Molecular recognition of morphine and fentanyl by the human µ-opioid receptor.

Morphine and fentanyl are among the most used opioid drugs that confer analgesia and unwanted side effects through both G protein and arrestin signaling pathways of µ-opioid receptor (µOR). Here, we report structures of the human µOR-G protein complexes bound to morphine and fentanyl, which uncover key differences in how they bind the receptor. We also report structures of µOR bound to TRV130, PZM21, and SR17018, which reveal preferential interactions of these agonists with TM3 side of the ligand-binding pocket rather than TM6/7 side. In contrast, morphine and fentanyl form dual interactions with both TM3 and TM6/7 regions. Mutations at the TM6/7 interface abolish arrestin recruitment of µOR promoted by morphine and fentanyl. Ligands designed to reduce TM6/7 interactions display preferential G protein signaling. Our results provide crucial insights into fentanyl recognition and signaling of µOR, which may facilitate rational design of next-generation analgesics.

Structural insights into the human D1 and D2 dopamine receptor signaling complexes.

The D1- and D2-dopamine receptors (D1R and D2R), which signal through Gs and Gi, respectively, represent the principal stimulatory and inhibitory dopamine receptors in the central nervous system. D1R and D2R also represent the main therapeutic targets for Parkinson's disease, schizophrenia, and many other neuropsychiatric disorders, and insight into their signaling is essential for understanding both therapeutic and side effects of dopaminergic drugs. Here, we report four cryoelectron microscopy (cryo-EM) structures of D1R-Gs and D2R-Gi signaling complexes with selective and non-selective dopamine agonists, including two currently used anti-Parkinson's disease drugs, apomorphine and bromocriptine. These structures, together with mutagenesis studies, reveal the conserved binding mode of dopamine agonists, the unique pocket topology underlying ligand selectivity, the conformational changes in receptor activation, and potential structural determinants for G protein-coupling selectivity. These results provide both a molecular understanding of dopamine signaling and multiple structural templates for drug design targeting the dopaminergic system.

Crystal Structures of Membrane Transporter MmpL3, an Anti-TB Drug Target.

Despite intensive efforts to discover highly effective treatments to eradicate tuberculosis (TB), it remains as a major threat to global human health. For this reason, new TB drugs directed toward new targets are highly coveted. MmpLs (Mycobacterial membrane proteins Large), which play crucial roles in transporting lipids, polymers and immunomodulators and which also extrude therapeutic drugs, are among the most important therapeutic drug targets to emerge in recent times. Here, crystal structures of mycobacterial MmpL3 alone and in complex with four TB drug candidates, including SQ109 (in Phase 2b-3 clinical trials), are reported. MmpL3 consists of a periplasmic pore domain and a twelve-helix transmembrane domain. Two Asp-Tyr pairs centrally located in this domain appear to be key facilitators of proton-translocation. SQ109, AU1235, ICA38, and rimonabant bind inside the transmembrane region and disrupt these Asp-Tyr pairs. This structural data will greatly advance the development of MmpL3 inhibitors as new TB drugs.

FMRP phosphorylation modulates neuronal translation through YTHDF1.

RNA-binding proteins (RBPs) control messenger RNA fate in neurons. Here, we report a mechanism that the stimuli-induced neuronal translation is mediated by phosphorylation of a YTHDF1-binding protein FMRP. Mechanistically, YTHDF1 can condense with ribosomal proteins to promote the translation of its mRNA targets. FMRP regulates this process by sequestering YTHDF1 away from the ribosome; upon neuronal stimulation, FMRP becomes phosphorylated and releases YTHDF1 for translation upregulation. We show that a new small molecule inhibitor of YTHDF1 can reverse fragile X syndrome (FXS) developmental defects associated with FMRP deficiency in an organoid model. Our study thus reveals that FMRP and its phosphorylation are important regulators of activity-dependent translation during neuronal development and stimulation and identifies YTHDF1 as a potential therapeutic target for FXS in which developmental defects caused by FMRP depletion could be reversed through YTHDF1 inhibition.

Structures of the human dopamine D3 receptor-Gi complexes.

The dopamine system, including five dopamine receptors (D1R-D5R), plays essential roles in the central nervous system (CNS), and ligands that activate dopamine receptors have been used to treat many neuropsychiatric disorders. Here, we report two cryo-EM structures of human D3R in complex with an inhibitory G protein and bound to the D3R-selective agonists PD128907 and pramipexole, the latter of which is used to treat patients with Parkinson's disease. The structures reveal agonist binding modes distinct from the antagonist-bound D3R structure and conformational signatures for ligand-induced receptor activation. Mutagenesis and homology modeling illuminate determinants of ligand specificity across dopamine receptors and the mechanisms for Gi protein coupling. Collectively our work reveals the basis of agonist binding and ligand-induced receptor activation and provides structural templates for designing specific ligands to treat CNS diseases targeting the dopaminergic system.

Hormone- and antibody-mediated activation of the thyrotropin receptor.

Thyroid-stimulating hormone (TSH), through activation of its G-protein-coupled thyrotropin receptor (TSHR), controls the synthesis of thyroid hormone-an essential metabolic hormone1-3. Aberrant signalling of TSHR by autoantibodies causes Graves' disease (hyperthyroidism) and hypothyroidism, both of which affect millions of patients worldwide4. Here we report the active structures of TSHR with TSH and the activating autoantibody M225, both bound to the allosteric agonist ML-1096, as well as an inactivated TSHR structure with the inhibitory antibody K1-707. Both TSH and M22 push the extracellular domain (ECD) of TSHR into an upright active conformation. By contrast, K1-70 blocks TSH binding and cannot push the ECD into the upright conformation. Comparisons of the active and inactivated structures of TSHR with those of the luteinizing hormone/choriogonadotropin receptor (LHCGR) reveal a universal activation mechanism of glycoprotein hormone receptors, in which a conserved ten-residue fragment (P10) from the hinge C-terminal loop mediates ECD interactions with the TSHR transmembrane domain8. One notable feature is that there are more than 15 cholesterols surrounding TSHR, supporting its preferential location in lipid rafts9. These structures also highlight a similar ECD-push mechanism for TSH and autoantibody M22 to activate TSHR, therefore providing the molecular basis for Graves' disease.

Structures of full-length glycoprotein hormone receptor signalling complexes.

Luteinizing hormone and chorionic gonadotropin are glycoprotein hormones that are related to follicle-stimulating hormone and thyroid-stimulating hormone1,2. Luteinizing hormone and chorionic gonadotropin are essential to human reproduction and are important therapeutic drugs3-6. They activate the same G-protein-coupled receptor, luteinizing hormone-choriogonadotropin receptor (LHCGR), by binding to the large extracellular domain3. Here we report four cryo-electron microscopy structures of LHCGR: two structures of the wild-type receptor in the inactive and active states; and two structures of the constitutively active mutated receptor. The active structures are bound to chorionic gonadotropin and the stimulatory G protein (Gs), and one of the structures also contains Org43553, an allosteric agonist7. The structures reveal a distinct 'push-and-pull' mechanism of receptor activation, in which the extracellular domain is pushed by the bound hormone and pulled by the extended hinge loop next to the transmembrane domain. A highly conserved 10-residue fragment (P10) from the hinge C-terminal loop at the interface between the extracellular domain and the transmembrane domain functions as a tethered agonist to induce conformational changes in the transmembrane domain and G-protein coupling. Org43553 binds to a pocket of the transmembrane domain and interacts directly with P10, which further stabilizes the active conformation. Together, these structures provide a common model for understanding the signalling of glycoprotein hormone receptors and a basis for drug discovery for endocrine diseases.

Structural insights into the lipid and ligand regulation of serotonin receptors.

Serotonin, or 5-hydroxytryptamine (5-HT), is an important neurotransmitter1,2 that activates the largest subtype family of G-protein-coupled receptors3. Drugs that target 5-HT1A, 5-HT1D, 5-HT1E and other 5-HT receptors are used to treat numerous disorders4. 5-HT receptors have high levels of basal activity and are subject to regulation by lipids, but the structural basis for the lipid regulation and basal activation of these receptors and the pan-agonism of 5-HT remains unclear. Here we report five structures of 5-HT receptor-G-protein complexes: 5-HT1A in the apo state, bound to 5-HT or bound to the antipsychotic drug aripiprazole; 5-HT1D bound to 5-HT; and 5-HT1E in complex with a 5-HT1E- and 5-HT1F-selective agonist, BRL-54443. Notably, the phospholipid phosphatidylinositol 4-phosphate is present at the G-protein-5-HT1A interface, and is able to increase 5-HT1A-mediated G-protein activity. The receptor transmembrane domain is surrounded by cholesterol molecules-particularly in the case of 5-HT1A, in which cholesterol molecules are directly involved in shaping the ligand-binding pocket that determines the specificity for aripiprazol. Within the ligand-binding pocket of apo-5-HT1A are structured water molecules that mimic 5-HT to activate the receptor. Together, our results address a long-standing question of how lipids and water molecules regulate G-protein-coupled receptors, reveal how 5-HT acts as a pan-agonist, and identify the determinants of drug recognition in 5-HT receptors.

Structure of Mpro from SARS-CoV-2 and discovery of its inhibitors.

A new coronavirus, known as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is the aetiological agent responsible for the 2019-2020 viral pneumonia outbreak of coronavirus disease 2019 (COVID-19)1-4. Currently, there are no targeted therapeutic agents for the treatment of this disease, and effective treatment options remain very limited. Here we describe the results of a programme that aimed to rapidly discover lead compounds for clinical use, by combining structure-assisted drug design, virtual drug screening and high-throughput screening. This programme focused on identifying drug leads that target main protease (Mpro) of SARS-CoV-2: Mpro is a key enzyme of coronaviruses and has a pivotal role in mediating viral replication and transcription, making it an attractive drug target for SARS-CoV-25,6. We identified a mechanism-based inhibitor (N3) by computer-aided drug design, and then determined the crystal structure of Mpro of SARS-CoV-2 in complex with this compound. Through a combination of structure-based virtual and high-throughput screening, we assayed more than 10,000 compounds-including approved drugs, drug candidates in clinical trials and other pharmacologically active compounds-as inhibitors of Mpro. Six of these compounds inhibited Mpro, showing half-maximal inhibitory concentration values that ranged from 0.67 to 21.4 µM. One of these compounds (ebselen) also exhibited promising antiviral activity in cell-based assays. Our results demonstrate the efficacy of our screening strategy, which can lead to the rapid discovery of drug leads with clinical potential in response to new infectious diseases for which no specific drugs or vaccines are available.

Bioactive compounds from Huashi Baidu decoction possess both antiviral and anti-inflammatory effects against COVID-19.

The coronavirus disease 2019 (COVID-19) pandemic is an ongoing global health concern, and effective antiviral reagents are urgently needed. Traditional Chinese medicine theory-driven natural drug research and development (TCMT-NDRD) is a feasible method to address this issue as the traditional Chinese medicine formulae have been shown effective in the treatment of COVID-19. Huashi Baidu decoction (Q-14) is a clinically approved formula for COVID-19 therapy with antiviral and anti-inflammatory effects. Here, an integrative pharmacological strategy was applied to identify the antiviral and anti-inflammatory bioactive compounds from Q-14. Overall, a total of 343 chemical compounds were initially characterized, and 60 prototype compounds in Q-14 were subsequently traced in plasma using ultrahigh-performance liquid chromatography with quadrupole time-of-flight mass spectrometry. Among the 60 compounds, six compounds (magnolol, glycyrrhisoflavone, licoisoflavone A, emodin, echinatin, and quercetin) were identified showing a dose-dependent inhibition effect on the SARS-CoV-2 infection, including two inhibitors (echinatin and quercetin) of the main protease (Mpro), as well as two inhibitors (glycyrrhisoflavone and licoisoflavone A) of the RNA-dependent RNA polymerase (RdRp). Meanwhile, three anti-inflammatory components, including licochalcone B, echinatin, and glycyrrhisoflavone, were identified in a SARS-CoV-2-infected inflammatory cell model. In addition, glycyrrhisoflavone and licoisoflavone A also displayed strong inhibitory activities against cAMP-specific 3',5'-cyclic phosphodiesterase 4 (PDE4). Crystal structures of PDE4 in complex with glycyrrhisoflavone or licoisoflavone A were determined at resolutions of 1.54 Å and 1.65 Å, respectively, and both compounds bind in the active site of PDE4 with similar interactions. These findings will greatly stimulate the study of TCMT-NDRD against COVID-19.

An inductive graph neural network model for compound-protein interaction prediction based on a homogeneous graph.

Identifying the potential compound-protein interactions (CPIs) plays an essential role in drug development. The computational approaches for CPI prediction can reduce time and costs of experimental methods and have benefited from the continuously improved graph representation learning. However, most of the network-based methods use heterogeneous graphs, which is challenging due to their complex structures and heterogeneous attributes. Therefore, in this work, we transformed the compound-protein heterogeneous graph to a homogeneous graph by integrating the ligand-based protein representations and overall similarity associations. We then proposed an Inductive Graph AggrEgator-based framework, named CPI-IGAE, for CPI prediction. CPI-IGAE learns the low-dimensional representations of compounds and proteins from the homogeneous graph in an end-to-end manner. The results show that CPI-IGAE performs better than some state-of-the-art methods. Further ablation study and visualization of embeddings reveal the advantages of the model architecture and its role in feature extraction, and some of the top ranked CPIs by CPI-IGAE have been validated by a review of recent literature. The data and source codes are available at https://github.com/wanxiaozhe/CPI-IGAE.

Structure of the glucagon receptor in complex with a glucagon analogue.

Class B G-protein-coupled receptors (GPCRs), which consist of an extracellular domain (ECD) and a transmembrane domain (TMD), respond to secretin peptides to play a key part in hormonal homeostasis, and are important therapeutic targets for a variety of diseases. Previous work has suggested that peptide ligands bind to class B GPCRs according to a two-domain binding model, in which the C-terminal region of the peptide targets the ECD and the N-terminal region of the peptide binds to the TMD binding pocket. Recently, three structures of class B GPCRs in complex with peptide ligands have been solved. These structures provide essential insights into peptide ligand recognition by class B GPCRs. However, owing to resolution limitations, the specific molecular interactions for peptide binding to class B GPCRs remain ambiguous. Moreover, these previously solved structures have different ECD conformations relative to the TMD, which introduces questions regarding inter-domain conformational flexibility and the changes required for receptor activation. Here we report the 3.0 Å-resolution crystal structure of the full-length human glucagon receptor (GCGR) in complex with a glucagon analogue and partial agonist, NNC1702. This structure provides molecular details of the interactions between GCGR and the peptide ligand. It reveals a marked change in the relative orientation between the ECD and TMD of GCGR compared to the previously solved structure of the inactive GCGR-NNC0640-mAb1 complex. Notably, the stalk region and the first extracellular loop undergo major conformational changes in secondary structure during peptide binding, forming key interactions with the peptide. We further propose a dual-binding-site trigger model for GCGR activation-which requires conformational changes of the stalk, first extracellular loop and TMD-that extends our understanding of the previously established two-domain peptide-binding model of class B GPCRs.

High-Throughput-Methyl-Reading (HTMR) assay: a solution based on nucleotide methyl-binding proteins enables large-scale screening for DNA/RNA methyltransferases and demethylases.

Epigenetic therapy has significant potential for cancer treatment. However, few small potent molecules have been identified against DNA or RNA modification regulatory proteins. Current approaches for activity detection of DNA/RNA methyltransferases and demethylases are time-consuming and labor-intensive, making it difficult to subject them to high-throughput screening. Here, we developed a fluorescence polarization-based 'High-Throughput Methyl Reading' (HTMR) assay to implement large-scale compound screening for DNA/RNA methyltransferases and demethylases-DNMTs, TETs, ALKBH5 and METTL3/METTL14. This assay is simple to perform in a mix-and-read manner by adding the methyl-binding proteins MBD1 or YTHDF1. The proteins can be used to distinguish FAM-labelled substrates or product oligonucleotides with different methylation statuses catalyzed by enzymes. Therefore, the extent of the enzymatic reactions can be coupled with the variation of FP binding signals. Furthermore, this assay can be effectively used to conduct a cofactor competition study. Based on the assay, we identified two natural products as candidate compounds for DNMT1 and ALKBH5. In summary, this study outlines a powerful homogeneous approach for high-throughput screening and evaluating enzymatic activity for DNA/RNA methyltransferases and demethylases that is cheap, easy, quick, and highly sensitive.

Molecular insights into differentiated ligand recognition of the human parathyroid hormone receptor 2.

The parathyroid hormone receptor 2 (PTH2R) is a class B1 G protein-coupled receptor (GPCR) involved in the regulation of calcium transport, nociception mediation, and wound healing. Naturally occurring mutations in PTH2R were reported to cause hereditary diseases, including syndromic short stature. Here, we report the cryogenic electron microscopy structure of PTH2R bound to its endogenous ligand, tuberoinfundibular peptide (TIP39), and a heterotrimeric Gs protein at a global resolution of 2.8 Å. The structure reveals that TIP39 adopts a unique loop conformation at the N terminus and deeply inserts into the orthosteric ligand-binding pocket in the transmembrane domain. Molecular dynamics simulation and site-directed mutagenesis studies uncover the basis of ligand specificity relative to three PTH2R agonists, TIP39, PTH, and PTH-related peptide. We also compare the action of TIP39 with an antagonist lacking six residues from the peptide N terminus, TIP(7-39), which underscores the indispensable role of the N terminus of TIP39 in PTH2R activation. Additionally, we unveil that a disease-associated mutation G258D significantly diminished cAMP accumulation induced by TIP39. Together, these results not only provide structural insights into ligand specificity and receptor activation of class B1 GPCRs but also offer a foundation to systematically rationalize the available pharmacological data to develop therapies for various disorders associated with PTH2R.

Constitutive signal bias mediated by the human GHRHR splice variant 1.

Alternative splicing of G protein-coupled receptors has been observed, but their functions are largely unknown. Here, we report that a splice variant (SV1) of the human growth hormone-releasing hormone receptor (GHRHR) is capable of transducing biased signal. Differing only at the receptor N terminus, GHRHR predominantly activates Gs while SV1 selectively couples to ß-arrestins. Based on the cryogenic electron microscopy structures of SV1 in the apo state or GHRH-bound state in complex with the Gs protein, molecular dynamics simulations reveal that the N termini of GHRHR and SV1 differentiate the downstream signaling pathways, Gs versus ß-arrestins. As suggested by mutagenesis and functional studies, it appears that GHRH-elicited signal bias toward ß-arrestin recruitment is constitutively mediated by SV1. The level of SV1 expression in prostate cancer cells is also positively correlated with ERK1/2 phosphorylation but negatively correlated with cAMP response. Our findings imply that constitutive signal bias may be a mechanism that ensures cancer cell proliferation.

Multi-instance learning of graph neural networks for aqueous pKa prediction.

MOTIVATION: The acid dissociation constant (pKa) is a critical parameter to reflect the ionization ability of chemical compounds and is widely applied in a variety of industries. However, the experimental determination of pKa is intricate and time-consuming, especially for the exact determination of micro-pKa information at the atomic level. Hence, a fast and accurate prediction of pKa values of chemical compounds is of broad interest. RESULTS: Here, we compiled a large-scale pKa dataset containing 16 595 compounds with 17 489 pKa values. Based on this dataset, a novel pKa prediction model, named Graph-pKa, was established using graph neural networks. Graph-pKa performed well on the prediction of macro-pKa values, with a mean absolute error around 0.55 and a coefficient of determination around 0.92 on the test dataset. Furthermore, combining multi-instance learning, Graph-pKa was also able to automatically deconvolute the predicted macro-pKa into discrete micro-pKa values. AVAILABILITY AND IMPLEMENTATION: The Graph-pKa model is now freely accessible via a web-based interface (https://pka.simm.ac.cn/). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Loops Mediate Agonist-Induced Activation of the Stimulator of Interferon Genes Protein.

The stimulator of interferon genes (STING) is an important therapeutic target for cancer diseases. The activated STING recruits downstream tank-binding kinase 1 (TBK1) to trigger several important immune responses. However, the molecular mechanism of how agonist molecules mediate the STING-TBK1 interactions remains elusive. Here, we performed molecular dynamics simulations to capture the conformational changes of STING and TBK1 upon agonist binding. Our simulations revealed that multiple helices (α5-α7) and especially three loops (loop 6, loop 8, and C-terminal tail) of STING participated in the allosteric mediation of the STING-TBK1 interactions. Consistent results were also observed in the simulations of the constitutive activating mutant of STING (R284S). We further identified α5 as a key region in this agonist-induced activation mechanism of STING. Free-energy perturbation calculations of multiple STING agonists demonstrated that an alkynyl group targeting α5 is a determinant for agonist activities. These results not only offer deeper insights into the agonist-induced allosteric mediation of STING-TKB1 interactions but also provide a guidance for future drug development of this important therapeutic target.

Structure of the full-length glucagon class B G-protein-coupled receptor.

The human glucagon receptor, GCGR, belongs to the class B G-protein-coupled receptor family and plays a key role in glucose homeostasis and the pathophysiology of type 2 diabetes. Here we report the 3.0 Å crystal structure of full-length GCGR containing both the extracellular domain and transmembrane domain in an inactive conformation. The two domains are connected by a 12-residue segment termed the stalk, which adopts a ß-strand conformation, instead of forming an α-helix as observed in the previously solved structure of the GCGR transmembrane domain. The first extracellular loop exhibits a ß-hairpin conformation and interacts with the stalk to form a compact ß-sheet structure. Hydrogen-deuterium exchange, disulfide crosslinking and molecular dynamics studies suggest that the stalk and the first extracellular loop have critical roles in modulating peptide ligand binding and receptor activation. These insights into the full-length GCGR structure deepen our understanding of the signalling mechanisms of class B G-protein-coupled receptors.

AlphaFold2 versus experimental structures: evaluation on G protein-coupled receptors.

As important drug targets, G protein-coupled receptors (GPCRs) play pivotal roles in a wide range of physiological processes. Extensive efforts of structural biology have been made on the study of GPCRs. However, a large portion of GPCR structures remain unsolved due to structural instability. Recently, AlphaFold2 has been developed to predict structure models of many functionally important proteins including all members of the GPCR family. Herein we evaluated the accuracy of GPCR structure models predicted by AlphaFold2. We revealed that AlphaFold2 could capture the overall backbone features of the receptors. However, the predicted models and experimental structures were different in many aspects including the assembly of the extracellular and transmembrane domains, the shape of the ligand-binding pockets, and the conformation of the transducer-binding interfaces. These differences impeded the use of predicted structure models in the functional study and structure-based drug design of GPCRs, which required reliable high-resolution structural information.

143D, a novel selective KRASG12C inhibitor exhibits potent antitumor activity in preclinical models.

The KRASG12C mutant has emerged as an important therapeutic target in recent years. Covalent inhibitors have shown promising antitumor activity against KRASG12C-mutant cancers in the clinic. In this study, a structure-based and focused chemical library analysis was performed, which led to the identification of 143D as a novel, highly potent and selective KRASG12C inhibitor. The antitumor efficacy of 143D in vitro and in vivo was comparable with that of AMG510 and of MRTX849, two well-characterized KRASG12C inhibitors. At low nanomolar concentrations, 143D showed biochemical and cellular potency for inhibiting the effects of the KRASG12C mutation. 143D selectively inhibited cell proliferation and induced G1-phase cell cycle arrest and apoptosis by downregulating KRASG12C-dependent signal transduction. Compared with MRTX849, 143D exhibited a longer half-life and higher maximum concentration (Cmax) and area under the curve (AUC) values in mouse models, as determined by tissue distribution assays. Additionally, 143D crossed the blood‒brain barrier. Treatment with 143D led to the sustained inhibition of KRAS signaling and tumor regression in KRASG12C-mutant tumors. Moreover, 143D combined with EGFR/MEK/ERK signaling inhibitors showed enhanced antitumor activity both in vitro and in vivo. Taken together, our findings indicate that 143D may be a promising drug candidate with favorable pharmaceutical properties for the treatment of cancers harboring the KRASG12C mutation.