Coordinated Research Pojects ARC Fundings

Lipids and microRNAs from Extracellular Vesicles in Multiple Sclerosis: understanding  disease pathophysiology to design innovative nanomedecines (EVIMS) 

Multiple sclerosis (MS) is the most common cause of non-traumatic neurological disorders of the central nervous system (CNS) affecting young adults. It is characterized by multifocal inflammatory demyelinating lesions and axonal loss. The study of novel bioactive molecules contained in extracellular vesicles (EVs), trafficking to and within the CNS, represents an unprecedented opportunity to unravel novel signaling pathways involved in disease pathogenesis, and should lead to the development of new therapeutic strategies. We hypothesized that specific miRNAs and bioactive lipid species, circulating in EVs, are dysregulated in MS and influence both inflammatory and demyelinating processes.  

Thus, this project aims to identify novel bioactive molecules, isolated from EVs, to study dysregulated pathways in MS that could be amenable to corrective therapeutic approaches using nanomedicines. The main objectives of the project were: (i) identify relevant miRNAs and lipids in EVs of MS patients in different phases of the disease compared to control subjects, (ii) study their impact on the immune response, neuroinflammation and myelin disruption/repair, and (iii) develop new nanomedicines that will protect and allow the non-invasive delivery of the identified miRNAs and lipids (or of molecules modulating their effects).  

WP1: We isolated and compared EVs from the cerebrospinal fluid (CSF) of MS patients, stratified according to disease activity, as well as control subjects (VvP). Although we encountered technical challenges in our EV isolation protocol, due to the low amounts of EVs in the CSF, we finally succeeded in performing a qPCR-based array on a subset of CSF samples. We then confirmed the upregulation of 2 EV-associated miRNAs from MS patients and validated their effect on inflammation (VvP & GM). MiR-21-5p and miR-632 were enriched in CSF-EVs from MS patients, independently of disease activity. Our in-silico analysis, coupled with proteomic analysis, also allowed us to identify and confirm CC2D1A as a target of miR-632, along with GSK3β, proteins that may mediate neuroinflammation and demyelination (Deltombe et al, manuscript under revision). In addition, we studied the effect of several MS-dysregulated miRNAs on the differentiation of CG-4 oligodendrocyte progenitor cells, indicating that MS-associated miRNA changes are not only biomarkers but can also interfere with remyelination processes. This supports WP1’s central hypothesis that miRNAs may contribute to demyelination/remyelination imbalance and therefore represent relevant therapeutic targets1.  

WP2: We then assessed the bioactivity of miR-632 in vitro, in a primary culture of glial cells, on neuro-inflammation (GM, VvP, AdR). We also tested miR-124 and miR181c, as they were identified as dysregulated in MS patient serum by VvP’s team2. MiR-632 demonstrated anti-inflammatory effects in primary glial cell cultures (Figure 1). MiR1243 and 181c also reduced BV2 cells (microglial cell line) activation. We encounter several limitations related to the sensitivity of the LC-MS method regarding the analysis of the lipids present in EV extracted from patients’ CSF. Nevertheless, we were able to characterize the lipid composition of EV isolated from mesenchymal stem cells in steady and activated states4 (GM&AdR). As planned in the project as an alternative strategy, GM then identified bioactive lipids of interest to be tested for their anti-inflammatory potential. Among these lipids, glycerol ester (PG-Gs) and ethanolamide (PG-EA) derivatives of the prostaglandins (i.e. PGD2, PGE2, PGF2α) were particularly explored. These lipids and their prostaglandins counterparts showed differential effects when tested using primary murine glial cell cultures. Among the tested lipids, PGD2-G decreased TNFα and Iba1 expression in the culture system used (GM&VVP). To assess the properties of this lipid in vivo in CNS diseases, AdR developed lipid nanoparticle formulations (LNC) aiming at improving CNS delivery of the molecule. These formulations were tested alone and in combination with calcitriol in vitro and in vivo models of MS5,6 (AdR&GM).

WP3: We developed a protocol for incorporating lipids into EV, initially using the lipid anandamide (AEA) to explore and optimize encapsulation conditions (GM&AdR). Our results showed that incubating EV with lipids at 37°C for 2 hours was the most efficient method, compared to the widely used saponin-based approach7. Importantly, we identified a critical but rarely discussed bias in the literature. Indeed, the typical saponin concentrations used exceed their critical micellar concentration, leading to lipid encapsulation in saponin micelles rather than in EV. Since these micelles cannot be separated from EV, this results in a systematic overestimation of drug loading in EV. To further refine the process, we compared post-loading purification methods to balance drug removal efficiency and EV recovery. Building on these findings, we encapsulated the bioactive lipid 15d-PGJ2 into EV isolated from microglial cells (GM&AdR). We first characterized EV derived from resting (M0), activated (M1), and pro-resolutive (M2) microglial cells, analyzing their size, number, surface markers, proteome, and RNA content (AdR&GM). Finally, we demonstrated that 15d-PGJ2-loaded M0 EV effectively modulate microglial activation in vitro (Auquière et al. paper in preparation).  

To encapsulate miR181c in EV, we first tested an endogenous approach by transducing BV2 cells with a commercial plasmid encoding an exomotif linked to miR181c (AdR, GM, VvP). While this increased miR181c expression in both cells and their derived EV, the yield was insufficient for therapeutic applications as only 1 miRNA molecule was loaded per 20000-400000 EV. We then shifted to an alternative strategy: encapsulating miR181c in lipid nanoparticles (LNP), a platform previously optimized for miR124 delivery to modulate neuroinflammation3 (AdR, GM, VvP). This approach successfully reduced inflammation in vivo when LNP-miR181c were locally injected into the brains of LPS-activated mice (Ma, Auquière et al. paper in preparation, Figure 2).

In conclusion, this work, based on a truly interactive multidisciplinary consortium, revealed new roles for miRNAs as crucial molecular players in (neuro-)inflammation regulation and glial physiology. Importantly, the project also aimed to provide a novel therapeutic strategy by developing new nanomedicines containing therapeutic lipids and miRNAs that will address unmet therapeutic needs in this disease. This work ultimately reinforced the partnership between the three research groups and promoted the creation of a translational multidisciplinary research pole for MS at UCLouvain.