In vivo CRISPRa decreases seizures and rescues cognitive deficits in a rodent model of epilepsy

Abstract

Epilepsy is a major health burden, calling for new mechanistic insights and therapies. CRISPR-mediated gene editing shows promise to cure genetic pathologies, although hitherto it has mostly been applied ex vivo. Its translational potential for treating non-genetic pathologies is still unexplored. Furthermore, neurological diseases represent an important challenge for the application of CRISPR, because of the need in many cases to manipulate gene function of neurons in situ. A variant of CRISPR, CRISPRa, offers the possibility to modulate the expression of endogenous genes by directly targeting their promoters. We asked if this strategy can effectively treat acquired focal epilepsy, focusing on ion channels because their manipulation is known be effective in changing network hyperactivity and hypersynchronziation. We applied a doxycycline-inducible CRISPRa technology to increase the expression of the potassium channel gene Kcna1 (encoding Kv1.1) in mouse hippocampal excitatory neurons. CRISPRa-mediated Kv1.1 upregulation led to a substantial decrease in neuronal excitability. Continuous video-EEG telemetry showed that AAV9-mediated delivery of CRISPRa, upon doxycycline administration, decreased spontaneous generalized tonic-clonic seizures in a model of temporal lobe epilepsy, and rescued cognitive impairment and transcriptomic alterations associated with chronic epilepsy. The focal treatment minimizes concerns about off-target effects in other organs and brain areas. This study provides the proof-of-principle for a translational CRISPR-based approach to treat neurological diseases characterized by abnormal circuit excitability.

Keywords: CRISPR; epilepsy; gene promoter; gene therapy; potassium channels.

Figure 1

CRISPRa increases endogenous Kcna1 expression reducing neuronal excitability in vitro. (A) Schematic representation of the CRISPRa approach to increasing Kcna1. Primary neurons were transduced with lentivirus 1 day after plating and experiments were performed after 10 days for qPCR and western blotting, and 14 days for electrophysiology. (B) Kcna1 mRNA expression normalized to the control LacZ sgRNA (blue) for the most effective sgRNAs and combinations of different sgRNAs tested in P19 cells (Supplementary Fig. 1). One-way ANOVA followed by Bonferroni multiple comparison test versus sgLacZ. (C and D) Western blots were used to determine the increase in K_v1.1 and glycosylated K_v1.1 (glyc) in neurons transduced either with dCas9A and sg19 (red), or sgLacZ (blue). Student’s t-test. (E) MA plots showing log₂ fold-change as a function of log₂ base mean expression of Kcna1-dCas9A treated neurons with respect to Ctrl-dCas9A. Kcna1 = pink; off-targets = green; all other genes = grey. (F) Messenger RNA expression relative to sgLacZ for each off-target. The expression level of sgLacZ is represented as the dashed line at 1. Multiple Student’s t-tests, each compared to control and corrected for multiple comparison (α = 0.0083). (G) Representative traces of recordings from neurons transduced either with Ctrl-dCas9A (sgLacZ, blue) or Kcna1-dCas9A (sg19, red) and injected with 300 pA steps in current clamp. (H) Average firing rates in response to different current injections for neurons transduced with ctrl-dCas9A or Kcna1-dCas9A. Two-way RM ANOVA. (I) Neuronal and action potential (AP) properties in neurons transduced with ctrl-dCas9A or Kcna1-dCas9A. Student’s t-test.

Figure 2

CRISPRa delivered with AAV9 increases endogenous Kcna1 expression and reduces CA1 pyramidal neuron excitability. (A) Schematic representation of the approach for ex vivo quantification of CRISPRa effects. Mice were injected with AAVs at approximately postnatal Day (P)60 and experiments were performed 3 weeks after. (B) Representative live image of Kcna1-dCas9A expression in the hippocampus (bregma −2.70; inset = CA1 region, red = tdTomato) (C) Representative traces from CA1 pyramidal neurons, transduced either with Ctrl-dCas9A (sgLacZ, blue) or Kcna1-dCas9A (sg19, red) in pryramidal neurons injected with 280 pA steps in current clamp. (D) Average firing rates in response to different current steps for neurons transduced with Ctrl-dCas9A or Kcna1-dCas9A. Two-way repeated measures ANOVA. (E) Neuronal and action potential (AP) properties in neurons transduced with Ctrl-dCas9A or Kcna1-dCas9A. Maximal firing rate, current threshold to elicit the first action potential, action potential width and resting membrane potential are shown. Student’s t-test. (F) Cumulative frequency (%) of the action potential widths in neurons injected with 280 pA of current. Kolmogorov-Smirnov test for cumulative distributions. (G) Activity clamp protocol to mimic 24 interictal bursts of synaptic input from an epileptic network in neurons transduced with Ctrl-dCas9A or Kcna1-dCas9A (left). Black arrows represent the action potentials missed in the Kcna1-dCas9A neurons. Number of action potentials for each burst showed as mean ± standard error of the mean (SEM) (middle). Two-way ANOVA. The histogram represents the average number of action potentials for each neuron in the 24 bursts (right). Student’s t-test.

Figure 3

CRISPRa-Kcna1 decreases number of seizures in a mouse model of acquired intractable temporal lobe epilepsy. (A) Schematic representation of the CRISPRa approach used in vivo to treat the intra-amygdala kainic acid (KA) focal model of temporal lobe epilepsy. Mice were injected with AAVs 1 week after kainic acid treatment. Baseline EEG recordings were started the following week and continued for 2 weeks. Two further weeks of EEG recordings were performed after doxycycline food supplementation. (B) Representative immunofluorescence 7 weeks after status epilepticus of neurons transduced with Ctrl-dCas9A 4 weeks after status epilepticus. Scale bars: DG = 250 μm; CA1 = 50 μm. (C) Quantitative PCR analysis of Kcna1 and dCas9 expression in the ipsilateral hippocampus relative to the contralateral hippocampus in epileptic mice transduced with either Ctrl-dCas9A or Kcna1-dCas9A at the end of the experiments. Student’s t-test. (D) Left: Raster plot showing all seizures before and after doxycycline administration in 13 mice treated with Ctrl-dCas9A and nine mice treated with Kcna1-dCas9A. Right: Pie charts showing the proportion of animals showing either more or fewer seizures after doxycycline food than during the baseline. Fisher’s exact test. (E) Number of seizures/day before and after doxycycline administration in control-dCas9 (n = 13) and Kcna1-dCas9 (n = 9) treated animals. Two-way ANOVA followed by Bonferroni multiple comparison test. Empty blue circles are animals that died during baseline and excluded from the analysis. (F) Cumulative plot of seizures normalized to the baseline in mice transduced with either ctrl-dCas9A or Kcna1-dCas9A. Two-way ANOVA.

Figure 4

CRISPRa-Kcna1 rescues cognitive deficits and mitigates transcriptomic changes in a mouse model of acquired intractable temporal lobe epilepsy. (A) Experimental plan for behaviour and transcriptomic analysis. Mice were injected with AAVs a week after kainic acid (KA) treatment. Baseline behavioural tests were performed 2 weeks after AAV injection. Further behavioural tests were performed 2 weeks after doxycycline food supplementation. At the end of the experiment hippocampi were collected for transcriptomic analysis. (B) Graphical representation of the Object Location Test (OLT) and Novel Object Recognition Test (NORT). (C) Discrimination index for sham control and epileptic animals before and after (green box) doxycycline in mice treated either with Ctrl-dCas9A or Kcna1-dCas9A. Two-way ANOVA followed by Bonferroni multiple comparison test. (D) Volcano plots showing statistical significance (−log₁₀P-value) as a function of fold-change in gene expression (log₂FC), comparing pairs of datasets as indicated above each plot. Genes showing P < 0.05 difference in expression (−log₁₀P-value > 1.3) are highlighted in red. (E) Venn diagram showing the fraction of differentially regulated genes in epileptic Ctrl-dCas9A-treated mice (E Ctrl-dCas9A) which were rescued in Kcna1-dCas9A-treated mice (E Kcna1-dCas9A). (F) Histogram displaying representative gene ontology (GO) categories functionally enriched among the 2742 genes that were differentially expressed in epileptic control mice (E Ctrl-dCas9A) versus treated mice (E Kcna1-dCas9A).