Amino Acid-Selective Segmental Isotope Labeling of Multidomain Proteins for Structural Biology

Efficient Segmental Isotope Labeling of Integral Membrane Proteins for High-Resolution NMR Studies

High-resolution structural NMR analyses of membrane proteins are challenging due to their large size, resulting in broad resonances and strong signal overlap. Among the isotope labeling methods that can remedy this situation, segmental isotope labeling is a suitable strategy to simplify NMR spectra and retain high-resolution structural information. However, protein ligation within integral membrane proteins is complicated since the hydrophobic protein fragments are insoluble, and the removal of ligation side-products is elaborate. Here, we show that a stabilized split-intein system can be used for rapid and high-yield protein trans-splicing of integral membrane proteins under denaturing conditions. This setup enables segmental isotope labeling experiments within folded protein domains for NMR studies. We show that high-quality NMR spectra of markedly reduced complexity can be obtained in detergent micelles and lipid nanodiscs. Of note, the nanodisc insertion step specifically selects for the ligated and correctly folded membrane protein and simultaneously removes ligation byproducts. Using this tailored workflow, we show that high-resolution NMR structure determination is strongly facilitated with just two segmentally isotope-labeled membrane protein samples. The presented method will be broadly applicable to structural and dynamical investigations of (membrane-) proteins and their complexes by solution and solid-state NMR but also other structural methods where segmental labeling is beneficial.

Unexpected dynamics in femtomolar complexes of binding proteins with peptides

Ultra-tight binding is usually observed for proteins associating with rigidified molecules. Previously, we demonstrated that femtomolar binders derived from the Armadillo repeat proteins (ArmRPs) can be designed to interact very tightly with fully flexible peptides. Here we show for ArmRPs with four and seven sequence-identical internal repeats that the peptide-ArmRP complexes display conformational dynamics. These dynamics stem from transient breakages of individual protein-residue contacts that are unrelated to overall unbinding. The labile contacts involve electrostatic interactions. We speculate that these dynamics allow attaining very high binding affinities, since they reduce entropic losses. Importantly, only NMR techniques can pick up these local events by directly detecting conformational exchange processes without complications from changes in solvent entropy. Furthermore, we demonstrate that the interaction surface of the repeat protein regularizes upon peptide binding to become more compatible with the peptide geometry. These results provide novel design principles for ultra-tight binders.

Chemoenzymatic Semisynthesis of Proteins

Protein semisynthesis-defined herein as the assembly of a protein from a combination of synthetic and recombinant fragments-is a burgeoning field of chemical biology that has impacted many areas in the life sciences. In this review, we provide a comprehensive survey of this area. We begin by discussing the various chemical and enzymatic methods now available for the manufacture of custom proteins containing noncoded elements. This section begins with a discussion of methods that are more chemical in origin and ends with those that employ biocatalysts. We also illustrate the commonalities that exist between these seemingly disparate methods and show how this is allowing for the development of integrated chemoenzymatic methods. This methodology discussion provides the technical foundation for the second part of the review where we cover the great many biological problems that have now been addressed using these tools. Finally, we end the piece with a short discussion on the frontiers of the field and the opportunities available for the future.

Small-Angle Neutron Scattering of RNA-Protein Complexes

Small-angle neutron scattering (SANS) provides structural information on biomacromolecules and their complexes in dilute solutions at the nanometer length scale. The overall dimensions, shapes, and interactions can be probed and compared to information obtained by complementary structural biology techniques such as crystallography, NMR, and EM. SANS, in combination with solvent H₂O/D₂O exchange and/or deuteration, is particularly well suited to probe the internal structure of RNA-protein (RNP) complexes since neutrons are more sensitive than X-rays to the difference in scattering length densities of proteins and RNA, with respect to an aqueous solvent. In this book chapter we provide a practical guide on how to carry out SANS experiments on RNP complexes, as well as possibilities of data analysis and interpretation.

Towards complete polypeptide backbone NH assignment via combinatorial labeling

Combinatorial selective isotope labeling is a valuable tool to facilitate polypeptide backbone resonance assignment in cases of low sensitivity or extensive chemical shift degeneracy. It involves recording of ¹⁵N-HSQC and 2D HN-projections of triple-resonance spectra on a limited set of samples containing different combinations of labeled and unlabeled amino acid types. Using labeling schemes in which the three backbone heteronuclei (amide nitrogen, α-carbon and carbonyl carbon) are enriched in ¹⁵N or ¹³C isotopes - individually as well as simultaneously - usually yields abundant amino-acid type information of consecutive residues i and i - 1. Although this results in a large number of anchor points that can be used in the sequential assignment process, for most amide signals the exact positioning of the corresponding residue the polypeptide sequence still relies on matching intra- and interresidual ¹³C chemical shifts obtained from 3D spectra. An obvious way to obtain more sequence-specific assignments directly with combinatorial labeling would be to increase the number of samples. This is, however, undesirable because of increased sample preparation efforts and costs. Irrespective of the number of samples, unambiguous assignments cannot be accomplished for i - 1/i pairs that are not unique in the sequence. Here we show that the ambiguity for non-unique pairs can be resolved by including information about the labeling state of residues i + 1 and i - 2. Application to a 35-residue peptide resulted in complete assignments of all detectable signals in the ¹⁵N HSQC which, due to its repetitive sequence and ¹³C chemical shift degeneracies, was difficult to achieve by other means. For a medium-sized protein (165 residues, rotational correlation time 8.2 ns) the improved protocol allowed the extent of backbone amide assignment to be expanded to 88% solely using a suite of 2D ¹H-¹⁵N correlated spectra.

Peptide binding affinity redistributes preassembled repeat protein fragments

Designed armadillo repeat proteins (dArmRPs) are modular peptide binders composed of N- and C-terminal capping repeats Y and A and a variable number of internal modules M that each specifically recognize two amino acids of the target peptide. Complementary fragments of dArmRPs obtained by splitting the protein between helices H1 and H2 of an internal module show conditional and specific assembly only in the presence of a target peptide (Michel, E., Plückthun, A., and Zerbe, O. (2018). Peptide-guided assembly of repeat protein fragments. Angew. Chem. Int. Ed. 57, 4576–4579). Here, we investigate dArmRP fragments that already spontaneously assemble with high affinity, e.g. those obtained from splits between entire modules or between helices H2 and H3. We find that the interaction of the peptide with the assembled fragments induces distal conformational rearrangements that suggest an induced fit on a global protein level. A population analysis of an equimolar mixture of an N-terminal and three C-terminal fragments with various affinities for the target peptide revealed predominant assembly of the weakest peptide binder. However, adding a target peptide to this mixture altered the population of the protein complexes such that the combination with the highest affinity for the peptide increased and becomes predominant when adding excess of peptide, highlighting the feasibility of peptide-induced enrichment of best binders from inter-modular fragment mixtures.

Segmental isotope labelling and solid-state NMR of a 12 × 59 kDa motor protein: identification of structural variability

Segmental isotope labelling enables the NMR study of an individual domain within a multidomain protein, but still in the context of the entire full-length protein. Compared to the fully labelled protein, spectral overlap can be greatly reduced. We here describe segmental labelling of the (double-) hexameric DnaB helicase from Helicobacter pylori using a ligation approach. Solid-state spectra demonstrate that the ligated protein has the same structure and structural order as the directly expressed full-length protein. We uniformly ¹³C/¹⁵N labeled the N-terminal domain (147 residues) of the protein, while the C-terminal domain (311 residues) remained in natural abundance. The reduced signal overlap in solid-state NMR spectra allowed to identify structural "hotspots" for which the structure of the N-terminal domain in the context of the oligomeric full-length protein differs from the one in the isolated form. They are located near the linker between the two domains, in an α-helical hairpin.

Segmental isotopic labeling of HIV-1 capsid protein assemblies for solid state NMR

Recent studies of noncrystalline HIV-1 capsid protein (CA) assemblies by our laboratory and by Polenova and coworkers (Protein Sci 19:716-730, 2010; J Mol Biol 426:1109-1127, 2014; J Biol Chem 291:13098-13112, 2016; J Am Chem Soc 138:8538-8546, 2016; J Am Chem Soc 138:12029-12032, 2016; J Am Chem Soc 134:6455-6466, 2012; J Am Chem Soc 132:1976-1987, 2010; J Am Chem Soc 135:17793-17803, 2013; Proc Natl Acad Sci USA 112:14617-14622, 2015; J Am Chem Soc 138:14066-14075, 2016) have established the capability of solid state nuclear magnetic resonance (NMR) measurements to provide site-specific structural and dynamical information that is not available from other types of measurements. Nonetheless, the relatively high molecular weight of HIV-1 CA leads to congestion of solid state NMR spectra of fully isotopically labeled assemblies that has been an impediment to further progress. Here we describe an efficient protocol for production of segmentally labeled HIV-1 CA samples in which either the N-terminal domain (NTD) or the C-terminal domain (CTD) is uniformly 15N,13C-labeled. Segmental labeling is achieved by trans-splicing, using the DnaE split intein. Comparisons of two-dimensional solid state NMR spectra of fully labeled and segmentally labeled tubular CA assemblies show substantial improvements in spectral resolution. The molecular structure of HIV-1 assemblies is not significantly perturbed by the single Ser-to-Cys substitution that we introduce between NTD and CTD segments, as required for trans-splicing.

A method for the unbiased and efficient segmental labelling of RNA-binding proteins for structure and biophysics

Most eukaryotic RNA regulators recognise their RNA and protein partners by the combinatorial use of several RNA binding domains. Inter-domain dynamics and interactions play a key role in recognition and can be analysed by techniques such as NMR or FRET, provided that the information relative to the individual interactions can be de-convoluted. Segmentally labelling the proteins by ligating labelled and unlabelled peptide chains allows one to filter out unwanted information and observe the labelled moieties only. Several strategies have been implemented to ligate two protein fragments, but multiple ligations, which are necessary to segmentally label proteins of more than two domains, are more challenging and often dependent on the structure and solubility of the domains. Here we report a method to ligate multiple protein segments that allows the fast, high yield labelling of both internal and end domains, depending on the requirements. We use TCEP and mercaptophenylacetic acid (MPAA) in an optimised reaction environment to achieve an efficient ligation of protein domains independently from their structure or solubility. We expect the method will provide a useful tool for the molecular study of combinatorial protein–RNA recognition in RNA regulation.

Acceleration of protein backbone NMR assignment by combinatorial labeling: Application to a small molecule binding study

Selective labeling with stable isotopes has long been recognized as a valuable tool in protein NMR to alleviate signal overlap and sensitivity limitations. In this study, combinatorial 15 N-, 13 Cα -, and 13 C'-selective labeling has been used during the backbone assignment of human cyclophilin D to explore binding of an inhibitor molecule. Using a cell-free expression system, a scheme that involves 15 N, 1-13 C, 2-13 C, fully 15 N/13 C, and unlabeled amino acids was optimized to gain a maximum of assignment information from three samples. This scheme was combined with time-shared triple-resonance NMR experiments, which allows a fast and efficient backbone assignment by giving the unambiguous assignment of unique amino acid pairs in the protein, the identity of ambiguous pairs and information about all 19 non-proline amino acid types. It is therefore well suited for binding studies where de novo assignments of amide 1 H and 15 N resonances need to be obtained, even in cases where sensitivity is the limiting factor.

Segmental Isotopic Labeling of Proteins for NMR Study Using Intein Technology

Segmental isotopic labeling of samples for NMR studies is attractive for large complex biomacromolecular systems, especially for studies of function-related protein-ligand interactions and protein dynamics (Goto and Kay, Curr Opin Struct Biol 10:585-592, 2000; Rosa et al., Molecules (Basel, Switzerland) 18:440, 2013; Hiroaki, Expert Opin Drug Discovery 8:523-536, 2013). Advantages of segmental isotopic labeling include selective examination of specific segment(s) within a protein by NMR, significantly reducing the spectral complexity for large proteins, and allowing for the application of a variety of solution-based NMR strategies. By utilizing intein techniques (Wood and Camarero, J Biol Chem 289:14512-14519, 2014; Paulus, Annu Rev Biochem 69:447-496, 2000), two related approaches can generally be used in the segmental isotopic labeling of proteins: expressed protein ligation (Muir, Annu Rev Biochem 72:249-289, 2003) and protein trans-splicing (Shah et al., J Am Chem Soc 134:11338-11341, 2012). Here, we describe general implementation and latest improvements of expressed protein ligation method for the production of segmental isotopic labeled NMR samples.

Isotope-labeling strategies for solution NMR studies of macromolecular assemblies

Proteins come together in macromolecular assemblies, recognizing and binding to each other through their structures, and operating on their substrates through their motions. Detailed characterization of these processes is particularly suited to NMR, a high-resolution technique sensitive to structure, dynamics, and interactions. Advances in isotope-labeling have enabled such studies to an ever-increasing range of systems. Here we highlight recent applications and bring to the fore the range of options to produce labeled proteins and to control the specific placement of isotopes. The increased labeling control and affordability, together with the possibility to combine strategies will further deepen and extend the range of protein assembly investigations.

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.

Cotranslational incorporation of non-standard amino acids using cell-free protein synthesis

Over the last years protein engineering using non-standard amino acids has gained increasing attention. As a result, improved methods are now available, enabling the efficient and directed cotranslational incorporation of various non-standard amino acids to equip proteins with desired characteristics. In this context, the utilization of cell-free protein synthesis is particularly useful due to the direct accessibility of the translational machinery and synthesized proteins without having to maintain a vital cellular host. We review prominent methods for the incorporation of non-standard amino acids into proteins using cell-free protein synthesis. Furthermore, a list of non-standard amino acids that have been successfully incorporated into proteins in cell-free systems together with selected applications is provided.

Time-shared experiments for efficient assignment of triple-selectively labeled proteins

Combinatorial triple-selective labeling facilitates the NMR assignment process for proteins that are subject to signal overlap and insufficient signal-to-noise in standard triple-resonance experiments. Aiming at maximum amino-acid type and sequence-specific information, the method represents a trade-off between the number of selectively labeled samples that have to be prepared and the number of spectra to be recorded per sample. In order to address the demand of long measurement times, we here propose pulse sequences in which individual phase-shifted transients are stored separately and recombined later to produce several 2D HN(CX) type spectra that are usually acquired sequentially. Sign encoding by the phases of (13)C 90° pulses allows to either select or discriminate against (13)C' or (13)C(α) spins coupled to (15)N. As a result, (1)H-(15)N correlation maps of the various isotopomeric species present in triple-selectively labeled proteins are deconvoluted which in turn reduces problems due to spectral overlap. The new methods are demonstrated with four different membrane proteins with rotational correlation times ranging from 18 to 52 ns.

Practical aspects of NMR signal assignment in larger and challenging proteins

NMR has matured into a technique routinely employed for studying proteins in near physiological conditions. However, applications to larger proteins are impeded by the complexity of the various correlation maps necessary to assign NMR signals. This article reviews the data analysis techniques traditionally employed for resonance assignment and describes alternative protocols necessary for overcoming challenges in large protein spectra. In particular, simultaneous analysis of multiple spectra may help overcome ambiguities or may reveal correlations in an indirect manner. Similarly, visualization of orthogonal planes in a multidimensional spectrum can provide alternative assignment procedures. We describe examples of such strategies for assignment of backbone, methyl, and nOe resonances. We describe experimental aspects of data acquisition for the related experiments and provide guidelines for preliminary studies. Focus is placed on large folded monomeric proteins and examples are provided for 37, 48, 53, and 81 kDa proteins.

Solution structure of the two RNA recognition motifs of hnRNP A1 using segmental isotope labeling: how the relative orientation between RRMs influences the nucleic acid binding topology

Human hnRNP A1 is a multi-functional protein involved in many aspects of nucleic-acid processing such as alternative splicing, micro-RNA biogenesis, nucleo-cytoplasmic mRNA transport and telomere biogenesis and maintenance. The N-terminal region of hnRNP A1, also named unwinding protein 1 (UP1), is composed of two closely related RNA recognition motifs (RRM), and is followed by a C-terminal glycine rich region. Although crystal structures of UP1 revealed inter-domain interactions between RRM1 and RRM2 in both the free and bound form of UP1, these interactions have never been established in solution. Moreover, the relative orientation of hnRNP A1 RRMs is different in the free and bound crystal structures of UP1, raising the question of the biological significance of this domain movement. In the present study, we have used NMR spectroscopy in combination with segmental isotope labeling techniques to carefully analyze the inter-RRM contacts present in solution and subsequently determine the structure of UP1 in solution. Our data unambiguously demonstrate that hnRNP A1 RRMs interact in solution, and surprisingly, the relative orientation of the two RRMs observed in solution is different from the one found in the crystal structure of free UP1 and rather resembles the one observed in the nucleic-acid bound form of the protein. This strongly supports the idea that the two RRMs of hnRNP A1 have a single defined relative orientation which is the conformation previously observed in the bound form and now observed in solution using NMR. It is likely that the conformation in the crystal structure of the free form is a less stable form induced by crystal contacts. Importantly, the relative orientation of the RRMs in proteins containing multiple-RRMs strongly influences the RNA binding topologies that are practically accessible to these proteins. Indeed, RRM domains are asymmetric binding platforms contacting single-stranded nucleic acids in a single defined orientation. Therefore, the path of the nucleic acid molecule on the multiple RRM domains is strongly dependent on whether the RRMs are interacting with each other. The different nucleic acid recognition modes by multiple-RRM domains are briefly reviewed and analyzed on the basis of the current structural information.