Poster Presentation Lorne Infection and Immunity 2022

Rabies against the machine: using artificial intelligence to find cellular and exosomal microRNA signatures of lyssavirus infection in a human stem-cell derived neural culture (#147)

Ryan Farr 1 , Nathan Godde 2 , Christopher Cowled 1 , Vinod Sundaramoorthy 2 , Diane Green 2 , Cameron Stewart 1 , John Bingham 2 , Carmel O'Brien 3 4 , Megan Dearnley 2
  1. Health and Biosecurity at the Australian Centre for Disease Preparedness (ACDP), CSIRO, Geelong, Victoria, Australia
  2. Australian Animal Health Laboratory at the Australian Centre for Disease Preparedness, CSIRO, Geelong, Victoria, Australia
  3. Manufacturing, CSIRO, Melbourne, Victoria, Australia
  4. Australian Regenerative Medicine Institute, Monash University, Clayton, Victoria, Australia

Rabies (lyssavirus) continues to have a significant impact on global mortality, especially in children under 15 years of age. This virus is extremely well-adapted to avoid host detection while travelling through the nerves to the brain. Until recently, research to fully understand the role of non-coding RNA in the pathogenesis of this neurotropic virus has been hampered by a lack of human models. Recently, we described the use of human stem cell-derived neural cultures to model lyssavirus infection [1]. Here, we have utilised this model to investigate the effect of lyssavirus infection on microRNA (miRNA) expression in human neural cells and the exosomes they secrete. We found 25 cellular and 16 exosomal miRNAs that were differentially expressed (FDR p-value <0.05) in response to infection with different lyssavirus strains. Supervised machine learning was then used to find combinations of cellular and exosomal miRNAs that are indicative of lyssavirus infection. A signature of 6 cellular miRNAs (miR-99b-5p, miR-346, miR-5701, miR-138-2-3p, miR-651-5p, and miR-7977) or 4 exosomal miRNAs (miR-25-3p, miR-26b-5p, miR-218-5p, miR-598-3p) could distinguish infected neural cells with almost 100% accuracy, giving an insight into the molecular pathogenesis pathways and highlighting potential biomarkers for rabies detection. Many of the miRNAs identified through machine learning, such as miR-5701 and miR-99b-5p, have established roles in regulating neuronal cell death, neuroprotection and neuroregeneration, and some have been found to be dysregulated in other neuronal disorders, such as Parkinson’s disease, Alzheimer’s disease, and amyotrophic lateral sclerosis (ALS). Interestingly, other miRNAs, such as miR-138-2, correlate with behavioural changes that are also seen in rabies infection, including panic, anxiety, confusion, impulsivity, and cognitive impairment. Together, this data lays the foundation for a more complete understanding of rabies molecular pathogenesis pathways, and highlights the use of exosomal miRNAs as biomarkers of rabies infection for improved detection and patient outcomes.

  1. 1. Sundaramoorthy, V., et al., Modelling Lyssavirus Infections in Human Stem Cell-Derived Neural Cultures. Viruses, 2020. 12(4).