Inflammation triggers synaptic alteration and degeneration in experimental autoimmune encephalomyelitis

Abstract

Neurodegeneration is the irremediable pathological event occurring during chronic inflammatory diseases of the CNS. Here we show that, in experimental autoimmune encephalomyelitis (EAE), a mouse model of multiple sclerosis, inflammation is capable in enhancing glutamate transmission in the striatum and in promoting synaptic degeneration and dendritic spine loss. These alterations occur early in the disease course, are independent of demyelination, and are strongly associated with massive release of tumor necrosis factor-alpha from activated microglia. CNS invasion by myelin-specific blood-borne immune cells is the triggering event, and the downregulation of the early gene Arc/Arg3.1, leading to the abnormal expression and phosphorylation of AMPA receptors, represents a culminating step in this cascade of neurodegenerative events. Accordingly, EAE-induced synaptopathy subsided during pharmacological blockade of AMPA receptors. Our data establish a link between neuroinflammation and synaptic degeneration and calls for early neuroprotective therapies in chronic inflammatory diseases of the CNS.

Figure 1.

EAE alters glutamatergic transmission in the striatum. a, b, Half-width and decay time of glutamatergic sEPSCs recorded from striatal neurons increased in the presymptomatic (10 dpi) and in the acute (25 dpi) phase of EAE. **p < 0.01. c, Cumulative distributions of sEPSC decay time recorded in HC and in the acute stage of EAE. The electrophysiological traces are examples of sEPSC mean peak obtained by group analysis. d, Scatter plot of sEPSC half-width and sEPSC decay time recorded from neurons of EAE mice at 25 dpi. Pearson's test revealed a significant positive correlation. e, f, The graphs show that the sEPSC rise time and amplitude were unchanged in EAE at 10 and 25 dpi. g, The frequency of sEPSCs was increased at both preclinical and clinical stages of EAE. **p < 0.01. h, The electrophysiological traces are examples of sEPSCs (downward deflections) recorded from striatal neurons in control conditions and 10 and 25 dpi with MOG. i, The graph shows that pharmacological inhibition of NMDA receptors with MK-801 failed to normalize sEPSC half-width in EAE mice. j, EAE pathological process causes changes in GluR1 protein composition of PSD preparations. Left, Immunoblot of PSD proteins obtained from hippocampus and striatum of EAE mice at 10 and 25 dpi and control group animals (n = 3 for each experiment). Right, Densitometric quantification changes in gray values and SEM (EAE group/control group values). The histogram shows that the expression of GluR1 subunit of AMPA receptors and its phosphorylation at the Ser845 in isolated striatal postsynaptic densities of presymptomatic (10 dpi) and symptomatic (25 dpi) EAE mice was increased. *p < 0.01. k, Representative density plot showing calcein AM fluorescence versus Annexin V fluorescence for large synaptosomal particles (data were collected for 20,000 particles from each sample; 10,000 events are plotted). Percentage of total particles is shown for each quadrant. Particles positive for both markers are in the top right quadrant. Increase in Annexin V labeling was observed in striatum synaptosomes taken from EAE mice (25 dpi). Background, Untreated P-2 double labeled for Annexin V and calcein AM.

Figure 2.

T lymphocytes alter synaptic transmission and cause microglia/macrophage activation in EAE. a, Inflammatory lesions in the striatum of EAE mice (20 dpi) were evaluated by double staining for CD3 (green in the box) or Iba1 (red in the box). b, The histogram shows that both rise and decay time of sEPSCs recorded from neurons in the presence of CD3⁺ cells were altered. The electrophysiological traces on the right are examples of sEPSC mean peak obtained by group analysis in control conditions (CD3⁺ cells from healthy controls) and in the presence of CD3⁺ cells extracted from EAE mice. **p < 0.01. c, Primary microglial cell cultures were generated starting from P2 C57BL/6 newborn. Primary microglia cells were shaken off and treated with either the Th1 or the Th2 mix for 24 h (n = 3 independent cell cultures). Then, total mRNA was extracted and used for the real-time PCR analysis. The histogram shows that Th1 mix treatment resulted in a dramatic upregulation of either Iba1 (4.3 ± 0.6; ***p < 0.001) or TNFα (3.15 ± 1.7; *p < 0.05) mRNA levels, markers of microglia/macrophage cell activation. Data represent the log2 of the fold changes (±SEM) with respect to the vehicle-treated cells. d, Striatal sections from HC and EAE mice (20, 30, and 60 dpi) were probed for IddU detection during 10 h of IddU administration. IddU⁺ cells (green) were distributed within either the dorsolateral or ventrolateral SVZ in both HC and EAE 60 dpi (arrows in first and fourth panels). However, EAE 20 dpi showed many IddU⁺ cells located within the striatal parenchyma (second panel). This phenotype was reduced at later time points as shown at either EAE 30 or 60 dpi. Scale bar, 150 μm. e, The mean ± SEM number of IddU⁺ cells were counted at each time point and plotted on histogram. ***p < 0.001. f, g, Cell counts revealed a mean number per section of 20 ± 2 (p < 0.0001) IddU⁺ cells at 8 dpi, 47 ± 2 (p < 0.0001) IddU⁺ cells at 14 dpi, 73 ± 16 (p < 0.01) IddU⁺ cells at 18 dpi, 280 ± 45 (p < 0.001) IddU⁺ cells at 20 dpi, 30 ± 14 (NS) IddU⁺ cells at 30 dpi, and 7 ± 2 (NS) IddU⁺ cells at 60 dpi. f, Confocal optical sections of parallel sections probed for Iba1 (red) and IddU (green) detection revealed that 45 ± 8% of the proliferating cells were Iba1⁺ at 8 dpi, 65 ± 8% of the proliferating cells were Iba1⁺ at either 12 or 14 dpi, 91 ± 3% of the proliferating cells were Iba1⁺ at either at 18 or 20 dpi, and 68 ± 11% of the proliferating cells were Iba1⁺ at either 25 or 30 dpi. Scale bar, 10 μm. h, Histogram showed mRNA fold changes ± SEM for Iba1 (4.3 ± 0.7; p < 0.01), CD45 (8.3 ± 2.7; p < 0.05), and TNFα (10.22 ± 0.9; p < 0.01). Total RNA was extracted from laser-captured microdissection of striatum, respectively, from HC and EAE 20 dpi brains and then used for quantitative real-time PCR. *p < 0.05; **p < 0.01.

Figure 3.

Activated microglia reproduces the synaptic defects of EAE mice. a, b, The histograms show that both sEPSC half-width and decay time increased in the presence of activated microglia (BV2 microglia and primary microglia) but not in the presence of nonactivated microglia. **p < 0.01. c, The electrophysiological traces are examples of sEPSC mean peak obtained by group analysis in the presence of activated and nonactivated BV2 microglia. d, Microglia failed to alter sEPSC rise time. e, Activated BV2 microglia failed to further increase sEPSC half-width and decay time in EAE mice.

Figure 4.

TNFα reproduces the synaptic defects of EAE mice. a, The electrophysiological traces are examples of sEPSC mean peak obtained by group analysis in control conditions and in the presence of TNFα. b, The histogram shows that both sEPSC half-width and decay time were increased in the presence of TNFα. **p < 0.01. c, TNFα failed to further increase sEPSC half-width and decay time in EAE mice. d, Activated microglia failed to alter sEPSC kinetic properties after the blockade of endogenous TNFα activity with TNFR–Ig.

Figure 5.

Arc/Arg3x.1 mRNA levels were downregulated in Th1-treated primary neuronal cell cultures and in EAE. a, Primary neurons obtained from CD1 E16.5 embryos were treated with a Th1 mix or saline [control (C)] for, respectively, 3, 6, and 24 h (n = 5 independent cell cultures). After the removal of the stimulus, neurons did not show any morphological changes with respect to untreated ones (data not shown). RT-PCR analysis showed that CD119 and p55 expression was maintained in Th1-treated neurons. b, Then, total extracts from Th1 at 3, 6, and 24 h and vehicle were applied on 10% SDS-PAGE, followed by Western analysis with, respectively, α-GluR1, α-pSer845 GluR1, α-PSD95, and α-TuJ1 as housekeeping gene. Histograms to the right of blots show fold induction (FI) over vehicle (±SD). GluR1 was significantly upregulated during either 6 or 24 h of Th1 treatment: 3 h, 1.6 ± 1.7, NS; 6 h, 1.7 ± 0.4, *p < 0.05; 24 h, 2.6 ± 0.6, *p < 0.05. GluR1 Ser845 was significantly upregulated after either 6 or 24 h of Th1 treatment: 3 h, 0.1 ± 0.01, NS; 6 h, 3.3 ± 0.3, *p < 0.05; 24 h, 3.5 ± 0.2, *p < 0.05. PSD95 was significantly upregulated during either 6 or 24 h of Th1 treatment: 3 h,1.0 ± 0.07, NS; 6 h, 3.1 ± 0.2, *p < 0.05; 24 h, 2.7 ± 0.5, *p < 0.05. c, Representative micrograph of vehicle-treated neurons stained for PSD95 (green). Application of Th1 treatment increases the number of PSD95-expressing synaptic boutons in cultured neurons. d, Arc/Arg3.1 mRNA levels were measured in neuronal cultures at 3, 6, 12, and 24 h after Th1 treatment by using a real-time RT-PCR assay and the mRNA levels, expressed as percentages ± SEM. *p < 0.05. Histogram in e shows bouton fold change over controls (3 h, 2.0 ± 1.1, *p < 0.05; 6 h, 2.5 ± 1.6, *p < 0.05; 24 h, 2.45 ± 1.1, *p < 0.05). f, HC and EAE brain coronal sections probed by in situ hybridization for Arc/Arg3.1 mRNA detection display Arc/Arg3.1 mRNA expression within the cortex and within the striatum. Right panels show Arc/Arg3.1 mRNA expression, respectively, in EAE 20 and 30 dpi brains. EAE basal neocortical Arc/Arg3.1 mRNA levels are conserved but were strongly downregulated within EAE striatum. Scale bars: c, 10 μm; f, 150 μm.

Figure 6.

Blockade of AMPA receptors ameliorates the clinical and synaptic deficits in EAE. a, Time course of the clinical score of EAE mice treated or not with NBQX, blocker of AMPA/kainite receptors. b, The histogram shows the quantification of spine density in the three experimental conditions. **p < 0.01 versus EAE; ^# p < 0.05 versus HC. c, Examples of single-section Golgi preparations showing dendrites from HC and EAE mice receiving intraperitoneal injections of NBQX or vehicle. Scale bar: all three panels, 8 μm.