Enhanced hippocampal LTP but normal NMDA receptor and AMPA receptor function in a rat model of CDKL5 deficiency disorder

Abstract

Background: Mutations in the X-linked gene cyclin-dependent kinase-like 5 (CDKL5) cause a severe neurological disorder characterised by early-onset epileptic seizures, autism and intellectual disability (ID). Impaired hippocampal function has been implicated in other models of monogenic forms of autism spectrum disorders and ID and is often linked to epilepsy and behavioural abnormalities. Many individuals with CDKL5 deficiency disorder (CDD) have null mutations and complete loss of CDKL5 protein, therefore in the current study we used a Cdkl5^-/y rat model to elucidate the impact of CDKL5 loss on cellular excitability and synaptic function of CA1 pyramidal cells (PCs). We hypothesised abnormal pre and/or post synaptic function and plasticity would be observed in the hippocampus of Cdkl5^-/y rats.

Methods: To allow cross-species comparisons of phenotypes associated with the loss of CDKL5, we generated a loss of function mutation in exon 8 of the rat Cdkl5 gene and assessed the impact of the loss of CDLK5 using a combination of extracellular and whole-cell electrophysiological recordings, biochemistry, and histology.

Results: Our results indicate that CA1 hippocampal long-term potentiation (LTP) is enhanced in slices prepared from juvenile, but not adult, Cdkl5^-/y rats. Enhanced LTP does not result from changes in NMDA receptor function or subunit expression as these remain unaltered throughout development. Furthermore, Ca²⁺ permeable AMPA receptor mediated currents are unchanged in Cdkl5^-/y rats. We observe reduced mEPSC frequency accompanied by increased spine density in basal dendrites of CA1 PCs, however we find no evidence supporting an increase in silent synapses when assessed using a minimal stimulation protocol in slices. Additionally, we found no change in paired-pulse ratio, consistent with normal release probability at Schaffer collateral to CA1 PC synapses.

Conclusions: Our data indicate a role for CDKL5 in hippocampal synaptic function and raise the possibility that altered intracellular signalling rather than synaptic deficits contribute to the altered plasticity.

Limitations: This study has focussed on the electrophysiological and anatomical properties of hippocampal CA1 PCs across early postnatal development. Studies involving other brain regions, older animals and behavioural phenotypes associated with the loss of CDKL5 are needed to understand the pathophysiology of CDD.

Keywords: AMPA receptor; CDKL5; NMDA receptor; hippocampus; intrinsic properties; rat; synaptic plasticity.

Fig. 1

Validation of Cdkl5^-/y rats. (A) Schematic of the Cdkl5 knockout strategy depicting the WT and null alleles. The null allele has a 10 bp (bp) deletion in exon 8 (region shown in blue in WT allele), leading to a frame shift and an in frame, premature STOP codon forming in exon 9 (*). (B) Genotyping results from male WT and Cdkl5^−/y animals. Higher band in WT and Cdkl5^−/y animals resulting from F1 and R primers product. Lower band in the WT samples resulting from F2 and R primer products is absent in the null samples due to the 10 bp deleted sequence. (C) Western blot showing the absence of CDKL5 in hippocampal and prefrontal cortex tissue preparations from WT and Cdkl5^−/y rats. (D) Quantification of CDKL5 western blot protein expression in hippocampal and prefrontal cortex preparations. (E) Western blot showing the absence of CDKL5 in hippocampal synaptosome preparations from WT and Cdkl5^−/y rats. (F) Quantification of CDKL5 western blot protein expression in hippocampal synaptosome preparations

Fig. 2

Typical excitability of CA1 pyramidal cells. (A) Representative traces of whole cell recordings from WT (black, upper) and Cdkl5^−/y (green, lower) CA1 pyramidal cells in response to subsequent 25 pA steps. Traces shown from − 100 pA to rheobase-1 and for the maximum firing frequency (I = 400 pA). (B, B’) Input resistance. (C, C’) rheobase current. (D, D’) Action potential discharge in response to 500 ms long 25 pA current steps up to 400 pA (Two-way ANOVA genotype effect F_16,208=0.12, p = 0.66). Data shown as mean ± SEM (WT – n = 26 cells/8 rats, Cdkl5^−/y – n = 24 cells/7 rats, data points represent single cells (B, C, D) or animal averages (B’, C’, D’)

Fig. 3

Hippocampal long term potentiation (LTP) in juvenile Cdkl5-/y rats. (A) Representative WT (upper) and Cdkl5^−/y (lower) fEPSP traces before (baseline) and after (post tetanus) LTP induction. (B) Time-course showing long term potentiation (LTP) in the hippocampal CA1 induced by two trains with 100 pulses at 100 Hz (20 s apart), resulting in a significant increase in LTP in Cdkl5^−/y rats when compared to WT. (C) LTP in the final 10 min of the recording relative to baseline (WT n = 8 rats; Cdkl5^−/y: n = 13 rats; *p < 0.05 Two tailed T test, data points represent animal averages)

Fig. 4

Unaltered NMDA receptor function and subunit composition in the hippocampus of P28-35 Cdkl5^-/y rats. (A) Representative traces of AMPA receptor and NMDA receptor-mediated currents evoked by stimulating Schafer collateral inputs to CA1. (B, B’) NMDAR/AMPAR ratio (p = 0.31 GLMM), (C, C’) Pharmacologically isolated NMDA receptor-mediated EPSC decay time (p = 0.78 Mann-Whitney U test performed on animal averages). Data shown as mean ± SEM (WT n = 10 rats / 14 cells; Cdkl5/y: n = 9 rats / 18 cells), data points represent individual cells (B, C) and respective animal averages (B’, C’). (D) Representative western blot images from synaptosome preparations probed for NMDA receptor subunits GluN1, GluN2A, GluN2B and respective Total Protein stain. (E–G) Quantification of protein expression level normalised to total protein and WT. Data shown as mean ± SEM. ns-p > 0.05 Two-tailed T test

Fig. 5

Typical NMDA receptor developmental trajectory in Cdkl5^-/y rats. (A) NMDAR/AMPAR ratio in WT and Cdkl5^−/y rats aged P7 to P22. (B) NMDAR decay time constant over development. (C) Representative traces of NMDA receptor-mediated EPSCs in the presence or absence (baseline) of the GluN2B antagonist Ro 25-6981. (D) NMDA receptor-mediated EPSC amplitude for individual cells before (full circles) and after (clear circles) Ro 25-6981 application, with recordings from each cell connected by a straight line across 3 age groups examined. (E) Percentage of NMDA receptor current blocked by RO 25-6981 based on cells shown in (D). All data shown as mean ± SEM, data points represent animal averages (except in D)

Fig. 6

Unaltered AMPA receptor-mediated EPSC I-V relationship in CA1 pyramidal cells of Cdkl5^-/y rats. (A) Representative traces of AMPA receptor-mediated currents from WT (upper, black) and Cdkl5^−/y (lower, green) recorded over a range of holding potentials (-80mV to + 40 mV) in the presence of 0.1mM spermine in the intracellular solution. (B, B’) I-V relationship AMPA receptor-mediated EPSC normalised to EPSC amplitude at -80 mV holding potential (Genotype effect: F_1,11=1.794, p = 0.21, Two-way ANOVA) (C) Rectification index calculated as the ratio of the difference EPSC amplitude between 0 mV and − 80 mV and 0 mV and + 40 mV. Data shown as mean ± SEM, data shown for individual cells (B, C) and animal averages (B’, C’) (WT n = 24 cells / 10 rats; Cdkl5^−/y: n = 27 cells / 11 rats))

Fig. 7

Reduced mEPSC frequency and typical PPR in CA1 pyramidal cells from Cdkl5^-/y rats. (A) Representative traces of mEPSC recordings from WT (left, black) and Cdkl5^−/y (right, green) rats. (B–B’) mEPSC frequency. (C-C’) mEPSC amplitude: WT n = 24 cells / 9 rats, Cdkl5^−/yn = 20 cells / 8 rats. (D) Cumulative distribution of inter-event interval. (E) Cumulative distribution of mEPSC amplitude. (F) Representative traces of EPSCs evoked by PP stimulation of Schafer collateral inputs to CA1 pyramidal cells from WT (upper, black) and Cdkl5^−/y rats (lower, green) at an interstimulus interval of 50 ms. (G) PPR of evoked (WT n = 18 cells / 7 rats, Cdkl5^−/yn = 9 cells / 5 rats) at paired-pulse intervals of 20, 50 and 100 ms. Data in bar charts shown as mean ± SEM (data points represent individual cells (B, C, G) or corresponding animal averages (B’, C’, G’)

Fig. 8

CA1 pyramidal cell morphology and spine density across multiple dendritic compartments. (A) Example reconstruction of CA1 pyramidal cells from WT and Cdkl5^−/y rats filled with biocytin during whole cell patch clamp recordings. (B) Sholl analysis of the dendritic arborisation (Two way ANOVA: Interaction: F _76,912 = 2.094, p < 0.001, genotype effect p = 0.38). (C) Total dendritic length, (D) total length of basal dendrites, (E) total length of apical dendrites. Data shown as mean ± SEM (WT - n = 14 cells/7 rats, Cdkl5^−/y - n = 14 cells/7 rats, data points represent animal averages, all p values > 0.05, Two tailed t-test))

Fig. 9

Spine density across dendritic compartments of CA1 pyramidal cells. (A) Representative segments of basal and apical (oblique and tuft) dendrites from CA1 pyramidal cells filled during whole-cell patch-clamp recordings. (B) Spine density in basal dendrites (WT: n = 12 cells/6 rats, Cdkl5^−/y: n = 12 cells / 7 rats). (C) Spine density in apical oblique dendrites (WT: n = 9 cells / 6 rats, Cdkl5^−/y: n = 12 cells / 6 rats). (D) Spine density in apical tuft dendrites (WT: n = 7 cells/4 rats, Cdkl5^−/y: n = 11 cells / 7 rats). Data shown as mean ± SEM, data points represent cell (B, C, D) or animal averages (B’, C’, D’). *p < 0.05, ns p > 0.05 LMM

Fig. 10

Minimal stimulation of CA3 inputs to CA1 pyramidal cells reveal no difference in silent synapses in Cdkl5^-/y rats. (A) Representative traces of EPSCs recorded at -70 mV and + 40 mV evoked by minimal stimulation of Schaffer collaterals. (B) Example time-course of synaptic responses throughout a single WT and Cdkl5^-/y recording upon Schafer collateral stimulation. (C) Response probability at -70 mV and + 40 mV (data shown as cells, values for each cell connected by a black line. Two-way ANOVA Genotype effect: F_1,46 = 5.16, p = 0.03, Holding potential effect: F_1,46=38.50, p < 0.0001). (D) Ratio of the response probability at + 40mV and − 70 mV following minimal stimulation of Schaffer collaterals. (WT n = 30 cells / 11 rats, Cdkl5^-/yn = 18 cells / 7 rats)