Summary: Researchers have found reverse transcriptase (RT) enzymatic activity in aging human neurons and in brains affected by Alzheimer’s disease. That activity appears to come from truncated variants of LINE1 (L1) elements—ancient mobile genetic sequences that can encode RT. Because several FDA-approved HIV medications target RT, these drugs could potentially be repurposed for clinical trials to slow or prevent Alzheimer’s disease progression.
The study detected abundant, previously unannotated monocistronic L1 transcripts that encode RT, with particularly strong expression in gray matter where neurons accumulate somatic changes over a lifetime. The findings link endogenous RT activity in the human brain to a plausible molecular source and identify a potential near-term therapeutic strategy using existing RT inhibitors.
Key facts
- Endogenous RT activity in brain: All examined cortical samples showed RT enzymatic activity, concentrated in neuron-rich gray matter and detectable even in non-diseased aged brains.
- Monocistronic LINE1 variants: Long-read sequencing revealed thousands of truncated L1 transcripts, many not annotated in the reference genome, and many encoding only the ORF2 region that contains RT.
- Therapeutic implication: Because approved HIV therapies already inhibit RT, researchers suggest prospective clinical trials to evaluate these drugs in early Alzheimer’s disease.
Source: Sanford Burnham Prebys
Background: Alzheimer’s disease is the leading cause of dementia and affects over one-tenth of Americans aged 65 and older. Incidence is projected to rise substantially in the coming decades, increasing the need for new treatment approaches that preserve cognition and reduce caregiver and health-system burden.
Several retrospective analyses of medical records recently suggested that people taking HIV reverse transcriptase inhibitors had a lower incidence of Alzheimer’s disease. To investigate whether endogenous RT activity exists in the aging human brain and how it might arise, Jerold Chun, MD, PhD, and colleagues at Sanford Burnham Prebys assayed cortical tissue from donors with and without Alzheimer’s disease. Their findings were published online May 14, 2025 in The Journal of Neuroscience.
RT is well known for enabling retroviruses such as HIV to copy RNA into DNA. Prior work from the Chun laboratory proposed that RT-mediated somatic recombination of the amyloid precursor protein (APP) gene could occur in neurons and contribute to sporadic Alzheimer’s disease. The current study directly searched for enzymatic RT activity and the RNA sources that could encode it.
Using enzymatic assays, targeted long-read (PacBio HiFi) sequencing, and spatial transcriptomics, the team examined cerebral cortex samples from 31 donors (Alzheimer’s disease and non-diseased). RT enzymatic activity was present in every brain sample, though overall RT levels tended to be lower in terminal Alzheimer’s samples, consistent with neuronal loss. Spatial data showed RT expression was higher in gray matter and correlated with neuronal ORF2 expression, supporting a neuronal origin for most activity.
Unexpectedly, intact bicistronic full-length L1 transcripts were extremely rare (<0.01% of sequences). Instead, researchers identified widespread monocistronic L1 transcripts that include either ORF1 or ORF2 alone. More than 550 distinct protein-encoding ORF2 variants were observed—over twice the number identified in the human reference genome (hg38)—and hundreds of L1-derived transcripts were not previously annotated.
Functional tests showed that many truncated ORF2 variants encode active reverse transcriptase, and activity levels varied dramatically across variants—more than a 50-fold range for RT activity and modest variation in endonuclease activity. Those results indicate that monocistronic ORF2 transcripts could drive RT-mediated somatic genomic changes in post-mitotic neurons, potentially contributing to genomic mosaicism as the brain ages and in Alzheimer’s disease.
Juliet Nicodemus, the study’s first author, noted that prevailing assumptions held L1 function required intact bicistronic mRNA; the long-read sequencing data changed that view by revealing many truncated, functional transcripts. Chun emphasized the need to characterize the different ORF2 variants and their enzymatic behaviors to enable targeted interventions.
Because several RT inhibitor drugs are FDA-approved and have established safety profiles for treating HIV, the authors propose prospective clinical trials to test whether these agents can reduce or delay Alzheimer’s progression in people with early disease. Such trials could offer a near-term way to evaluate RT inhibition as a therapeutic strategy while further mechanistic research continues.
Additional contributors to the study include Christine S. Liu, Linnea Ransom, Valerie Tan, William Romanow, and Natalia Jimenez from Sanford Burnham Prebys. Funding sources included the National Institutes of Health, the National Institute on Aging, the Bruce Ford & Anne Smith Bundy Foundation, and the Larry L. Hillblom Foundation.
About this neuropharmacology and Alzheimer’s disease research news
Author: Greg Calhoun
Source: Sanford Burnham Prebys
Contact: Greg Calhoun – Sanford Burnham Prebys
Image: The image is credited to Neuroscience News
Original Research: Closed access. “Sequence diversity and encoded enzymatic differences of monocistronic L1 ORF2 mRNA variants in the aged normal and Alzheimer’s disease brain” by Jerold Chun et al., Journal of Neuroscience
Abstract
Sequence diversity and encoded enzymatic differences of monocistronic L1 ORF2 mRNA variants in the aged normal and Alzheimer’s disease brain
Endogenous reverse transcriptase (RT) activity has been inferred from somatic retroinsertion events in the brain, but direct characterization of enzymatic sources remained incomplete. L1 (LINE-1) retrotransposons typically express two proteins from a bicistronic transcript—ORF1 (RNA-binding) and ORF2 (containing RT and endonuclease domains)—to enable retrotransposition. Using enzymatic assays, targeted PacBio HiFi long-read sequencing, and quantitative spatial transcriptomics on cortical samples from Alzheimer’s disease and aged non-diseased donors, this study found that full-length bicistronic L1 transcripts were exceedingly rare. Instead, monocistronic ORF1 and ORF2 transcripts were widespread, with ORF2 expression enriched in neurons and gray matter. Over 550 protein-encoding ORF2 variants were detected, many not present in the reference genome. Functional overexpression experiments demonstrated a broad range of RT and endonuclease activities among ORF2 variants, supporting their potential to drive RT-mediated somatic recombination and genomic mosaicism in the aging and diseased human brain.