Selective Isotopic Unlabeling of Proteins Using Metabolic Precursors: Application to NMR Assignment of Intrinsically Disordered Proteins

Structural response of G protein binding to the cyclodepsipeptide inhibitor FR900359 probed by NMR spectroscopy

The cyclodepsipeptide FR900359 (FR) and its analogs are able to selectively inhibit the class of Gq proteins by blocking GDP/GTP exchange. The inhibitor binding site of Gq has been characterized by X-ray crystallography, and various binding and functional studies have determined binding kinetics and mode of inhibition. Here we investigate isotope-labeled FR bound to the membrane-anchored G protein heterotrimer by solid-state nuclear magnetic resonance (ssNMR) and in solution by liquid-state NMR. The resulting data allowed us to identify regions of the inhibitor which show especially pronounced effects upon binding and revealed a generally rigid binding mode in the cis conformation under native-like conditions. The inclusion of the membrane environment allowed us to show a deep penetration of FR into the lipid bilayer illustrating a possible access mode of FR into the cell. Dynamic nuclear polarization (DNP)-enhanced ssNMR was used to observe the structural response of specific segments of the Gα subunit to inhibitor binding. This revealed rigidification of the switch I binding site and an allosteric response in the α5 helix as well as suppression of structural changes induced by nucleotide exchange due to inhibition by FR. Our NMR studies of the FR-G protein complex conducted directly within a native membrane environment provide important insights into the inhibitors access via the lipid membrane, binding mode, and structural allosteric effects.

Analysis of the homodimeric structure of a D-Ala-D-Ala metallopeptidase, VanX, from vancomycin-resistant bacteria

Bacteria that have acquired resistance to most antibiotics, particularly those causing nosocomial infections, create serious problems. Among these, the emergence of vancomycin-resistant enterococci was a tremendous shock, considering that vancomycin is the last resort for controlling methicillin-resistant Staphylococcus aureus. Therefore, there is an urgent need to develop an inhibitor of VanX, a protein involved in vancomycin resistance. Although the crystal structure of VanX has been resolved, its asymmetric unit contains six molecules aligned in a row. We have developed a structural model of VanX as a stable dimer in solution, primarily utilizing nuclear magnetic resonance (NMR) residual dipolar coupling. Despite the 46 kDa molecular mass of the dimer, the analyses, which are typically not as straightforward as those of small proteins around 10 kDa, were successfully conducted. We assigned the main chain using an amino acid-selective unlabeling method. Because we found that the zinc ion-coordinating active sites in the dimer structure were situated in the opposite direction to the dimer interface, we generated an active monomer by replacing an amino acid at the dimer interface. The monomer consists of only 202 amino acids and is expected to be used in future studies to screen and improve inhibitors using NMR.

Improved spectral resolution of [13C,1H]-HSQC spectra of aromatic amino acid residues in proteins produced by cell-free synthesis from inexpensive 13C-labelled precursors

Cell-free protein synthesis using eCells allows production of amino acids from inexpensive 13C-labelled precursors. We show that the metabolic pathway converting pyruvate, glucose and erythrose into aromatic amino acids is maintained in eCells. Judicious choice of 13C-labelled starting material leads to proteins, where the sidechains of aromatic amino acids display [13C,1H]-HSQC cross-peaks free of one-bond 13C-13C couplings. Selective 13C-labelling of tyrosine and phenylalanine residues is achieved simply by using different compositions of the reaction buffers.

Cell-free synthesis of proteins with selectively 13C-labelled methyl groups from inexpensive precursors

The novel eCell system maintains the activity of the entire repertoire of metabolic Escherichia coli enzymes in cell-free protein synthesis. We show that this can be harnessed to produce proteins with selectively 13 C-labelled amino acids from inexpensive 13 C-labelled precursors. The system is demonstrated with selective 13 C labelling of methyl groups in the proteins ubiquitin and peptidyl-prolyl cis-trans isomerase B. Starting from 3-13 C-pyruvate, 13 C-HSQC cross-peaks are obtained devoid of one-bond 13 C-13 C scalar couplings. Starting from 2-13 C-methyl-acetolactate, single methyl groups of valine and leucine are labelled. Labelling efficiencies are 70 % or higher, and the method allows us to produce perdeuterated proteins with protonated methyl groups in a residue-selective manner. The system uses the isotope-labelled precursors sparingly and is readily scalable.

Specific isotopic labelling and reverse labelling for protein NMR spectroscopy: using metabolic precursors in sample preparation

The study of protein structure, dynamics and function by NMR spectroscopy commonly requires samples that have been enriched ('labelled') with the stable isotopes 13C and/or 15N. The standard approach is to uniformly label a protein with one or both of these nuclei such that all C and/or N sites are in principle 'NMR-visible'. NMR spectra of uniformly labelled proteins can be highly complicated and suffer from signal overlap. Moreover, as molecular size increases the linewidths of NMR signals broaden, which decreases sensitivity and causes further spectral congestion. Both effects can limit the type and quality of information available from NMR data. Problems associated with signal overlap and signal broadening can often be alleviated though the use of alternative, non-uniform isotopic labelling patterns. Specific isotopic labelling 'turns on' signals at selected sites while the rest of the protein is NMR-invisible. Conversely, specific isotopic unlabelling (also called 'reverse' labelling) 'turns off' selected signals while the rest of the protein remains NMR-visible. Both approaches can simplify NMR spectra, improve sensitivity, facilitate resonance assignment and permit a range of different NMR strategies when combined with other labelling tools and NMR experiments. Here, we review methods for producing proteins with enrichment of stable NMR-visible isotopes, with particular focus on residue-specific labelling and reverse labelling using Escherichia coli expression systems. We also explore how these approaches can aid NMR studies of proteins.

Selective labeling and unlabeling strategies in protein solid-state NMR spectroscopy

Selective isotope labeling is central in NMR experiments and often allows to push the limits on the systems investigated. It has the advantage to supply additional resolution by diminishing the number of signals in the spectra. This is particularly interesting when dealing with the large protein systems which are currently becoming accessible to solid-state NMR studies. Isotope labeled proteins for NMR experiments are most often expressed in E. coli systems, where bacteria are grown in minimal media supplemented with ¹⁵NH₄Cl and ¹³C-glucose as sole source of nitrogen and carbon. For amino acids selective labeling or unlabeling, specific amino acids are supplemented in the minimal medium. The aim is that they will be incorporated in the protein by the bacteria. However, E. coli amino-acid anabolism and catabolism tend to interconnect different pathways, remnant of a subway system. These connections lead to inter conversion between amino acids, called scrambling. A thorough understanding of the involved pathways is thus important to obtain the desired labeling schemes, as not all combinations of amino acids are adapted. We present here a detailed overview of amino-acid metabolism in this context. Each amino-acid pathway is described in order to define accessible combinations for ¹³C or ¹⁵N specific labeling or unlabeling. Using as example the ABC transporter BmrA, a membrane protein of 600 residues, we demonstrate how these strategies can be applied. Indeed, even though there is no size limit in solid-state NMR, large (membrane) proteins are still a challenge due to heavy signal overlap. To initiate resonance assignment in these large systems, we describe how selectively labeled samples can be obtained with the addition of labeled or unlabeled amino acids in the medium. The reduced spectral overlap enabled us to identify typical spectral fingerprints and to initiate sequential assignment using the more sensitive 2D DARR experiments with long mixing time showing inter-residue correlations.

Amino Acid Selective 13C Labeling and 13C Scrambling Profile Analysis of Protein α and Side-Chain Carbons in Escherichia coli Utilized for Protein Nuclear Magnetic Resonance

Amino acid selective isotope labeling is an important nuclear magnetic resonance technique, especially for larger proteins, providing strong bases for the unambiguous resonance assignments and information concerning the structure, dynamics, and intermolecular interactions. Amino acid selective ¹⁵N labeling suffers from isotope dilution caused by metabolic interconversion of the amino acids, resulting in isotope scrambling within the target protein. Carbonyl ¹³C atoms experience less isotope scrambling than the main-chain ¹⁵N atoms do. However, little is known about the side-chain ¹³C atoms. Here, the ¹³C scrambling profiles of the Cα and side-chain carbons were investigated for ¹⁵N scrambling-prone amino acids, such as Leu, Ile, Tyr, Phe, Thr, Val, and Ala. The level of isotope scrambling was substantially lower in ¹³Cα and ¹³C side-chain labeling than in ¹⁵N labeling. We utilized this reduced scrambling-prone character of ¹³C as a simple and efficient method for amino acid selective ¹³C labeling using an Escherichia coli cold-shock expression system and high-cell density fermentation. Using this method, the ¹³C labeling efficiency was >80% for Leu and Ile, ∼60% for Tyr and Phe, ∼50% for Thr, ∼40% for Val, and 30-40% for Ala. ¹H-¹⁵N heteronuclear single-quantum coherence signals of the ¹⁵N scrambling-prone amino acid were also easily filtered using ¹⁵N-{¹³Cα} spin-echo difference experiments. Our method could be applied to the assignment of the 55 kDa protein.

Protein labeling strategies for liquid-state NMR spectroscopy using cell-free synthesis

Preparation of a protein sample for liquid-state nuclear magnetic resonance (NMR) spectroscopy analysis requires optimization of many parameters. This review describes labeling strategies for obtaining assignments of protein resonances. Particular emphasis is placed on the advantages of cell-free protein production, which enables exclusive labeling of the protein of interest, thereby simplifying downstream processing steps and increasing the availability of different labeling strategies for a target protein. Furthermore, proteins can be synthesized in milligram yields, and the open nature of the cell-free system allows the addition of stabilizers, scrambling inhibitors or hydrophobic solubilization environments directly during the protein synthesis, which is especially beneficial for membrane proteins. Selective amino acid labeling of the protein of interest, the possibility of addressing scrambling issues and avoiding the need for labile amino acid precursors have been key factors in enabling the introduction of new assignment strategies based on different labeling schemes as well as on new pulse sequences. Combinatorial selective labeling methods have been developed to reduce the number of protein samples necessary to achieve a complete backbone assignment. Furthermore, selective labeling helps to decrease spectral overlap and overcome size limitations for solution NMR analysis of larger complexes, oligomers, intrinsically disordered proteins and membrane proteins.

Modern Technologies of Solution Nuclear Magnetic Resonance Spectroscopy for Three-dimensional Structure Determination of Proteins Open Avenues for Life Scientists

Nuclear magnetic resonance (NMR) spectroscopy is a powerful technique for structural studies of chemical compounds and biomolecules such as DNA and proteins. Since the NMR signal sensitively reflects the chemical environment and the dynamics of a nuclear spin, NMR experiments provide a wealth of structural and dynamic information about the molecule of interest at atomic resolution. In general, structural biology studies using NMR spectroscopy still require a reasonable understanding of the theory behind the technique and experience on how to recorded NMR data. Owing to the remarkable progress in the past decade, we can easily access suitable and popular analytical resources for NMR structure determination of proteins with high accuracy. Here, we describe the practical aspects, workflow and key points of modern NMR techniques used for solution structure determination of proteins. This review should aid NMR specialists aiming to develop new methods that accelerate the structure determination process, and open avenues for non-specialist and life scientists interested in using NMR spectroscopy to solve protein structures.

Backbone resonance assignments of the m1A22 tRNA methyltransferase TrmK from Bacillus subtilis

RNA modification is a post-transcriptional process by which certain nucleotides are altered after their initial incorporation into an RNA chain. Transfer RNAs (tRNAs) is the most heavily modified class of RNA molecules. These modifications expand the chemical and functional diversity of tRNAs and enhance their structural stability. To date, more than 100 modifications have been identified, the majority of which are specific from one domain of life. However, few modifications are extensively present in the three domains of life. Among those, the m(1)A nucleotide, which consists in the methylation at position 1 of the adenine aromatic ring, is found in tRNAs and ribosomal RNAs. In tRNAs, the m(1)A modification occurs at position 9, 14, 22, 57 and 58. The enzyme TrmK catalyzes the m(1)A formation at position 22. Here we report the backbone (1)H, (15)N and (13)C chemical shift assignments of TrmK from Bacillus subtilis obtained by heteronuclear multidimensional NMR spectroscopy as well as its secondary structure in solution as predicted by TALOS+. These assignments of TrmK pave the way for interaction studies with its tRNA substrates.

Chemical shift assignments of the fibronectin III like domains 7-8 of type VII collagen

Type VII collagen (Col7) is important for skin stability. This is underlined by the severe skin blistering phenotype in the Col7 related diseases dystrophic epidermolysis bullosa and epidermolysis bullosa acquisita (EBA). Col7 has a large N-terminal non-collagenous domain (NC1) that is followed by the triple helical collagenous domain. The NC1 domain has subdomains with homology to adhesion molecules and mediates important interactions within the extracellular matrix. An 185 amino acid long part of the NC1-subdomain termed fibronectin III like domains 7 and 8 (FNIII7-8) was investigated. Antibodies against this region are pathogenic in a mouse model of EBA and one reported missense mutations of Col7 lies within these domains. The nearly complete NMR resonance assignment of recombinant FNIII7-8 of Col7 is reported.

An extended combinatorial 15N, 13Cα, and 13C' labeling approach to protein backbone resonance assignment

Solution NMR studies of α-helical membrane proteins are often complicated by severe spectral crowding. In addition, hydrophobic environments like detergent micelles, isotropic bicelles or nanodiscs lead to considerably reduced molecular tumbling rates which translates into line-broadening and low sensitivity. Both difficulties can be addressed by selective isotope labeling methods. In this publication, we propose a combinatorial protocol that utilizes four different classes of labeled amino acids, in which the three backbone heteronuclei (amide nitrogen, α-carbon and carbonyl carbon) are enriched in (15)N or (13)C isotopes individually as well as simultaneously. This results in eight different combinations of dipeptides giving rise to cross peaks in (1)H-(15)N correlated spectra. Their differentiation is achieved by recording a series of HN-detected 2D triple-resonance spectra. The utility of this new scheme is demonstrated with a homodimeric 142-residue membrane protein in DHPC micelles. Restricting the number of selectively labeled samples to three allowed the identification of the amino-acid type for 77 % and provided sequential information for 47 % of its residues. This enabled us to complete the backbone resonance assignment of the uniformly labeled protein merely with the help of a 3D HNCA spectrum, which can be collected with reasonable sensitivity even for relatively large, non-deuterated proteins.

NMR approaches for structural analysis of multidomain proteins and complexes in solution

NMR spectroscopy is a key method for studying the structure and dynamics of (large) multidomain proteins and complexes in solution. It plays a unique role in integrated structural biology approaches as especially information about conformational dynamics can be readily obtained at residue resolution. Here, we review NMR techniques for such studies focusing on state-of-the-art tools and practical aspects. An efficient approach for determining the quaternary structure of multidomain complexes starts from the structures of individual domains or subunits. The arrangement of the domains/subunits within the complex is then defined based on NMR measurements that provide information about the domain interfaces combined with (long-range) distance and orientational restraints. Aspects discussed include sample preparation, specific isotope labeling and spin labeling; determination of binding interfaces and domain/subunit arrangements from chemical shift perturbations (CSP), nuclear Overhauser effects (NOEs), isotope editing/filtering, cross-saturation, and differential line broadening; and based on paramagnetic relaxation enhancements (PRE) using covalent and soluble spin labels. Finally, the utility of complementary methods such as small-angle X-ray or neutron scattering (SAXS, SANS), electron paramagnetic resonance (EPR) or fluorescence spectroscopy techniques is discussed. The applications of NMR techniques are illustrated with studies of challenging (high molecular weight) protein complexes.

α-Ketoacids as precursors for phenylalanine and tyrosine labelling in cell-based protein overexpression

(13)C-α-ketoacid metabolic precursors of phenylalanine and tyrosine effectively enter the metabolism of a protein overexpressing E. coli strain to label Phe- and Tyr-residues devoid of any cross-labelling. The methodology gives access to highly selective labelling patterns as valuable tools in protein NMR spectroscopy without the need of (15)N-chiral amino acid synthesis using organic chemistry.

Resonance assignment for a particularly challenging protein based on systematic unlabeling of amino acids to complement incomplete NMR data sets

NMR-based structure determination of a protein requires the assignment of resonances as indispensable first step. Even though heteronuclear through-bond correlation methods are available for that purpose, challenging situations arise in cases where the protein in question only yields samples of limited concentration and/or stability. Here we present a strategy based upon specific individual unlabeling of all 20 standard amino acids to complement standard NMR experiments and to achieve unambiguous backbone assignments for the fast precipitating 23 kDa catalytic domain of human aprataxin of which only incomplete standard NMR data sets could be obtained. Together with the validation of this approach utilizing the protein GB1 as a model, a comprehensive insight into metabolic interconversion ("scrambling") of NH and CO groups in a standard Escherichia coli expression host is provided.

Amino acid selective labeling and unlabeling for protein resonance assignments

Structural characterization of proteins by NMR spectroscopy begins with the process of sequence specific resonance assignments in which the (1)H, (13)C and (15)N chemical shifts of all backbone and side-chain nuclei in the polypeptide are assigned. This process requires different isotope labeled forms of the protein together with specific experiments for establishing the sequential connectivity between the neighboring amino acid residues. In the case of spectral overlap, it is useful to identify spin systems corresponding to the different amino acid types selectively. With isotope labeling this can be achieved in two ways: (i) amino acid selective labeling or (ii) amino acid selective 'unlabeling'. This chapter describes both these methods with more emphasis on selective unlabeling describing the various practical aspects. The recent developments involving combinatorial selective labeling and unlabeling are also discussed.

Selective isotope labelling of leucine residues by using α-ketoacid precursor compounds

You can have one without the other: A new metabolic precursor compound can selectively introduce (13)C and (2)H patterns at leucine residues in proteins in cell-based expression systems without interfering with valine metabolism. The protocol provides selectively labelled macromolecules well suited for probing structure and dynamics in high-molecular-weight proteins by NMR spectroscopy.

Combination of ¹⁵N reverse labeling and afterglow spectroscopy for assigning membrane protein spectra by magic-angle-spinning solid-state NMR: application to the multidrug resistance protein EmrE

Magic-angle-spinning (MAS) solid-state NMR spectroscopy has emerged as a viable method to characterize membrane protein structure and dynamics. Nevertheless, the spectral resolution for uniformly labeled samples is often compromised by redundancy of the primary sequence and the presence of helical secondary structure that results in substantial resonance overlap. The ability to simplify the spectrum in order to obtain unambiguous site-specific assignments is a major bottleneck for structure determination. To address this problem, we used a combination of (15)N reverse labeling, afterglow spectroscopic techniques, and frequency-selective dephasing experiments that dramatically improved the ability to resolve peaks in crowded spectra. This was demonstrated using the polytopic membrane protein EmrE, an efflux pump involved in multidrug resistance. Residues preceding the (15)N reverse labeled amino acid were imaged using a 3D NCOCX afterglow experiment and those following were recorded using a frequency-selective dephasing experiment. Our approach reduced the spectral congestion and provided a sensitive way to obtain chemical shift assignments for a membrane protein where no high-resolution structure is available. This MAS methodology is widely applicable to the study of other polytopic membrane proteins in functional lipid bilayer environments.