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
. 2022 Dec 20;13(1):49.
doi: 10.1186/s13229-022-00528-z.

Experience-dependent changes in hippocampal spatial activity and hippocampal circuit function are disrupted in a rat model of Fragile X Syndrome

Antonis Asiminas  1   2   3   4 , Sam A Booker  5   6   7 , Owen R Dando  5   6   7   8 , Zrinko Kozic  5   6 , Daisy Arkell  5   6   7 , Felicity H Inkpen  5   6   7 , Anna Sumera  5   6   7 , Irem Akyel  5   6 , Peter C Kind #  5   6   7   9 , Emma R Wood #  10   11   12   13
Affiliations

Experience-dependent changes in hippocampal spatial activity and hippocampal circuit function are disrupted in a rat model of Fragile X Syndrome

Antonis Asiminas et al. Mol Autism. .

Abstract

Background: Fragile X syndrome (FXS) is a common single gene cause of intellectual disability and autism spectrum disorder. Cognitive inflexibility is one of the hallmarks of FXS with affected individuals showing extreme difficulty adapting to novel or complex situations. To explore the neural correlates of this cognitive inflexibility, we used a rat model of FXS (Fmr1-/y).

Methods: We recorded from the CA1 in Fmr1-/y and WT littermates over six 10-min exploration sessions in a novel environment-three sessions per day (ITI 10 min). Our recordings yielded 288 and 246 putative pyramidal cells from 7 WT and 7 Fmr1-/y rats, respectively.

Results: On the first day of exploration of a novel environment, the firing rate and spatial tuning of CA1 pyramidal neurons was similar between wild-type (WT) and Fmr1-/y rats. However, while CA1 pyramidal neurons from WT rats showed experience-dependent changes in firing and spatial tuning between the first and second day of exposure to the environment, these changes were decreased or absent in CA1 neurons of Fmr1-/y rats. These findings were consistent with increased excitability of Fmr1-/y CA1 neurons in ex vivo hippocampal slices, which correlated with reduced synaptic inputs from the medial entorhinal cortex. Lastly, activity patterns of CA1 pyramidal neurons were dis-coordinated with respect to hippocampal oscillatory activity in Fmr1-/y rats.

Limitations: It is still unclear how the observed circuit function abnormalities give rise to behavioural deficits in Fmr1-/y rats. Future experiments will focus on this connection as well as the contribution of other neuronal cell types in the hippocampal circuit pathophysiology associated with the loss of FMRP. It would also be interesting to see if hippocampal circuit deficits converge with those seen in other rodent models of intellectual disability.

Conclusions: In conclusion, we found that hippocampal place cells from Fmr1-/y rats show similar spatial firing properties as those from WT rats but do not show the same experience-dependent increase in spatial specificity or the experience-dependent changes in network coordination. Our findings offer support to a network-level origin of cognitive deficits in FXS.

PubMed Disclaimer

Conflict of interest statement

P.C.K. is a senior editor for Molecular Autism. The other authors declare no financial or non-financial competing interests.

Figures

Fig. 1
Fig. 1
WT and Fmr1−/y rats move similar distances and at similar speeds while foraging in a novel environment over the two days, but Fmr1−/y rats visit more of the environment than WT rats. A Schematic of the recording protocol (Top) and example trajectories from a WT (black) and an Fmr1−/y rat (red). B Total path length decreased across the three sessions within a day for both WT and Fmr1−/y rats, but there was no difference between genotypes or across days (session-in-day p < 0.0001; day p = 0.069; genotype p = 0.967). C Both WT and Fmr1−/y rats showed good coverage of the environment in all recording sessions. Fmr1−/y rats visited a significantly higher proportion of the environment than WT rats (*genotype p = 0.001). There was also a decrease in exploration across sessions within each day, but not across days (session-in day p = 0.001; day p = 0.615). Data points depict rat group means; error bars depict SEM; Statistical analyses used 3-way ANOVA with genotype (between subjects), day (within subjects) and session-in day (within subjects); nWT = 7 rats, n Fmr1−/y = 7 rats. Pale yellow and pale purple backgrounds denote data from Day 1 and Day 2, respectively
Fig. 2
Fig. 2
Example firing rate maps and activity measures. Example firing rate maps from 4 WT (left) and 4 Fmr1−/y (right) CA1 pyramidal cells, each from a different animal. Top: Spike waveforms (black lines) from the tetrode channel with the highest amplitude waveforms across the 6 recording sessions. Red solid lines indicate the mean waveforms. Dotted red lines indicate the standard error. Middle: movement trajectory (black path) and superimposed action potentials of the cell (red dots) across the 6 sessions in the novel environment. Mean session firing rates (Hz) and burst probability (Burst P) are stated above each plot. Bottom: Smoothed firing rate maps of the same cell, with warmer colours indicating higher firing rates. Spatial information (bits/spk), sparsity, place field size (cm2) and percentage of visited bins in which the cells fired (% Active) for each session is stated above the firing rate map. Pale yellow and pale purple backgrounds denote data from Day 1 and Day 2, respectively
Fig. 3
Fig. 3
WT but not Fmr1−/y CA1 pyramidal neurons exhibit experience-dependent changes in mean firing rate and burst probability. A The mean firing rate of CA1 pyramidal neurons decreased significantly between the first and second day of exploration in the novel environment in WT but not in Fmr1−/y rats (genotype x day interaction p = 0.006; post hoc comparisons Day 1 vs Day 2, WT: p < 0.0001; Fmr1−/y: p = 0.131). There were no significant differences in mean firing rate between genotypes on either Day 1 or Day 2 (post hoc comparisons p’s > 0.05). B Burst probability of pyramidal neurons decreased significantly between Day 1 and Day 2 in WT, but not in Fmr1−/y rats (genotype x day interaction p = 0.005; post hoc comparisons Day 1 vs Day 2: WT: p < 0.0001; Fmr1−/y p = 0.924). There were no significant differences between genotypes on either day (post hoc comparisons p’s > 0.05). Data represent cell means and SEMs. Statistical analyses used linear mixed effect (LME) modelling with genotype, day and session-in-day as fixed factors, cell and rat as random factors, followed by Tukey comparisons on emmeans for significant interactions. NWT-D1 = 222 cells, NWT-D2 = 207 cells, NKO-D1 = 211 cells, NKO-D2 = 205 cells. Pale yellow and pale purple backgrounds denote data from Day 1 and Day 2, respectively
Fig. 4
Fig. 4
Experience-dependent refinement of spatial coding in the CA1 pyramidal cells is impaired in Fmr1−/y rats. A The spatial information of CA1 pyramidal cells increased between the first and second day of exploration in the novel environment in WT but not in in Fmr1−/y rats, with no significant difference between genotypes on either day (genotype x day interaction p = 0.005; post hoc comparisons Day 1 vs Day 2: WT p < 0.001; Fmr1−/y p = 0. 129; WT vs Fmr1−/y: p > 0.05 on both days). B The spatial sparsity of CA1 pyramidal cell firing decreased significantly between Day 1 and Day 2 in WT but not Fmr1−/y rats, with no differences between genotypes on either day (genotype x day interaction p = 0.013; post hoc comparisons Day 1 vs Day 2: WT p < 0.001; Fmr1−/y p = 0.091; WT vs Fmr1−/y: p > 0.05 on both days). C The size of CA1 pyramidal cell place fields decreased significantly between Day 1 and Day 2 in WT but not Fmr1−/y rats, with no differences between genotypes on either day (genotype x day interaction p = 0.017; post hoc comparisons Day 1 vs Day 2: WT p < 0.001; Fmr1−/y p = 0.051; WT vs Fmr1−/y: p > 0.05 on both days). D The proportion of visited pixels in which CA1 pyramidal cells fired (% active bins) did not differ significantly across days or between genotypes (genotype x day interaction p = 0.070). Data represent cell means and SEMs. Statistical analyses using LME modelling and Tukey post hoc tests as in Fig. 2. NWT-D1 = 222, NWT-D2 = 207, NKO-D1 = 211, NKO-D2 = 205. Pale yellow and pale purple backgrounds denote data from Day 1 and Day 2, respectively
Fig. 5
Fig. 5
The stability of CA1 pyramidal cell firing rate maps does not differ between WT and Fmr1−/y rats. A Example firing rate maps from 4 WT (left) and 4 Fmr1−/y (right) CA1 pyramidal cells, each from a different animal. Smoothed firing rate maps of the same cell, with warmer colours indicating higher firing rates. R values from Pearson correlations (Fisher z-transformed) between the firing rate maps of consecutive exploration sessions are indicated below the arrows between sessions. B Mean Pearson correlation coefficients (Fisher z-transformed) between firing rate maps for consecutive sessions for the population of WT and Fmr1−/y pyramidal cells. Firing rate map stability did not differ significantly between WT and Fmr1−/y cells (genotype p = 0.094). Both genotypes had less stable maps between days (Session 3–4 comparison) than between sessions of the same day (post hoc tests on S3-S4 vs all other comparisons, #p’s < 0.05). Data represent cell means and SEM. Statistical analyses used LME modelling with genotype and session comparison (i.e. S1-S2, S2-S3, S3-S4, S4-S5 and S5-S6) as fixed factors, and cell and rat as random factors, followed by Tukey post hoc comparisons on emmeans for significant main effect of session. S1vsS2: NWT = 194, NKO = 167, S2vsS3: NWT = 191, NKO = 179, S3vsS4: NWT = 121, NKO = 152, S4vsS5: NWT = 190, NKO = 161, S5vsS6: NWT = 188, NKO = 169. Pale yellow and pale purple backgrounds denote data from Day 1 and Day 2, respectively
Fig. 6
Fig. 6
Fmr1−/y CA1 pyramidal neurons display increased excitability, which correlates with reduced synaptic inputs from the medial entorhinal cortex. A Representative traces from CA1 pyramidal neurons in the dorsal hippocampus from WT and Fmr1−/y rats, in response to depolarizing current injections (0–400 pA, 25 pA steps, 500 ms duration). B Action potential discharge in 500 ms compared to injected current for all recorded CA1 pyramidal neurons in WT and Fmr1−/y rats. Data shown as the mean response recorded per rat, with total number of neurons indicated. C The slope of the curve in (B) quantified for each neuron. Individual neuron data are shown overlain as filled circles, and the number of tested neurons shown below in parentheses. Average resting membrane potential (D) and input resistance (E), measured from the zero-current potential. Average data are plotted for the rheobase current (F), the voltage threshold (G), and medium afterhyperpolarization (mAHP) amplitude (H), the latter two measured from the first action potential at rheobase. I Visualization of the AIS in flattened confocal z-stacks after immunofluorescent labelling for AnkyrinG (AnkG, green pseudocolour) and merged with NeuN (blue pseudocolour) from WT and Fmr1−/y rats. Scale bars shown: 20 µm. J Quantification of AIS length, displayed as the average of individual animals. K Representative traces from cell attached recordings (upper traces) and whole cell recordings (lower traces) following stimulation of str. lacunosum moleculare to distal CA1, from WT and Fmr1−/y rats. EPSP data is shown as the average traces in response to stimulation intensities. L Average data for EPSP amplitude in response to the same stimulation intensities, delivered to the Schaffer collateral (SC), and M temporoammonic (TA) pathways. Average spike probability, measured in cell-attached recordings, for SC (N) and TA (O) paths. For L–O, all graphs display the result from 2-way ANOVA for genotype shown above the chart and number of tested neurons indicated in parentheses. All data is shown as mean ± SEM. Statistics shown as: ns—p > 0.05, *—p < 0.05, and ***—p < 0.001 from GLMM analysis, except panels B, K, L, M, N which are the result of 2-way ANOVA for genotype
Fig. 7
Fig. 7
No differences between WT and Fmr1−/y rats in the power of hippocampal oscillatory activity. A Top: Example CA1 LFP traces bandpass filtered for theta (6–12 Hz) from a WT (black) and an Fmr1−/y (red) rat. Bottom: Box and whisker plots depicting theta power (6–12 Hz) for each session. The middle line represents rat median, upper and lower end of the box represents 95th and 5th percentile, whiskers represent maximum and minimum values. There was a significant decrease in theta power across sessions within a day (p = 0.021) but no significant difference between genotypes (p = 0.367) or days (p = 0.581) and no significant interactions. B Slow gamma (30–45 Hz) (same layout as in A) exhibited no significant differences between genotypes (p = 0.184), days (p = 0.735) or sessions within a day (p = 0.817) and no significant interactions. C Medium gamma (55–100 Hz) (same layout as in A) exhibited no significant differences between genotypes (p = 0.103), days (p = 0.521) or sessions within a day (p = 0.265) and no significant interactions. Statistical analysis used 3-way ANOVA (genotype, day, session in day). NWT = 7, NKO = 7. Pale yellow and pale purple backgrounds denote data from Day 1 and Day 2, respectively
Fig. 8
Fig. 8
Fmr1−/y CA1 pyramidal neurons are less phase locked to gamma oscillations than WT neurons. A Representative firing phase distribution of a single principal neuron along a theta cycle. Same for slow gamma oscillations (D) and medium gamma oscillations (G). B Average Mean Vector Length (MVL), quantifying the strength of phase locking to theta oscillations across the six recording sessions. MVL decreased significantly in cells of both genotypes across sessions within a day (session-in-day effect p = 0.026), but there were no differences between genotypes or days (p’s > 0.05). C The proportion of significantly phase-locked (Rayleigh p < 0.05) pyramidal neurons to theta oscillations across six recording sessions. There were no differences between genotypes, and no genotype x session interaction (p’s > 0.05). However, there is a significant main effect of session (Log-likelihood ratio: session effect p = 0.029), with theta phase locking higher in sessions 1 and 3 compared to sessions 5 and 6 (two proportion z-test, session 1 vs session 5 #p < 0.01; session 1 vs session 6 #p < 0.01; session 2 vs session 5 #p < 0.05; session 2 vs session 6 #p < 0.01) E WT neurons exhibit higher MVLs (stronger phase locking) to slow gamma compared to Fmr1−/y neurons (LME: main effect of genotype *p = 0.022), with no significant differences between days or sessions within a day (p’s > 0.05). Further tests indicate significant differences between genotypes only in sessions 1 and 3 (Tukey post hoc, session 1 +p < 0.001; session 3 +p = 0.02). F A higher proportion of WT pyramidal neurons is significantly phase locked to slow gamma oscillations compared to Fmr1−/y neurons (Log-likelihood ratio: genotype effect p = 0.035) and sessions different between one another (Log-likelihood ratio: session effect p < 0.001). While there was no significant genotype x session interaction (Log-likelihood ratio: genotype x session p = 0.053), the proportion of slow gamma phase locked neurons was higher in WT that rats in Sessions 1 and 3 of Day 1, but not for any other session (two proportion z-test, WT vs Fmr1−/y Session 1: +p < 0.001; Session 3: +p = 0.023) a significantly higher proportion of WT neurons were phase-locked in session 1 than any other session (two proportion z-test WT, session 1 vs session 2 #p < 0.001; session 1 vs session 3 #p = 0.014; session 1 vs session 4 #p < 0.001; session 1 vs session 5 #p < 0.001; session 1 vs session 6 #p < 0.001). H There were no significant differences in MVL with respect to medium gamma between genotypes, days or sessions within a day (LME all p’s > 0.05). I However, a significantly higher proportion of WT than Fmr1−/y neurons was significantly phase locked to medium gamma oscillations (Log-likelihood ratio: genotype effect p = 0.013). NWT-D1 = 222, NWT-D2 = 207, NKO-D1 = 211, NKO-D2 = 205. Pale yellow and pale purple backgrounds denote data from Day 1 and Day 2, respectively
Fig. 9
Fig. 9
Theta and gamma phase preferences of CA1 pyramidal neurons differ between WT and Fmr1−/y rats. A Schematic depiction of oscillation phases. The oscillation troughs were defined as 0°. B Mean preferred theta phase for significantly phase-locked (Rayleigh p < 0.05) CA1 pyramidal neurons during the six recording sessions. Fmr1−/y neurons fired during the ascending phase of theta on both days, whereas WT neurons fired significantly earlier (at the trough of theta) on Day 1, and shifted to the ascending phase of theta on Day 2 (Harrison-Kanji test: main effect of genotype p < 0.001; genotype x day interaction p = 0.0025; Watson-Williams post hoc tests Day 1 vs Day 2: WT p < 0.001, Fmr1−/y p > 0.05; WT vs Fmr1−/y: Day 1 *p < 0.001, Day 2 *p < 0.001). C Distribution of preferred theta phase for unique pyramidal neurons recorded from WT (Top-black) and Fmr1−/y (Bottom-red) rats during Day 1 (left) and Day 2 (right). D Same as (B) for slow gamma phase. Fmr1−/y pyramidal neurons fired earlier during the descending phase of the oscillation compared to WT (Harrison-Kanji test: main effect of genotype: p = 0.003) (Watson-Williams post hoc tests, Day 1 *p < 0.05; Day 2 p = 0.103). This effect was driven by differences between genotypes during Day 1 (). E Same as (C) for slow gamma oscillations. F Same as (B) for medium gamma phase. No differences between genotypes in the preferred medium gamma phase (p > 0.05). G Same as (C) for medium gamma oscillations. Theta: NWT-D1 = 214, NWT-D2 = 197, NKO-D1 = 199, NKO-D2 = 187, SGamma: NWT-D1 = 89, NWT-D2 = 51, NKO-D1 = 65, NKO-D2 = 47, SGamma: NWT-D1 = 78, NWT-D2 = 63, NKO-D1 = 60, NKO-D2 = 56. Pale yellow and pale purple backgrounds denote data from Day 1 and Day 2, respectively
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Colak D, Zaninovic N, Cohen MS, Rosenwaks Z, Yang WY, Gerhardt J, et al. Promoter-bound trinucleotide repeat mRNA drives epigenetic silencing in fragile X syndrome. Science. 2014;343(6174):1002–1005. doi: 10.1126/science.1245831. - DOI - PMC - PubMed
    1. Pieretti M, Zhang F, Fu YH, Warren ST, Oostra BA, Caskey CT, et al. Absence of expression of the FMR-1 gene in fragile X syndrome. Cell. 1991;66(4):817–822. doi: 10.1016/0092-8674(91)90125-I. - DOI - PubMed
    1. Verkerk AJMH, Pieretti M, Sutcliffe JS, Fu YH, Kuhl DPA, Pizzuti A, et al. Identification of a gene (FMR-1) containing a CGG repeat coincident with a breakpoint cluster region exhibiting length variation in fragile X syndrome. Cell. 1991;65(5):905–914. doi: 10.1016/0092-8674(91)90397-H. - DOI - PubMed
    1. Coffee B, Keith K, Albizua I, Malone T, Mowrey J, Sherman SL, et al. Incidence of Fragile X Syndrome by newborn screening for methylated FMR1 DNA. Am J Hum Genet. 2009;85(4):503–514. doi: 10.1016/j.ajhg.2009.09.007. - DOI - PMC - PubMed
    1. Hunter J, Rivero-Arias O, Angelov A, Kim E, Fotheringham I, Leal J. Epidemiology of fragile X syndrome: a systematic review and meta-analysis. Am J Med Genet A. 2014;164A(7):1648–1658. doi: 10.1002/ajmg.a.36511. - DOI - PubMed

Publication types

Substances