Protein Science最新文献

In recent years, significant advancements have been made in deep learning-based computational modeling of proteins, with DeepMind's AlphaFold2 standing out as a landmark achievement. These computationally modeled protein structures not only provide atomic coordinates but also include self-confidence metrics to assess the relative quality of the modeling, either for individual residues or the entire protein. However, these self-confidence scores are not always reliable; for instance, poorly modeled regions of a protein may sometimes be assigned high confidence. To address this limitation, we introduce Equivariant Quality Assessment Folding (EQAFold), an enhanced framework that refines the Local Distance Difference Test prediction head of AlphaFold to generate more accurate self-confidence scores. Our results demonstrate that EQAFold outperforms the standard AlphaFold architecture and recent model quality assessment protocols in providing more reliable confidence metrics. Source code for EQAFold is available at https://github.com/kiharalab/EQAFold_public.

Cytoplasmic dynein is a motor protein that plays a role in a number of cellular processes including retrograde transport. In many cases, dynein needs to interact with another protein, dynactin, to be fully active. An important step in the assembly of the dynein/dynactin complex is the interaction between the N-terminal portion of the intermediate chain (IC) subunit of dynein and the coiled-coil 1B (CC1B) region of the p150^Glued subunit of dynactin. Despite evidence for this interaction from binding studies, the exact location of where these proteins bind has remained elusive due to the dynamic nature of the interaction and the presence of intrinsically disordered regions in IC. By using intermolecular paramagnetic relaxation enhancements, we have been able to constrain the location of IC binding on p150^Glued to a position that is different from what has recently been hypothesized in a model of the dynein/dynactin complex based on cryo-electron microscopy (cryo-EM) data and AlphaFold predictions. In addition, although phosphorylation is important for regulating dynein/dynactin interactions, we show that a phosphomimetic mutation of IC is not sufficient to alter binding with p150^Glued.

The ester bond crosslink discovered within bacterial adhesin proteins offers a captivating insight into the convergent evolution of enzyme-like machinery. Crystal structures reveal a putative catalytic triad comprising an acid-base-nucleophile combination and an oxyanion-like site that suggests a serine protease-like mechanism drives the crosslinking process. We now provide confirmation of the mechanism, revealing functional catalytic dyads or triads, and the recapitulation of protease machinery from a Pseudomonas bacterium and a human cytomegalovirus related only by convergent evolution. Molecular dynamics simulations suggest how a conservative threonine-to-serine mutation of the nucleophile induces hydrolysis and eliminates the ester bond crosslink. Collectively, our structural, functional, and computational efforts detail the molecular intricacies of intramolecular ester bond formation and underscore the convergent evolutionary adaptations of bacteria in exploiting enzyme-like machinery to protect essential adhesin proteins from the mechanical, biological, and chemical hostilities of their replicative niche.

Monoubiquitinated histone H2B at K123 in yeast (K120 in humans) is a transient modification that is both attached and removed during transcription. H2B is ubiquitinated in yeast by the E2/E3 pair, Rad6/Bre1, and deubiquitinated by two enzymes, Ubp8 and Ubp10. Previous studies had shown that Ubp10 has higher activity on ubiquitinated H2A/H2B dimers than on intact nucleosomes, but that activity on nucleosomes is higher in the presence of the histone chaperone, FACT. By contrast, the Ubp8 complex has equal activity on both histone substrates and is unaffected by FACT. We report here the results of single-molecule FRET experiments showing that FACT unwraps DNA and evicts ubiquitinated H2A/H2B dimers, the preferred substrate of Ubp10. To explore the basis for the differing activity of Ubp10 on ubiquitinated H2A/H2B dimers and nucleosomes, we employed crosslinking mass spectrometry combined with structural modeling. These studies revealed that Ubp10 forms a different set of interactions with H2A/H2B in free versus nucleosomal states. Acidic stretches within the N-terminal intrinsically disordered region (IDR) of Ubp10 interact extensively with H2A/H2B heterodimers, whereas this portion of Ubp10 interacts more with the tails of histones H3 and H4 in the nucleosome. The importance of these interactions for affinity is consistent with binding studies showing the IDR is necessary for substrate interactions. Structural modeling using the crosslinks as constraints suggests that the complex formed by Ubp10 with free H2A/H2B dimers could not be formed within a nucleosome due to steric clash with the DNA, H3, and H4, thereby explaining its low activity on ubiquitinated nucleosomes.

The activities of [NiFe]-hydrogenase enzymes, which are critical to many microbes, require insertion of a Ni(II) ion into the bimetallic catalytic center. Delivery of Ni(II) to [NiFe]-hydrogenases depends, in part, on the metallochaperone HypB, which lies at the center of a Ni(II) transfer pathway that includes the metal storage protein SlyD and the metallochaperone HypA. SlyD is a source of Ni(II) ions for HypB, whereas Ni(II) from HypB is transferred to HypA. In this work, we examine how the intrinsically disordered N-terminal extension (NTE) of HypB modulates the action of the HypB GTPase domain (G-domain). The NTE contains a high-affinity Ni(II) binding site, while the G-domain contains a lower affinity Ni(II) binding site that is affected by binding of guanine nucleotides. The HypB G-domain is also affected by SlyD and provides Ni(II) to HypA. Our NMR data show that, although disordered, the HypB NTE possesses residual structure and makes transient interactions with the HypB G-domain and with SlyD. A set of common residues in the center of the NTE are affected by SlyD and G-domain binding, and also by binding of Ni(II) to the high-affinity site located at the N terminus of the protein. The NTE interacts with residues in or near the Ni(II)- and GDP-binding sites in the G-domain, which are also affected when SlyD binds the NTE. Thus, the data showcase a complex interaction network between HypB and SlyD, and provide molecular details regarding how the HypB NTE links the activities of the HypB G-domain and SlyD.

Outer membrane β-barrel proteins (OMPs) are channels found in the outer membranes of Gram-negative bacteria characterized by a stable and diverse barrel architecture, which has made them attractive for nanopore sensing applications. Here, we systematically investigated the feasibility of expanding outer membrane protein G (OmpG) from its native 14-stranded β-barrel to an enhanced conductance variant by independently duplicating each of its seven hairpin units and inserting them downstream of their endogenous positions. Most combinations did not increase pore diameter, but duplication of the terminal seventh hairpin exhibited a rare population of pores with enhanced conductance, suggesting barrel enlargement. Further engineering efforts to optimize the terminal β-turn sequence have resulted in up to 50% of pores with increased conductance. Importantly, the enlarged pores retained the sensing functionality of the original scaffold, highlighting the potential of this approach for developing β-barrel OMP sensors with tunable dimensions.

The UBR family of ubiquitin ligases binds to N-termini of their targets (known as N-degron) to induce their ubiquitination and degradation via a conserved domain known as UBR-box. UBR1 and UBR2 share the highest sequence homology among the family, and substantial structural studies were previously performed for substrate binding by the UBR-boxes of UBR1 and UBR2. Here, we describe a new pocket in the UBR-boxes of UBR1 and UBR2 for binding the second residues of N-degrons through determining five co-crystal structures of the UBR-boxes with various N-degron peptides. Together with binding affinities measured by fluorescence polarization, we show that the two highly homologous UBR-boxes can interact with the second residue of an N-degron differently. In addition, the UBR-boxes undergo different conformational changes when binding N-degrons. Furthermore, we demonstrate that the sidechain of the third amino acid of an N-degron has no contribution to binding the UBR-boxes. These findings represent a new conceptual advancement for the UBR E3 ligases and the new insights described here can be leveraged for developing their selective ligands for research and potential therapies.

Coiled coil structures, first proposed by Crick in the 1950s, are protein structural motifs found across diverse biological systems. Honeybee silk was among the earliest identified coiled coils, with X-ray diffraction studies in the 1960s revealing its characteristic helical packing. Decades of research have provided insights into silk composition and formation, yet the molecular details of its coiled coil assembly and final structure remained unresolved. In this study, we generated a structural model of the tetrameric coiled coil using AlphaFold and validated it with crosslinking mass spectrometry and medium-resolution cryo-electron microscopy. The model reveals that the four proteins (F1-F4) adopt an antiparallel configuration in a defined clockwise arrangement (F1-F3-F2-F4). Furthermore, we experimentally investigated the formation of this coiled coil complex using biochemical techniques, including blue-native PAGE and circular dichroism spectroscopy. The sum of these experimental results and the structural predictions has allowed for the elucidation of key transitional steps in the assembly pathway, suggesting molecular interactions that may drive tetramer formation. These findings support a stepwise assembly model in which F2 and F4 form a stable core, F3 binds to the complex, and F1 initiates formation of the final, highly ordered structure. These structural insights establish a framework for understanding and directing coiled coil assembly, the fundamental building block of honeybee silk. By resolving this structure and its assembly process, this work lays the foundation for future rational design of functional sequences and materials with tailored properties.

Type Six Secretion Systems (T6SS) are molecular machines that export toxic effector proteins into bacterial competitors or eukaryotic cells. Pseudomonas aeruginosa's H1-T6SS secretes Tse1, which contains a disulfide bond between cysteines at positions 7 and 148, linking its N- and C-terminal regions. The role of this disulfide bond in Tse1 activity and mechanism of action during bacterial competition is unknown. In this study, we investigated the role of the C7-C148 disulfide bond within Tse1. First, NMR spectroscopy experiments suggest a redox-active instead of a structural disulfide bond. Moreover, while the presence of this bond did not alter Tse1's amidase activity or toxicity in Escherichia coli, substituting cysteines C7 or C148 in P. aeruginosa strains affected the bacterium's capacity to lyse prey cells. Secretome analysis showed that the Tse1C148S variant was not secreted via the H1-T6SS, whereas the Tse1C7S variant was secreted. These findings suggest that cysteine 148 is likely important for Tse1's assembly with the T6SS machinery, while cysteine 7 appears to be involved in its disassembly, potentially through the formation of the disulfide bond. This study points to a potential redox regulation mechanism during the assembly and disassembly of Tse1 with Hcp1, consistent with a "bridge of delivery" model.

Building an accurate atomic structure model of a protein into a cryo-electron microscopy (cryo-EM) map at worse than 3 Å resolution is difficult. To facilitate this task, we devised a method for assigning the amino acid residue sequence to the backbone fragments traced in an input cryo-EM map (EMSequenceFinder). EMSequenceFinder relies on a Bayesian scoring function for ranking 20 standard amino acid residue types at a given backbone position, based on the fit to a density map, map resolution, and secondary structure propensity. The fit to a density is quantified by a convolutional neural network that was trained on ~5.56 million amino acid residue densities extracted from cryo-EM maps at 3-10 Å resolution and corresponding atomic structure models deposited in the Electron Microscopy Data Bank (EMDB). We benchmarked EMSequenceFinder by predicting the sequences of 58,044 distinct ɑ-helix and β-strand fragments, given the fragment backbone coordinates fitted in their density maps. EMSequenceFinder identifies the correct sequence as the best-scoring sequence in 77.8% of these cases. We also assessed EMSequenceFinder on separate datasets of cryo-EM maps at resolutions from 4 to 6 Å. The accuracy of EMSequenceFinder (58%) was better than that of three tested state-of-the-art methods, including findMysequence (45%), ModelAngelo (27%), and sequence_from_map in Phenix (12.9%). We further illustrate EMSequenceFinder by threading the Severe Acute Respiratory Syndrome Coronavirus 2 Non-Structural Protein 2 sequence into eight cryo-EM maps at resolutions from 3.7 to 7.0 Å. EMSequenceFinder is implemented in our open-source Integrative Modeling Platform (IMP) program. Thus, it is expected to be helpful for integrative structure modeling based on a cryo-EM map and other information, such as models of protein complex components and chemical crosslinks between them. EMSequenceFinder is available as part of our open-source IMP distribution at https://integrativemodeling.org/.