Automated assignment of methyl NMR spectra from large proteins

Labeling of methyl groups: a streamlined protocol and guidance for the selection of 2H precursors based on molecular weight

A streamlined one-day protocol is described to produce isotopically methyl-labeled protein with high levels of deuterium for NMR studies. Using this protocol, the D2O and 2H-glucose content of the media and protonation level of ILV labeling precursors (ketobutyrate and ketovalerate) were varied. The relaxation rate of the multiple-quantum (MQ) state that is present during the HMQC-TROSY pulse sequence was measured for different labeling schemes and this rate was used to predict upper limits of molecular weights for various labeling schemes. The use of deuterated solvents (D2O) or deuterated glucose is not required to obtain 1H-13C correlated NMR spectra of a 50 kDa homodimeric protein that are suitable for assignment by mutagenesis. High quality spectra of 100-150 kDa proteins, suitable for most applications, can be obtained without the use of deuterated glucose. The proton on the β-position of ketovalerate appears to undergo partial exchange with deuterium under the growth conditions used in this study.

Time-optimized protein NMR assignment with an integrative deep learning approach using AlphaFold and chemical shift prediction

Chemical shift assignment is vital for nuclear magnetic resonance (NMR)-based studies of protein structures, dynamics, and interactions, providing crucial atomic-level insight. However, obtaining chemical shift assignments is labor intensive and requires extensive measurement time. To address this limitation, we previously proposed ARTINA, a deep learning method for automatic assignment of two-dimensional (2D)-4D NMR spectra. Here, we present an integrative approach that combines ARTINA with AlphaFold and UCBShift, enabling chemical shift assignment with reduced experimental data, increased accuracy, and enhanced robustness for larger systems, as presented in a comprehensive study with more than 5000 automated assignment calculations on 89 proteins. We demonstrate that five 3D spectra yield more accurate assignments (92.59%) than pure ARTINA runs using all experimentally available NMR data (on average 10 3D spectra per protein, 91.37%), considerably reducing the required measurement time. We also showcase automated assignments of only 15N-labeled samples, and report improved assignment accuracy in larger synthetic systems of up to 500 residues.

Fully Automated Characterization of Protein-Peptide Binding by Microfluidic 2D NMR

We demonstrate an automated microfluidic nuclear magnetic resonance (NMR) system that quantitatively characterizes protein-ligand interactions without user intervention and with minimal sample needs through protein-detected heteronuclear 2D NMR spectroscopy. Quantitation of protein-ligand interactions is of fundamental importance to the understanding of signaling and other life processes. As is well-known, NMR provides rich information both on the thermodynamics of binding and on the binding site. However, the required titrations are laborious and tend to require large amounts of sample, which are not always available. The present work shows how the analytical power of NMR detection can be brought in line with the trend of miniaturization and automation in life science workflows.

AssignSLP_GUI, a software tool exploiting AI for NMR resonance assignment of sparsely labeled proteins

Not all proteins are amenable to uniform isotopic labeling with 13C and 15N, something needed for the widely used, and largely deductive, triple resonance assignment process. Among them are proteins expressed in mammalian cell culture where native glycosylation can be maintained, and proper formation of disulfide bonds facilitated. Uniform labeling in mammalian cells is prohibitively expensive, but sparse labeling with one or a few isotopically enriched amino acid types is an option for these proteins. However, assignment then relies on accessing the best match between a variety of measured NMR parameters and predictions based on 3D structure, often from X-ray crystallography. Finding this match is a challenging process that has benefitted from many computational tools, including trained neural nets for chemical shift prediction, genetic algorithms for searches through a myriad of assignment possibilities, and now AI-based prediction of high-quality structures for protein targets. AssignSLP_GUI, a new version of a software package for assignment of resonances from sparsely-labeled proteins, uses many of these tools. These tools and new additions to the package are highlighted in an application to a sparsely-labeled domain from a glycoprotein, CEACAM1.

Combining NMR, SAXS and SANS to characterize the structure and dynamics of protein complexes

Understanding the structure and dynamics of biological macromolecules is essential to decipher the molecular mechanisms that underlie cellular functions. The description of structure and conformational dynamics often requires the integration of complementary techniques. In this review, we highlight the utility of combining nuclear magnetic resonance (NMR) spectroscopy with small angle scattering (SAS) to characterize these challenging biomolecular systems. NMR can assess the structure and conformational dynamics of multidomain proteins, RNAs and biomolecular complexes. It can efficiently provide information on interaction surfaces, long-distance restraints and relative domain orientations at residue-level resolution. Such information can be readily combined with high-resolution structural data available on subcomponents of biomolecular assemblies. Moreover, NMR is a powerful tool to characterize the dynamics of biomolecules on a wide range of timescales, from nanoseconds to seconds. On the other hand, SAS approaches provide global information on the size and shape of biomolecules and on the ensemble of all conformations present in solution. Therefore, NMR and SAS provide complementary data that are uniquely suited to investigate dynamic biomolecular assemblies. Here, we briefly review the type of data that can be obtained by both techniques and describe different approaches that can be used to combine them to characterize biomolecular assemblies. We then provide guidelines on which experiments are best suited depending on the type of system studied, ranging from fully rigid complexes, dynamic structures that interconvert between defined conformations and systems with very high structural heterogeneity.

An NMR look at an engineered PET depolymerase

Plastic environmental pollution is a major issue that our generation must face to protect our planet. Plastic recycling has the potential not only to reduce the pollution but also to limit the need for fossil-fuel-based production of new plastics. Enzymes capable of breaking down plastic could thereby support such a circular economy. Polyethylene terephthalate (PET) degrading enzymes have recently attracted considerable interest and have been subjected to intensive enzyme engineering to improve their characteristics. A quadruple mutant of Leaf-branch Compost Cutinase (LCC) was identified as a most efficient and promising enzyme. Here, we use NMR to follow the initial LCC enzyme through its different mutations that lead to its improved performance. We experimentally define the two calcium-binding sites and show their importance on the all-or-nothing thermal unfolding process, which occurs at a temperature of 72°C close to the PET glass transition temperature. Using various NMR probes such as backbone amide, methyl group, and histidine side-chain resonances, we probe the interaction of the enzymes with mono-(2-hydroxyethyl)terephthalic acid. The latter experiments are interpreted in terms of accessibility of the active site to the polymer chain.

Motional clustering in supra-τc conformational exchange influences NOE cross-relaxation rate

Biomolecular spin relaxation processes, such as the NOE, are commonly modeled by rotational τ_c-tumbling combined with fast motions on the sub-τ_c timescale. Motions on the supra-τ_c timescale, in contrast, are considered to be completely decorrelated to the molecular tumbling and therefore invisible. Here, we show how supra-τ_c dynamics can nonetheless influence the NOE build-up between methyl groups. This effect arises because supra-τ_c motions can cluster the fast-motion ensembles into discrete states, affecting distance averaging as well as the fast-motion order parameter and hence the cross-relaxation rate. We present a computational approach to estimate methyl-methyl cross-relaxation rates from extensive (>100×τ_c) all-atom molecular dynamics (MD) trajectories on the example of the 723-residue protein Malate Synthase G. The approach uses Markov state models (MSMs) to resolve transitions between metastable states and thus to discriminate between sub-τ_c and supra-τ_c conformational exchange. We find that supra-τ_c exchange typically increases NOESY cross-peak intensities. The methods described in this work extend the theory of modeling sub-μs dynamics in spin relaxation and thus contribute to a quantitative estimation of NOE cross-relaxation rates from MD simulations, eventually leading to increased precision in structural and functional studies of large proteins.

Assignment of Ala, Ile, LeuproS, Met, and ValproS methyl groups of the protruding domain of murine norovirus capsid protein VP1 using methyl-methyl NOEs, site directed mutagenesis, and pseudocontact shifts

The protruding domain (P-domain) of the murine norovirus (MNV) capsid protein VP1 is essential for infection. It mediates receptor binding and attachment of neutralizing antibodies. Protein NMR studies into interactions of the P-domain with ligands will yield insights not easily available from other biophysical techniques and will extend our understanding of MNV attachment to host cells. Such studies require at least partial NMR assignments. Here, we describe the assignment of about 70% of the Ala, Ile, LeuproS, Met, and ValproS methyl groups. An unfavorable distribution of methyl group resonance signals prevents complete assignment based exclusively on 4D HMQC-NOESY-HMQC experiments, yielding assignment of only 55 out of 100 methyl groups. Therefore, we created point mutants and measured pseudo contact shifts, extending and validating assignments based on methyl-methyl NOEs. Of note, the P-domains are present in two different forms caused by an approximate equal distribution of trans- and cis-configured proline residues in position 361.

NMR Assignment of Methyl Groups in Immobilized Proteins Using Multiple-Bond 13C Homonuclear Transfers, Proton Detection, and Very Fast MAS

In nuclear magnetic resonance spectroscopy of proteins, methyl protons play a particular role as extremely sensitive reporters on dynamics, allosteric effects, and protein–protein interactions, accessible even in high-molecular-weight systems approaching 1 MDa. The notorious issue of their chemical shift assignment is addressed here by a joint use of solid-state 1H-detected methods at very fast (nearly 100 kHz) magic-angle spinning, partial deuteration, and high-magnetic fields. The suitability of a series of RF schemes is evaluated for the efficient coherence transfer across entire 13C side chains of methyl-containing residues, which is key for establishing connection between methyl and backbone 1H resonances. The performance of ten methods for recoupling of either isotropic 13C–13C scalar or anisotropic dipolar interactions (five variants of TOBSY, FLOPSY, DIPSI, WALTZ, RFDR, and DREAM) is evaluated experimentally at two state-of-the-art magic-angle spinning (55 and 94.5 kHz) and static magnetic field conditions (18.8 and 23.5 T). Model isotopically labeled compounds (alanine and Met-Leu-Phe tripeptide) and ILV-methyl and amide-selectively protonated, and otherwise deuterated chicken α-spectrin SH3 protein are used as convenient reference systems. Spin dynamics simulations in SIMPSON are performed to determine optimal parameters of these RF schemes, up to recently experimentally attained spinning frequencies (200 kHz) and B0 field strengths (28.2 T). The concept of linearization of 13C side chain by appropriate isotope labeling is revisited and showed to significantly increase sensitivity of methyl-to-backbone correlations. A resolution enhancement provided by 4D spectroscopy with non-uniform (sparse) sampling is demonstrated to remove ambiguities in simultaneous resonance assignment of methyl proton and carbon chemical shifts.

Large Chaperone Complexes Through the Lens of Nuclear Magnetic Resonance Spectroscopy

Molecular chaperones are the guardians of the proteome inside the cell. Chaperones recognize and bind unfolded or misfolded substrates, thereby preventing further aggregation; promoting correct protein folding; and, in some instances, even disaggregating already formed aggregates. Chaperones perform their function by means of an array of weak protein-protein interactions that take place over a wide range of timescales and are therefore invisible to structural techniques dependent upon the availability of highly homogeneous samples. Nuclear magnetic resonance (NMR) spectroscopy, however, is ideally suited to study dynamic, rapidly interconverting conformational states and protein-protein interactions in solution, even if these involve a high-molecular-weight component. In this review, we give a brief overview of the principles used by chaperones to bind their client proteins and describe NMR methods that have emerged as valuable tools to probe chaperone-substrate and chaperone-chaperone interactions. We then focus on a few systems for which the application of these methods has greatly increased our understanding of the mechanisms underlying chaperone functions. Expected final online publication date for the Annual Review of Biophysics, Volume 51 is May 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Progress toward automated methyl assignments for methyl-TROSY applications

Methyl-TROSY spectroscopy has extended the reach of solution-state NMR to supra-molecular machineries over 100 kDa in size. Methyl groups are ideal probes for studying structure, dynamics, and protein-protein interactions in quasi-physiological conditions with atomic resolution. Successful implementation of the methodology requires accurate methyl chemical shift assignment, and the task still poses a significant challenge in the field. In this work, we outline the current state of technology for methyl labeling, data collection, data analysis, and nuclear Overhauser effect (NOE)-based automated methyl assignment approaches. We present MAGIC-Act and MAGIC-View, two Python extensions developed as part of the popular NMRFAM-Sparky package, and MAGIC-Net a standalone structure-based network analysis program. MAGIC-Act conducts statistically driven amino acid typing, Leu/Val pairing guided by 3D HMBC-HMQC, and NOESY cross-peak symmetry checking. MAGIC-Net provides model-based NOE statistics to aid in selection of a methyl labeling scheme. The programs provide a versatile, semi-automated framework for rapid methyl assignment.

Optimized precursor to simplify assignment transfer between backbone resonances and stereospecifically labelled valine and leucine methyl groups: application to human Hsp90 N-terminal domain

Methyl moieties are highly valuable probes for quantitative NMR studies of large proteins. Hence, their assignment is of the utmost interest to obtain information on both interactions and dynamics of proteins in solution. Here, we present the synthesis of a new precursor that allows connection of leucine and valine pro-S methyl moieties to backbone atoms by linear ¹³C-chains. This optimized ²H/¹³C-labelled acetolactate precursor can be combined with existing ¹³C/²H-alanine and isoleucine precursors in order to directly transfer backbone assignment to the corresponding methyl groups. Using this simple approach leucine and valine pro-S methyl groups can be assigned using a single sample without requiring correction of ¹H/²H isotopic shifts on ¹³C resonances. The approach was demonstrated on the N-terminal domain of human HSP90, for which complete assignment of Ala-β, Ile-δ₁, Leu-δ₂, Met-ε, Thr-γ and Val-γ₂ methyl groups was obtained.

Evaluation of Multi-Objective Optimization Algorithms for NMR Chemical Shift Assignment

An automated NMR chemical shift assignment algorithm was developed using multi-objective optimization techniques. The problem is modeled as a combinatorial optimization problem and its objective parameters are defined separately in different score functions. Some of the heuristic approaches of evolutionary optimization are employed in this problem model. Both, a conventional genetic algorithm and multi-objective methods, i.e., the non-dominated sorting genetic algorithms II and III (NSGA2 and NSGA3), are applied to the problem. The multi-objective approaches consider each objective parameter separately, whereas the genetic algorithm followed a conventional way, where all objectives are combined in one score function. Several improvement steps and repetitions on these algorithms are performed and their combinations are also created as a hyper-heuristic approach to the problem. Additionally, a hill-climbing algorithm is also applied after the evolutionary algorithm steps. The algorithms are tested on several different datasets with a set of 11 commonly used spectra. The test results showed that our algorithm could assign both sidechain and backbone atoms fully automatically without any manual interactions. Our approaches could provide around a 65% success rate and could assign some of the atoms that could not be assigned by other methods.

NMR spectroscopy captures the essential role of dynamics in regulating biomolecular function

Biomolecules are in constant motion. To understand how they function, and why malfunctions can cause disease, it is necessary to describe their three-dimensional structures in terms of dynamic conformational ensembles. Here, we demonstrate how nuclear magnetic resonance (NMR) spectroscopy provides an essential, dynamic view of structural biology that captures biomolecular motions at atomic resolution. We focus on examples that emphasize the diversity of biomolecules and biochemical applications that are amenable to NMR, such as elucidating functional dynamics in large molecular machines, characterizing transient conformations implicated in the onset of disease, and obtaining atomic-level descriptions of intrinsically disordered regions that make weak interactions involved in liquid-liquid phase separation. Finally, we discuss the pivotal role that NMR has played in driving forward our understanding of the biomolecular dynamics-function paradigm.

Nearest-neighbor NMR spectroscopy: categorizing spectral peaks by their adjacent nuclei

Methyl-NMR enables atomic-resolution studies of structure and dynamics of large proteins in solution. However, resonance assignment remains challenging. The problem is to combine existing structural informational with sparse distance restraints and search for the most compatible assignment among the permutations. Prior classification of peaks as either from isoleucine, leucine, or valine reduces the search space by many orders of magnitude. However, this is hindered by overlapped leucine and valine frequencies. In contrast, the nearest-neighbor nuclei, coupled to the methyl carbons, resonate in distinct frequency bands. Here, we develop a framework to imprint additional information about passively coupled resonances onto the observed peaks. This depends on simultaneously orchestrating closely spaced bands of resonances along different magnetization trajectories, using principles from control theory. For methyl-NMR, the method is implemented as a modification to the standard fingerprint spectrum (the 2D-HMQC). The amino acid type is immediately apparent in the fingerprint spectrum. There is no additional relaxation loss or an increase in experimental time. The method is validated on biologically relevant proteins. The idea of generating new spectral information using passive, adjacent resonances is applicable to other contexts in NMR spectroscopy.

Silkworm Pupae Function as Efficient Producers of Recombinant Glycoproteins with Stable-Isotope Labeling

Baculovirus-infected silkworms are promising bioreactors for producing recombinant glycoproteins, including antibodies. Previously, we developed a method for isotope labeling of glycoproteins for nuclear magnetic resonance (NMR) studies using silkworm larvae reared on an artificial diet containing ¹⁵N-labeled yeast crude protein extract. Here, we further develop this method by introducing a technique for the expression of isotope-labeled glycoproteins by silkworm pupae, which has several potential advantages relative to larvae-based techniques in terms of production yield, ease of handling, and storage. Here, we fed fifth instar larvae an artificial diet with an optimized composition containing [methyl-¹³C]methionine, leading to pupation. Nine-day-old pupae were then injected with recombinant Bombyx mori nucleopolyhedrovirus (BmNPV) bacmid for expression of recombinant human immunoglobulin G (IgG). From the whole-body homogenates of pupae, 0.35 mg/pupa of IgG was harvested, which is a yield that is five times higher than can be obtained from larvae. Recombinant IgG, thus prepared, exhibited mainly three kinds of pauci-mannose-type oligosaccharides and had a ¹³C-enrichment ratio of approximately 80%. This enabled selective observation of NMR signals originating from the methionyl methyl group of IgG, confirming its conformational integrity. These data demonstrate the utility of silkworm pupae as factories for producing recombinant glycoproteins with amino-acid-selective isotope labeling.