THE ROLE OF CELLULAR AND EXOSOMAL NEURAL-DERIVED LONG NON CODING RNA (LNCRNA) IN MULTIPLE SCLEROSIS: POTENTIAL BIOMARKERS OF DISEASE SUSCEPTIBILITY AND PROGRESSION

Long non-coding RNAs (lncRNAs) are a novel class of transcripts that are pervasively transcribed in the genome. Several lines of evidence correlate dysregulation of different lncRNAs to human diseases including neurological and autoimmune disorders, but their expression has not been exhaustively investigated in MS so far. The main aim of this study was to identify a specific signature of cellular and neural-derived exosomal lncRNA expression. Regarding lncRNA expression levels from Peripheral Blood Mononuclear Cells (PBMC), we studied a discovery cohort of MS patients who were compared against controls. Results were validated in a larger cohort and further replicated in an independent Belgian population. LncRNA PCR arrays from System Bioscience (SBI) containing 90 common lncRNAs were used to screen lncRNA expression levels in PBMC from 5 patients with Relapsing Remitting (RR)-MS, 5 with Primary Progressive (PP)-MS and 5 age-matched controls. Results were validated by real time PCR in a further independent Italian cohort consisting of 30 PBMC samples from MS patients and 30 controls. Best hits were replicated using droplet digital PCR in a Belgian cohort consisting of 24 MS patients and 23 controls. In particular, in the Italian validation cohort ANRIL, TUG1, XIST (p<0.0001) and SOX2OT (p<0.001) were strongly down-regulated in RR-MS versus controls, while GOMAFU, HULC (p<0.0001) and BACE-1AS (p<0.001) showed a robust down-regulation both in RR and Progressive MS in comparison with controls. NRON and TUG1 downregulation in MS patients, compared with controls (p<0.05 and p<0.0001 respectively), was confirmed in the Belgian population. In addition, a protocol for the extraction and characterisation of neural-derived exosomes has been developed in order to investigate exosomal lncRNA expression levels. Using two types of commercial arrays, the human RT2 lncFinder array (QIAGEN) and the human RT2 lncRNA inflammation response and autoimmunity array (QIAGEN), generalised deregulation in exosomal lncRNA was observed. Moreover, the expression pattern of these molecules was different in RR-MS and in PP-MS. Precisely, results from the human RT2 lncFinder array (QIAGEN) analysis led to the identification of 7 most significantly deregulated lncRNAs, precisely AIRN (5.30-fold increase over controls, p=0.04); FAS-AS1 (4.76-fold increase over controls, p=0.02); HOTAIR (4.47-fold increase over controls, p=0.03); NAMA (13.24-fold increase over controls, p=0.01); TRERNA1 (5.84-fold increase over controls, p=0.01) and HOXA-AS2 (0.56-fold increase over controls, p=0.04). Six lncRNA were significantly deregulated in the RR-MS subgroup, precisely AIRN (10.77-fold increase over controls, p=0.04); DLX6-AS1 (46.95-fold increase over controls, p=0.01); FAS-AS1 (11.37-fold increase over controls, p=0.001); HOTAIR (9.31-fold increase over controls; p=0.02); and TRERNA1 (6.61-fold increase over controls, p=0.003). In PP-MS only SOX-2OT showed a significant upregulation (8.95-fold increase over controls, p=0.02). When we used the array containing lncRNA linked with inflammation and autoimmunity, MZF-AS1 (0.47-fold decrease over controls, p=0.03), CEP83-AS1 (0.15-fold decrease over controls, p=0.02), RP11-282O18.3 (0.27-fold decrease over controls, p=0.02), RP11-84C13.1 (0.28-fold decrease over controls, p=0.04), SNHG7 (0.064-fold decrease over controls, p=0.04) and TP73-AS1 (0.48-fold decrease over controls, p=0.04) were significantly downregulated in MS, regardless of the subtype, while RP11-38P22.2 (19.5-fold increase over controls, p=0.04) showed an upregulation. Considering the disease subgroups, RR-MS patients showed a significant downregulation in RP11-363G2.4 (0.07-fold decrease over controls, p=0.008) and in TP73-AS1 (0.76-fold decrease over controls, p=0.02), while RP11-38P22.2 levels were upregulated (22.32-fold increase over controls, p=0.04). We found a general downregulation in lncRNA expression analysed in PP-MS, in particular FGF14-IT1 (0.08-fold decrease over controls, p=0.007) and RP11-282O18.3 (0.14-fold decrease over controls, p=0.04) were significantly altered. Some important forms of dysregulation were observed, considering the expression levels of lncRNAs known to be involved in brain function and in neurological and autoimmune disorders. The rationale of this study might then be used to set up a future study with the purpose of selecting potential biomarkers for disease aggressiveness and possible response to therapy.