Extracellular RNA splice events in cerebrospinal fluid as candidate biomarkers of myotonic dystrophy type 1
Preeti Kumari1, Ningyan Hu1, Alex Sizemore1, Lauren Sullivan1, Brigham Mckee1, Parker Conquest1, Thurman Wheeler1
1Massachusetts General Hospital
Objective:
To evaluate extracellular RNA (exRNA) in cerebrospinal fluid (CSF) as a source of CNS-derived splice events that may serve as biomarker of DM1 disease activity.
Background:
Alternative splicing is mis-regulated in post-mortem DM1 central nervous system (CNS) tissue. In past few years Cerebrospinal Fluid (CSF) biomarkers have been used to characterize disease severity and prognosis of the disease and helpful tool for to measure changes after treatment. exRNA in CSF provides a source of CNS-derived splice events in living individuals.
Design/Methods:
We examined fresh CSF samples from unaffected controls (N=14) from the MGH CSF biobank and DM1 individuals with mild CNS involvement (N=2). To separate EVs from cells contained in CSF, we used low speed centrifugation followed by filtration of the supernatant. EV size and concentration were measured using microscopy and particle tracking analysis software (NanoSight). We extracted exRNA from the EV pellet, produced cDNA, and quantified gene expression and splice events (% exon inclusion) by droplet digital PCR.
Results:
Mean particle diameter is about 175 nm in CSF and serum vs. 220 nm in urine. In CSF cells, total RNA content is 10 - 100-fold higher than exRNA. Normalized expression of both DMPK and CNBP was 50 - 60% higher in CSF exRNA than in CSF cells. Splicing is significantly different in CSF exRNA vs CSF cells. Alternative splicing of GOLGA4 exon 23 inclusion was 65% in DM1 and 40% in UA (P<0.01). Differential splicing of NUMA1, NCOR2 and MBNL2 was non-significant between groups.
Conclusions:
Quantification of CNS-derived alternative splice products in CSF exRNA is feasible. Removal of leukocytes and erythrocytes normally found in CSF will enhance accurate quantification of CNS-derived splice events. GOLGA4 is a candidate biomarker of early CNS involvement in DM1, while NUMA1, NCOR2, and MBNL2 splice events are candidate biomarkers of disease severity.