Basic Science and Pathogenesis

Background: Herpes simplex virus (HSV-1) has been associated with molecular and cellular signatures associated with Alzheimer's disease (AD). We explored the use of both recent single-cell and bulk transcriptomics technologies in dissecting the molecular and cellular virus-human interactions with HSV-1 infected cerebral organoids (2D and 3D). We compared the results with our previous observations from bulk RNA sequencing and discovered novel insights into HSV-1 induced AD-associated molecular pathology that were made possible by each transcriptomics technology.

Method: We used recent emerging technologies such as single-cell spatial RNA sequencing (STOmics Stereo-seq) and single-cell non-spatial RNA sequencing (Parse Evercode). In addition, we have data generated from our previous work using bulk RNA sequencing.

Result: Previously, we found that differentially expressed genes in HSV-1 infected dissociated cells from cerebral organoids (2D cOrgs) using bulk RNA sequencing data were exclusively enriched for AD-associated genes implicated through genome-wide association studies (GWAS), but not for genes associated with other neurodegenerative or autoimmune diseases. UMAP analyses on single-cell non-spatial RNA sequencing data with GFP-tagged HSV-1 infected 2D cOrgs clustered by viral transcripts [Fig. 1] revealed huge differences in the proportions of true late viral transcripts but not of leaky late viral transcripts between clusters, while similar clustering of 3D cOrgs [Fig. 2] showed distinct proportions of both true late and leaky late transcripts. Pseudobulk analyses showed no enrichment for AD-associated GWAS genes among the differentially expressed genes for HSV-1 infected 2D cOrgs versus uninfected 2D cOrgs, but revealed significant enrichment for GWAS genes associated with autoimmune diseases such as Type 1 diabetes and multiple sclerosis. Similar analyses of HSV-1 infected 3D cOrgs versus uninfected 3D cOrgs did not show enrichment in GWAS gene lists across the 21 common neurodegenerative, neuropsychiatric or autoimmune diseases that we had surveyed.

Conclusion: Taken together, our use of multi-transcriptomics technologies has enabled us to gain course to fine molecular granularity into neuroinflammation-induced AD viral-human gene interactions and a careful dissection of the multi-transcriptomics data enables us to gain insights into biomarkers and gene targets for AD therapeutics development in the near future.