Date Published: March 8, 2018
Publisher: Public Library of Science
Author(s): David E. Timm, Valorie Bowman, Russell Madsen, Charles Rauch, Dimitrios Fotiadis.
Protein arginine methyl transferase 5 (PRMT5) is a signaling protein and histone modifying enzyme that is important in many cellular processes, including regulation of eukaryotic gene transcription. Reported here is a 3.7 Å structure of PRMT5, solved in complex with regulatory binding subunit MEP50 (methylosome associated protein 50, WDR77, p44), by single particle (SP) cryo-Electron Microscopy (cryo-EM) using micrographs of particles that are visibly crowded and aggregated. Despite suboptimal micrograph appearance, this cryo-EM structure is in good agreement with previously reported crystal structures of the complex, which revealed a 450 kDa hetero-octameric assembly having internal D2 symmetry. The catalytic PRMT5 subunits form a core tetramer and the MEP50 subunits are arranged peripherally in complex with the PRMT5 N-terminal domain. The cryo-EM reconstruction shows good side chain definition and shows a well-resolved peak for a bound dehydrosinefungin inhibitor molecule. These results demonstrate the applicability of cryo-EM in determining structures of human protein complexes of biomedical significance and suggests cryo-EM could be further utilized to understand PRMT5 interactions with other biologically important binding proteins and ligands.
PRMT5 (capsuleen, Dart5, Hsl7, Jbp1, Skb1) is a key arginine methyl transferase that is involved in a plethora of important cellular functions and clinical disease states. This enzyme catalyzes the non-processive transfer of a methyl group from a S-adenosyl methionine co-factor (SAM, Fig 1) to the ω-nitrogen atoms in the Arg guanidinium group to generate S-adenosyl homocysteine (SAH, Fig 1) and monomethyl Arg and symmetric dimethyl Arg products . Vertebrate PRMT5 is almost always found bound to the adapter protein MEP50 , which has been shown to enhance methyl transferase activity for substrates like histones H2A, H3 and H4 by a mechanism that likely involves enhanced substrate binding and recognition via MEP50 interactions with substrates . Histone methylation by PRMT5 is critical to epigenetic regulation of gene transcription and modulates expression of gene targets such as cyclin E1 , the Rb tumor suppressor  and ribosomal proteins . PRMT5 also impacts gene expression by methylating and thus regulating the activity of important transcription factors such as E2F-1 , NF-κB  and the p53 tumor suppressor . Additional important proliferative signals are influenced by PRMT5 methylation of the EGF receptor  and RAF  within the RAS-ERK pathway downstream of the EGF receptor.
Negative stain imaging (Fig 2A) was initially used to determine suitability of the PRMT5:MEP50 sample for EM analyses, to optimize buffer components and to estimate the protein concentration required to yield good particle densities in cryo-EM grids. Grids obtained with NaCl concentrations of 150 mM yielded images showing dispersed particles of ~15 nm and having an apparent ‘donut hole’ at the center of the particles (Fig 2A inset). Inclusion of low mM concentrations of the SAM competitive inhibitor, dehydrosinefungin [16, 26], was associated with increased particle density and reduced aggregation in negative stained grids, so the inhibitor was included as a buffer component for cryo-EM grid preparation. Protein concentrations of 0.23–0.58 mg/mL worked well for negative staining. Aggregation was significantly more prevalent in cryo-grids that were screened, even at lower dilutions. A grid prepared at a protein concentration of 0.38 mg/mL was chosen for high resolution cryo-imaging. Some small filamentous aggregates of ~15 nm width are observed at a low frequency, with or without dehydrosinefungin, in most negative stained and cryo- images (Fig 2A and 2B), indicating a slight propensity for the complex to associate in a side to side manner under these buffer conditions. However, it is unclear if this association is of any biologic relevance. The cryo-electron micrographs appear sub-optimal in having crowded particle densities, significant amounts of protein aggregation and some ice and/or ethane contamination (Fig 2B).
Recent advances in the field of biological cryo-EM have led to much well-founded enthusiasm for this next revolution in structural biology. Our biological understanding of molecular mechanisms and structures for entities ranging from the atomic to sub-cellular sizes has been and will continue to be impacted in a dramatic manner by this high resolution imaging technology. However, the application of cryo-EM in published drug discovery or structure-based drug design reports remains largely absent. At the time of this paper’s preparation, there are fewer than 100 SP cryo-EM maps present in the EMDB for human samples at a sufficient resolution (< 4 Å) to resolve low molecular weight ligands from surrounding protein groups within a binding site. Also, while there are notable examples of small ligands resolved within cryo-EM structures [28–31, 47], they represent a very small fraction of existing high resolution maps. There also remains a notable absence of high resolution cryo-EM structures for therapeutic proteins, including monoclonal antibodies (mAbs) or antigen binding fragments (Fabs) derived from therapeutic mAbs. Source: http://doi.org/10.1371/journal.pone.0193205