A backbone-centred energy function of neural networks for protein design.

A protein backbone structure is designable if a substantial number of amino acid sequences exist that autonomously fold into it1,2. It has been suggested that the designability of backbones is governed mainly by side chain-independent or side chain type-insensitive molecular interactions3-5, indicating an approach for designing new backbones (ready for amino acid selection) based on continuous sampling and optimization of the backbone-centred energy surface. However, a sufficiently comprehensive and precise energy function has yet to be established for this purpose. Here we show that this goal is met by a statistical model named SCUBA (for Side Chain-Unknown Backbone Arrangement) that uses neural network-form energy terms. These terms are learned with a two-step approach that comprises kernel density estimation followed by neural network training and can analytically represent multidimensional, high-order correlations in known protein structures. We report the crystal structures of nine de novo proteins whose backbones were designed to high precision using SCUBA, four of which have novel, non-natural overall architectures. By eschewing use of fragments from existing protein structures, SCUBA-driven structure design facilitates far-reaching exploration of the designable backbone space, thus extending the novelty and diversity of the proteins amenable to de novo design.

Combined prediction and design reveals the target recognition mechanism of an intrinsically disordered protein interaction domain.

An increasing number of protein interaction domains and their targets are being found to be intrinsically disordered proteins (IDPs). The corresponding target recognition mechanisms are mostly elusive because of challenges in performing detailed structural analysis of highly dynamic IDP-IDP complexes. Here, we show that by combining recently developed computational approaches with experiments, the structure of the complex between the intrinsically disordered C-terminal domain (CTD) of protein 4.1G and its target IDP region in NuMA can be dissected at high resolution. First, we carry out systematic mutational scanning using dihydrofolate reductase-based protein complementarity analysis to identify essential interaction regions and key residues. The results are found to be highly consistent with an α/ß-type complex structure predicted by AlphaFold2 (AF2). We then design mutants based on the predicted structure using a deep learning protein sequence design method. The solved crystal structure of one mutant presents the same core structure as predicted by AF2. Further computational prediction and experimental assessment indicate that the well-defined core structure is conserved across complexes of 4.1G CTD with other potential targets. Thus, we reveal that an intrinsically disordered protein interaction domain uses an α/ß-type structure module formed through synergistic folding to recognize broad IDP targets. Moreover, we show that computational prediction and experiment can be jointly applied to segregate true IDP regions from the core structural domains of IDP-IDP complexes and to uncover the structure-dependent mechanisms of some otherwise elusive IDP-IDP interactions.

Molecular basis for C-degron recognition by CRL2APPBP2 ubiquitin ligase.

E3 ubiquitin ligases determine the specificity of eukaryotic protein degradation by selective binding to destabilizing protein motifs, termed degrons, in substrates for ubiquitin-mediated proteolysis. The exposed C-terminal residues of proteins can act as C-degrons that are recognized by distinct substrate receptors (SRs) as part of dedicated cullin-RING E3 ubiquitin ligase (CRL) complexes. APPBP2, an SR of Cullin 2-RING ligase (CRL2), has been shown to recognize R-x-x-G/C-degron; however, the molecular mechanism of recognition remains elusive. By solving several cryogenic electron microscopy structures of active CRL2APPBP2 bound with different R-x-x-G/C-degrons, we unveiled the molecular mechanisms underlying the assembly of the CRL2APPBP2 dimer and tetramer, as well as C-degron recognition. The structural study, complemented by binding experiments and cell-based assays, demonstrates that APPBP2 specifically recognizes the R-x-x-G/C-degron via a bipartite mechanism; arginine and glycine, which play critical roles in C-degron recognition, accommodate distinct pockets that are spaced by two residues. In addition, the binding pocket is deep enough to enable the interaction of APPBP2 with the motif placed at or up to three residues upstream of the C-end. Overall, our study not only provides structural insight into CRL2APPBP2-mediated protein turnover but also serves as the basis for future structure-based chemical probe design.

Structure of GDP-bound Rab7 Q67L in complex with ORP1L.

Molecular processes are orchestrated by various proteins that promote early endosomes to become late endosomes and eventually fuse with lysosomes, guaranteeing the degradation of the content. Rab7, which is localized to late endosomes, is one of the most well-known GTPases. ORP1L is recruited by Rab7 to facilitate the fusion of late endosomes and lysosomes. Here, we present the structure of GDP-bound Rab7 Q67L with ORP1L. Structural analysis, supported by biochemical and ITC binding experiments, not only provides structural insight into the interactions between the ORP1L ANK domain and Rab7 but also suggests that the GTPase activity of Rab7 does not interfere with its ORP1L-binding capacity.

Reactions of Cisplatin with Thioredoxin-1 Regulate Intracellular Redox Homeostasis.

Cisplatin is a widely used anticancer drug. In addition to inducing DNA damage, increased levels of reactive oxygen species (ROS) play a significant role in cisplatin-induced cell death. Thioredoxin-1 (Trx1), a redox regulatory protein that can scavenge ROS, has been found to eliminate cisplatin-induced ROS, while elevated Trx1 levels are associated with cisplatin resistance. However, it is unknown whether the effect of Trx1 on the cellular response to cisplatin is due to its direct reaction and how this reaction influences the activity of Trx1. In this work, we performed detailed studies of the reaction between Trx1 and cisplatin. Trx1 is highly reactive to cisplatin, and the catalytic motif of Trx1 (CGPC) is the primary binding site of cisplatin. Trx1 can bind up to 6 platinum moieties, resulting in the structural alteration and oligomerization of Trx1 depending on the degree of platination. Platination of Trx1 inhibits its interaction with ASK1, a Trx1-binding protein that regulates cell apoptosis. Furthermore, the reaction with cisplatin suppresses drug-induced ROS generation, which could be associated with drug resistance. This study provides more insight into the mechanism of action of cisplatin.

Structural studies of SALL family protein zinc finger cluster domains in complex with DNA reveal preferential binding to an AATA tetranucleotide motif.

The Spalt-like 4 transcription factor (SALL4) plays an essential role in controlling the pluripotent property of embryonic stem cells via binding to AT-rich regions of genomic DNA, but structural details on this binding interaction have not been fully characterized. Here, we present crystal structures of the zinc finger cluster 4 (ZFC4) domain of SALL4 (SALL4ZFC4) bound with different dsDNAs containing a conserved AT-rich motif. In the structures, two zinc fingers of SALL4ZFC4 recognize an AATA tetranucleotide. We also solved the DNA-bound structures of SALL3ZFC4 and SALL4ZFC1. These structures illuminate a common preference for the AATA tetranucleotide shared by ZFC4 of SALL1, SALL3, and SALL4. Furthermore, our cell biology experiments demonstrate that the DNA-binding activity is essential for SALL4 function as DNA-binding defective mutants of mouse Sall4 failed to repress aberrant gene expression in Sall4-/- mESCs. Thus, these analyses provide new insights into the mechanisms of action underlying SALL family proteins in controlling cell fate via preferential targeting to AT-rich sites within genomic DNA during cell differentiation.

Structural basis for METTL6-mediated m3C RNA methylation.

RNA modifications play important roles in mediating the biological functions of RNAs. 3-methylcytidine (m3C), albeit less abundant, is found to exist extensively in tRNAs, rRNAs and mRNAs. Human METTL6 is a m3C methyltransferase for tRNAs, including tRNASER(UGA). We solved the structure of human METTL6 in the presence of S-adenosyl-L-methionine and found by enzyme assay that recombinant human METTL6 is active towards tRNASER(UGA). Structural analysis indicated the detailed interactions between S-adenosyl-L-methionine and METTL6, and suggested potential tRNA binding region on the surface of METTL6. The structural research, complemented by biochemistry enzyme assay, will definitely shed light on the design of potent inhibitors for METTL6 in near future.

Fragment-Based Discovery of AF9 YEATS Domain Inhibitors.

YEATS (YAF9, ENL, AF9, TAF14, SAS5) family proteins recognize acylated histones and in turn regulate chromatin structure, gene transcription, and stress signaling. The chromosomal translocations of ENL and mixed lineage leukemia are considered oncogenic drivers in acute myeloid leukemia and acute lymphoid leukemia. However, known ENL YEATS domain inhibitors have failed to suppress the proliferation of 60 tested cancer cell lines. Herein, we identified four hits from the NMR fragment-based screening against the AF9 YEATS domain. Ten inhibitors of new chemotypes were then designed and synthesized guided by two complex structures and affinity assays. The complex structures revealed that these inhibitors formed an extra hydrogen bond to AF9, with respect to known ENL inhibitors. Furthermore, these inhibitors demonstrated antiproliferation activities in AF9-sensitive HGC-27 cells, which recapitulated the phenotype of the CRISPR studies against AF9. Our work will provide the basis for further structured-based optimization and reignite the campaign for potent AF9 YEATS inhibitors as a precise treatment for AF9-sensitive cancers.

Structural basis for RNA 3'-end recognition by the PIWIL2 PAZ domain.

PIWI family proteins are important members of Argonaute family that play an essential role in spermatogenesis and development when loaded with piRNAs. Here we solved the crystal structure of the human PIWIL2 PAZ domain and found its PAZ domain adopts a canonical PAZ fold. We furhter built a homology model of PIWIL2 bound to 2 nt 3' overhangs. We found that PIWIL2 utilizes a deep hydrophobic concave to accommodate the 2 nt at 3'-end of RNAs. The recognition of 2 nt 3' overhangs by PIWIL2 is conserved in other human PIWIL proteins, implicating the evolutionarily conserved role of PAZ domain in binding to target RNAs.

Structural insights into SMCR8 C-degron recognition by FEM1B.

C-degrons play critical roles in targeting the receptor proteins of Cullin-RING E3 ligase complexes to initiate protein degradation. FEM1 proteins, including FEM1A, FEM1B, and FEM1C, act as the receptors to specifically recognize Arg/C-degrons to enable CRL2-mediated protein turnover. Very few substrates have been identified for FEM1B, except CDK5R1. We found that CRL2FEM1B also recognizes the C-degron of an SMCR8 isoform, and uncovered the recognition of SMCR8 by FEM1B through presenting the structure of FEM1B bound to SMCR8. Our work provides insights into the role of CRL2FEM1B in regulating the lifetime of SMCR8, a critical autophagy regulator.

Structural and biochemical insights into the recognition of RNA helicase CGH-1 by CAR-1 in C. elegans.

A protein-RNA complex containing the RNA helicase CGH-1 and a germline specific RNA-binding protein CAR-1 is involved in various aspects of function in C. elegans. However, the structural basis for the assembly of this protein complex remains unclear. Here, we elucidate the molecular basis of the recognition of CGH-1 by CAR-1. Additionally, we found that the ATPase activity of CGH-1 is stimulated by NTL-1a MIF4G domain in vitro. Furthermore, we determined the structures of the two RecA-like domains of CGH-1 by X-ray crystallography at resolutions of 1.85 and 2.40 Å, respectively. Structural and biochemical approaches revealed a bipartite interface between CGH-1 RecA2 and the FDF-TFG motif of CAR-1. NMR and structure-based mutations in CGH-1 RecA2 or CAR-1 attenuated or disrupted CGH-1 binding to CAR-1, assessed by ITC and GST-pulldown in vitro. These findings provide insights into a conserved mechanism in the recognition of CGH-1 by CAR-1. Together, our data provide the missing physical links in understanding the assembly and function of CGH-1 and CAR-1 in C. elegans.

Increasing the efficiency and accuracy of the ABACUS protein sequence design method.

MOTIVATION: The ABACUS (a backbone-based amino acid usage survey) method uses unique statistical energy functions to carry out protein sequence design. Although some of its results have been experimentally verified, its accuracy remains improvable because several important components of the method have not been specifically optimized for sequence design or in contexts of other parts of the method. The computational efficiency also needs to be improved to support interactive online applications or the consideration of a large number of alternative backbone structures. RESULTS: We derived a model to measure solvent accessibility with larger mutual information with residue types than previous models, optimized a set of rotamers which can approximate the sidechain atomic positions more accurately, and devised an empirical function to treat inter-atomic packing with parameters fitted to native structures and optimized in consistence with the rotamer set. Energy calculations have been accelerated by interpolation between pre-determined representative points in high-dimensional structural feature spaces. Sidechain repacking tests showed that ABACUS2 can accurately reproduce the conformation of native sidechains. In sequence design tests, the native residue type recovery rate reached 37.7%, exceeding the value of 32.7% for ABACUS1. Applying ABACUS2 to designed sequences on three native backbones produced proteins shown to be well-folded by experiments. AVAILABILITY AND IMPLEMENTATION: The ABACUS2 sequence design server can be visited at http://biocomp.ustc.edu.cn/servers/abacus-design.php. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Solution structure of TbUfm1 from Trypanosoma brucei and its binding to TbUba5.

Ubiquitin-like proteins are conserved in eukaryotes and involved in numerous cellular processes. Ufm1 is proved to play important roles in endoplasmic reticulum homeostasis, vesicle transportation and embryonic development. Enzyme cascade of Ufm1 is similar to that of ubiquitin. Mature Ufm1 is activated and conjugated to substrates by assistance of Ufm1 activating enzyme Uba5 (E1), Ufm1 conjugating enzyme Ufc1 (E2), and Ufm1 ligating enzyme Ufl1 (E3). Here, we determined the solution structure of TbUfm1 from Trypanosoma brucei (T. brucei) by NMR spectroscopy and explored the interactions between TbUfm1 and TbUba5/TbUfc1/TbUfl1. TbUfm1 adopts a typical ß-grasp fold, which partially wraps a central α-helix and the other two helixes. NMR chemical shift perturbation indicated that TbUfm1 interacts with TbUba5 via a hydrophobic pocket formed by α1α2ß1ß2. Although the structure and Uba5-interaction mode of TbUfm1 are conserved in Ufm1 proteins, there are also some differences, which might reflect the potential diversity of Ufm1 in evolution and biological functions.

Multi-functional regulator MapZ controls both positioning and timing of FtsZ polymerization.

The tubulin-like GTPase protein FtsZ, which forms a discontinuous cytokinetic ring at mid-cell, is a central player to recruit the division machinery to orchestrate cell division. To guarantee the production of two identical daughter cells, the assembly of FtsZ, namely Z-ring, and its precise positioning should be finely regulated. In Streptococcus pneumoniae, the positioning of Z-ring at the division site is mediated by a bitopic membrane protein MapZ (mid-cell-anchored protein Z) through direct interactions between the intracellular domain (termed MapZ-N (the intracellular domain of MapZ)) and FtsZ. Using nuclear magnetic resonance titration experiments, we clearly assigned the key residues involved in the interactions. In the presence of MapZ-N, FtsZ gains a shortened activation delay, a lower critical concentration for polymerization and a higher cooperativity towards GTP hydrolysis. On the other hand, MapZ-N antagonizes the lateral interactions of single-stranded filaments of FtsZ, thus slows down the formation of highly bundled FtsZ polymers and eventually maintains FtsZ at a dynamic state. Altogether, we conclude that MapZ is not only an accelerator to trigger the polymerization of FtsZ, but also a brake to tune the velocity to form the end-product, FtsZ bundles. These findings suggest that MapZ is a multi-functional regulator towards FtsZ that controls both the precise positioning and proper timing of FtsZ polymerization.

Solution structure of TbTFIIS2-2 PWWP domain from Trypanosoma brucei and its binding to H4K17me3 and H3K32me3.

Posttranslational modifications (PTMs) of core histones, such as histone methylation, play critical roles in a variety of biological processes including transcription regulation, chromatin condensation and DNA repair. In T. brucei, no domain recognizing methylated histone has been identified so far. TbTFIIS2-2, as a potential transcription elongation factors in T. brucei, contains a PWWP domain in the N-terminus which shares low sequence similarity compared with other PWWP domains and is absent from other TFIIS factors. In the present study, the solution structure of TbTFIIS2-2 PWWP domain was determined by NMR spectroscopy. TbTFIIS2-2 PWWP domain adopts a global fold containing a five-strand ß-barrel and two C-terminal α-helices similar to other PWWP domains. Moreover, through systematic screening, we revealed that TbTFIIS2-2 PWWP domain is able to bind H4K17me3 and H3K32me3. Meanwhile, we identified the critical residues responsible for the binding ability of TbTFIIS2-2 PWWP domain. The conserved cage formed by the aromatic amino acids in TbTFIIS2-2 PWWP domain is essential for its binding to methylated histones.

Combinatorial readout of unmodified H3R2 and acetylated H3K14 by the tandem PHD finger of MOZ reveals a regulatory mechanism for HOXA9 transcription.

Histone acetylation is a hallmark for gene transcription. As a histone acetyltransferase, MOZ (monocytic leukemia zinc finger protein) is important for HOX gene expression as well as embryo and postnatal development. In vivo, MOZ forms a tetrameric complex with other subunits, including several chromatin-binding modules with regulatory functions. Here we report the solution structure of the tandem PHD (plant homeodomain) finger (PHD12) of human MOZ in a free state and the 1.47 Å crystal structure in complex with H3K14ac peptide, which reveals the structural basis for the recognition of unmodified R2 and acetylated K14 on histone H3. Moreover, the results of chromatin immunoprecipitation (ChIP) and RT-PCR assays indicate that PHD12 facilitates the localization of MOZ onto the promoter locus of the HOXA9 gene, thereby promoting the H3 acetylation around the promoter region and further up-regulating the HOXA9 mRNA level. Taken together, our findings suggest that the combinatorial readout of the H3R2/K14ac by PHD12 might represent an important epigenetic regulatory mechanism that governs transcription and also provide a clue of cross-talk between the MOZ complex and histone H3 modifications.

Structure and allosteric coupling of type â ¡ antitoxin CopASO.

Toxin-antitoxin (TA) systems play critical roles in the environment adaptation of bacteria. Allosteric coupling between the N-terminal DNA-binding domain and the C-terminal toxin-binding domain of antitoxins contributes to conditional cooperativity in the functioning of type II TA. Herein, using circular dichroism (CD), nuclear magnetic resonance (NMR), X-ray crystallography, and size exclusion chromatography (SEC), the structure and DNA binding of CopASO, a newly identified type II antitoxin in Shewanella oneidensis, were investigated. Our data show that CopASO is a typical RHH antitoxin with an ordered N-terminal domain and a disordered C-terminal domain, and furthermore indicate that the C-terminal domain facilitates DNA binding of the N-terminal domain, which in turn induces the C-terminal domain to fold and associate.

Modulation of the aggregation of an amyloidogenic sequence by flanking-disordered region in the intrinsically disordered antigen merozoite surface protein 2.

The abundant Plasmodium falciparum merozoite surface protein MSP2, a potential malaria vaccine candidate, is an intrinsically disordered protein with some nascent secondary structure present in its conserved N-terminal region. This relatively ordered region has been implicated in both membrane interactions and amyloid-like aggregation of the protein, while the significance of the flanking-disordered region is unclear. In this study, we show that aggregation of the N-terminal conserved region of MSP2 is influenced in a length- and sequence-dependent fashion by the disordered central variable sequences. Intriguingly, MSP2 peptides containing the conserved region and the first five residues of the variable disordered regions aggregated more rapidly than a peptide corresponding to the conserved region alone. In contrast, MSP2 peptides extending 8 or 12 residues into the disordered region aggregated more slowly, consistent with the expected inhibitory effect of flanking-disordered sequences on the aggregation of amyloidogenic ordered sequences. Computational analyses indicated that the helical propensity of the ordered region of MSP2 was modulated by the adjacent disordered five residues in a sequence-dependent manner. Nuclear magnetic resonance and circular dichroism spectroscopic studies with synthetic peptides confirmed the computational predictions, emphasizing the correlation between aggregation propensity and conformation of the ordered region and the effects thereon of the adjacent disordered region. These results show that the effects of flanking-disordered sequences on a more ordered sequence may include enhancement of aggregation through modulation of the conformational properties of the more ordered sequence.

Solution structure of TbCentrin4 from Trypanosoma brucei and its interactions with Ca2+ and other centrins.

Centrin is a conserved calcium-binding protein that plays an important role in diverse cellular biological processes such as ciliogenesis, gene expression, DNA repair and signal transduction. In Trypanosoma brucei, TbCentrin4 is mainly localized in basal bodies and bi-lobe structure, and is involved in the processes coordinating karyokinesis and cytokinesis. In the present study, we solved the solution structure of TbCentrin4 using NMR (nuclear magnetic resonance) spectroscopy. TbCentrin4 contains four EF-hand motifs consisting of eight α-helices. Isothermal titration calorimetry experiment showed that TbCentrin4 has a strong Ca2+ binding ability. NMR chemical shift perturbation indicated that TbCentrin4 binds to Ca2+ through its C-terminal domain composed of EF-hand 3 and 4. Meanwhile, we revealed that TbCentrin4 undergoes a conformational change and self-assembly induced by high concentration of Ca2+ Intriguingly, localization of TbCentrin4 was dispersed or disappeared from basal bodies and the bi-lobe structure when the cells were treated with Ca2+in vivo, implying the influence of Ca2+ on the cellular functions of TbCentrin4. Besides, we observed the interactions between TbCentrin4 and other Tbcentrins and revealed that the interactions are Ca2+ dependent. Our findings provide a structural basis for better understanding the biological functions of TbCentrin4 in the relevant cellular processes.

Cooperation of Escherichia coli Hfq hexamers in DsrA binding.

Hfq is a bacterial post-transcriptional regulator. It facilitates base-pairing between sRNA and target mRNA. Hfq mediates DsrA-dependent translational activation of rpoS mRNA at low temperatures. rpoS encodes the stationary-phase σ factor σ(S), which is the central regulator in general stress response. However, structural information on Hfq-DsrA interaction is not yet available. Although Hfq is reported to hydrolyze ATP, the ATP-binding site is still unknown. Here, we report a ternary crystal complex structure of Escherichia coli Hfq bound to a major Hfq recognition region on DsrA (AU(6)A) together with ADP, and a crystal complex structure of Hfq bound to ADP. AU(6)A binds to the proximal and distal sides of two Hfq hexamers. ADP binds to a purine-selective site on the distal side and contacts conserved arginine or glutamine residues on the proximal side of another hexamer. This binding mode is different from previously postulated. The cooperation of two different Hfq hexamers upon nucleic acid binding in solution is verified by fluorescence polarization and solution nuclear magnetic resonance (NMR) experiments using fragments of Hfq and DsrA. Fluorescence resonance energy transfer conducted with full-length Hfq and DsrA also supports cooperation of Hfq hexamers upon DsrA binding. The implications of Hfq hexamer cooperation have been discussed.