Background

Ependyma forms the epithelial interface boarding the ventricular walls and the spinal cord’s central canal, maintaining crucial central nervous system functions such as the regulation of cerebrospinal fluid (CSF) circulation by synchronous ciliary beating and the monitoring of molecular exchanges between CSF and parenchyma. These functions are dependent of the cellular coupling via gap junction channels.

Neuromyelitis Optica (NMO) is associated with autoantibodies (NMO-IgG), directed against aquaporin 4 (AQP4). NMO-IgG are present in patient’s serum and CSF during attacks, and are known to trigger astrocyte dysfunction leading to demyelination and axonal loss. Interestingly, ependymal cells also express AQP4 and evidences of ependymal alteration have been reported in NMO. Indeed, MRI abnormalities of the ventricles have been reported, and pathological analyses showed AQP4 loss in the ventricular walls of NMO patients.

Objectives

Our aim is to evaluate if NMO-IgG target ependymal cells, and lead to ependymal morphological changes and dysfunctions.

Methods

We used purified NMO-IgG from AQP4 antibodies positive patients’ plasma (NMO-IgG AQP4⁺). IgG from healthy donors (CTRL-IgG) and IgG from AQP4-antibodies negative patients (NMO-IgG AQP4^-) were used as controls. We evaluated ependymal cells on two models: primary ependymal cell cultures and cultured wholemount dissections (“en-face” view of the entire ependyma) of adult rat lateral ventricular walls. We first looked at the effect of different IgGs on the expression of AQP4 and Connexin43, the main gap junction channel expressed by ependymal cells. Then, we assessed the ciliary beating of ependymal cells by analyzing the diffusion of india ink deposited on wholemount preparations. Coupling of ependymal cells in culture was evaluated by measuring the diffusion of Lucifer Yellow (LY), a gap junction specific dye incorporated in cells by scrape loading.

Results

We showed that NMO-IgG AQP4⁺ exposure induced: 1) the delocalization of AQP4 membrane expression and morphological changes of ependymal cells from primary cultures and wholemounts, by contrast different IgG controls had no effect; 2) morphological changes of cilia in primary cultures and alteration of the ciliary dynamic; 3) the alteration of Connexin43 expression and function, demonstrated by the decrease of LY spreading after scrape loading.

Conclusions

These results suggest that NMO-IgG directly induce dysfunctions of ependymal cells, through the perturbation of AQP4 and Cx43 expression and functions.

Background

To predict the clinical course of myelin oligodendrocyte glycoprotein (MOG)-antibody (Ab)-associated disease (MOGAD) is essential to guide treatment recommendations.

Objectives

We aimed to 1) compare clinical features and disease course, and 2) to evaluate the association of MOG-Ab dynamics and relapses, between children and adults with MOGAD.

Methods

Retrospective study evaluating clinical features of 98 children and 266 adults with MOGAD, between January 2014 and September 2019. To analyse relapses over the whole disease course, a Cox regression analysis for recurrent time-to-event data was performed, introducing treatment as time-dependent covariate. To evaluate dynamics, delta mean fluorescence intensity ratio signal (ΔMFIratio) of MOG-Ab was measured in patients with a minimum time elapsed between two samples of 4 months.

Results

Median age at onset of symptoms was 10.9 (interquartile range 5.4-14.3) years in children and 36.2 (27.7-47.6) in adults. Isolated optic neuritis was the most frequent clinical presentation both in children (40.8%) and adults (55.9%), p=0.013, and acute disseminated encephalomyelitis syndrome was more frequent in children (36.7% vs. 5.6%; p<0.001). Compared to adults, children displayed a better recovery (EDSS ≥3.0 at last follow-up reached only by 10 of 97 [10.3%] vs. 66/247 [26.7%], p<0.001).

In the multivariate analysis, adults were at higher risk of relapse than children (Hazard ratio 1.41, 95%Confidence interval [CI] 1.12-1.78; p=0.003). Among the 124 participants evaluated for MOG-Ab dynamics, 36.3% became seronegative, 60.5% decrease and 3.2% increase the ΔMFIratio. At two years, 64.2% (95%CI 40.9-86.5) of non-relapsing children became MOG-Ab negative compared to 14.1% (95%CI 4.7-38.3) of relapsing ones, log-rank p<0.001, with no differences observed between non-relapsing and relapsing adults, log-rank p=0.280.

Conclusions

MOGAD differs in its clinical presentation at onset, showing a progressive shift in the clinical features across age-groups. Compared to children, adults have a higher risk of relapses and a worse functional recovery. Finally, children with monophasic disease became MOG-Ab negative earlier than relapsing ones, but not in adults. Considering these differences, management and treatment guidelines should be considered independently in children and adults.