Open Access

In integrative structural biology/hybrid modeling approaches, we integrate structural models of macromolecules and experimental data to obtain faithful representations of the structures underlying the data. For example, in ensemble refinement by reweighting we first generate structural ensembles of flexible and dynamic biological macromolecules in molecular simulations. In a subsequent reweighting step, we refine the statistical weights of the structures to strike a balance between the information provided by simulations and by experimental data. For the "Bayesian inference of ensembles" approach (BioEn), we present two complementary methods to solve the underlying challenging high-dimensional optimization problem. We systematically investigate reliability, accuracy, and efficiency of these methods and integrate molecular dynamics simulations of the disordered peptide Ala-5 and NMR J-couplings. We provide an open-source library free of charge at https://github.com/bio-phys/BioEn.

Ensemble refinement produces structural ensembles of flexible and dynamic biomolecules by integrating experimental data and molecular simulations. Here we present two efficient numerical methods to solve the computationally challenging maximum-entropy problem arising from a Bayesian formulation of ensemble refinement. Recasting the resulting constrained weight optimization problem into an unconstrained form enables the use of gradient-based algorithms. In two complementary formulations that differ in their dimensionality, we optimize either the log-weights directly or the generalized forces appearing in the explicit analytical form of the solution. We first demonstrate the robustness, accuracy, and efficiency of the two methods using synthetic data. We then use NMR J-couplings to reweight an all-atom molecular dynamics simulation ensemble of the disordered peptide Ala-5 simulated with the AMBER99SB*-ildn-q force field. After reweighting, we find a consistent increase in the population of the polyproline-II conformations and a decrease of α-helical-like conformations. Ensemble refinement makes it possible to infer detailed structural models for biomolecules exhibiting significant dynamics, such as intrinsically disordered proteins, by combining input from experiment and simulation in a balanced manner.

Four different structural models, which all fit the same X-ray powder pattern, were obtained in the structure determination of 4,11-di­fluoro­quinacridone (C20H10N2O2F2) from unindexed X-ray powder data by a global fit. The models differ in their lattice parameters, space groups, Z, Z′, molecular packing and hydrogen bond patterns. The molecules form a criss-cross pattern in models A and B, a layer structure built from chains in model C and a criss-cross arrangement of dimers in model D. Nevertheless, all models give a good Rietveld fit to the experimental powder pattern with acceptable R-values. All molecular geometries are reliable, except for model D, which is slightly distorted. All structures are crystallochemically plausible, concerning density, hydrogen bonds, intermolecular distances etc. All models passed the checkCIF test without major problems; only in model A a missed symmetry was detected. All structures could have probably been published, although 3 of the 4 structures were wrong. The investigation, which of the four structures is actually the correct one, was challenging. Six methods were used: (1) Rietveld refinements, (2) fit of the crystal structures to the pair distribution function (PDF) including the refinement of lattice parameters and atomic coordinates, (3) evaluation of the colour, (4) lattice-energy minimizations with force fields, (5) lattice-energy minimizations by two dispersion-corrected density functional theory methods, and (6) multinuclear CPMAS solid-state NMR spectroscopy (1H, 13C, 19F) including the comparison of calculated and experimental chemical shifts. All in all, model B (perhaps with some disorder) can probably be considered to be the correct one. This work shows that a structure determination from limited-quality powder data may result in totally different structural models, which all may be correct or wrong, even if they are chemically sensible and give a good Rietveld refinement. Additionally, the work is an excellent example that the refinement of an organic crystal structure can be successfully performed by a fit to the PDF, and the combination of computed and experimental solid-state NMR chemical shifts can provide further information for the selection of the most reliable structure among several possibilities.

The evolution of cell-free protein synthesis (CFPS) over recent decades has made it a widely used system for expressing membrane proteins (MPs). Unlike traditional methods, CFPS allows direct and translocon-independent expression of MPs within lipid membranes, such as liposomes or nanodiscs (NDs), without the need for detergent solubilization. This open nature of CF systems enables customization of the experimental environment, including expression conditions, choice of nanoparticles (NPs), lipid composition, and addition of stabilizing molecules. Membrane scaffold protein (MSP)-based NDs emerged as a gold standard for cotranslational solubilization of MPs using the CF-system. This approach allowed not only biochemical characterization, but also structural studies of MPs and even GPCRs. However, to solubilize MPs inside nanoparticles via the traditional reconstitution route, apart from MSPs other scaffolds were successfully implemented, e.g. the saposin A (commercially known as Salipro) scaffold system or the synthetic styrene maleic acid lipid particles (SMALPs). In this study the potential of saposin A-based nanoparticles (SapNPs) was explored for cotranslational MP solubilization. Three strategies for applying SapNPs in CF systems were investigated: preassembly, (i) coassembly (ii), and coexpression (iii). (i) Preassembly involved forming SapNPs before CF expression and adding them to the CF reaction. In coassembly mode SapA and lipids were mixed in the CF reaction for spontaneous assembly with the synthesized MP. In coexpression mode lipids were added to the CF reaction while coexpressing SapA with the MP target. Proteorhodopsin (PR) served as a model protein to evaluate these strategies due to its ability to oligomerize and straightforward quantification using the cofactor retinal. Preassembled SapNPs provided homogeneous, aggregate-free particles yielding up to 200 µM solubilized PR inside in the CF reaction. Coassembly was also successfully applied to produce PR/SapNP complexes at slightly lower yields, however the system was prone to produce soluble aggregates at too high PR template concentrations and overall needed more adjustments. Coexpression resulted in PR yields below 20 µM and was not considered viable for MP production. Finally, the preassembled SapNPs were used to produce functional G-protein coupled receptor probes. Despite lower overall performance compared to MSP-based systems, SapNPs showed potential as an alternative in CF systems for specific MPs. The second optimization approach was directed at the CF lysate itself. CF synthesis for NMR analysis benefits from selective labeling schemes enabled by truncated amino acid (AA) metabolic pathways in lysates, reducing spectral ambiguity. However, residual enzymatic AA conversions persist, leading to label dilution and ambiguous NMR spectra. This study aimed to eliminate these residual activities in the E. coli A19 strain, generating optimized CF lysates for NMR applications. The approach involved cumulative gene deletions of the most problematic scrambling enzymes. The new strain, “Stablelabel,” included deletions and modifications in genes asnA, ansA, ansB, glnA, aspC, and ilvE, effectively eliminating background activities of L-Asn, L-Asp, and conversions of L-Glu to L-Asp and L-Gln. However, residual conversion of L-Gln to L-Glu persisted due to glutaminase activity of several glutaminases using the inhibitor 6 diazo-5-oxo-L-norleucine (DON). Stablelabel showed a slightly slower growth than A19, and an overall good performance with 2.7 mg/mL GFP expressed in the reaction mixture (RM) compared to the parental A19 strain with 3.5 mg/mL. Furthermore, the strain was successfully applied to demonstrate methyl group labeling of MPs using preconverted L-val and L-leu from their respective precursors 2-ketoisovalerate and 4-methyl-2-oxovalerate. In this study, lipid nanoparticle particle-and strain engineering vividly demonstrated the potential of CFPS systems and their versatility. While the SapNP system requires further engineering to potentially reach the efficiency of the well-studied MSP NDs, this study provides an example of nanoparticle characterization allowing new insights into NP behavior in CF systems. Furthermore, it was shown that strain engineering is a straightforward solution to tailor CF lysates to the individual requirements. After this thesis was submitted, Stablelabel in fact was successfully applied for backbone assignment of casein kinase 1, thereby demonstrating its suitability to express complex targets for NMR studies.

A plethora of modified nucleotides extends the chemical and conformational space for natural occurring RNAs. tRNAs constitute the class of RNAs with the highest modification rate. The extensive modification modulates their overall stability, the fidelity and efficiency of translation. However, the impact of nucleotide modifications on the local structural dynamics is not well characterized. Here we show that the incorporation of the modified nucleotides in tRNAfMet from Escherichia coli leads to an increase in the local conformational dynamics, ultimately resulting in the stabilization of the overall tertiary structure. Through analysis of the local dynamics by NMR spectroscopic methods we find that, although the overall thermal stability of the tRNA is higher for the modified molecule, the conformational fluctuations on the local level are increased in comparison to an unmodified tRNA. In consequence, the melting of individual base pairs in the unmodified tRNA is determined by high entropic penalties compared to the modified. Further, we find that the modifications lead to a stabilization of long-range interactions harmonizing the stability of the tRNA’s secondary and tertiary structure. Our results demonstrate that the increase in chemical space through introduction of modifications enables the population of otherwise inaccessible conformational substates.