Summary: Scientists at Washington University School of Medicine have developed a lab method to convert patient skin cells into mature neurons that retain age-related features, enabling accurate study of late-onset Alzheimer’s disease. These directly reprogrammed neurons reproduce key Alzheimer’s hallmarks—amyloid beta accumulation, tau pathology and neuronal loss—allowing detailed analysis of aging-related mechanisms and testing of potential therapies.
Using this cellular reprogramming approach, researchers identified age-associated changes in retrotransposable elements in the genome that appear to contribute to disease progression. Suppressing these so-called “jumping genes” with the antiviral drug lamivudine (3TC) reduced pathology in lab-grown neurons, supporting early intervention strategies that target aging-linked mechanisms.
Key Facts:
- Researchers directly converted patient skin cells into neurons to model late-onset Alzheimer’s disease (LOAD) in the lab.
- Age-related changes in retrotransposable elements were linked to pathological features of LOAD.
- Lamivudine (3TC), an inhibitor of retrotransposable element activity, reduced amyloid, tau and neuronal death in LOAD-derived neuron models.
Source: WUSTL
Researchers at Washington University School of Medicine in St. Louis report a practical way to capture aging effects in neurons without a brain biopsy, advancing study of the most common form of Alzheimer’s disease. Their method makes it possible to examine how aging contributes to late-onset disease and to test interventions directly in patient-derived neural tissue.
The team converted skin cells from people with late-onset Alzheimer’s into neurons using a reprogramming protocol that preserves cellular age. Late-onset Alzheimer’s typically appears after age 65 and represents the majority of clinical cases. The new neuronal models reproduce hallmark neuropathology—amyloid beta deposits, insoluble tau aggregates and progressive neuronal death—within a three-dimensional culture context.
To better mimic brain tissue, the converted neurons were cultured either on a thin gel layer or allowed to self-assemble into compact, three-dimensional spheroids. When researchers compared spheroids made from patients with sporadic late-onset Alzheimer’s, inherited early-onset Alzheimer’s, and cognitively healthy age-matched donors, distinct differences emerged. Spheroids from late-onset patients rapidly developed amyloid plaques and tau tangles, activated inflammatory gene programs and then exhibited neuron loss—closely mirroring clinical disease progression. Older healthy donors showed modest amyloid accumulation, demonstrating that the model captures aging effects while distinguishing disease-driven pathology.
The team also evaluated the timing of treatments that interfere with amyloid formation. Early intervention—before plaques formed—significantly reduced amyloid accumulation, tau pathology and downstream neurodegeneration. Treatments applied later, after deposits had already appeared, had limited impact, underscoring the importance of early detection and therapy in Alzheimer’s.
A central molecular insight from the study implicates retrotransposable elements (RTEs)—segments of DNA capable of changing position in the genome—in LOAD pathogenesis. Activity of these elements increases with age in both healthy older donors and patients, but appears to contribute more strongly to pathology in Alzheimer’s-derived neurons. Pharmacological suppression of RTE activity with lamivudine (3TC) reduced amyloid and tau accumulation, lessened neuronal death and decreased markers of DNA damage and inflammation in LOAD spheroids. Notably, lamivudine did not reduce pathology in models of inherited, early-onset Alzheimer’s, highlighting distinct molecular drivers between sporadic, age-related and familial forms of the disease.
Senior author Andrew Yoo, PhD, emphasizes that this model provides a patient-specific, age-preserving platform for studying how aging increases vulnerability to neurodegeneration and for identifying therapeutic targets unique to late-onset Alzheimer’s.
The researchers plan to extend these studies by building more complex spheroids that include multiple brain cell types, such as neurons and glia, to better capture cell–cell interactions that influence disease progression. By combining direct neuronal reprogramming with three-dimensional culture and targeted interventions, the work opens new opportunities to test treatments that address age-related mechanisms in Alzheimer’s and to pursue personalized therapeutic strategies.
About this Alzheimer’s disease research news
Author: Jessica Church
Source: WUSTL
Contact: Jessica Church – WUSTL
Image: The image is credited to Neuroscience News
Original Research: Closed access.
“Modeling late-onset Alzheimer’s disease neuropathology via direct neuronal reprogramming” by Andrew Yoo et al. Science
Abstract
Modeling late-onset Alzheimer’s disease neuropathology via direct neuronal reprogramming
INTRODUCTION
Extracellular amyloid-β (Aβ) accumulation, insoluble tau formation and neuronal loss are core neuropathological hallmarks of Alzheimer’s disease (AD). Most experimental AD models have focused on rare genetic mutations that cause early-onset, autosomal dominant AD (ADAD). Modeling the age-associated neuropathology of sporadic late-onset AD (LOAD), which accounts for over 95% of cases, has been challenging because many platforms lose age-related cellular features. Induced pluripotent stem cell–derived neurons tend to reset to a fetal-like state, while direct reprogramming of patient somatic cells can preserve cellular aging signatures.
Here, researchers used brain-enriched microRNAs (miR-9/9* and miR-124) as efficient reprogramming factors to generate LOAD neurons that retain age-related properties in a three-dimensional environment, providing a robust platform to capture late-onset, age-associated AD phenotypes.
RATIONALE
Neurons produced by 3D direct conversion of fibroblasts from LOAD patients preserve both genetic background and cellular age. The authors hypothesized that miRNA-induced LOAD neurons would reproduce late-onset neuropathological features driven by aging, enabling study of age-linked molecular mechanisms and intervention testing.
RESULTS
As proof of concept, the team generated cortical neurons from fibroblasts of individuals with ADAD and LOAD using miR-9/9*-124 combined with neuronal transcription factors. Neurons were cultured in two 3D formats: a thin Matrigel layer and high-density, self-assembled neuronal spheroids. Compared with age-matched control neurons from cognitively normal donors, 3D AD neurons displayed extracellular Aβ accumulation, seed-competent and insoluble tau, dystrophic neurites and neurodegeneration. Applying the same strategy to fibroblasts from LOAD patients produced hallmark AD neuropathology as well.
Inhibiting APP processing early in reprogramming lowered subsequent Aβ deposition, tauopathy and neurodegeneration, whereas late treatment after Aβ deposits formed was ineffective. LOAD neurons also showed gene expression changes linked to neuroinflammation. Both aged healthy control and LOAD neurons (ages 66–90) exhibited altered expression of retrotransposable elements relative to younger control neurons (ages 36–61). Treating LOAD neurons with lamivudine (3TC) to disrupt age-associated RTE dysregulation reduced Aβ and tau pathology, neuronal death and DNA damage, alongside changes in inflammation-related gene expression.
CONCLUSION
miRNA-induced, directly reprogrammed neurons cultured in 3D form a feasible and sufficient human model for late-onset Alzheimer’s neuropathology. These neurons provide a patient-based platform to investigate how aging increases susceptibility to neurodegeneration in LOAD and to identify aging-related mechanisms and therapeutic targets. Future work will expand these models to include additional aging processes, examine AD risk genes expressed in neurons and incorporate other brain cell types to better reflect in vivo pathology in patient-derived neural tissues.