Patient-Derived PSEN1 Cerebral Organoids Revealed Parallel Development of Amyloid-β Accumulation and Network Dysfunction

1. Problem Statement

Alzheimer's Disease (AD) is characterized by the accumulation of amyloid-β (Aβ) plaques and progressive neurodegeneration. While animal models have advanced understanding, species-specific differences limit their translational value. A critical gap in current research is the temporal relationship between Aβ dysregulation (specifically the shift toward the toxic Aβ42 isoform) and the onset of neuronal network dysfunction (hyperexcitability and hypersynchrony). It remains unclear whether Aβ accumulation drives early network hyperactivity or if these phenomena occur independently. Furthermore, the transition from early hyperexcitability to the later hypoexcitability observed in advanced AD needs better characterization in a human-relevant model.

2. Methodology

The study utilized a human-induced pluripotent stem cell (hiPSC)-derived cerebral organoid model to investigate these dynamics over an extended maturation period (Differentiation Day 60 to Day 139).

• Cell Lines:
  • AD Model: hiPSCs from a patient carrying the familial AD-causing PSEN1 A246E mutation.
  • Control: hiPSCs from a healthy, unrelated wild-type (WT) donor.
• Organoid Differentiation: Organoids were generated using a modified Lancaster protocol, maturing through neuroepithelial organization to a cortical-like identity. They were maintained in BrainPhys medium to support synaptic activity.
• Electrophysiology (MEA):
  • Technique: Multielectrode Array (MEA) recordings were performed twice weekly from DD60 to DD139.
  • Parameters: Spontaneous activity was analyzed for spike parameters (active electrodes, frequency, amplitude), burst parameters (burst count, duration, intraburst frequency), and synchrony (Global Synchrony Index - GSI).
• Pathological Characterization:
  • Immunohistochemistry: Staining for Aβ, neuronal markers (NeuN, MAP2, TUJ1), and progenitor markers (SOX2, PAX6) to assess morphology and Aβ plaque formation.
  • Biochemistry: ELISA was used to quantify secreted Aβ40 and Aβ42 in conditioned media to calculate the Aβ42/40 ratio.
  • Molecular Biology: qPCR analyzed gene expression of neuronal and glial markers.
• Statistical Analysis: Non-parametric tests (Mann-Whitney, Kruskal-Wallis) were used for group comparisons, and Spearman's rank correlation was used to link Aβ levels with electrophysiological metrics.

3. Key Contributions

• Temporal Mapping: The study provides a high-resolution timeline of network dysfunction in human AD organoids, capturing the transition from hyperexcitability to hypoexcitability.
• Correlation of Biomarkers with Function: It establishes a direct, positive correlation between Aβ42/40 ratios, Aβ aggregate size, and network hyperexcitability/hypersynchrony specifically during the early stages of the disease model.
• Validation of the Model: It confirms that PSEN1 A246E organoids recapitulate the "amyloid cascade" (Aβ accumulation preceding tau pathology) and the specific electrophysiological phenotype (early hyperexcitability) seen in human patients and animal models.

4. Key Results

A. Pathological Phenotype

• Aβ Accumulation: AD organoids showed a progressive increase in intracellular Aβ signal and the formation of Aβ aggregates starting around DD90, whereas WT organoids showed minimal accumulation.
• Aβ42/40 Ratio: AD organoids exhibited a consistently elevated Aβ42/40 ratio from DD60 through DD130, confirming the pathogenic shift in APP processing driven by the PSEN1 mutation.
• Tau Pathology: Tau pathology was sparse and did not show a clear correlation with electrophysiological changes in this timeframe, supporting the hypothesis that Aβ drives early network dysfunction.

B. Electrophysiological Dynamics

• Hyperexcitability Phase (DD60–DD100):
  • AD organoids displayed a transient peak in network activity (peaking between DD72–DD82) significantly higher than WT controls.
  • Active Electrodes: The percentage of active electrodes was significantly higher in AD organoids, indicating more neurons were recruited into the network.
  • Synchrony: The Global Synchrony Index (GSI) was significantly elevated in AD organoids, indicating hypersynchronous firing.
  • Bursting: AD organoids showed robust bursting behavior with higher intraburst spike frequencies and longer burst durations.
• Hypoexcitability Phase (Post-DD100):
  • Following the peak, AD organoids exhibited a gradual decline in activity, eventually falling below or stabilizing at lower levels compared to the initial peak. This mirrors the transition from early hyperexcitability to late-stage synaptic failure and cognitive decline seen in patients.

C. Correlation Analysis

• Positive Correlation: In AD organoids, there was a significant positive correlation between:
  • Normalized Aβ42/40 ratio and the percentage of active electrodes ($r_s = 0.94$).
  • Normalized Aβ42/40 ratio and GSI ($r_s = 1.00$).
  • Aβ aggregate size and GSI ($r_s = 0.94$).
• Absence in Controls: No such correlations were observed in WT organoids, suggesting the link is specific to the AD pathology.

5. Significance and Implications

• Mechanistic Insight: The findings strongly support the Amyloid Cascade Hypothesis, suggesting that Aβ dysregulation is a primary driver of early network hyperexcitability and hypersynchrony, occurring before significant neurodegeneration or tau pathology.
• Clinical Relevance: The observed trajectory (early hyperexcitability $\rightarrow$ late hypoexcitability) aligns with clinical observations in AD patients, where seizures and epileptiform activity are more common in preclinical/prodromal stages, while cognitive decline and network disconnection dominate later stages.
• Translational Model: These patient-derived organoids serve as a robust translational platform for:
  • Investigating the mechanisms of Aβ-induced excitability.
  • Screening for therapeutics that target early network dysfunction (e.g., anti-epileptiform drugs) before irreversible neuronal loss occurs.
  • Understanding the specific impact of PSEN1 variants on human neural networks, which animal models may not fully replicate.

In conclusion, this study demonstrates that PSEN1 A246E cerebral organoids faithfully model the parallel development of Aβ accumulation and network dysfunction, providing a critical tool for dissecting the early pathophysiology of Alzheimer's disease.