infected DG SOM cells with Cre-dependent tracer AAV and NI GABAergic cells with flippase-dependent tracer AAV in double transgenic SOM-Cre/vGAT-Flp mice (n = 4 mice). The representative image shows the injection site in NI. Scale bar: 200 μm. (B) SOM cells (red) receive multiple gephyrin-labeled (white) perisomatic synapses (yellow arrowheads) from NI GABAergic fibers (green). Scale bar: 10 μm. (C) Infection of NI GABAergic cells with either control (“CTRL mice”) or excitatory opsin-containing AAVs (“ChR2 mice”) and bilateral optic fiber implantation over the DG. The representative image shows the injection site in NI. Scale bar: 200 μm. After handling, the mice spent 10 min in a novel environment, where only 1 side of the DG was light illuminated. (D) Representative images show c-Fos labeling (red) of non-illuminated (non) and illuminated side (illumi.) of the DG, 40 min after exploration. Images show both sides for both the CTRL- and ChR2-mice. Scale bar: 200 μm. (E) Differences in the density of c-Fos positive cells (cells/mm2) in DG GC layer without and with illumination (medians, interquartile ranges, and individual data). (F) Sketch showing the experimental arrangement: silicon probes inserted into the DG and an optical fiber lowered above the NI to light activate the ChR2-expressing NI GABAergic neurons in vGAT-Cre mice (n = 4 mice). Photomicrographs depict the track of the silicone probe in the DG and the position of the optical fiber above the NI. Scale bar: 200 μm for DG and 500 μm for NI images. (G) Sample putative excitatory unit from DG with increasing activity upon light activation of the NI (top). Red dots indicate the average firing rate in the 60-s long periods. Vertical blue shadings denote the laser stimulation periods alternating with baseline periods. PSTH of the same unit (cell) with the corresponding average PSTH (bottom). (H) Summarized data showing all putative excitatory units (n = 3) with increasing activity upon NI light stimulation. Vertical blue shadings denote the laser stimulation periods (boxes indicate the median, lower and upper quartiles, red circles depict the individual data). (I) Sample putative inhibitory unit from DG with decreasing activity upon light activation of the NI (top). Blue dots indicate the average firing rate in the 60-s long periods. Vertical blue shadings denote the laser stimulation periods alternating with baseline periods. PSTH of the same unit with the corresponding average PSTH (bottom) (for details, see S3 Extended Data). The data underlying this figure can be found in S1 Data. AAV, adeno-associated virus; ChR2, channelrhodopsin 2; DG, dentate gyrus; GABA, γ-aminobutyric acid; GC, granule cell; NI, nucleus incertus; PSTH, peristimulus time histogram; SOM, somatostatin; vGAT, vesicular GABA transporter.
Fig 4
Fig 4. Reactivation of NI fibers in DG or NI cells activated during memory acquisition recalls fear memory.
(A) Infection of NI GABAergic cells with either control (“CTRL mice”) or excitatory opsin-containing AAVs (“ChR2 mice”) and bilateral optic fiber implantation over the DG. Representative images: labeling of NI cells and their fibers in DG. Optic fiber is blue. Scale bars: 200 μm. Day 6: baseline freezing behavior of mice without and with light illumination. Day 7: All mice received foot shocks aligned with light illumination. Day 8 in a novel environment B: 2 min OFF 3 min ON 2 min OFF light cycle revealed that optical reactivation of NI fibers could recall fear memory in ChR2 but not in CTRL mice. (B) Individual percentages of time spent with freezing behavior during the light OFF-ON-OFF cycles for each mouse on day 8. (C) Freezing time difference between the 2 groups during the light ON period on day 8 (medians and interquartile ranges). (D) Significant difference in the changes in freezing behavior between the first light OFF and ON periods for both groups (medians and interquartile ranges; for details, see S4 Extended Data). (E) After infecting NI GABAergic cells with excitatory opsin-containing (“ChR2-mice”) or control AAVs (“CTRL-mice”), we implanted an optic fiber over the NI. Representative image: injection site and optic fiber position (blue). Scale bar 200 μm. Day 6: baseline freezing behaviors of mice were assessed without and with light illumination. Then, in a novel environment B, all mice received 4 foot shocks precisely aligned with light illuminations. Day 8 in environment C: 2 min OFF 3 min ON 2 min OFF light cycle demonstrated that optical reactivation of NI GABAergic somata could recall fear memory. (F) Freezing time differences between the 2 groups during the light ON period on day 7 (medians and interquartile ranges). (G) Individual percentages of time spent with freezing behavior during the light OFF-ON-OFF cycles for each mouse on day 7 (for details, see S4 Extended Data). The data underlying this figure can be found in S1 Data. AAV, adeno-associated virus; ChR2, channelrhodopsin 2; DG, dentate gyrus; GABA, γ-aminobutyric acid; NI, nucleus incertus; vGAT, vesicular GABA transporter.
Fig 5
Fig 5. NI GABAergic cells are activated during contextual fear memory recall, and their activity is necessary for fear memory retrieval.
(A) After handling and preexposure to environment A on day 6, vGAT/ZsGreen mice were placed into environment A on day 7, where they received 4 foot shocks. (B) Day 8: Home-mice were sacrificed immediately after taking them out of their home cages. Recent-mice (Rec) were placed back in environment A for 5 min. Later, we analyzed the c-Fos activity (red) of GABAergic cells (green) in NI. Scale bar: 40 μm. (C) Thirty days later, we tested Remote-mice (Rem) as well. (D) Differences in density (cells/mm2) of c-Fos positive GABAergic cells in NI (medians and interquartile ranges). (E) Contextual fear responses of Recent- and Remote-mice during the 5-min recall period (medians and interquartile ranges). (F) Significant correlation (Spearman-rank correlation) between the density of c-Fos labeling of NI GABAergic cells and DG GCs in Recent-mice. (G) After infecting NI GABAergic cells with ArchT-containing (ArchT mice) or control AAVs (CTRL mice), we implanted an optic fiber over the NI. Representative image: injection site and optic fiber position (orange). Scale bar 200 μm. After handling, mice received 4 foot shocks. On the next 2 days, we analyzed the freezing behavior of these mice without (day 7: “CFC7”) and then with (day 8: “CFC8”) light illumination of the NI. (H) Individual percentages of time spent with freezing behavior during the light OFF (CFC7) and light ON cycles (CFC8) for each mouse on days 7 and 8, respectively. (I) Significant difference in the changes in freezing behavior between the light OFF (CFC7) and ON (CFC8) periods (medians and interquartile ranges; for details, see S5 Extended Data). The data underlying this figure can be found in S1 Data. AAV, adeno-associated virus; ArchT, archaerhodopsin T-3; DG, dentate gyrus; GABA, γ-aminobutyric acid; GC, granule cell; Home, homecage; NI, nucleus incertus; Rec, recent; Rem, remote; vGAT, vesicular GABA transporter.
Fig 6
Fig 6. Cortical memory centers strongly and specifically innervate NI and target hippocampus-projecting NI cells.
(A–C) The anterograde tracer AAV was injected into the mPFC (A, n = 2 mice), the ACC (B, n = 3 mice), or the RSC (C, n = 3 mice) bilaterally, and the retrograde tracer FG was injected into the hippocampus (HIPP) of the vGluT1-Cre mice bilaterally. (D–G) Representative injection sites in the HIPP (D), mPFC (E), ACC (F), and RSC (G). Br.: Bregma position, PL: Prelimbic cortex, IL: Infralimbic cortex. Scale bars: 200 μm for (D), 500 μm for (E–G). (H–J) Innervation pattern (red) of vGluT1-positive cells from the mPFC (H), the ACC (I), or the RSC (J) in the NI (HIPP-projecting retrogradely labeled cells are green). Scale bar: 200 μm. (K–M) Pairs of confocal images show that vGluT1-positive fibers (red) from the mPFC (K), the ACC (L), and the RSC (M) establish Homer-1 labeled (white) synaptic contacts (yellow arrowheads) on FG-positive neurons (green). Scale bar: 10 μm. (N) The experimental arrangement: silicon probes inserted into the DG and an optical fiber lowered above the NI to light activate the ChR2-expressing afferent fibers arriving from the ACC in vGluT1-Cre mice (n = 4 mice). Photomicrographs show the injection site in the ACC, the track of the silicone probe in the DG, and the position of the optical fiber above the NI. Scale bars: 200 μm for every images. (O) Sample putative excitatory unit from DG with increasing activity upon optogenetical activation of the ACC afferent fibers in NI (top). Red dots indicate the average firing rate in the 60-s long periods. Vertical blue shadings show the laser stimulation periods alternating with baseline periods. PSTH of the same unit with the corresponding average PSTH (bottom). (P) Summary graph of all putative excitatory units (n = 15) with increasing activity after the stimulation of ACC fibers in the NI. Vertical blue shadings show the laser stimulation periods (boxes indicate the median, lower and upper quartiles, red circles depict the individual data). (Q) Sample putative inhibitory unit from DG with decreasing activity upon light activation of the ACC afferent fibers in NI (top). Blue dots indicate the average firing rate in the 60-s long periods. Vertical blue shadings denote the laser stimulation periods alternating with baseline periods. PSTH of the same unit with the corresponding average PSTH (bottom). (For details, see S6 Extended Data). The data underlying this figure can be found in S1 Data. AAV, adeno-associated virus; ACC, anterior cingulate cortex; Br, bregma; ChR2, channelrhodopsin 2; DG, dentate gyrus; FG, FluoroGold; HIPP, hippocampus; mPFC, medial prefrontal cortex; NI, nucleus incertus; RSC, retrosplenial cortex; vGluT1, vesicular glutamate transporter 1.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Teyler TJ, Rudy JW. The Hippocampal Indexing Theory and Episodic Memory: Updating the Index. Hippocampus. 2007;17:1158–1169. doi: 10.1002/hipo.20350 - DOI - PubMed
    1. Goode TD, Tanaka KZ, Sahay A, McHugh TJ. An Integrated Index: Engrams, Place Cells, and Hippocampal Memory. Neuron. 2020;107:805–820. doi: 10.1016/j.neuron.2020.07.011 - DOI - PMC - PubMed
    1. Silva AJ, Zhou Y, Rogerson T, Shobe J, Balaji J. Molecular and cellular approaches to memory allocation in neural circuits. Science (80-). 2009;326:391–395. doi: 10.1126/science.1174519 - DOI - PMC - PubMed
    1. Josselyn SA, Köhler S, Frankland PW. Finding the engram. Nat Rev Neurosci. 2015;16:521–534. doi: 10.1038/nrn4000 - DOI - PubMed
    1. Josselyn SA, Tonegawa S. Memory engrams: Recalling the past and imagining the future. Science (80-). 2020;367. doi: 10.1126/science.aaw4325 - DOI - PMC - PubMed

Publication types

Grants and funding

This work was supported by the Hungarian National Research, Development and Innovation Office (OTKA K119521); the Frontline Research Excellence Program, by the Hungarian National Research, Development and Innovation Office (NRDI Fund 133837), the Hungarian Brain Research Program [NAP2.0 (2017-1.2.1-NKP-2017-00002) and NAP3.0 (NAP2022-I-1/2022)], European Union project RRF-2.3.1-21-2022-00004 within the framework of the Artificial Intelligence National Laboratory; European Union project RRF-2.3.1-21-2022-00011 within the framework of the Translational Neuroscience National Laboratory to G.N. The National Academy of Scientist Education Program of the National Biomedical Foundation under the sponsorship of the Hungarian Ministry of Culture and Innovation (FEIF/646-4/2021- ITM_SZERZ) to G.N and R.Z.S. The New National Excellence Program of the Ministry of Innovation, Hungary, UNKP-20-3-SE-31 and UNKP-21-3-SE-9 and the Semmelweis 250+ Excellence PhD Fellowship, EFOP-3.6.3-VEKOP-16-2017-00009 to K.Z. The National Research, Development and Innovation Office, Hungary, FK129019 to A.M.B. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.