Protein structure dynamics by crosslinking mass spectrometry

Crosslinking mass spectrometry captures protein structures in solution. The crosslinks reveal spatial proximities as distance restraints, but do not easily reveal which of these restraints derive from the same protein conformation. This superposition can be reduced by photo-crosslinking, and adding information from protein structure models, or quantitative crosslinking reveals conformation-specific crosslinks. As a consequence, crosslinking MS has proven useful already in the context of multiple dynamic protein systems. We foresee a breakthrough in the resolution and scale of studying protein dynamics when crosslinks are used to guide deep-learning-based protein modelling. Advances in crosslinking MS, such as photoactivatable crosslinking and in-situ crosslinking, will then reveal protein conformation dynamics in the cellular context, at a pseudo-atomic resolution, and plausibly in a time-resolved manner.     

A deep learning model for predicting optimal distance range in crosslinking mass spectrometry data

Macromolecular assemblies play an important role in all cellular processes. While there has recently been significant progress in protein structure prediction based on deep learning, large protein complexes cannot be predicted with these approaches. The integrative structure modeling approach characterizes multi-subunit complexes by computational integration of data from fast and accessible experimental techniques. Crosslinking mass spectrometry is one such technique that provides spatial information about the proximity of crosslinked residues. One of the challenges in interpreting crosslinking datasets is designing a scoring function that, given a structure, can quantify how well it fits the data. Most approaches set an upper bound on the distance between Cα atoms of crosslinked residues and calculate a fraction of satisfied crosslinks. However, the distance spanned by the crosslinker greatly depends on the neighborhood of the crosslinked residues. Here, we design a deep learning model for predicting the optimal distance range for a crosslinked residue pair based on the structures of their neighborhoods. We find that our model can predict the distance range with the area under the receiver-operator curve of 0.86 and 0.7 for intra- and inter-protein crosslinks, respectively. Our deep scoring function can be used in a range of structure modeling applications.     

Photoactivatable Reagents Based on Aryl(trifluoromethyl)diazirines: Synthesis and Application for Studying Nucleic Acid–Protein Interactions

Recent data on the synthesis of photoactivatable derivatives of nucleic acids and proteins on the basis of aryl(trifluoromethyl)diazirines—analogs of nucleosides, nucleotides, oligonucleotides, as well as amino acids and peptides—are reviewed. The synthesis of bi- and polyfunctional photoactivatable reagents, including those containing a cleavable function, designed for postsynthetic modification of biopolymers is described. Data are given on the use of the photoactivatable derivatives for studying nucleic acid–protein interactions by the method of photoaffinity labeling. Special consideration is paid to the results obtained by the authors" team in cooperation with other researchers as well as graduate students of the Chemistry of Natural Products Chair, Chemical Faculty, Moscow State University and destined to solve various scientific tasks in the domain of nucleic acid–protein recognition with the use of photoaffinity crosslinking.     

Protein flexibility is key to cisplatin crosslinking in calmodulin

Chemical crosslinking in combination with Fourier transform ion cyclotron resonance mass spectrometry (FTICR MS) has significant potential for studying protein structures and protein-protein interactions. Previously, cisplatin has been shown to be a crosslinker and crosslinks multiple methionine (Met) residues in apo-calmodulin (apo-CaM). However, the inter-residue distances obtained from nuclear magnetic resonance structures are inconsistent with the measured distance constraints by crosslinking. Met residues lie too far apart to be crosslinked by cisplatin. Here, by combining FTICR MS with a novel computational flexibility analysis, the flexible nature of the CaM structure is found to be key to cisplatin crosslinking in CaM. It is found that the side chains of Met residues can be brought together by flexible motions in both apo-CaM and calcium-bound CaM (Ca(4) -CaM). The possibility of cisplatin crosslinking Ca(4) -CaM is then confirmed by MS data. Therefore, flexibility analysis as a fast and low-cost computational method can be a useful tool for predicting crosslinking pairs in protein crosslinking analysis and facilitating MS data analysis. Finally, flexibility analysis also indicates that the crosslinking of platinum to pairs of Met residues will effectively close the nonpolar groove and thus will likely interfere with the binding of CaM to its protein targets, as was proved by comparing assays for cisplatin-modified/unmodified CaM binding to melittin. Collectively, these results suggest that cisplatin crosslinking of apo-CaM or Ca(4) -CaM can inhibit the ability of CaM to recognize its target proteins, which may have important implications for understanding the mechanism of tumor resistance to platinum anticancer drugs.     

Distance restraints from crosslinking mass spectrometry: Mining a molecular dynamics simulation database to evaluate lysine–lysine distances

Integrative structural biology attempts to model the structures of protein complexes that are challenging or intractable by classical structural methods (due to size, dynamics, or heterogeneity) by combining computational structural modeling with data from experimental methods. One such experimental method is chemical crosslinking mass spectrometry (XL‐MS), in which protein complexes are crosslinked and characterized using liquid chromatography‐mass spectrometry to pinpoint specific amino acid residues in close structural proximity. The commonly used lysine‐reactive N‐hydroxysuccinimide ester reagents disuccinimidylsuberate (DSS) and bis(sulfosuccinimidyl)suberate (BS³) have a linker arm that is 11.4 Å long when fully extended, allowing C_α (alpha carbon of protein backbone) atoms of crosslinked lysine residues to be up to ～24 Å apart. However, XL‐MS studies on proteins of known structure frequently report crosslinks that exceed this distance. Typically, a tolerance of ～3 Å is added to the theoretical maximum to account for this observation, with limited justification for the chosen value. We used the Dynameomics database, a repository of high‐quality molecular dynamics simulations of 807 proteins representative of diverse protein folds, to investigate the relationship between lysine–lysine distances in experimental starting structures and in simulation ensembles. We conclude that for DSS/BS³, a distance constraint of 26–30 Å between C_α atoms is appropriate. This analysis provides a theoretical basis for the widespread practice of adding a tolerance to the crosslinker length when comparing XL‐MS results to structures or in modeling. We also discuss the comparison of XL‐MS results to MD simulations and known structures as a means to test and validate experimental XL‐MS methods.     

The hydroxypyridinium crosslinks of skeletal collagens: Their measurement,properties and a proposed pathway of formation

A method is described for quantifying both reducible and mature crosslinking amino acids of collagen. The main crosslinking residue in cartilage, dentine and mature bone was the 3-hydroxypyridinium compound identified by Fujimoto et al. (1–3). Adult articular cartilage contained about one residue per collagen molecule, over forty times the content of the reducible crosslinks. We propose that hydroxypyridinium residues are formed by spontaneous interaction of two residues of hydroxylysino-5-ketonorleucine. This reaction explains the disappearance of reducible crosslinks at maturity and provides a novel mechanism for lateral crosslinking within and between fibrils which may account for some of the unique physical properties of hard tissue collagens.     

Actin‐induced dimerization of palladin promotes actin‐bundling

A subset of actin binding proteins is able to form crosslinks between two or more actin filaments, thus producing structures of parallel or networked bundles. These actin crosslinking proteins interact with actin through either bivalent binding or dimerization. We recently identified two binding sites within the actin binding domain of palladin, an actin crosslinking protein that plays an important role in normal cell adhesion and motility during wound healing and embryonic development. In this study, we show that actin induces dimerization of palladin. Furthermore, the extent of dimerization reflects earlier comparisons of actin binding and bundling between different domains of palladin. On the basis of these results we hypothesized that actin binding may promote a conformational change that results in dimerization of palladin, which in turn may drive the crosslinking of actin filaments. The proximal distance between two actin binding sites on crosslinking proteins determines the ultrastructural properties of the filament network, therefore we also explored interdomain interactions using a combination of chemical crosslinking experiments and actin cosedimentation assays. Limited proteolysis data reveals that palladin is less susceptible to enzyme digestion after actin binding. Our results suggest that domain movements in palladin are necessary for interactions with actin and are induced by interactions with actin filaments. Accordingly, we put forth a model linking the structural changes to functional dynamics.     

Crosslinking of Chitosan with Dialdehyde Derivatives of Nucleosides and Nucleotides. Mechanism and Comparison with Glutaraldehyde

In medical and pharmaceutical applications, chitosan is used as a component of hydrogels–macromolecular networks swollen in water. Chemical hydrogels are formed by covalent links between the crosslinking reagents and amino functionalities of chitosan. To date, the most commonly used chitosan crosslinkers are dialdehydes, such as glutaraldehyde (GA). We have developed novel GA like crosslinkers with additional functional groups–dialdehyde derivatives of uridine (oUrd) and nucleotides (oUMP and oAMP)–leading to chitosan-based biomaterials with new properties. The process of chitosan crosslinking was investigated in details and compared to crosslinking with GA. The rates of crosslinking with oUMP, oAMP, and GA were essentially the same, though much higher than in the case of oUrd. The remarkable difference in the crosslinking properties of nucleoside and nucleotide dialdehydes can be clearly attributed to the presence of the phosphate group in nucleotides that participates in the gelation process through ionic interactions with the amino groups of chitosan. Using NMR spectroscopy, we have not observed the formation of aldimine bonds. It can be concluded that the real number of crosslinks needed to cause gelation of chitosan chains may be less than 1%.     

Analysis of EDTA-chelatable proteins from DNA-protein crosslinks induced by a carcinogenic chromium(VI) in cultured intact human cells

DNA-protein crosslinks (DPCs) were induced in intact human leukemic T-lymphocyte MOLT4 cells or isolated nuclei by treatment with potassium chromate, chromium(III) chloride hexahydrate or x-rays. The proteins complexed to DNA were analyzed by two-dimensional SDS-polyacrylamide gel electrophoresis (PAGE). A group of identical non-histone proteins was crosslinked to DNA by any of the three treatments, except that a 51 kDa basic protein was additionally complexed to DNA when either potassium chromate or chromium(III) chloride hexahydrate was the crosslinking agent. Treatment of chromate-induced DNA-protein crosslinks with EDTA or thiourea followed by ultracentifugation dissociated the major proteins from the complex indicating that these proteins were crosslinked to DNA by direct participation of a EDTA-chelatable form of chromium such as Cr(III) through sulfur containing amino acid residues. The 51 kDa protein was not seen in the post-EDTA pellet but was present in the post-thiourea pellet, indicating that it was also crosslinked to DNA by Cr(III) through non-sulfur-containing amino acids. Digestion of x-rays-induced DPCs by DNase I also revealed this protein on two-dimensional gels indicating that the same protein was also crosslinked by oxidative mechanisms. The involvement of oxidative mechanisms in the crosslinking process was indicated as the majority of the proteins in chromate-induced DPCs were resistant to EDTA and thiourea treatment, and were found to crosslink to DNA when x-rays were used as the crosslinking agent. These results suggest that the chromate-induced DPCs are formed by the generation of reactive oxygen species during the intracellular chromate reduction as well as by the biologically generated Cr(III). About 19% of DNA-protein crosslinks actually involve Cr(III) crosslinking DNA to proteins, about 14% involve Cr(III) crosslinking DNA to proteins through non-sulfhydryl containing moieties and about 5% involve Cr(III) crosslinking DNA to sulfhydryl groups on proteins. The remaining 81% of DNA-protein crosslinks appear to be oxidatively crosslinked out of which about 45% appear to be through sulfhydryl groups and another 36% appear to be through non-sulfhydryl groups.     

Increasing the chemical space of proteins in living cells via genetic code expansion

In recent years it has become possible to genetically encode an expanded set of designer amino acids with tailored chemical and physical properties (dubbed unnatural amino acids, UAAs) into proteins in living cells by expanding the genetic code. Together with developments in chemistries that are amenable to and selective within physiological settings, these strategies have started to have a big impact on biological studies, as they enable exciting in cellulo applications. Here we highlight recent advances to covalently stabilize transient protein–protein interactions and capture enzyme substrate-complexes in living cells using proximity-triggered and residue-selective photo-induced crosslinking approaches. Furthermore, we describe recent efforts in controlling enzyme activity with photocaged UAAs and in extending their application to a variety of enzymatic scaffolds. In addition, we discuss the site-specific incorporation of UAAs mimicking post-translational modifications (PTMs) and approaches to generate natively-linked ubiquitin–protein conjugates to probe the role of PTMs in modulating complex cellular networks.     

The structure of the thioredoxin-triosephosphate isomerase complex provides insights into the reversible glutathione-mediated regulation of triosephosphate isomerase

Protein-protein interactions are crucial for many biological functions. The redox interactome encompasses numerous weak transient interactions in which thioredoxin plays a central role. Proteomic studies have shown that thioredoxin binds to numerous proteins belonging to various cellular processes, including energy metabolism. Thioredoxin has cross talk with other redox mechanisms involving glutathionylation and has functional overlap with glutaredoxin in deglutathionylation reactions. In this study, we have explored the structural and biochemical interactions of thioredoxin with the glycolytic enzyme, triosephosphate isomerase. Nuclear magnetic resonance chemical shift mapping methods and molecular dynamics-based docking have been applied in deriving a structural model of the thioredoxin-triosephosphate isomerase complex. The spatial proximity of active site cysteine residues of thioredoxin to reactive thiol groups on triosephosphate isomerase provides a direct link to the observed deglutathionylation of cysteine 217 in triosephosphate isomerase, thereby reversing the inhibitory effect of S-glutathionylation of triosephosphate isomerase.     

PyXlinkViewer: A flexible tool for visualization of protein chemical crosslinking data within the PyMOL molecular graphics system

Chemical crosslinking‐mass spectrometry (XL‐MS) is a valuable technique for gaining insights into protein structure and the organization of macromolecular complexes. XL‐MS data yield inter‐residue restraints that can be compared with high‐resolution structural data. Distances greater than the crosslinker spacer‐arm can reveal lowly populated “excited” states of proteins/protein assemblies, or crosslinks can be used as restraints to generate structural models in the absence of structural data. Despite increasing uptake of XL‐MS, there are few tools to enable rapid and facile mapping of XL‐MS data onto high‐resolution structures or structural models. PyXlinkViewer is a user‐friendly plugin for PyMOL v2 that maps intra‐protein, inter‐protein, and dead‐end crosslinks onto protein structures/models and automates the calculation of inter‐residue distances for the detected crosslinks. This enables rapid visualization of XL‐MS data, assessment of whether a set of detected crosslinks is congruent with structural data, and easy production of high‐quality images for publication.     

Site-specific crosslinks of yeast U6 snRNA to the pre-mRNA near the 5' splice site.

We have introduced a single photochemical crosslinking reagent into specific sites in the central domain of U6 to identify the sites that are in close proximity to the pre-mRNA substrate. Four distinct U6 snRNAs were synthesized with a single 4-thiouridine (4-thioU) at positions 46, 51, 54, and 57, respectively. Synthetic U6 RNA containing the 4-thioU modifications can functionally reconstitute splicing activity in cell-free yeast splicing extracts depleted of endogenous U6 snRNA. Upon photoactivation with UV (>300 nm), 4-thioU at position 46 forms crosslinks to pre-mRNA near the 5' splice site at nt +4, +5, +6, and +7 in the intron, whereas 4-thioU at position 51 crosslinks to the pre-mRNA at positions -2, -1, +1, +2, +3, and at the invariant G in the lariat intermediate. All crosslinks are dependent on the presence of ATP and the splicing substrate. The two crosslinks to the pre-mRNA from position 46 and 51 of U6 can also occur in prp2 heat-inactivated yeast splicing extracts blocked immediately prior to the first chemical step. Significantly, the crosslink from position 51 can undergo subsequent splicing when the mutant extract is complemented with functional Prp2 protein in a chase experiment, indicating that the crosslink reflects a functional interaction that is maintained during the first step. The crosslink to lariat intermediate appears when the mutant spliceosomes are complemented with functional Prp2 protein added exogenously. This experiment is a paradigm for future studies in which different mutant extracts are used to establish the stage in assembly at which particular RNA-RNA interactions defined by unique crosslinks occur.     

Site-specific modification and RNA crosslinking of the RNA-binding domain of PKR

RNA-dependent protein kinase (PKR) is an interferon-induced, RNA-activated enzyme that phosphorylates and inhibits the function of the translation initiation factor eIF-2. PKR is activated in vitro by binding RNA molecules with extensive duplex structure. To further define the nature of the RNA regulation of PKR, we have prepared and characterized site-specifically modified proteins consisting of the PKR 20 kDa RNA-binding domain (RBD). Here we show that the two cysteines found naturally in this domain can be altered by site-directed mutagenesis without loss of RNA binding affinity or the RNA-regulated kinase activity. Introduction of cysteine residues at other sites in the PKR RBD allows for site-specific modification with thiol-selective reagents. PKR RBD mutants reacted selectively with a maleimide to introduce a photoactivatable crosslinking aryl azide at three different positions in the protein. RNA crosslinking efficiency was found to be dependent on the amino acid modified, suggesting differences in access to the RNA from these positions in the protein. One of the amino acid modifications that led to crosslinking of the RNA is located at a residue known to be an autophosphorylation site, suggesting that autophosphorylation at this site could influence the RNA binding properties of PKR. The PKR RBD conjugates described here and other similar reagents prepared via these methods are applicable to future studies of PKR–RNA complexes using techniques such as photocrosslinking, fluorescence resonance energy transfer and affinity cleaving.     

Biochemical characterization and helix stabilizing properties of HSNP-C' from the thermoacidophilic archaeon Sulfolobus acidocaldarius

Helix stabilizing nucleoid protein HSNP-C' from the thermophilic archaeon Sulfolobus acidocaldarius has been characterized with respect to its interactions with nucleic acids by gel retardation assay, affinities to immobilized matrices, electron microscopy, and fluorescence titration. The amino acids implicated in the DNA binding site of the protein have been shown by selectively modifying specific amino acyl functional groups and looking at their effects on the DNA binding properties of the protein. Lysine, arginine, tryptophan, and tyrosine residues of the protein HSNP-C' were modified with pyridoxal-5-phosphate; 2,3-butanedione; BNPS-skatole; and tetranitromethane, respectively. The modification of residues was assessed according to standard procedures. The effect of the chemical modification on the function of the protein HSNP-C' with respect to DNA protein interactions was studied and the results indicate the definite involvement of tyrosines and also the significant involvement of the flanking tryptophan residues in the DNA binding domain on the protein.     

Human protein aging: modification and crosslinking through dehydroalanine and dehydrobutyrine intermediates

Nonenzymatic post‐translational modification (PTM) of proteins is a fundamental molecular process of aging. The combination of various modifications and their accumulation with age not only affects function, but leads to crosslinking and protein aggregation. In this study, aged human lens proteins were examined using HPLC–tandem mass spectrometry and a blind PTM search strategy. Multiple thioether modifications of Ser and Thr residues by glutathione (GSH) and its metabolites were unambiguously identified. Thirty‐four of 36 sites identified on 15 proteins were found on known phosphorylation sites, supporting a mechanism involving dehydroalanine (DHA) and dehydrobutyrine (DHB) formation through β‐elimination of phosphoric acid from phosphoserine and phosphothreonine with subsequent nucleophilic attack by GSH. In vitro incubations of phosphopeptides demonstrated that this process can occur spontaneously under physiological conditions. Evidence that this mechanism can also lead to protein–protein crosslinks within cells is provided where five crosslinked peptides were detected in a human cataractous lens. Nondisulfide crosslinks were identified for the first time in lens tissue between βB2‐ & βB2‐, βA4‐ & βA3‐, γS‐ & βB1‐, and βA4‐ & βA4‐crystallins and provide detailed structural information on in vivo crystallin complexes. These data suggest that phosphoserine and phosphothreonine residues represent susceptible sites for spontaneous breakdown in long‐lived proteins and that DHA‐ and DHB‐mediated protein crosslinking may be the source of the long‐sought after nondisulfide protein aggregates believed to scatter light in cataractous lenses. Furthermore, this mechanism may be a common aging process that occurs in long‐lived proteins of other tissues leading to protein aggregation diseases.     

Uncovering post-translational modification-associated protein–protein interactions

In living systems, the chemical space and functional repertoire of proteins are dramatically expanded through the post-translational modification (PTM) of various amino acid residues. These modifications frequently trigger unique protein–protein interactions (PPIs) – for example with reader proteins that directly bind the modified amino acid residue – which leads to downstream functional outcomes. The modification of a protein can also perturb its PPI network indirectly, for example, through altering its conformation or subcellular localization. Uncovering the network of unique PTM-triggered PPIs is essential to fully understand the roles of an ever-expanding list of PTMs in our biology. In this review, we discuss established strategies and current challenges associated with this endeavor.     

Delineation of bacterial luciferase aldehyde site by bifunctional labeling reagents

Previously we have established that a highly reactive cysteinyl group on the alpha subunit is at the aldehyde site of the (alpha beta) dimeric Vibrio harveyi luciferase. Three isomeric bifunctional reagents have been synthesized and used to further delineate the luciferase aldehyde site. These probes differ in their relative positions of and distances between the two functional groups active in chemical and photochemical labelings, respectively. Each of the probes can effectively and reversibly inactivate luciferase by forming a disulfide linkage primarily to the reactive cysteinyl residue. Upon subsequent photolysis, a diazoacetate arm in each probe was activated for photochemical labeling of amino acid residues within reach. After reductive regeneration of the reactive cysteinyl residue, 0.35-0.40 probe per dimeric luciferase was found to have been photochemically incorporated, correlating well with the degree of irreversible enzyme inactivation. Low but significant amounts of the three isomeric probes initially attached to the alpha reactive cysteine through a disulfide have been found to photochemically tag certain residues on beta. The latter residues are estimated to be no more than 8-11 A away from the alpha reactive cysteine. Thus the reactive cysteinyl residue, and hence the aldehyde site, must be at or near the alpha beta subunit interface. Furthermore, the structural integrity of the microenvironment surrounding this reactive cysteinyl residue is crucial to luciferase activity. An HPLC method for the isolation of luciferase alpha and beta subunits has also been developed.     

Assessment of chemical-crosslink-assisted protein structure modeling in CASP13

With the advance of experimental procedures obtaining chemical crosslinking information is becoming a fast and routine practice. Information on crosslinks can greatly enhance the accuracy of protein structure modeling. Here, we review the current state of the art in modeling protein structures with the assistance of experimentally determined chemical crosslinks within the framework of the 13th meeting of Critical Assessment of Structure Prediction approaches. This largest-to-date blind assessment reveals benefits of using data assistance in difficult to model protein structure prediction cases. However, in a broader context, it also suggests that with the unprecedented advance in accuracy to predict contacts in recent years, experimental crosslinks will be useful only if their specificity and accuracy further improved and they are better integrated into computational workflows.     

Tethering polypeptides through bifunctional PEG cross-linking agents to probe protein function: application to ATP synthase

Chemical crosslinking mediated by short bifunctional reagents has been widely used for determining physical relationships among polypeptides in multisubunit proteins, but less often for functional studies. Here we introduce the approach of tethering polypeptides by using bifunctional reagents containing a lengthy, flexible PEG linker as a form of crosslinking especially suited to functional analyses. The rotary molecular motor ATP synthase was used as a model subject. Single cysteine residues were introduced into selected positions of ATP synthase epsilon subunit, a component of the rotor subcomplex of the enzyme, and the unrelated maltose binding protein (MBP), then the two purified recombinant proteins were crosslinked by means of a dimaleimido-PEG cross-linking agent. Following purification, the epsilon-PEG-MBP was incorporated into membrane-bound ATP synthase by reconstitution with epsilon-depleted F(1)-ATPase and membrane vesicles that had been stripped of endogenous F(1). ATP synthase reconstituted using epsilon-PEG-MBP had reduced ATP hydrolytic activity that was uncoupled from the pumping of H(+), indicating the physical blockage of rotation of the gammaepsilonc(10) rotor by the conjugated MBP, whereas enzyme reconstituted with epsilon-PEG was normal. These results directly demonstrate the feasibility of studying mechanistic features of molecular motors through PEG-based conjugation of unrelated proteins. Since tethering polypeptides provides a means of maintaining proximity without directly specifying or modifying interactions, application of the general method to other types of protein functional studies is envisioned.