Abstract
Spt-Ada-Gcn5-Acetyltransferase (SAGA) is a conserved multi-subunit complex that activates RNA polymerase II-mediated transcription by acetylating and deubiquitinating nucleosomal histones and by recruiting TATA box binding protein (TBP) to DNA. The prototypical yeast Saccharomyces cerevisiae SAGA contains 19 subunits that are organized into Tra1, core, histone acetyltransferase, and deubiquitination modules. Recent cryo-electron microscopy studies have generated high-resolution structural information on the Tra1 and core modules of yeast SAGA. However, the two catalytical modules were poorly resolved due to conformational flexibility of the full assembly. Furthermore, the high sample requirement created a formidable barrier to further structural investigations of SAGA. Here, we report a workflow for isolating/stabilizing yeast SAGA and preparing cryo-EM specimens at low protein concentration using a graphene oxide support layer. With this procedure, we were able to determine a cryo-EM reconstruction of yeast SAGA at 3.1 Å resolution and examine its conformational landscape with the neural network-based algorithm cryoDRGN. Our analysis revealed that SAGA adopts a range of conformations with its HAT module and central core in different orientations relative to Tra1.
Introduction
The Spt-Ada-Gcn5-Acetyltransferase (SAGA) complex is a massive (1.8 MDa), 19-subunit molecular machine essential for regulating gene expression in eukaryotes. It performs two primary functions: modifying histones through acetylation and deubiquitination, and recruiting the TATA-binding protein (TBP) to DNA. Despite its biological significance, SAGA has presented significant challenges for structural biology.
First, standard cryo-electron microscopy (cryo-EM) typically requires high protein concentrations, necessitating massive 100-liter yeast cultures to obtain sufficient native material. Second, SAGA is highly dynamic; its catalytic modules (HAT and DUB) are tethered to a central core by flexible linkers. This inherent flexibility often results in blurred images in traditional cryo-EM processing, obscuring the fine details of active sites.
Methods and Materials
To overcome these barriers, a two-pronged approach involving optimized specimen preparation and advanced computational analysis was employed.
Specimen Preparation and Graphene Oxide Support
Instead of extreme protein concentration, the researchers utilized Lacey carbon grids coated with a thin layer of graphene oxide (GO). The GO layer acts as a support that adsorbs SAGA particles even at low concentrations, allowing for a 25-fold reduction in the required yeast culture scale (down to 4 liters). Additionally, Gradient Fixation (GraFix)—combining a glycerol gradient with limited glutaraldehyde crosslinking—was used to stabilize the complex during purification.
Data Collection and cryoDRGN Analysis
A primary dataset of 17,955 cryo-EM movies was collected on a Titan Krios microscope. From an initial pool of over 1.1 million particle images, a high-quality subset of 542,311 particles was refined. To address SAGA’s flexibility, the team employed cryoDRGN, a deep neural network-based algorithm. Unlike traditional discrete 3D classification, cryoDRGN learns a continuous "latent space" of conformations, enabling the visualization of SAGA as a moving ensemble rather than a static snapshot. The resulting maps were deposited in the EMDB under accession codes EMD-26792, EMD-26803, and EMD-26804.
Results
High-Resolution Structural Insights
The team achieved a consensus reconstruction at 3.1 Å resolution. Local refinements further improved the resolution of the Tra1 module to 2.9 Å and the core to 3.2 Å, permitting the visualization of individual amino acid side chains. These structures were validated by docking existing atomic models (PDB: 6t9j and 6t9k) into the new density maps with minimal adjustments.
The Conformational Landscape
Analysis of the conformational landscape revealed a continuous range of motion rather than discrete states. Key findings include:
- Hinge Mechanism: The Tra1 module (the "head") is the most rigid component, anchored to the core via a flexible hinge involving Taf12 and Spt20. This allows the complex to reposition itself relative to the DNA-binding head.
- 50 Å Range of Motion: The "tail" region, containing the HAT catalytic module, exhibits a movement range of up to 50 Å, suggesting an active search mechanism for targets within the chromatin environment.
- Spt8 and TBP Loading: Observations of the Spt8 subunit retracting from Tra1 likely represent a critical step in the transfer of TBP to DNA during transcription initiation.
Landscape analysis showed that 60% of particles exist in an intermediate "curved" state, while 17% are "arched" (tail retracted) and 23% are in a "donut" configuration (tail tucked in).
Discussion
This study successfully mapped the continuous conformational landscape of the SAGA complex, providing a more nuanced view of its dynamic nature. The use of graphene oxide support layers significantly lowers the barrier for studying rare or low-yield endogenous complexes.
However, despite these advancements, the HAT and DUB catalytic modules did not reach high enough resolution for atomic-level modeling within the full assembly due to their extreme flexibility. Future research will likely focus on capturing SAGA in complex with substrates, such as the nucleosome, to stabilize these flexible modules and achieve higher-resolution imaging of the active sites.
Summary updated