TL;DR
Researchers have developed a novel cryo-EM image classification method to dissect the dynamic conformational states of the human γ-secretase complex. This technique revealed three distinct conformations of the catalytic subunit, one of which features an unexpected transmembrane helix, and provided insights into how substrate binding and inhibitor DAPT rigidify the enzyme, leading to a new model for substrate entry.
What Problem Does This Paper Address?
The γ-secretase complex is a crucial intramembrane protease involved in cleaving proteins like Notch (implicated in cancer) and amyloid precursor protein (APP), the aberrant cleavage of which leads to Alzheimer's disease. Understanding its mechanism is vital for developing targeted therapies. Previous cryo-EM studies revealed significant disorder in presenilin (PS1), the catalytic subunit, hindering a detailed understanding of its plasticity and substrate interaction. This paper aims to overcome this challenge by developing and applying advanced image processing techniques to capture and analyze the dynamic conformational landscape of γ-secretase.
Methods
The core methodological innovation is a masked cryo-EM image classification procedure with residual signal subtraction. This technique addresses the problem of structural heterogeneity in cryo-EM data by:
- Masking: Focusing refinement on a specific region of interest (PS1 subunit) while effectively ignoring variability in other parts.
- Signal Subtraction: Subtracting the signal from known, stable parts of the complex (e.g., nicastrin, Aph-1, Pen-2) from experimental images. This isolates the signal from the more flexible region (PS1) and reduces noise.
- 3D Auto-refinement: Applying standard 3D refinement to the processed particles to obtain high-resolution maps of distinct conformations.
This approach was applied to both the apo-state (unbound) γ-secretase and a complex with the dipeptidic inhibitor DAPT.
Datasets
The paper utilizes cryo-EM data collected from purified human γ-secretase complexes.
- Apo-state data: A dataset of 400,000 particles was used for initial masked classification.
- DAPT-bound data: A dataset of 51,366 particles was used for structure determination of the inhibitor-bound complex.
No new benchmark datasets were introduced.
Key Insights and Ideas
- Conformational Plasticity is Key: The catalytic subunit (PS1) of γ-secretase is not static but exists in multiple distinct conformations, highlighting its dynamic nature.
- An Unexpected Helix: One of the identified apo-state conformations features an additional transmembrane helix not previously attributed to the known γ-secretase components. This suggests potential interactions with other membrane proteins or regulatory mechanisms.
- Inhibitor-Induced Rigidity: Binding of the DAPT inhibitor significantly rigidifies the PS1 subunit, particularly TM2 and TM6, and their associated linkers. This suggests a mechanism by which inhibitors stabilize specific conformations.
- Substrate Entry Model: The observed conformational changes, especially the rigidification upon DAPT binding and the presence of the novel helix, form the basis for a new model proposing that substrate enters the transmembrane domain through a cavity formed by TM2, TM3, and TM5 of PS1, with TM2 acting as a "lateral gate."
Key Results
- Three Apo-State Conformations: The masked classification successfully resolved three distinct conformations of the γ-secretase complex, primarily differing in the orientation of TM6 and the ordering of TM2.
- Novel Helix Identification: An unaccounted-for, kinked α-helix was identified in the cavity between TM2, TM3, and TM5 of PS1 in one of the apo-state conformations. Mass spectrometry and Western blot analyses suggest this helix may originate from co-purifying proteins.
- DAPT-Bound Structure: A 4.2 Å resolution structure of γ-secretase in complex with DAPT was determined. This structure showed significant ordering of PS1, particularly TM2 and TM6, compared to the apo-state.
- Reduced Mobility: Conformational mobility in TM2 and TM6 of PS1 was greatly reduced upon DAPT binding, and also upon the appearance of the additional helix in one of the apo-state conformations.
- Active Site Proximity: In the DAPT-bound structure, the catalytic aspartates were brought within 8.0 Å of each other, suggesting a conformation conducive to catalysis.
Limitations and Future Work
- Limited Resolution: While improved, the resolution of the DAPT-bound structure (4.2 Å) still limits detailed side-chain analysis.
- Identity of the Novel Helix: The exact identity and origin of the additional transmembrane helix remain speculative, though evidence points to co-purifying proteins.
- Substrate Mimicry: The hypothesis that the novel helix mimics substrate binding needs further experimental validation.
The authors suggest that future work should focus on:
- Determining cryo-EM structures in complex with various substrates, substrate analogues, and other modulators to further elucidate the mechanism.
- Utilizing the developed masked classification approach to study conformational shifts induced by different factors, which is crucial for understanding the enzyme's functioning.
- Investigating the role of the novel helix and its potential interactions with other membrane proteins.
Summary updated