Our group participates in the EU COST action BM1403 "Native Mass Spectrometry and Related Methods for Structural Biology".


Novel MS-Cleavable Cross-Linkers for Protein Structure Characterization

Chemical cross-linking combined with a subsequent enzymatic digestion and mass spectrometric analysis of the created cross-linked products presents an alternative approach for assessing low-resolution protein structures and for gaining insight into protein interfaces.  This project aims to design innovative amine-reactive, N-hydroxy succinimide (NHS) cross-linkers. Our collision-induced dissociative cross-linkers lead to the formation of indicative fragment ions and constant neutral losses in MS/MS spectra, allowing an unambiguous identification of cross-linked peptides. ESI and MALDI mass spectrometry can be used for analyzing the created cross-linked products. 

Our MS-cleavable, urea-based linker is shown below, offering the possibility to identify cross-linked products in an efficient and robust manner. This project is conducted in cooperation with Dr. Mathias Schäfer, University of Cologne (DFG project Si 867/15-1 and 15-2). 

Our cross-linking reagent is now commercially available at Thermo Fisher Scientific under the name DSBU (disuccinimidyl dibutyric urea):



Structure of Full-Length p53 Tumor Suppressor Probed by Chemical Cross-Linking and Mass Spectrometry

The tumor suppressor p53 presents a great challenge for 3D-structural analysis due to its inherent flexibility. In this project, our aim is to gain insights into the structure of full-length wild-type human p53 in solution by chemical cross-linking/mass spectrometry (MS). This approach allows us obtaining structural information of DNA-free p53 in solution without making use of the ultrastable quadruple p53 variant. The cross-links within one p53 monomer are in good agreement with the SAXS-based model of full-length p53. Our cross-linking data between different p53 molecules in the tetramer however indicate a large degree of flexibility in the C-terminal regulatory domain of full-length p53 in the absence of DNA. The cross-links suggest that the C-terminal regulatory domains are much closer to each other, resulting in a more compact arrangement of the p53 tetramer than perceived by the SAXS model.


Investigation of Laminin/Nidogen Interaction by Chemical Cross-Linking and Mass Spectrometry

The heterotrimeric (α,β,γ subunits), 800 kDa glycoprotein laminin is the main non-collagenous component of basement membranes, controlling cellular activities, such as cell adhesion, migration, differentiation, and apoptosis. Laminin is essential for basement membrane formation through interactions with itself and other basement membrane components. We perform detailed structural investigation of nidogen 1/laminin γ1 complexes using full-length nidogen-1 and a number of laminin γ1 variants. The interactions of nidogen-1 with laminin variants γ1 LEb2–4, γ1 LEb2–4 N836D, γ1 short arm, γ1 short arm N836D, and γ1 short arm ΔLEb3 are investigated by applying a combination of chemical cross-linking, high-resolution MS, and computational modeling. Two complementary cross-linking strategies are pursued to analyze solution structures of lamininγ1 variants and nidogen-1. The majority of distance information is obtained with the homobifuncational amine-reactive cross-linker bis(sulfosuccinimidyl)glutarate. In a second approach, UV-induced cross-linking is performed after incorporation of the diazirine-containing unnatural amino acids photo-leucine and photo-methionine into lamininγ1 LEb2–4, lamininγ1 short arm, and nidogen-1. Our results indicate that Asp-836 within lamininγ1 LEb3 domain is not essential for complex formation. Cross-links between lamininγ1 short arm and nidogen-1 are found in all protein regions, evidencing several additional contact regions apart from the known interaction site. Computational modeling based on the cross-linking constraints indicates the existence of a conformational ensemble of both the individual proteins and the nidogen-1/laminin γ1 complex. This finding implies different modes of interaction resulting in several distinct protein-protein interfaces.

This project is conducted in cooperation with Prof. Mats Paulsson, Dr. Frank Zaucke (University of Cologne) and Prof. Jens Meiler (Vanderbilt University, Nashville, USA) and was financially supported by the Deutsche Forschungsgemeinschaft, DFG (SPP 1086, Si 867/7-1) and the Graduiertenkolleg GRK 1026, Martin Luther University Halle-Wittenberg. Laminin/nidogen interaction studies are funded by the BMBF (ProNet-T3 project).



Identification and Structural Characterization of Protein-Protein Interactions in the Formate Metabolism of Escherichia coli by Mass Spectrometry

Protein-protein-interactions in the formate metabolism of Escherichia coli are identified and structurally characterized by mass spectrometry (MS)-based methods. The formate dehydrogenases Fdh-O, Fdh-N, and Fdh-H, are associated with the inner membrane and have their respective active site located towards the periplasm. Our aim is to shed light on the interactions of Fdh-N and Fdh-O in the periplasmic space that might be involved in stabilizing these respiratory enzymes after maturation and translocation across the cytoplasmic membrane. For this purpose, chemical cross-linking using a heterobifunctional amine/photo-reactive reagent followed by MS analysis of the cross-linked products is performed.

A second aspect of this project deals with the structural characterization of the FocA/PflB complex and its regulatory effect on formate transport. The formate-nitrate transporters (FNT) form a superfamily of pentameric membrane channels that translocate monovalent anions across biological membranes. FocA translocates formate bidirectionally, but the mechanisms underlying the control of translocation and substrate specificity remain unclear. Chemical cross-linking with a homobifunctional cross-linker is performed for deriving detailed 3D-structural data of the FocA/PflB complex. In order to determine a binding orientation of PflB to FocA, models of the complexes are generated using the software suite Rosetta (cooperation with Prof. Jens Meiler, Vanderbilt University, Nashville).

In conclusion, chemical cross-linking and MS in combination with computational modeling using Rosetta is a highly promising approach for the identification and structural characterization of multisubunit membrane protein complexes.

This project is conducted in cooperation with Prof. Dr. Gary Sawers, Institute of Microbiology, Martin-Luther University Halle-Wittenberg.

Monitoring Conformational Changes in Peroxisome Proliferator-Activated Receptor alpha (PPARα) by a Genetically Encoded Photo-Amino Acid, Cross-Linking, and Mass Spectrometry

The peroxisome proliferator-activated receptor alpha (PPARα) belongs to the nuclear receptor family that controls the expression of genes involved in fatty acid metabolism. PPARα promotes fatty acid catabolism in the liver and sceletal muscle. Disorders in fatty acid metabolism might lead to obesity, cardiovascular diseases, type II diabetes, and atherosclerosis. In this project, we investigate conformational changes in peroxisome proliferator-activated receptor α (PPARα) upon ligand binding. Using E. coli cells with a special tRNA/aminoacyl-tRNA synthetase pair, PPARα variants can be created, in which defined amino acids are site-specifically replaced by the genetically encoded photo-reactive amino acid para-benzoylphenylalanine (Bpa). PPARα variants are subjected to UV-induced cross-linking, both in the absence and presence of ligands. After the photo-cross-linking reaction, reaction mixtures are enzymatically digested and peptides are analyzed by MS. The inter-residue distances disclosed by the photo-chemical cross-links serve to monitor conformational changes in PPARα upon agonist and antagonist binding. The data obtained with our strategy underline the potential of genetically encoded internal photo-cross-linkers in combination with MS as an alternative method to monitor in-solution 3D-structures of drug targets.


Structural Insights into Retinal Guanylylcyclase/GCAP-2 Interaction by Cross-Linking and Mass Spectrometry

The retinal guanylylcyclases ROS-GC 1 and 2 are regulated via the intracellular site by guanylylcyclase-activating proteins (GCAPs). The mechanism of how GCAPs activate their target proteins remains elusive as exclusively structures of non-activating calcium-bound GCAP-1 and -2 are available. In this work, we apply a combination of chemical cross-linking with amine-reactive cross-linkers and photo-affinity labeling followed by MS analysis of the created cross-linked products to study the interaction between N-terminally myristoylated GCAP-2 and a peptide derived from the catalytic domain of full length ROS-GC 1. Based on the distance constraints imposed by the cross-links we are creating structural models of the calcium-loaded complex between myristoylated GCAP-2 and the GC peptide.

This project is conducted in cooperation with Prof. Daniel Huster (University of Leipzig), Dr. Christian Lange, and Prof. Alfred Blume (Halle). The project was funded by the DFG (Si 867/13-1).


Investigating Calmodulin/Munc13 Interaction

The efficacy of synaptic transmission between neurons can be transiently altered during neuronal network activity. This phenomenon of short-term plasticity is a key determinant of network properties, is involved in many physiological processes such as motor control, sound localization, or sensory adaptation, and is critically dependent on cytosolic calcium concentration. Due to their essential function in synaptic vesicle priming and in the modulation of synaptic strength, Munc13 proteins are key regulators of presynaptic short-term plasticity. However, the underlying molecular mechanisms and the identity of the calcium sensor/effector complexes involved are unclear. All four Munc13 isoforms share a common domain structure, including a calmodulin (CaM) binding site in their otherwise divergent N-termini. By combining chemical cross-linking, photoaffinity labeling, and mass spectrometry, we show that all neuronal Munc13 isoforms exhibit similar CaM binding modes. Moreover, we demonstrate that the 1-5-8-26 CaM binding motif discovered in Munc13-1 cannot be induced in the classical CaM target skMLCK (skeletal muscle myosin light chain kinase), indicating unique features of the Munc13 CaM binding motif.

This project is conducted in cooperation with Dr. Olaf Jahn, Max-Planck Institute for Experimental Medicine, Göttingen, and was financially supported by the Graduiertenkolleg GRK 1026 (Conformational Transitions in Macromolecular Interactions) at the Martin-Luther University Halle-Wittenberg until 2014.