Circulating microRNAs as prognostic biomarkers of diabetes reversal in NOD mice treated with proinsulin- and IL-10 secreting Lactococcus Lactis in combination with low-dose anti-CD3

Background

Oral administration of live genetically modified Lactococcus lactis (LL) secreting human proinsulin and IL-10 together with a short-course of low-dose anti-CD3 has been demonstrated to reverse T1D in NOD mice. Although LL therapy was successful in around 60% of NOD mice, 40% were metabolically non-responsive. Peripheral biomarkers to identify treatment-responsive populations may improve clinical translation of the LL therapy. 

Methods

TaqMan array cards were used to assess microRNA profiles in plasma samples collected at diabetes onset in glycemia-matched NOD mice prior to therapy, ultimately comparing 6 LL therapy responders and 6 non-responders. MicroRNAs differentially expressed in the profiling experiment were then validated by microRNA single assays in a second cohort of 15 responders and 15 non-responders.

Results

4 microRNAs were upregulated in plasma of LL therapy non-responders vs responders at disease onset: miR-34a, miR-125a-3p, miR-193b, and miR-365. The combination of miR-193b and miR-365 showed good prediction power (sensitivity: 86%; specificity: 80%). In silico pathway analysis, based on inferred microRNA target genes, identified possible metabolism-related signalling pathways predicting LL therapeutic success. The microRNA prognostic signature was specific for LL therapy, because the 4 microRNAs did not differ between plasma samples taken at disease onset from responders and non-responders given non-LL therapy (either anti-CD3 mono- or anti-CD3+IL-1Ra). Circulating microRNA signatures identified at disease onset in LL responders may be used to identify T1D patients that would benefit of this promising antigen-based immunotherapy.