Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Dec 11;11:442.
doi: 10.3389/fnmol.2018.00442. eCollection 2018.

MicroRNA-22 Controls Aberrant Neurogenesis and Changes in Neuronal Morphology After Status Epilepticus

Affiliations
Free PMC article

MicroRNA-22 Controls Aberrant Neurogenesis and Changes in Neuronal Morphology After Status Epilepticus

Edward H Beamer et al. Front Mol Neurosci. .
Free PMC article

Abstract

Prolonged seizures (status epilepticus, SE) may drive hippocampal dysfunction and epileptogenesis, at least partly, through an elevation in neurogenesis, dysregulation of migration and aberrant dendritic arborization of newly-formed neurons. MicroRNA-22 was recently found to protect against the development of epileptic foci, but the mechanisms remain incompletely understood. Here, we investigated the contribution of microRNA-22 to SE-induced aberrant adult neurogenesis. SE was induced by intraamygdala microinjection of kainic acid (KA) to model unilateral hippocampal neuropathology in mice. MicroRNA-22 expression was suppressed using specific oligonucleotide inhibitors (antagomir-22) and newly-formed neurons were visualized using the thymidine analog iodo-deoxyuridine (IdU) and a green fluorescent protein (GFP)-expressing retrovirus to visualize the dendritic tree and synaptic spines. Using this approach, we quantified differences in the rate of neurogenesis and migration, the structure of the apical dendritic tree and density and morphology of dendritic spines in newly-formed neurons.SE resulted in an increased rate of hippocampal neurogenesis, including within the undamaged contralateral dentate gyrus (DG). Newly-formed neurons underwent aberrant migration, both within the granule cell layer and into ectopic sites. Inhibition of microRNA-22 exacerbated these changes. The dendritic diameter and the density and average volume of dendritic spines were unaffected by SE, but these parameters were all elevated in mice in which microRNA-22 was suppressed. MicroRNA-22 inhibition also reduced the length and complexity of the dendritic tree, independently of SE. These data indicate that microRNA-22 is an important regulator of morphogenesis of newly-formed neurons in adults and plays a role in supressing aberrant neurogenesis associated with SE.

Keywords: epilepsy; microRNA-22; mouse model; neurogenesis; status epilepticus.

Figures

Figure 1
Figure 1
Increased miR-22 levels in the contralateral hippocampus during epilepsy in mice. (A) Schematic showing intraamygdala injection of kainic acid (KA) into the ipsilateral amygdala to induce status epilepticus (SE). (B) Graph showing increased levels of miR-22 in the contralateral dentate gyrus (DG) 2 weeks following SE (n = 4 per group, Student’s t-test: *t = 3.483, p = 0.0131). (C) In situ hybridization showing miR-22 in the DG of the contralateral hippocampus in non-epileptic control vehicle-injected mice and in mice subjected to SE. Of note, miR-22 levels seem to be mainly increased in the subgranular zone in the DG of the contralateral hippocampus during epilepsy (6 weeks following SE). Scale bar = 50 μm. (D) MiR-22 expression in the hippocampus in a patient suffering from temporal lobe epilepsy (TLE). MiR-22-positive cells were mainly localized to the DG (arrows and magnification). snRNA U6 was used as positive control. Scale bar = 1 mm. (E) Graph showing decreased miR-22 levels in the hippocampus (n = 4 per group Student’s t-test: **t = 4.955, p = 0.0026). (F) In situ hybridization showing decreased miR-22 levels in the subgranular zone of the DG 6 weeks following intra-cerebro-ventricular (i.c.v.) injection of Ant22. Scale bar = 50 μm. (G) Decreased P2X7 receptor expression in the contralateral hippocampus of epileptic mice 14 days following SE when compared to control mice (n = 4 per group, Students t-test: *t = 2.734, p = 0.034). (H) Lower BzATP-provoked currents detected by patch clamp from green fluorescent protein (GFP)-positive cells present in the contralateral DG of the hippocampus of epileptic P2rx7-GFP expressing mice when compared to GFP-positive cells from the ipsilateral DG 14 days following SE (n = 19 cells for ipsi- and 17 cells from contralateral hippocampus from nine mice per group, Student’s t-test, *t = 2.673, p = 0.0115). Scale bar = 50 μm. (I) Increased P2X7 receptor expression mainly localized to the subgranular zone of the DG of the contralateral hippocampus (arrows) in mice treated with Ant22 6 weeks following SE when compared to scramble-treated epileptic mice. n = 5, Scale bar = 50 μm.
Figure 2
Figure 2
Increased neurogenesis 6 weeks following SE in mice with suppressed miR-22 expression. (A) Graph and representative images showing more Doublecortin-(DCX)-positive cells in the hippocampus of epileptic mice. DCX-positive cell numbers were even higher in epileptic mice treated with Ant22 when compared to epileptic mice treated with Scr (n = 3 (Scr Cont), 4 Ant22 (Cont and KA) and 6 (Scr KA), one-way ANOVA: F(3,15) = 41.01, p < 0.0001; post hoc test (Bonferroni): ***t = 4.979, p < 0.001 (scramble/veh vs. scramble/KA); ***t = 9.726 (scramble/veh vs. antagomir/KA); ###t = 5.437, p < 0.001). Scale bar = 50 μm. (B) Graph and representative images showing more IdU-positive cells in epileptic mice 6 weeks following SE compared to control mice. The number of IdU-positive cells was also increased in epileptic mice treated with Ant22 when compared to Scr-treated epileptic mice (n = 3 (Scr Cont), 4 Ant22 (Cont and KA) and 6 (Scr KA), One-way ANOVA: F(3,12) = 2.951 p < 0.0001 post hoc tests (multiple-comparisons Bonferroni): **t = 5.308, p < 0.01; ***t = 8.285, p < 0.001; #t = 2.978, p < 0.05). Blue: nuclear marker DAPI, Red: IdU-positive cells. Scale bar = 50 μm. (C) Graph and representative image showing percentage of NeuN-positive IdU cells in the DG of the contralateral hippocampus 6 weeks following intraamygdala KA injection. Red: NeuN, Green: IdU (n = 3 (Scr Cont), 4 Ant22 (Cont and KA) and 6 (Scr KA). Scale bar = 20 μm.
Figure 3
Figure 3
Viral transfection of postsynaptic density protein 95 (PSD95)-GFP of scramble and antagomir-22 treated mice following SE. (A) Schematic time-course of experimental design. Twenty four hours following i.c.v. injection of Ant22/Scr, mice were injected with intraamygdala KA to induce SE which was terminated with lorazepam (Lz). One week following SE, mice were injected into the contralateral DG with PSD95-GFP expressing retrovirus and brain analyzed 6 weeks later. (B) Representative images of the contralateral DG from Scr and Ant22-treated control and epileptic mice 6 weeks following transfection with PSD95-GFP. Virus was injected 1 week post-SE. Note, strong increase in GFP-positive cells in the contralateral DG in Ant22-treated epileptic mice. Scale bar = 200 μm. (C) Graph showing more PSD-95-GFP-positive cells in the contralateral DG in Ant22 treated mice (two-Way ANOVA: Scr vs. Ant22 (***F(1,16) = 25.93, p < 0.001); Veh vs. KA (***F(1,16) = 91.810, p < 0.001), interaction (***F(1,16) = 20.28, p < 0.001). Post hoc tests (Tukey’s HSD): Veh/Scr vs. Veh/Ant22: n.s, p = 0.975; *Veh/Scr vs. KA/Scr: p = 0.012; ***KA/Scr vs. KA/Ant22: p < 0.001. (D) Representative image of migration calculation and graph showing mean (± SEM) distance traveled by PSD95-GFP-positive newly-formed neurons as percentage of width of granule cell layer (two-way ANOVA: Ant22 vs. Scr (**F(1,16) = 15.94, p = 0.001); Veh vs. KA (***F(1,16) = 94.88, p < 0.0001), interaction (**F(1,16) = 9.65, p = 0.007). Post hoc tests (Tukey’s HSD): Veh/Scr vs. Veh/Ant22: n.s, p = 0.922; **Veh/Scr vs. KA/Scr: p = 0.001, ***KA/Scr vs. KA/Ant22: p < 0.001. (E) Example of PSD95-GFP-positive, newly-formed neuron in ectopic location in dentate hilus in an epileptic Ant22-treated mouse 6 weeks following transfection with PSD95-GFP virus and graph showing percentage of PSD95-GFP-positive, newly-formed neurons found ectopically in dentate hilus (Student’s t-test: KA/Scr vs. KA/Ant22: *t8 = 2.02, p = 0.039. (F) Graph showing increased distance traveled across the granule cell layer by IdU-positive neurons 6 weeks following SE (two-Way ANOVA: Scr vs. Ant22: *F(1,16) = 8.32, p = 0.011; Veh vs. KA: ***F(1,16) = 92.83, p < 0.001; Interaction: *F(1,16) = 8.208, p = 0.011. Post hoc tests (Tukey’s HSD): Veh/Scr vs. Veh/Ant22: n.s, p = 1.000; **Veh/Scr vs. KA/Scr: p = 0.001; **KA/Scr vs. KA/Ant22: p = 0.004. (G) IdU-positive cells found ectopically in the contralateral hilus of the DG of epileptic Scr-treated mice 6 weeks following SE. Ectopic contralateral hilar cell count was exacerbated in epileptic Ant22-treated mice. Student’s t-test: KA/Scr vs. KA/Ant22: *t8 = 2.52, p = 0.018. (H) Representative image of PSD95-GFP-positive neurons in the contralateral DG of epileptic Ant22 mice 6 weeks following virus infection. Note, newly-formed neurons are all DCX-negative. DCX in red, DAPI in blue and PSD95-GFP transfected cells in green. Scale bar = 50 μm.
Figure 4
Figure 4
Ant22 reduces complexity of the dendritic tree, independently of SE. (A) Representative images of dendritic trees of PSD95-GFP-positive neurons of the four treatment groups (Scr and Ant22 control and epileptic mice 6 weeks following virus injection). Scale bar = 50 μm. (B) Dendritic length (combined length of entire dendritic tree) of all four treatment groups. Two-Way ANOVA: Scr vs. Ant22: F(1,204) = 12.19, p < 0.001; Veh vs. KA: n.s, F(1,204) = 0.97, p = 0.319; Interaction: n.s. F(1,204) = 0.00, p = 986. Post hoc tests (Tukey’s HSD): Veh/Scr vs. Veh/Ant22: n.s. p = 0.066; Veh/Scr vs. KA/Scr: n.s. p = 0.867; KA/Scr vs. KA/Ant22: n.s. p = 0.070. (C) Cartoon illustrating Sholl analysis. Concentric circles at 50 μm intervals. Number of branches crossing at each interval are counted. (D) Sholl analysis of dendritic trees from four treatment groups. Branching was analyzed every 50 μm. (E) Schoenen Ramification Index (SRI) of dendritic trees from the four treatment groups showing decreased branching in GFP-positive cells in Ant22-treated mice. Two-way ANOVA: Scr. vs. Ant22: *F(1,197) = 5.100, p = 0.025; Veh. vs. KA: F(1,197) = 2.767, p = 0.098, n.s; Interaction: F(1,197) = 1.54, p = 0.216, n.s). Post hoc tests (Tukey’s HSD): Veh/Scr vs. Veh/Ant22: n.s. p = 0.172; Veh/Scr vs. KA/Scr: n.s. p = 0.062; KA/Scr vs. KA/Ant22: n.s. p = 0.99. (F) Effect of Ant22-treatment and SE on dendritic diameter (two-Way ANOVA: Scr vs. Ant22: n.s. F(1,16) = 2.19, p = 0.157; Veh vs. KA: n.s. F(1,16) = 4.20, p = 0.0527; Interaction: n.s. F(1,16) = 4.33, p = 0.054; Post hoc tests (Tukey’s HSD): Veh/Scr vs. Veh/Ant22: n.s. p = 0.977; Veh/Scr vs. KA/Scr: n.s. p = 1.000; KA/Scr vs. KA/Ant22: n.s. p = 0.095; *Veh/Ant22 vs. KA/Ant22: p = 0.044. (G) Graph showing increase in dendritic diameter in Ant22-treated mice following SE at higher dendritic orders (Repeated Measures Two-way ANOVA: F3, 85 = 14.45, p < 0.001; post hoc test (Bonferroni): ##Scr/KA vs. Ant22/KA, t = 3.882, p < 0.001).
Figure 5
Figure 5
MiR-22 inhibition increases spine density and volume on dendrites of newly-formed neurons following SE. (A) Representative images showing increased spine density in Ant22-treated epileptic mice when compared to Scr-treated epileptic mice. Dendrites are shown in red and spines in green (arrows). Scale bar = 20 μm. (B) Graph showing that SE increases spine density and that this is exacerbated by Ant22 treatment (two-way ANOVA: Scr vs. Ant22: n.s. F(1,36) = 2.34, p = 0.135; Veh vs. KA: F(1,36) = 11.26, p = 0.002; Interaction: n.s. F(1,36) = 2.47, p = 0.125). Post hoc tests (Tukey’s HSD): Veh/Scr vs. Veh/Ant22: n.s. p = 1.000; Veh/Scr vs. KA/Scr: n.s. p = 0.593; KA/Scr vs. KA/Ant22: n.s. p = 0.144; **Veh/Ant22 vs. KA/Ant22: p = 0.007. (C) Elevated spine density in Ant22-treatd mice following SE most pronounced at higher dendritic orders: Repeated-measures ANOVA: F(3,198) = 92.16, p < 0.0001, post hoc test (Bonferroni; scramble KA vs. antagomir KA): #t = 2.840, p < 0.05 (order ii); #t = 3.112 (order iii), p < 0.05; ###t = 4.384, p < 0.001 (order iv); ##t = 3.069, p < 0.01 (order v); ###t = 4.421, p < 0.001 (order vi). (D) Graph showing average spine volume increased following KA-induced SE in Ant22-treated mice, compared with Scr-treated controls (two-Way ANOVA: Scr vs. Ant22: F(1,26) = 5.87 p = 0.023; Veh vs. KA: n.s. F(1,26) = 2.933, p = 0.099; Interaction: F(1,26) = 5.66, p = 0.025. Post hoc tests (Tukey’s HSD): Veh/Scr vs. Veh/Ant22: n.s. p = 1.000; Veh/Scr vs. KA/Scr: n.s. p = 0.910; *KA/Scr vs. KA/Ant22: p = 0.011; *Veh/Ant22 vs. KA/Ant22: p = 0.048. (E) Increased spine volume in KA and Ant22-treated mice consistent across all dendritic orders (Repeated Measures ANOVA: F(3, 80) = 32.26, p < 0.0001; post hoc test (Bonferroni; scramble KA vs. antagomir/KA): #t = 2.840, p < 0.05 (order ii); #t = 3.112, p < 0.05 (order iii); ###t = 4.384, p < 0.001 (order iv); ##t = 3.609, p < 0.01 (order v); ###t = 4.421, p < 0.001 (order vi)). (F) Examples of mushroom-shaped and non-mushroom shaped spines. Spines with a head diameter greater than 0.4 μm and a head/neck ratio greater than 1.2 (Bourne and Harris, 2011) were considered “mushroom-shaped.” Scale bar = 0.5 μm. (G) Ratio of mushroom/non-mushroom shaped spines increased following KA-induced SE in Ant22-treated mice, compared with Scr-treated controls (two-way ANOVA: Scr vs. Ant22: F(1,36) = 4.95, p = 0.032; Veh vs. KA: F(1,36) = 10.22, p = 0.003; Interaction: F(1,36) = 4.35, p = 0.044. Post hoc tests (Tukey’s HSD): Veh/Scr vs. Veh/Ant22: n.s. p = 1.000; Veh/Scr vs. KA/Scr: n.s. p = 0.861; *KA/Scr vs. KA/Ant22: p = 0.021; **Veh/Ant22 vs. KA/Ant22: p = 0.003. (H) Increased density of mushroom-shaped spines in mice following treatment with Ant22 and 6 weeks following KA-induced SE most marked at higher dendritic orders (Repeated Measures ANOVA: F(3,198) = 59.24, p < 0.0001; post hoc test (Bonferroni; scramble/KA vs. antagomir/KA): #t = 2.827, p < 0.05 (order iii); ###t = 4.351, p < 0.001 (order iv); ###t = 6.069, p < 0.001 (order v); ###t = 7.737, p < 0.001 (order vi)).

Similar articles

See all similar articles

Cited by 4 articles

References

    1. Althaus A. L., Sagher O., Parent J. M., Murphy G. G. (2015). Intrinsic neurophysiological properties of hilar ectopic and normotopic dentate granule cells in human temporal lobe epilepsy and a rat model. J. Neurophysiol. 113, 1184–1194. 10.1152/jn.00835.2014 - DOI - PMC - PubMed
    1. Araya R., Eisenthal K. B., Yuste R. (2006). Dendritic spines linearize the summation of excitatory potentials. Proc. Natl. Acad. Sci. U S A 103, 18799–18804. 10.1073/pnas.0609225103 - DOI - PMC - PubMed
    1. Arisi G. M., Garcia-Cairasco N. (2007). Doublecortin-positive newly born granule cells of hippocampus have abnormal apical dendritic morphology in the pilocarpine model of temporal lobe epilepsy. Brain Res. 1165, 126–134. 10.1016/j.brainres.2007.06.037 - DOI - PubMed
    1. Bartel D. P. (2004). MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116, 281–297. 10.1016/S0092-8674(04)00045-5 - DOI - PubMed
    1. Berenguer J., Herrera A., Vuolo L., Torroba B., Llorens F., Sumoy L., et al. . (2013). MicroRNA 22 regulates cell cycle length in cerebellar granular neuron precursors. Mol. Cell. Biol. 33, 2706–2717. 10.1128/mcb.00338-13 - DOI - PMC - PubMed

LinkOut - more resources

Feedback