Systematic In Vivo Analysis of the Intrinsic Determinants of Amyloid β Pathogenicity

Leila M. Luheshi, Gian Gaetano Tartaglia, Ann‐Christin Brorsson, Amol Pawar, Ian E Watson, Fabrizio Chiti, Michele Vendruscolo, David A. Lomas, Christopher M. Dobson, Damian C. CrowtherPLoS 2007 · DOI

Abstract

Protein aggregation into amyloid fibrils and protofibrillar aggregates is associated with a number of the most common neurodegenerative diseases. We have established, using a computational approach, that knowledge of the primary sequences of proteins is sufficient to predict their in vitro aggregation propensities. Here we demonstrate, using rational mutagenesis of the Aβ42 peptide based on such computational predictions of aggregation propensity, the existence of a strong correlation between the propensity of Aβ42 to form protofibrils and its effect on neuronal dysfunction and degeneration in a Drosophila model of Alzheimer disease. Our findings provide a quantitative description of the molecular basis for the pathogenicity of Aβ and link directly and systematically the intrinsic properties of biomolecules, predicted in silico and confirmed in vitro, to pathogenic events taking place in a living organism.

Introduction

Alzheimer’s disease is inextricably linked to the aggregation of the Amyloid $\beta$ (A$\beta_{42}$) peptide into insoluble plaques. While biophysicists have become proficient at predicting how mutations affect protein aggregation in vitro, the transition to a complex living brain (in vivo) introduces significant variables. Cellular environments contain "quality control" machinery, such as molecular chaperones and proteases, that can mitigate aggregation. This study addresses whether the intrinsic physicochemical properties of a protein are sufficient to predict its pathogenicity in a living system, or if cellular factors override these basic rules. Furthermore, the research seeks to identify which specific stage of the aggregation process—early intermediate "protofibrils" or final "mature fibrils"—is the primary driver of toxicity.

Methods and Datasets

The researchers employed a multidisciplinary approach bridging computational modeling, in vitro biophysics, and in vivo genetics.

In Silico and In Vitro Analysis

An established algorithm ($Z_{agg}$) was used to calculate the aggregation propensity of all 798 possible single-point mutations of the A$\beta_{42}$ peptide. From this library, 17 variants spanning a wide range of predicted propensities (e.g., F20E, E22G, M35F) were selected. To validate these predictions, the team performed in vitro kinetic assays using Thioflavin T fluorescence and electron microscopy to visualize aggregate morphology.

Drosophila Modeling

The selected variants were expressed in the central nervous system of Drosophila melanogaster. Toxicity was quantified by measuring median survival (longevity) and climbing ability (locomotor performance). Wild-type A$\beta_{42}$ and the non-toxic A$\beta_{40}$ peptide served as benchmarks for high and low toxicity, respectively.

Key Results and Insights

Sequence as a Predictor of Pathogenicity

The study's most striking finding is that the primary amino acid sequence remains the dominant predictor of toxicity, even within the complex environment of the fly brain. A simple physics-based calculation of aggregation propensity correlated strongly ($r = 0.75$) with fly longevity. Pathogenic effects proved incredibly sensitive to sequence; a change in aggregation propensity of less than 15% could shift a fly's lifespan from 30 days to just 8 days.

Survival curves of Drosophila expressing Aβ42 variants. Panel A compares the low-aggregation F20E variant to wild-type (WT). Panel B compares the I31E/E22G variant, which bypasses protofibril formation, to the highly toxic E22G variant.

The "Protofibril" Hypothesis

While general aggregation propensity ($Z_{agg}$) was a strong predictor, the researchers identified a key outlier: the I31E/E22G variant. Although predicted to aggregate rapidly, it was relatively non-toxic in flies. Investigation revealed that this variant bypasses the disordered intermediate "protofibril" stage, forming stable mature fibrils directly.

This led to the development of a refined algorithm, $Z_{tox}$, specifically tuned to predict the propensity to form protofibrils. This model showed an even higher correlation with fly survival ($r = 0.83$), suggesting that intermediate, disordered aggregates are the true pathogenic species. These intermediates may be more reactive with cellular membranes or harder for cellular machinery to clear than mature fibrils.

In Vivo Evidence

Immunohistochemistry confirmed these findings. Non-toxic variants, such as F20E, showed no A$\beta$ deposition in the brain by day 20. In contrast, toxic variants exhibited massive accumulation and brain vacuolation (tissue degradation), directly linking the predicted biophysical properties to physical neurodegeneration.

Discussion and Future Directions

This study demonstrates that the toxicity of A$\beta$ variants can be predicted with remarkable accuracy using computational algorithms based on primary sequence. While Drosophila is an excellent high-throughput model, future work is needed to determine if the absolute toxicity thresholds identified here translate directly to mammalian systems with different lifespans and clearance pathways.

The authors suggest that this "in silico to in vivo" pipeline could be applied to other neurodegenerative diseases involving protein misfolding, such as Parkinson's ($\alpha$-synuclein) or Huntington's (polyglutamine), to identify universal structural determinants of protein toxicity.

Summary updated

Abstract

Protein aggregation into amyloid fibrils and protofibrillar aggregates is associated with a number of the most common neurodegenerative diseases. We have established, using a computational approach, that knowledge of the primary sequences of proteins is sufficient to predict their in vitro aggregation propensities. Here we demonstrate, using rational mutagenesis of the Abeta42 peptide based on such computational predictions of aggregation propensity, the existence of a strong correlation between the propensity of Abeta42 to form protofibrils and its effect on neuronal dysfunction and degeneration in a Drosophila model of Alzheimer disease. Our findings provide a quantitative description of the molecular basis for the pathogenicity of Abeta and link directly and systematically the intrinsic properties of biomolecules, predicted in silico and confirmed in vitro, to pathogenic events taking place in a living organism.
BibTeX
@inproceedings{luheshi2007systematic,
  author = {Leila M. Luheshi and Gian Gaetano Tartaglia and Ann‐Christin Brorsson and Amol Pawar and Ian E Watson and Fabrizio Chiti and Michele Vendruscolo and David A. Lomas and Christopher M. Dobson and Damian C. Crowther},
  title = {{Systematic In Vivo Analysis of the Intrinsic Determinants of Amyloid β Pathogenicity}},
  booktitle = {PLoS},
  year = {2007},
  doi = {10.1371/journal.pbio.0050290},
}
Figure