LemonCatcher is created, a protein superglue that spontaneously forms an amide bond to the LemonTag peptide under these harsh acidic and cold quench conditions, even at -20 °C, and SelQueX makes possible selective characterization of protein structural dynamics within the complex cellular milieu.
Abstract
Proteins are dynamic molecules which sensitively adapt according to their environment. Hydrogen-Deuterium eXchange Mass Spectrometry (HDX-MS) provides unique insights into protein conformational processes. However, existing methodology cannot selectively enrich proteins post-labeling because D-to-H back exchange must be minimized by rapid processing at pH 2.3-3.0 and 0 °C, where affinity purification fails. Here, we create LemonCatcher, a protein superglue that spontaneously forms an amide bond to the LemonTag peptide under these harsh acidic and cold quench conditions, even at -20 °C. Engineering of a bead-coupled LemonCatcher purification system introduces fast and selective quench-capture HDX-MS (SelQueX) on LemonTagged fusion proteins. We demonstrate targeted measurement of protein dynamics in living bacterial cells, revealing ligand-induced conformational changes in maltose-binding protein. Moreover, probing a stalled membrane protein nascent-chain supports a role for the ribosome in maintaining partially unfolded folding intermediates. Thus, SelQueX makes possible selective characterization of protein structural dynamics within the complex cellular milieu.
Determining the higher order structure (HOS) of proteins and protein complexes is central to understanding their functions, dynamics, and interactions. Traditional structural biology approaches, such as X-ray crystallography and nuclear magnetic resonance, provide high-resolution snapshots but often require large amounts of homogeneous samples and may miss dynamic or heterogeneous states. Mass spectrometry (MS) has become an indispensable tool for sensitive and rapid analysis of intact proteins and assemblies under native or near-native conditions. This review discusses the major MS-based strategies for probing HOS. Native mass spectrometry (nMS) preserves non-covalent interactions and exhibits characteristic charge-state distributions that report on folding, while native top-down fragmentation and ion mobility spectrometry provide sequence-specific and conformation-specific information. Hydrogen–deuterium exchange MS measures backbone amide exchange rates to map regions of solvent accessibility, ligand binding, and allosteric regulation in solution. Covalent labeling MS irreversibly modifies solvent-accessible side chains, allowing epitope mapping and detection of subtle conformational changes, while fast photochemical oxidation of proteins offers microsecond snapshots of transient structures. Chemical cross-linking MS applies bifunctional reagents to capture proximity between residues or subunits, providing distance restraints for integrative modeling and proteome-wide interaction mapping. We outline recent advances in instrumentation, software, labeling chemistry and in-cell techniques across these modalities, and we illustrate their applications to characterizing membrane proteins, large assemblies, therapeutic antibodies, intrinsically disordered proteins, and protein–ligand complexes. Together, these tools offer complementary insights into HOS that are reshaping structural biology, biopharmaceutical development and mechanistic studies.
Dulakshi Herath, Kaitlyn N. Walls, Ashlyn N. Dollar et al.· In Analysis· 0 citations
Membrane proteins constitute over 30% of the human proteome and represent more than 60% of drug targets, making them of critical interest in pharmaceutical discovery. Membrane proteins function within complex lipid environments that actively regulate their structure, dynamics, and activity. Hydrogen–deuterium exchange mass spectrometry (HDX-MS) has emerged as a powerful approach for probing membrane protein dynamics in solution and native-like assemblies, including nanodiscs, SMALPs, and liposomes, paving the way to study integral membrane proteins within the context of living cells. In this perspective, we discuss HDX-MS analysis of membrane protein–lipid interactions, particularly key challenges such as lipid-induced ion suppression and chromatographic interference, and highlight recent advances in lipid removal strategies, subzero chromatography, mass spectrometry/ion mobility, and the integration of molecular dynamics simulations and artificial intelligence. We further emphasise the synergy of HDX-MS with lipidomics and native MS to better understand the interplay between lipid composition, binding stoichiometry, and structural dynamics. Together, these approaches establish an emerging multidimensional framework for understanding membrane protein–lipid interplay under physiologically relevant conditions.
Artemis Lioupi, Agni F. M. Gavriilidou, Argyris Politis· Chemical Science· 0 citations
Properly folded proteins are essential for studying protein function and developing therapeutic applications. However, when recombinant proteins are overexpressed in bacterial systems such as Escherichia coli, they often accumulate as inclusion bodies, insoluble protein aggregates in non-native conformations. Here, we report a refolding protocol for mouse Frizzled8 cysteine-rich domain (mouse FZD8 CRD) recovered from inclusion bodies, yielding a soluble, folded CRD preparation suitable for NMR analysis. Because mouse FZD8 CRD contains multiple conserved cysteine residues that must form an appropriate intramolecular disulfide-bonding pattern, the refolding strategy combines denaturant-mediated solubilization, dilution into a redox-buffered refolding solution, concentration, dialysis, lyophilization, and HPLC purification. Lyophilization followed by HPLC purification enables the resolution of the desired CRD species from aggregated and misfolded products. In this workflow, HPLC served as a key purification step, providing the resolution needed to enrich a folded, multi-disulfide-bond-containing CRD species from closely related misfolded species. NMR spectroscopy supported the presence of a folded, conformationally homogeneous CRD preparation. This workflow was also applied to mouse secreted Frizzled-related protein 3 cysteine-rich domain (mouse sFRP3 CRD), suggesting that it may be useful for other Frizzled-type CRDs with conserved cysteine-rich architectures, although additional validation will be required before broader application to more distantly related cysteine-rich proteins.
Ho-Jin Lee, Jie J. Zheng· International Journal of Mol...· 0 citations
Native mass spectrometry (nMS) is a powerful label-free method for detecting biomolecular complexes, resolving stoichiometry, and quantifying affinity (Kd). However, signal overlap in heterogeneous systems often limits its accuracy. Charge detection (CD)-nMS, which independently measures the mass-to-charge ratio and charge of individual ions, overcomes this challenge, enabling quantitative analysis of polydisperse and high molecular weight complexes with unresolved charge states. Here, we systematically validate CD-nMS for precise Kd determination using soluble protein–ligand complexes of known Kd and extend its application to quantify glycan ligand binding by a highly glycosylated immune lectin. We then demonstrate the implementation of slow mixing mode (SLOMO), a time-resolved mixing method that corrects for nonuniform response factors of interacting species, using CD-nMS to enable robust quantification of protein–protein interactions. Finally, we apply SLOMO-CD-nMS to directly detect and quantify bacterial toxin binding to glycolipids embedded in membrane-like assemblies, a capability not accessible with conventional nMS. These measurements uncovered previously unrecognized assembly pathways and demonstrate, for the first time, that SLOMO-CD-nMS can resolve and quantify multivalent lectin engagement with glycolipids in a native-like membrane context. Collectively, these results establish CD-nMS, alone or in combination with SLOMO, as a broadly applicable assay for quantitative characterization of complex biomolecular interactions across soluble, glycosylated, and membrane-associated systems.
Ziyu Zhang, D. Bui, Ling Han et al.· Analytical Chemistry· 0 citations
Proteins are intrinsically dynamic molecules that continuously explore conformational ensembles to execute biological functions. Conventional structural biology methods rely on in vitro reconstitution of purified components and therefore capture predominantly static snapshots, often overlooking the regulatory roles of the cellular microenvironment, such as molecular crowding, weak interaction networks, and post-translational modifications. This limitation has driven an urgent need to transition from in vitro reconstruction to in vivo characterization within living cells. Nuclear magnetic resonance (NMR) spectroscopy provides atomic-resolution insights into structure and motions spanning multiple timescales, yet its application is constrained by molecular weight limits, isotopic labeling requirements, and inherently low throughput. Cross-linking mass spectrometry (XL-MS) complements NMR by delivering sparse but long-range spatial restraints without an upper molecular weight limit. The integration of NMR and XL-MS establishes a powerful synergistic framework that bridges atomic-resolution local structures and large-scale interaction topologies, thereby enabling comprehensive characterization of protein dynamic conformations and interaction networks in native environments. Here, we review how this integrative strategy advances the understanding of intrinsically disordered proteins, multi-domain proteins, and dynamic protein-protein interaction networks in native cellular environments. We further discuss emerging technological frontiers, including hyperpolarized NMR, photo-cross-linking, organelle-resolved analysis, and artificial intelligence-guided integrative modeling, which together promise to transform our ability to resolve the true functional states of proteins inside cells.
Zhou Gong, Qun Zhao, Min Sun et al.· Magnetic Resonance Letters· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduAug 27, 2026
A new machine-learning framework aims to improve the success rate of computational protein design while moving away from results that reproduce sequences found in nature.