This article was generated entirely by AI without human authorship. Please read with discretion, or have an AI verify its accuracy.
AI TranslationSimplified ChineseEnglish
Key Insights
This issue features 14 papers covering GPCR signaling and drug discovery, sweet and bitter taste receptors, multi-crystal data processing, protein design, and bile acid transport.
GPCR Conformations, Signaling Regulation, and Drug Discovery (5 papers)
01 · Designing rigid protein fiducials to visualize GPCR conformational states
Authors: Alina A. Vo, Yuan-En Sun, Justin G. English, Michael J. Robertson
Without signaling partners such as G proteins, small, dynamic GPCRs are difficult to align reliably in cryo-EM, limiting direct observation of ligand-induced conformations. This study combines generative protein design with deep-learning predictions of conformational ensembles to design protein fiducials that provide rigid references at different fusion sites. The authors validate the approach by determining inactive-state structures of four pharmacologically relevant GPCRs. They also design an extracellular fiducial for the β2-adrenergic receptor, allowing key intracellular structural changes to be visualized without a G protein. Comparisons with recent co-folding prediction models reveal substantial shortcomings in current methods for predicting ligand-induced conformations. The work uses engineered fiducials to expand the range of observable receptor states and underscores the need for experimental structures to test predictions.
Highlight · Designed rigid fiducials expand the range of GPCR conformational states that can be visualized without signaling partners.
Authors: Kai Wang, Snezana T. Dimova, John N. Hanson, Abhishek Thakur, Youwei Xu, Yini Liu, Amy E. Moritz, Alexander Goldberg, Bing Xie, Jayachandra Rayadurgam, Feijun Wang, Ashley N. Nilson, Kathryn D. Luderman, R. Benjamin Free, Wen Hu, Jian Zhang, H. Eric Xu, Kevin J. Frankowski, Yue Wang, Lei Shi, David R. Sibley, Youwen Zhuang
GPCRs contain multiple allosteric sites, but determining whether these sites work together to enhance signaling requires both structural and functional evidence. Focusing on the dopamine D1 receptor, the authors determine cryo-EM structures with multiple positive allosteric modulators simultaneously occupying distinct pockets and observe marked functional synergy. Under the conditions tested, engaging all three sites increases dopamine potency by more than a thousandfold. Molecular dynamics analysis further suggests that occupancy at multiple sites restricts intracellular loop motion and stabilizes activation-associated structures, an effect that no single modulator achieves. The findings provide a mechanistic framework for cooperation among spatially distributed allosteric sites. This large change in pharmacological potency in an experimental system should not be equated directly with improved clinical efficacy or safety.
Highlight · Multiple allosteric pockets in the D1 receptor can work together to stabilize active conformations, markedly amplifying signaling responses in experiments.
A key question in pharmacology is whether the sustained antidepressant effects of psychedelics can be separated from their perception-altering properties at the level of receptor signaling. Combining structure-guided design of biased ligands with animal experiments, the authors propose that β-arrestin signaling through the 5-HT2A receptor contributes to longer-lasting antidepressant-like behavioral effects in mice, while ligand dissociation kinetics influence receptor occupancy and responses across signaling pathways. The study reports that specific experimental dosing conditions preserve the relevant behavioral effects without detectable disturbances in the hallucinogen-associated behaviors assessed, with observations in mice and cynomolgus monkeys. The work offers mechanistic clues for disentangling receptor signaling and behavioral outcomes. However, animal behavioral measures are not equivalent to human subjective experiences and cannot establish treatment regimens for humans.
Highlight · Animal studies implicate receptor kinetics and β-arrestin signaling in separating sustained antidepressant-like effects from behavioral disturbances.
04 · mechanoLight: A genetically encoded fluorescent biosensor for visualizing 5-HT2AR mechanotransduction in live cells
Proceedings of the National Academy of Sciences · Publication/issue date: 2026-10-27
Authors: In-Yeop Baek, Hyunbin Kim, Duwan Kwon, Hahnbeom Park, Jihye Seong
GPCRs can respond not only to chemical ligands but also to mechanical stimuli, yet distinguishing these modes of activation in living cells is difficult. mechanoLight is a red fluorescent sensor that reports mechanical activation of the 5-HT2A receptor. It is relatively sensitive to conformational changes induced by fluid shear stress but responds only weakly to serotonin. By contrast, the existing psychLight2 sensor preferentially reports ligand-induced activation. Comparisons of linker regions, structural modeling, and simulations suggest that sensor flexibility contributes to this difference. Using downstream signaling and mutational analyses, the authors also examine the structural basis of receptor mechanotransduction and extend the approach to other mechanosensitive GPCRs. The study provides a tool for distinguishing modes of receptor activation, while highlighting that fluorescent probes may selectively report particular conformations rather than all receptor activity.
Highlight · mechanoLight preferentially reports conformations associated with mechanical stimuli, providing a tool to distinguish different modes of GPCR activation.
Authors: Sijie Huang, Heng Liu, Yujin Wu, Joao M. Braz, Divya Kranthi, Brendan W. Hall, Xinyue Zhang, Dmytro S. Radchenko, Yurii S. Moroz, John J. Irwin, Allan I. Basbaum, H. Eric Xu, William C. Wetsel, Brian K. Shoichet
TAAR1 regulates monoamine signaling and offers a direction for antipsychotic drug research beyond conventional dopamine and serotonin targets. The study docks approximately 65 million molecules, tests 55 highly ranked candidates, and identifies 14 agonists. The authors note that the compact active-state pocket favors agonist screening, but a model optimized for the initial hit rate may be less useful for subsequent improvements in affinity. Further optimization yields candidates with nanomolar affinity, and cryo-EM supports their predicted binding poses. Three compounds with relatively high brain exposure show activity in a mouse prepulse inhibition model, with no catalepsy observed in the assessments performed. The findings link structure-based screening, optimization, and animal behavioral validation, but performance in these models does not directly establish efficacy or safety in patients with schizophrenia.
Highlight · Structure-based screening of TAAR1 identifies agonists active in animal models and reveals trade-offs between hit discovery and subsequent optimization.
The human sweet taste receptor is a heterodimer of TAS1R2 and TAS1R3 that recognizes structurally diverse sweet compounds, but its complete structure had long remained elusive. This study determines structures of the full-length receptor in ligand-free and sucralose-bound states, revealing an asymmetric dimer and locating the main sucralose-binding site in the Venus flytrap domain of TAS1R2. Mutational and molecular dynamics analyses further support this mode of sweetener recognition, while comparisons between states reveal conformational changes that propagate to other regions after ligand binding. The work provides a molecular basis for understanding signal transmission in class C chemosensory GPCRs and may help compare how different sweeteners are recognized. Its implications are primarily mechanistic: binding structures alone cannot establish a sweetener’s overall nutritional or metabolic effects.
Highlight · Full-length sweet taste receptor structures reveal an asymmetric dimer and show how TAS1R2 recognizes sucralose.
TAS2R16 is notably selective for bitter compounds containing a β-glucoside moiety, but the structural basis for recognizing this shared chemical feature and triggering signaling had been unclear. The study determines cryo-EM structures of salicin-activated human TAS2R16 in complexes with gustducin, Gi1, and Gi2, describing the ligand’s binding mode and the interactions between its glucoside moiety and the receptor. Molecular docking and mutational experiments further support recognition of specific chemical groups and link this process to activation and downstream G-protein coupling. Comparisons across signaling partners also reveal differences in receptor coupling. The work provides a structural basis for receptor selectivity toward bitter glycosides, although potential applications still require appropriate functional and physiological testing.
Highlight · Salicin-bound TAS2R16 structures explain how this bitter taste receptor selectively recognizes the glucoside moiety.
08 · Bitter taste TAS2R14 activation by intracellular tastants and cholesterol
Nature · Publication/issue date: 2024-07
Authors: Xiaolong Hu, Weizhen Ao, Mingxin Gao, Lijie Wu, Yuan Pei, Shenhui Liu, Yiran Wu, Fei Zhao, Qianqian Sun, Junlin Liu, Longquan Jiang, Xin Wang, Yan Li, Qiwen Tan, Jie Cheng, Fan Yang, Chi Yang, Jinpeng Sun, Tian Hua, Zhi-Jie Liu
TAS2R14 recognizes a wide range of bitter compounds and is also found in tissues outside the mouth. Cryo-EM structures of the receptor bound to different agonists and G proteins reveal a ligand arrangement unlike that of typical class A GPCRs: cholesterol occupies the region usually corresponding to the orthosteric pocket, while the agonists studied bind to intracellular pockets. Drawing on mutational experiments and molecular dynamics, the authors discuss how multiple binding sites jointly support broad ligand recognition and receptor activation, and compare coupling to gustducin and Gi1. The findings suggest that bitter taste receptors may use cooperation among multiple sites for chemosensation; a canonical binding mode established for one GPCR should not be assumed to apply across an entire receptor family. The structural observations also provide a basis for future tests of how membrane lipids affect sensory function.
Highlight · Cholesterol and intracellular agonist-binding pockets together underpin the multisite recognition mechanism of TAS2R14.
Bitter taste perception depends on the TAS2R family, but a lack of experimental structures had limited understanding of its recognition and activation mechanisms. This study determines cryo-EM structures of human TAS2R46 complexed with engineered gustducin in ligand-free and ligand-bound states. The structures reveal an architecture distinct from other GPCRs, activation-associated structural features, and a state pre-coupled to its signaling partner. Dynamics also vary across regions: the extracellular portion is more variable, while the intracellular portion is relatively stable, suggesting that chemical recognition and signal transmission may be coordinated through conformational changes at different levels. The work provides an early structural reference for the bitter taste receptor family and supports subsequent comparisons of receptor selectivity. Findings obtained with engineered signaling partners still need to be corroborated by studies of native cellular function.
Highlight · Structures of TAS2R46 in two states provide a reference for understanding the distinctive recognition and activation mechanisms of bitter taste receptors.
Authors: J. Foadi, P. Aller, Y. Alguel, A. Cameron, D. Axford, R. L. Owen, W. Armour, D. G. Waterman, S. Iwata, G. Evans
Synchrotron microbeams allow protein crystals smaller than ten micrometers to be used for structural studies. However, microcrystals and sensitive crystals are especially susceptible to radiation damage, often requiring data collection from multiple crystals followed by merging. Selecting and combining data from different crystals can therefore be time-consuming. This methods paper introduces BLEND, which uses clustering to help select datasets suitable for joint processing and automates parts of the multi-crystal analysis workflow. The authors demonstrate its use in determining the structure of a novel membrane protein, showing how computational selection can reduce the burden of manually comparing dataset combinations. The work focuses on organizing and selecting multi-crystal data, rather than improving quality simply by adding more crystals; successful merging still depends on sufficient consistency among datasets.
Highlight · BLEND uses clustering to select multi-crystal datasets, reducing manual work during selection and merging.
A single microcrystal often cannot provide a complete high-resolution diffraction dataset. Merging small-wedge data from multiple crystals offers a solution, but processing large numbers of incomplete datasets is challenging. KAMO integrates established programs such as XDS and CCP4 into an automated workflow. It processes individual datasets, groups results with similar unit-cell parameters, selects the space group and resolves indexing ambiguity, then performs clustering, outlier rejection, and merging before generating reports. Validation using simulated data and experimental data from hundreds of crystals demonstrates largely automated production of merged structure-factor amplitudes. The main contribution is to connect previously separate decisions and computational steps, easing the data-processing burden for challenging samples that yield only microcrystals while retaining quality control for differences between crystals and anomalous observations.
Highlight · KAMO brings the processing, clustering, and merging of data from multiple microcrystals into an automated analysis workflow.
12 · Design of intrinsically disordered region binding proteins
Science · Publication/issue date: 2025-07-17
Authors: Kejia Wu, Hanlun Jiang, Derrick R. Hicks, Caixuan Liu, Edin Muratspahić, Theresa A. Ramelot, Yuexuan Liu, Kerrie McNally, Sebastian Kenny, Andrei Mihut, Amit Gaur, Brian Coventry, Wei Chen, Asim K. Bera, Alex Kang, Stacey Gerben, Mila Ya-Lan Lamb, Analisa Murray, Xinting Li, Madison A. Kennedy, Wei Yang, Zihao Song, Gudrun Schober, Stuart M. Brierley, John O’Neill, Michael H. Gelb, Gaetano T. Montelione, Emmanuel Derivery, David Baker
Intrinsically disordered regions lack stable three-dimensional structures and vary widely in sequence and conformational preferences, making them longstanding challenges for the design of specific protein binders. This study presents a binder-design method in which target disordered peptides bind in a range of extended conformations, with their side chains fitting into complementary pockets. Testing 39 diverse unstructured targets, the authors obtain designs for 34 with affinities ranging from roughly 100 picomolar to 100 nanomolar, screening an average of about 22 candidates per target. Cross-binding experiments support selectivity for the intended targets, and the study also demonstrates intracellular activity and detection applications. The results show that the absence of a fixed native fold does not preclude the design of specific recognition, although performance must still be validated for each target and application setting.
Highlight · A design strategy targeting extended peptide conformations expands high-affinity recognition to diverse intrinsically disordered regions.
Protein design typically requires natural templates or the manual translation of functional goals into geometric and sequence constraints. Pinal seeks to generate candidate proteins directly from natural-language descriptions, linking intended function to sequence and structure. The authors present a model trained on large-scale protein–text pairs. Computational evaluations assess predicted foldability, alignment with the descriptions, and sequence diversity, while experiments demonstrate the intended functions in four tasks involving fluorescent proteins, two enzyme classes, and a metabolic H protein. In the latter case, a designed H protein yields a product titer 1.7 times that obtained with the corresponding natural protein in a multienzyme carbon-fixation system. The study supports natural language as a starting point for candidate generation, but computational alignment still requires experimental validation, and a few successful examples do not establish that arbitrary functional descriptions can be reliably realized.
Highlight · Pinal uses natural language as the starting point for generating candidate proteins, with functional validation across four task categories.
Authors: Kang Xu, Bo-Rui Li, GaoXing Yu- Huang, Fang-Fang Wang, Jie Wang, Bo Wang, Chen-Yang Chu, Guo-Ting Chen, Cong-Zhao Zhou, Yuxing Chen, Wen-Tao Hou
OSTα/β returns reabsorbed bile acids to the portal circulation, but how it transports them across the membrane had been unclear. Cryo-EM structures of the human transporter, including a taurocholate-bound state, reveal a seven-transmembrane architecture distinct from typical SLC transporters. The substrate occupies a lateral groove on the protein surface rather than a central cavity. Polar residues lining the groove participate in substrate recognition, with mutational and transport experiments supporting the importance of these contacts. Simulations also suggest that substrate movement along the surface groove is energetically favorable. On this basis, the authors propose a facilitated lateral diffusion mechanism involving relatively small conformational changes. Together, the structural, functional, and simulation results offer an explanation distinct from the typical alternating-access model, adding a new molecular perspective on bile acid homeostasis.
Highlight · A surface groove in OSTα/β supports bile acid transport across the membrane through facilitated lateral diffusion.
This issue features 14 papers covering GPCR conformations and signaling regulation, drug discovery, sweet and bitter taste receptors, multi-crystal data processing, protein design, and bile acid transport. Combining structural analysis, functional assays, and computational design, the studies reveal cooperative interactions among multiple receptor sites and mechanisms of ligand recognition. They also advance mechanosensing probes, automated microcrystal data processing, proteins that bind disordered regions, and natural-language-guided protein design, while proposing a mechanism for OSTα/β-mediated lateral diffusion of bile acids. The computational predictions, animal behavioral findings, and structural insights require further experimental and clinical validation.
This issue features 14 papers covering GPCR signaling and drug discovery, sweet and bitter taste receptors, multi-crystal data processing, protein design, and bile acid transport.
GPCR Conformations, Signaling Regulation, and Drug Discovery (5 papers)
01 · Designing rigid protein fiducials to visualize GPCR conformational states
Nature Communications · Publication/issue date: 2026-10-24
Authors: Alina A. Vo, Yuan-En Sun, Justin G. English, Michael J. Robertson
Without signaling partners such as G proteins, small, dynamic GPCRs are difficult to align reliably in cryo-EM, limiting direct observation of ligand-induced conformations. This study combines generative protein design with deep-learning predictions of conformational ensembles to design protein fiducials that provide rigid references at different fusion sites. The authors validate the approach by determining inactive-state structures of four pharmacologically relevant GPCRs. They also design an extracellular fiducial for the β2-adrenergic receptor, allowing key intracellular structural changes to be visualized without a G protein. Comparisons with recent co-folding prediction models reveal substantial shortcomings in current methods for predicting ligand-induced conformations. The work uses engineered fiducials to expand the range of observable receptor states and underscores the need for experimental structures to test predictions.
Highlight · Designed rigid fiducials expand the range of GPCR conformational states that can be visualized without signaling partners.
DOI: 10.1038/s41467-026-78942-7
02 · Multiple allosteric sites allow for synergistic enhancement of GPCR signaling
Science Advances · Publication/issue date: 2026-10-21
Authors: Kai Wang, Snezana T. Dimova, John N. Hanson, Abhishek Thakur, Youwei Xu, Yini Liu, Amy E. Moritz, Alexander Goldberg, Bing Xie, Jayachandra Rayadurgam, Feijun Wang, Ashley N. Nilson, Kathryn D. Luderman, R. Benjamin Free, Wen Hu, Jian Zhang, H. Eric Xu, Kevin J. Frankowski, Yue Wang, Lei Shi, David R. Sibley, Youwen Zhuang
GPCRs contain multiple allosteric sites, but determining whether these sites work together to enhance signaling requires both structural and functional evidence. Focusing on the dopamine D1 receptor, the authors determine cryo-EM structures with multiple positive allosteric modulators simultaneously occupying distinct pockets and observe marked functional synergy. Under the conditions tested, engaging all three sites increases dopamine potency by more than a thousandfold. Molecular dynamics analysis further suggests that occupancy at multiple sites restricts intracellular loop motion and stabilizes activation-associated structures, an effect that no single modulator achieves. The findings provide a mechanistic framework for cooperation among spatially distributed allosteric sites. This large change in pharmacological potency in an experimental system should not be equated directly with improved clinical efficacy or safety.
Highlight · Multiple allosteric pockets in the D1 receptor can work together to stabilize active conformations, markedly amplifying signaling responses in experiments.
DOI: 10.1126/sciadv.aeh3401
03 · Sustained antidepressant-like effects of psychedelics through 5-HT2AR–β-arrestin signaling
Nature Neuroscience · Publication/issue date: 2026-10-23
Authors: Licong He, Lingjie Tang, Dongmei Cao, Wenwen Duan, Jing Yu, Pei Zhang, Huiqiong Li, Yumeng Wang, Bing Meng, Huan Wang, Yue Zhou, Wenhao Li, Xiaoyu Li, Zhijie Liu, Lei Zhang, Jianjun Cheng, Sheng Wang
A key question in pharmacology is whether the sustained antidepressant effects of psychedelics can be separated from their perception-altering properties at the level of receptor signaling. Combining structure-guided design of biased ligands with animal experiments, the authors propose that β-arrestin signaling through the 5-HT2A receptor contributes to longer-lasting antidepressant-like behavioral effects in mice, while ligand dissociation kinetics influence receptor occupancy and responses across signaling pathways. The study reports that specific experimental dosing conditions preserve the relevant behavioral effects without detectable disturbances in the hallucinogen-associated behaviors assessed, with observations in mice and cynomolgus monkeys. The work offers mechanistic clues for disentangling receptor signaling and behavioral outcomes. However, animal behavioral measures are not equivalent to human subjective experiences and cannot establish treatment regimens for humans.
Highlight · Animal studies implicate receptor kinetics and β-arrestin signaling in separating sustained antidepressant-like effects from behavioral disturbances.
DOI: 10.1038/s41593-026-02497-5
04 · mechanoLight: A genetically encoded fluorescent biosensor for visualizing 5-HT2AR mechanotransduction in live cells
Proceedings of the National Academy of Sciences · Publication/issue date: 2026-10-27
Authors: In-Yeop Baek, Hyunbin Kim, Duwan Kwon, Hahnbeom Park, Jihye Seong
GPCRs can respond not only to chemical ligands but also to mechanical stimuli, yet distinguishing these modes of activation in living cells is difficult. mechanoLight is a red fluorescent sensor that reports mechanical activation of the 5-HT2A receptor. It is relatively sensitive to conformational changes induced by fluid shear stress but responds only weakly to serotonin. By contrast, the existing psychLight2 sensor preferentially reports ligand-induced activation. Comparisons of linker regions, structural modeling, and simulations suggest that sensor flexibility contributes to this difference. Using downstream signaling and mutational analyses, the authors also examine the structural basis of receptor mechanotransduction and extend the approach to other mechanosensitive GPCRs. The study provides a tool for distinguishing modes of receptor activation, while highlighting that fluorescent probes may selectively report particular conformations rather than all receptor activity.
Highlight · mechanoLight preferentially reports conformations associated with mechanical stimuli, providing a tool to distinguish different modes of GPCR activation.
DOI: 10.1073/pnas.2610816123
05 · Structure based discovery of antipsychotic-like TAAR1 agonists
Nature Communications · Publication/issue date: 2026-09-17
Authors: Sijie Huang, Heng Liu, Yujin Wu, Joao M. Braz, Divya Kranthi, Brendan W. Hall, Xinyue Zhang, Dmytro S. Radchenko, Yurii S. Moroz, John J. Irwin, Allan I. Basbaum, H. Eric Xu, William C. Wetsel, Brian K. Shoichet
TAAR1 regulates monoamine signaling and offers a direction for antipsychotic drug research beyond conventional dopamine and serotonin targets. The study docks approximately 65 million molecules, tests 55 highly ranked candidates, and identifies 14 agonists. The authors note that the compact active-state pocket favors agonist screening, but a model optimized for the initial hit rate may be less useful for subsequent improvements in affinity. Further optimization yields candidates with nanomolar affinity, and cryo-EM supports their predicted binding poses. Three compounds with relatively high brain exposure show activity in a mouse prepulse inhibition model, with no catalepsy observed in the assessments performed. The findings link structure-based screening, optimization, and animal behavioral validation, but performance in these models does not directly establish efficacy or safety in patients with schizophrenia.
Highlight · Structure-based screening of TAAR1 identifies agonists active in animal models and reveals trade-offs between hit discovery and subsequent optimization.
DOI: 10.1038/s41467-026-77484-2
Sweet and Bitter Taste Receptors (4 papers)
06 · Structural and functional characterization of human sweet taste receptor
Nature · Publication/issue date: 2025-09
Authors: Zongjun Shi, Weixiu Xu, Lijie Wu, Xiaolei Yue, Shenhui Liu, Wei Ding, Jinyi Zhang, Bing Meng, Lianghao Zhao, Xiaoyan Liu, Junlin Liu, Zhi-Jie Liu, Tian Hua
The human sweet taste receptor is a heterodimer of TAS1R2 and TAS1R3 that recognizes structurally diverse sweet compounds, but its complete structure had long remained elusive. This study determines structures of the full-length receptor in ligand-free and sucralose-bound states, revealing an asymmetric dimer and locating the main sucralose-binding site in the Venus flytrap domain of TAS1R2. Mutational and molecular dynamics analyses further support this mode of sweetener recognition, while comparisons between states reveal conformational changes that propagate to other regions after ligand binding. The work provides a molecular basis for understanding signal transmission in class C chemosensory GPCRs and may help compare how different sweeteners are recognized. Its implications are primarily mechanistic: binding structures alone cannot establish a sweetener’s overall nutritional or metabolic effects.
Highlight · Full-length sweet taste receptor structures reveal an asymmetric dimer and show how TAS1R2 recognizes sucralose.
DOI: 10.1038/s41586-025-09302-6
07 · Structural basis of β-glucopyranoside salicin recognition by a human bitter taste GPCR
Cell Reports · Publication/issue date: 2025-05-27
Authors: Xin Wang, Cui Zhou, Weizhen Ao, Lijie Wu, Yiran Wu, Weixiu Xu, Shenhui Liu, Qiwen Tan, Ling Wang, Fei Zhao, Junlin Liu, Yuan Pei, Suwen Zhao, Tian Hua
TAS2R16 is notably selective for bitter compounds containing a β-glucoside moiety, but the structural basis for recognizing this shared chemical feature and triggering signaling had been unclear. The study determines cryo-EM structures of salicin-activated human TAS2R16 in complexes with gustducin, Gi1, and Gi2, describing the ligand’s binding mode and the interactions between its glucoside moiety and the receptor. Molecular docking and mutational experiments further support recognition of specific chemical groups and link this process to activation and downstream G-protein coupling. Comparisons across signaling partners also reveal differences in receptor coupling. The work provides a structural basis for receptor selectivity toward bitter glycosides, although potential applications still require appropriate functional and physiological testing.
Highlight · Salicin-bound TAS2R16 structures explain how this bitter taste receptor selectively recognizes the glucoside moiety.
DOI: 10.1016/j.celrep.2025.115604
08 · Bitter taste TAS2R14 activation by intracellular tastants and cholesterol
Nature · Publication/issue date: 2024-07
Authors: Xiaolong Hu, Weizhen Ao, Mingxin Gao, Lijie Wu, Yuan Pei, Shenhui Liu, Yiran Wu, Fei Zhao, Qianqian Sun, Junlin Liu, Longquan Jiang, Xin Wang, Yan Li, Qiwen Tan, Jie Cheng, Fan Yang, Chi Yang, Jinpeng Sun, Tian Hua, Zhi-Jie Liu
TAS2R14 recognizes a wide range of bitter compounds and is also found in tissues outside the mouth. Cryo-EM structures of the receptor bound to different agonists and G proteins reveal a ligand arrangement unlike that of typical class A GPCRs: cholesterol occupies the region usually corresponding to the orthosteric pocket, while the agonists studied bind to intracellular pockets. Drawing on mutational experiments and molecular dynamics, the authors discuss how multiple binding sites jointly support broad ligand recognition and receptor activation, and compare coupling to gustducin and Gi1. The findings suggest that bitter taste receptors may use cooperation among multiple sites for chemosensation; a canonical binding mode established for one GPCR should not be assumed to apply across an entire receptor family. The structural observations also provide a basis for future tests of how membrane lipids affect sensory function.
Highlight · Cholesterol and intracellular agonist-binding pockets together underpin the multisite recognition mechanism of TAS2R14.
DOI: 10.1038/s41586-024-07569-9
09 · Structural basis for strychnine activation of human bitter taste receptor TAS2R46
Science · Publication/issue date: 2022-09-16
Authors: Weixiu Xu, Lijie Wu, Shenhui Liu, Xiao Liu, Xiaoling Cao, Cui Zhou, Jinyi Zhang, You Fu, Yu Guo, Yiran Wu, Qiwen Tan, Ling Wang, Junlin Liu, Longquan Jiang, Zhongbo Fan, Yuan Pei, Jingyi Yu, Jianjun Cheng, Suwen Zhao, Xiaojiang Hao, Zhi-Jie Liu, Tian Hua
Bitter taste perception depends on the TAS2R family, but a lack of experimental structures had limited understanding of its recognition and activation mechanisms. This study determines cryo-EM structures of human TAS2R46 complexed with engineered gustducin in ligand-free and ligand-bound states. The structures reveal an architecture distinct from other GPCRs, activation-associated structural features, and a state pre-coupled to its signaling partner. Dynamics also vary across regions: the extracellular portion is more variable, while the intracellular portion is relatively stable, suggesting that chemical recognition and signal transmission may be coordinated through conformational changes at different levels. The work provides an early structural reference for the bitter taste receptor family and supports subsequent comparisons of receptor selectivity. Findings obtained with engineered signaling partners still need to be corroborated by studies of native cellular function.
Highlight · Structures of TAS2R46 in two states provide a reference for understanding the distinctive recognition and activation mechanisms of bitter taste receptors.
DOI: 10.1126/science.abo1633
Multi-crystal Diffraction Data Processing (2 papers)
10 · Clustering procedures for the optimal selection of data sets from multiple crystals in macromolecular crystallography
Acta Crystallographica Section D: Biological Crystallography · Publication/issue date: 2013-08-01
Authors: J. Foadi, P. Aller, Y. Alguel, A. Cameron, D. Axford, R. L. Owen, W. Armour, D. G. Waterman, S. Iwata, G. Evans
Synchrotron microbeams allow protein crystals smaller than ten micrometers to be used for structural studies. However, microcrystals and sensitive crystals are especially susceptible to radiation damage, often requiring data collection from multiple crystals followed by merging. Selecting and combining data from different crystals can therefore be time-consuming. This methods paper introduces BLEND, which uses clustering to help select datasets suitable for joint processing and automates parts of the multi-crystal analysis workflow. The authors demonstrate its use in determining the structure of a novel membrane protein, showing how computational selection can reduce the burden of manually comparing dataset combinations. The work focuses on organizing and selecting multi-crystal data, rather than improving quality simply by adding more crystals; successful merging still depends on sufficient consistency among datasets.
Highlight · BLEND uses clustering to select multi-crystal datasets, reducing manual work during selection and merging.
DOI: 10.1107/S0907444913012274
11 · KAMO: towards automated data processing for microcrystals
Acta Crystallographica Section D: Structural Biology · Publication/issue date: 2018-05-01
Authors: K. Yamashita, K. Hirata, M. Yamamoto
A single microcrystal often cannot provide a complete high-resolution diffraction dataset. Merging small-wedge data from multiple crystals offers a solution, but processing large numbers of incomplete datasets is challenging. KAMO integrates established programs such as XDS and CCP4 into an automated workflow. It processes individual datasets, groups results with similar unit-cell parameters, selects the space group and resolves indexing ambiguity, then performs clustering, outlier rejection, and merging before generating reports. Validation using simulated data and experimental data from hundreds of crystals demonstrates largely automated production of merged structure-factor amplitudes. The main contribution is to connect previously separate decisions and computational steps, easing the data-processing burden for challenging samples that yield only microcrystals while retaining quality control for differences between crystals and anomalous observations.
Highlight · KAMO brings the processing, clustering, and merging of data from multiple microcrystals into an automated analysis workflow.
DOI: 10.1107/S2059798318004576
Protein Design (2 papers)
12 · Design of intrinsically disordered region binding proteins
Science · Publication/issue date: 2025-07-17
Authors: Kejia Wu, Hanlun Jiang, Derrick R. Hicks, Caixuan Liu, Edin Muratspahić, Theresa A. Ramelot, Yuexuan Liu, Kerrie McNally, Sebastian Kenny, Andrei Mihut, Amit Gaur, Brian Coventry, Wei Chen, Asim K. Bera, Alex Kang, Stacey Gerben, Mila Ya-Lan Lamb, Analisa Murray, Xinting Li, Madison A. Kennedy, Wei Yang, Zihao Song, Gudrun Schober, Stuart M. Brierley, John O’Neill, Michael H. Gelb, Gaetano T. Montelione, Emmanuel Derivery, David Baker
Intrinsically disordered regions lack stable three-dimensional structures and vary widely in sequence and conformational preferences, making them longstanding challenges for the design of specific protein binders. This study presents a binder-design method in which target disordered peptides bind in a range of extended conformations, with their side chains fitting into complementary pockets. Testing 39 diverse unstructured targets, the authors obtain designs for 34 with affinities ranging from roughly 100 picomolar to 100 nanomolar, screening an average of about 22 candidates per target. Cross-binding experiments support selectivity for the intended targets, and the study also demonstrates intracellular activity and detection applications. The results show that the absence of a fixed native fold does not preclude the design of specific recognition, although performance must still be validated for each target and application setting.
Highlight · A design strategy targeting extended peptide conformations expands high-affinity recognition to diverse intrinsically disordered regions.
DOI: 10.1126/science.adr8063
13 · Programmable design of functional proteins from natural language
Nature · Publication/issue date: 2026-10-21
Authors: Fengyuan Dai, Shiyang You, Yudian Zhu, Yuan Gao, Lihao Fu, Xibin Zhou, Jin Su, Chentong Wang, Yuliang Fan, Qingsong Tan, Xiaoxiao Ma, Xianjun Deng, Letong Yu, Hui Qian, Yan He, Yitao Ke, Zhe Zhang, Chenchen Han, Xing Chang, Liangzhen Zheng, Sheng Wang, Yuan Ji, Kexin Hao, Longxing Cao, Yajie Wang, Anping Zeng, Shunzhi Wang, Chunjun Zhan, Jing Zhao, Tong Si, Jianming Liu, Hongyuan Lu, Fajie Yuan
Protein design typically requires natural templates or the manual translation of functional goals into geometric and sequence constraints. Pinal seeks to generate candidate proteins directly from natural-language descriptions, linking intended function to sequence and structure. The authors present a model trained on large-scale protein–text pairs. Computational evaluations assess predicted foldability, alignment with the descriptions, and sequence diversity, while experiments demonstrate the intended functions in four tasks involving fluorescent proteins, two enzyme classes, and a metabolic H protein. In the latter case, a designed H protein yields a product titer 1.7 times that obtained with the corresponding natural protein in a multienzyme carbon-fixation system. The study supports natural language as a starting point for candidate generation, but computational alignment still requires experimental validation, and a few successful examples do not establish that arbitrary functional descriptions can be reliably realized.
Highlight · Pinal uses natural language as the starting point for generating candidate proteins, with functional validation across four task categories.
DOI: 10.1038/s41586-026-11205-z
Other Papers (1 paper)
14 · Structural basis of bile acid lateral diffusion facilitated by the OSTα/β transporter
Nature Communications · Publication/issue date: 2026-10-23
Authors: Kang Xu, Bo-Rui Li, GaoXing Yu- Huang, Fang-Fang Wang, Jie Wang, Bo Wang, Chen-Yang Chu, Guo-Ting Chen, Cong-Zhao Zhou, Yuxing Chen, Wen-Tao Hou
OSTα/β returns reabsorbed bile acids to the portal circulation, but how it transports them across the membrane had been unclear. Cryo-EM structures of the human transporter, including a taurocholate-bound state, reveal a seven-transmembrane architecture distinct from typical SLC transporters. The substrate occupies a lateral groove on the protein surface rather than a central cavity. Polar residues lining the groove participate in substrate recognition, with mutational and transport experiments supporting the importance of these contacts. Simulations also suggest that substrate movement along the surface groove is energetically favorable. On this basis, the authors propose a facilitated lateral diffusion mechanism involving relatively small conformational changes. Together, the structural, functional, and simulation results offer an explanation distinct from the typical alternating-access model, adding a new molecular perspective on bile acid homeostasis.
Highlight · A surface groove in OSTα/β supports bile acid transport across the membrane through facilitated lateral diffusion.
DOI: 10.1038/s41467-026-78517-6