Abstract
Reverse transcriptase (RT) activity in the human brain has been inferred through somatic retroinsertion/retrotransposition events, however actual endogenous enzymatic activities and sources remain unclear. L1 (LINE-1) retrotransposons bicistronically express ORF2, containing RT and endonuclease (EN) domains, and RNA binding protein ORF1, together enabling L1 retrotransposition and contributing to somatic genomic mosaicism (SGM). Here, we assessed endogenous RT activities and L1 mRNA diversity from cerebral cortical samples of 31 Alzheimer’s disease (AD) and non-diseased (ND) brains (both sexes) using enzymatic functional assays, targeted PacBio HiFi long-read sequencing, and quantitative spatial transcriptomics. Expected bicistronic, full-length L1 transcripts were absent from most samples, constituting <0.01% of L1 sequences, of which >80% were non-coding. Monocistronic ORF1 and ORF2 transcripts were identified across all samples, consistent with quantitative spatial transcriptomics that identified discordant ORF2 and ORF1 expression in neurons. All brains had RT activity, with AD samples showing less activity, consistent with neuronal loss of terminal AD vs. aged ND donors. Brain RT activity was higher in grey matter and correlated with increased neuronal ORF2 expression, further supporting neuronal contributions. Remarkably, >550 protein-encoding, polyA+ ORF2 sequence variants were identified, over 2x more than identified in the human reference genome (hg38). Experimental overexpression of full-length and truncated ORF2 variants revealed ∼50-fold RT and ∼1.3-fold EN activity ranges, supporting endogenous functional capacity of monocistronic ORF2 variants in the human brain. The vast sequence diversity of monocistronic ORF2 mRNAs could underlie functional differences in RT-mediated somatic gene recombination/retroinsertion and resulting genomic mosaicism in the normal and diseased brain.
Significance Statement Human brain reverse transcriptase activity has been inferred through the “copy-and-paste life-cycle” of L1, which can generate genomic mosaicism via self-retrotransposition via a full-length L1 mRNA. However, their presence in aged and Alzheimer’s disease neurons remains unclear. We examined aged normal and Alzheimer’s brains for reverse transcriptase activity in prefrontal and medial-temporal cortices and its relationship to L1 via enzymatic activity assays and targeted PacBio sequencing. Reverse transcriptase activity was pervasive, however full-length L1 was largely absent. Instead, hundreds of different, truncated, novel L1 mRNA variants were identified, and experimental sampling revealed diverse reverse transcriptase activities. These data implicate truncated L1 variants as a source of functionally diverse and novel reverse transcriptases in the normal and Alzheimer's disease brain.
Footnotes
Dr. Chun has an employment relationship with Neurocrine Biosciences, Inc., a company that may potentially benefit from the research results. Dr. Chun’s relationship with Neurocrine Biosciences, Inc. has been reviewed and approved by Sanford Burnham Prebys Medical Discovery Institute in accordance with its Conflict of Interest Policies.
Research reported in this publication was supported by the National Institute on Aging of the NIH under Awards R01 AG065541 and NIH R01 AG071465 and by non-federal funds from The Bruce Ford & Anne Smith Bundy Foundation and the Larry L. Hillblom Foundation (JC) and NIH T32 GM007198-42S1 and R01 AG065541-02/03S1 (JN). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Brain specimens were obtained from Dalhousie, Goizueta Alzheimer’s Disease Research Center at Emory University, NIH Neurobiobank (Sepulveda), Southwest Dementia Brain Bank, University of California San Diego ADRC, and Washington University. We also thank the donors and families who shared these precious brain materials. We thank Danielle Jones, Nicole Coufal, MD PhD, and Carter Palmer, PhD for their discussions and input; Laura Wolszon, PhD for her efforts to source and obtain human specimens; Kang Liu, PhD at the Sanford Burnham Prebys Medical Discovery Institute Genomics Core for RIN analysis of brain samples.
Author Contributions: JN designed and conducted all experiments; JN & CSL analysed the data; CSL & JN did bioinformatic analyses of sequencing data; LR aided in human post-mortem tissue sample selection and preparation; VT & WR aided in RT activity assays and cloning RT gene constructs; NJ aided in transfecting cells for studies of function in ORF2 variants; JN & JC wrote the manuscript; JC conceived the study, supervised experiments, and secured funding.
This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license, which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.
