Comparative analysis of biophysical methods for monitoring protein proximity induction in the development of small molecule degraders.

BACKGROUND: Targeted protein degradation relies on inducing proximity between an E3 ubiquitin ligase and a target protein, and subsequent proteasomal degradation of the latter. Biophysical methods allow the measurement of the ternary complex formation by recombinant target and E3 ligase proteins in the presence of molecular glues and bifunctional degraders. The development of new chemotypes of degraders mediating ternary complex formation of unknown dimensions and geometries requires the use of different biophysical approaches. METHODS: The TR-FRET and AlphaLISA platforms have been applied to study molecular glues and bifunctional degraders. The performance of the label-based proximity assays was compared with the BLI method, which is a label-free, sensor-based approach. RESULTS: We present and compare two commonly used assays to monitor proximity induction, AlphaLISA and TR-FRET. The LinkScape system consisting of the CaptorBait peptide and the CaptorPrey protein is a novel method of protein labeling compatible with TR-FRET assay. CONCLUSIONS: The TR-FRET and AlphaLISA proximity assays enable detection of ternary complexes formed between an E3 Ligase, a target protein and a small molecule degrader. Experiments with different chemotypes of GSPT1 degraders showed that ALphaLISA was more susceptible to chemotype-dependent interference than TR-FRET assay. GENERAL SIGNIFICANCE: The discovery and optimization of small-molecule inducers of ternary complexes is greatly accelerated by using biophysical assays. The LinkScape-based TR-FRET assay is an alternative to antibody-based proximity assays due to the CaptorPrey's subnanomolar affinity to the CaptorBait-tagged protein target, and the 10-fold lower molecular weight of the CaptorPrey protein compared to the antibody.

Mechanical stress impairs pheromone signaling via Pkc1-mediated regulation of the MAPK scaffold Ste5.

Cells continuously adapt cellular processes by integrating external and internal signals. In yeast, multiple stress signals regulate pheromone signaling to prevent mating under unfavorable conditions. However, the underlying crosstalk mechanisms remain poorly understood. Here, we show that mechanical stress activates Pkc1, which prevents lysis of pheromone-treated cells by inhibiting polarized growth. In vitro Pkc1 phosphorylates conserved residues within the RING-H2 domains of the scaffold proteins Far1 and Ste5, which are also phosphorylated in vivo. Interestingly, Pkc1 triggers dispersal of Ste5 from mating projections upon mechanically induced stress and during cell-cell fusion, leading to inhibition of the MAPK Fus3. Indeed, RING phosphorylation interferes with Ste5 membrane association by preventing binding to the receptor-linked Gßγ protein. Cells expressing nonphosphorylatable Ste5 undergo increased lysis upon mechanical stress and exhibit defects in cell-cell fusion during mating, which is exacerbated by simultaneous expression of nonphosphorylatable Far1. These results uncover a mechanical stress-triggered crosstalk mechanism modulating pheromone signaling, polarized growth, and cell-cell fusion during mating.

Application of Dirichlet process mixture model to the identification of spin systems in protein NMR spectra.

Analysis of structure, function and interactions of proteins by NMR spectroscopy usually requires the assignment of resonances to the corresponding nuclei in protein. This task, although automated by methods such as FLYA or PINE, is still frequently performed manually. To facilitate the manual sequence-specific chemical shift assignment of complex proteins, we propose a method based on Dirichlet process mixture model (DPMM) that performs automated matching of groups of signals observed in NMR spectra to corresponding nuclei in protein sequence. The model has been extensively tested on 80 proteins retrieved from the BMRB database and has shown superior performance to the reference method.

NMRNet: a deep learning approach to automated peak picking of protein NMR spectra.

Motivation: Automated selection of signals in protein NMR spectra, known as peak picking, has been studied for over 20 years, nevertheless existing peak picking methods are still largely deficient. Accurate and precise automated peak picking would accelerate the structure calculation, and analysis of dynamics and interactions of macromolecules. Recent advancement in handling big data, together with an outburst of machine learning techniques, offer an opportunity to tackle the peak picking problem substantially faster than manual picking and on par with human accuracy. In particular, deep learning has proven to systematically achieve human-level performance in various recognition tasks, and thus emerges as an ideal tool to address automated identification of NMR signals. Results: We have applied a convolutional neural network for visual analysis of multidimensional NMR spectra. A comprehensive test on 31 manually annotated spectra has demonstrated top-tier average precision (AP) of 0.9596, 0.9058 and 0.8271 for backbone, side-chain and NOESY spectra, respectively. Furthermore, a combination of extracted peak lists with automated assignment routine, FLYA, outperformed other methods, including the manual one, and led to correct resonance assignment at the levels of 90.40%, 89.90% and 90.20% for three benchmark proteins. Availability and implementation: The proposed model is a part of a Dumpling software (platform for protein NMR data analysis), and is available at https://dumpling.bio/. Supplementary information: Supplementary data are available at Bioinformatics online.

Interaction prediction in structure-based virtual screening using deep learning.

We introduce a deep learning architecture for structure-based virtual screening that generates fixed-sized fingerprints of proteins and small molecules by applying learnable atom convolution and softmax operations to each molecule separately. These fingerprints are further non-linearly transformed, their inner product is calculated and used to predict the binding potential. Moreover, we show that widely used benchmark datasets may be insufficient for testing structure-based virtual screening methods that utilize machine learning. Therefore, we introduce a new benchmark dataset, which we constructed based on DUD-E, MUV and PDBBind databases.

Posttranslational modifications of intact proteins detected by NMR spectroscopy: application to glycosylation.

Posttranslational modifications (PTMs) are an integral part of the majority of proteins. The characterization of structure and function of PTMs can be very challenging especially for glycans. Existing methods to analyze PTMs require complicated sample preparations and suffer from missing certain modifications, the inability to identify linkage types and thus chemical structure. We present a direct, robust, and simple NMR spectroscopy method for the detection and identification of PTMs in proteins. No isotope labeling is required, nor does the molecular weight of the studied protein limit the application. The method can directly detect modifications on intact proteins without sophisticated sample preparation. This approach is well suited for diagnostics of proteins derived from native organisms and for the quality control of biotechnologically produced therapeutic proteins.

Computer vision-based automated peak picking applied to protein NMR spectra.

MOTIVATION: A detailed analysis of multidimensional NMR spectra of macromolecules requires the identification of individual resonances (peaks). This task can be tedious and time-consuming and often requires support by experienced users. Automated peak picking algorithms were introduced more than 25 years ago, but there are still major deficiencies/flaws that often prevent complete and error free peak picking of biological macromolecule spectra. The major challenges of automated peak picking algorithms is both the distinction of artifacts from real peaks particularly from those with irregular shapes and also picking peaks in spectral regions with overlapping resonances which are very hard to resolve by existing computer algorithms. In both of these cases a visual inspection approach could be more effective than a 'blind' algorithm. RESULTS: We present a novel approach using computer vision (CV) methodology which could be better adapted to the problem of peak recognition. After suitable 'training' we successfully applied the CV algorithm to spectra of medium-sized soluble proteins up to molecular weights of 26 kDa and to a 130 kDa complex of a tetrameric membrane protein in detergent micelles. Our CV approach outperforms commonly used programs. With suitable training datasets the application of the presented method can be extended to automated peak picking in multidimensional spectra of nucleic acids or carbohydrates and adapted to solid-state NMR spectra. AVAILABILITY AND IMPLEMENTATION: CV-Peak Picker is available upon request from the authors. CONTACT: gsw@mol.biol.ethz.ch; michal.walczak@mol.biol.ethz.ch; adam.gonczarek@pwr.edu.pl SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Intramolecular donor strand complementation in the E. coli type 1 pilus subunit FimA explains the existence of FimA monomers as off-pathway products of pilus assembly that inhibit host cell apoptosis.

Type 1 pili are filamentous organelles mediating the attachment of uropathogenic Escherichia coli to epithelial cells of host organisms. The helical pilus rod consists of up to 3000 copies of the main structural subunit FimA that interact via donor strand complementation, where the incomplete Ig-like fold of FimA is completed by insertion of the N-terminal extension (donor strand) of the following FimA subunit. Recently, it was shown that FimA also exists in a monomeric, assembly-incompetent form and that FimA monomers act as inhibitors of apoptosis in infected host cells. Here we present the NMR structure of monomeric wild-type FimA with its natural N-terminal donor strand complementing the Ig fold. Compared to FimA subunits in the assembled pilus, intramolecular self-complementation in the monomer stabilizes the FimA fold with significantly less interactions, and the natural FimA donor strand is inserted in the opposite orientation. In addition, we show that a motif of two glycine residues in the FimA donor strand, separated by five residues, is the prerequisite of the alternative, parallel donor strand insertion mechanism in the FimA monomer and that this motif is preserved in FimA homologs of many enteroinvasive pathogens. We conclude that FimA is a unique case of a protein with alternative, functionally relevant folding possibilities, with the FimA polymer forming the highly stable pilus rod and the FimA monomer promoting pathogen propagation by apoptosis suppression of infected epithelial target cells.

The RING domain of the scaffold protein Ste5 adopts a molten globular character with high thermal and chemical stability.

Ste5 is a scaffold protein that controls the pheromone response of the MAP-kinase cascade in yeast cells. Upon pheromone stimulation, Ste5 (through its RING-H2 domain) interacts with the ß and γ subunits of an activated heterodimeric G protein and promotes activation of the MAP-kinase cascade. With structural and biophysical studies, we show that the Ste5 RING-H2 domain exists as a molten globule under native buffer conditions, in yeast extracts, and even in denaturing conditions containing urea (7 M). Furthermore, it exhibits high thermal stability in native conditions. Binding of the Ste5 RING-H2 domain to the physiological Gß/γ (Ste4/Ste18) ligand is accompanied by a conformational transition into a better folded, more globular structure. This study reveals novel insights into the folding mechanism and recruitment of binding partners by the Ste5 RING-H2 domain. We speculate that many RING domains may share a similar mechanism of substrate recognition and molten-globule-like character.