DOI: 10.1002/chem.201605143

DOI: 10.1002/chem.201700694

Chemistry – A European Journal. 2017 Mar 9; Volume 23; Issue 24

Conformational Analysis of an Antibacterial Cyclodepsipeptide Active against Mycobacterium tuberculosis by a Combined ROE and RDC Analysis

 

Maic Fredersdorfa, Michael Kurza, Armin Bauera, Marc-Olivier Ebertc, Carla Riglingc, Laurie Lannesd, Christina Marie Thieleb*

a Sanofi-Aventis GmbH, Department of Chemistry Industriepark Hoechst, 65926 Frankfurt am Main, Germany

b Technische Universität Darmstadt, Clemens-Schöpf-Institut für Organische Chemie und Biochemie, Alarich-Weiss-Straße 4, 64287 Darmstadt, Germany

c ETH Zürich, Laboratorium für Organische Chemie, D-CHAB, Vladimir-Prelog-Weg 3, 8093 Zürich, Switzerland

d Present address: Center for Biomolecular Magnetic Resonance (BMRZ), Institute of Organic Chemistry and Chemical Biology, Johann Wolfgang Goethe Universität Frankfurt, Max-von-Laue-Straße 7, 60438 Frankfurt, Germany

* Correspondence should be addressed to Christina Marie Thiele, Technische Universität Darmstadt, Clemens-Schöpf-Institut für Organische Chemie und Biochemie, Alarich-Weiss-Straße 4, 64287 Darmstadt, Germany

 

Abstract

Griselimycin (GM) and methylgriselimycin (MGM), naturally produced by microorganisms of the genus Streptomyces, are cyclic depsipeptides composed of ten amino acids. They exhibit antibacterial activity against Mycobacterium species by inhibiting the sliding clamp of prokaryotic DNA polymerase III and are therefore considered as potential anti-tuberculosis drugs. The difference between the peptides is the presence of L-(R)-4-methyl-proline in MGM instead of L-proline in GM at position 8 of the amino acid sequence. Methylation increases both metabolic stability and activity of MGM compared to GM. To get deeper insight into the structure-activity relationship, the solution structure of the cyclic part of MGM was determined using rotating-frame nuclear Overhauser effect (ROE) distance restraints and residual dipolar couplings (RDC). The structure of MGM in solution is compared to the structure of GM in a co-crystal with DNA polymerase III subunit beta. As a result, a highly defined structural model of MGM is obtained, which shows related characteristics to the bound GM.

PMID: 28106929

http://dx.doi.org/10.1002/chem.201605143

 

Supplement

The actual global tuberculosis (TB) report of the World Health Organization (WHO) describes that 6.3 million new TB cases were notified to national authorities in 2016 and the treatment success rate for people newly diagnosed with TB was 83% in 2015.[1] However, due to resistance to rifampicin, the most effective first-line drug, new TB drugs have to be developed to overcome this issue. Griselimycin (GM) as well as methylgriselimycin (MGM) represent two promising peptides which show high activity against M. tuberculosis by inhibiting the DNA polymerase sliding clamp DnaN as shown in Figure 1. The mode of action of GM and MGM including all biological and biophysical experimental findings are described in detail in a recent research article in Science.[2]

Although there are previous nuclear magnetic resonance (NMR)[3] and X-ray crystallography studies performed on the conformation of GM, the three dimensional structure of MGM has not been studied so far. Due to the similarity of both peptides, only L-(R)-proline in GM is replaced by L-(R)-4-methyl-proline, we investigate the cyclic part of the MGM structure by NMR and compare the result to the crystal structure of GM bound to the target protein (sliding clamp). Both analytical techniques are complementary methods for the determination of molecular structure and dynamics. Apart from determining a three dimensional model of MGM via NMR, it is therefore of high interest, whether both techniques lead to the same conformational characteristics of the slightly different peptides or if major changes, e.g. due to methodological limitations (X-ray → crystal structure vs. NMR → solution structure) or binding effects from the ligand to the target protein, occur.

 

 

Figure 1. The sliding clamp (grey) of Mycobacterium tuberculosis in complex with griselimycin (purple) measured by X-ray diffraction with a resolution of 2.173 Å (5agu, DOI: 10.2210/pdb5AGU/pdb).

 

After the determination of the connectivity of the atoms of MGM by conventional NMR methods, the three-dimensional model is determined by using the complementary information from the rotating-frame nuclear Overhauser effect (ROE)[4] and residual dipolar couplings (RDCs)[5]. Whereas ROE measurements are routinely used in NMR spectroscopy for the determination of the three dimensional structure, the latter parameter (RDC) is not applied very often because the sample preparation and subsequent data evaluation process are usually more time-consuming and more demanding in comparison to the ROE only (at least until now). Due to the wealth of additional structural information which is encoded in RDCs we have happily taken on this additional effort.

In general, for the determination of RDCs, an anisotropic environment is necessary. Therefore MGM was introduced into a chemically cross-linked PDMS gel swollen in an organic solvent. This is one way to generate anisotropy in a NMR-tube.[6]  The prerequisite for this is the solubility of the compound in the solvent as well as the compatibility with the gel; both is the case for MGM. With this anisotropically oriented sample it is possible to obtain high quality RDCs which are used in combination with the ROEs to calculate a precise structural model of MGM. The initial structure calculation of MGM is performed by using the software XPOLR-NIH.[7] In addition, all obtained structures are cross-checked by using the order tensor approach using a self-written module of the software hotFCHT.[8] The structure with the best quality criteria is compared co-crystal structure of griselimycin (Figure 2).

 

 

Figure 2. Comparison of the NMR solution structure (black) with the co crystal structure of griselimycin bound to the sliding clamp of Mycobacterium tuberculosis (green), Figure 1. The overall structure of the backbone is very similar but local differences especially in the methyl-proline area are observable. Purple: Distances, Red: Torsion angles

 

Conclusion

A well-defined three-dimensional model of the cyclic part of MGM could be determined using a set of ROE- and RDC-restraints. By comparing the solution structure of MGM with the co-crystal structure of GM we find that the overall structure of the backbone is similar but significant local differences are visible. Thus, for this example the hypothesis that the most active molecule already prefers a conformation in solution which is similar to its conformation in the complex seems to be confirmed.

 

References:

[1] http://www.who.int/tb/publications/C2_2017GLOBAL_FACTSHEET.pdf?ua=1

[2] A. Kling, P. Lukat, D. V. Almeida, A. Bauer, E. Fontaine, S. Sordello, N. Zaburannyi, J. Herrmann, S. C. Wenzel, C. König, N. C. Ammerman, M. Belén Barrio, K. Borchers, F. Bordon-Pallier, M. Brönstrup, G.  Courtemanche, M. Gerlitz, M. Geslin, P. Hammann, D. W. Heinz, H. Hoffmann, S. Klieber, M. Kohlmann, M. Kurz, C. Lair, H. Matter, E. Nuermberger, S. Tyagi, L. Fraisse, J. H. Grosset, S. Lagrange, R. Müller, Science 2015, 348, 1106-1112

[3] M. T. Cung, B. Vitoux, P. Demange, M. Marraud, Molecular Conformations and Biological Interactions (Eds.: P. Balaram, S. Ramaseshan), Indian Academy of Sciences, Bangalore, 1991, 611-625

[4] D. Neuhaus, The Nuclear Overhauser Effect in Structural and Conformational Analysis 2000, 2nd Edition, Wiley-VCH Verlag, ISBN 978-0-471-24675-6

[5] F. Kramer, M. V. Deshmukh, H. Kessler, S. J. Glaser, Concepts Magn. Reson. Part A 2004, 21 A, 10-21

[6] Y.E. Moskalenko, V. Bagutski, C. M. Thiele, Chem. Commun. 2017, 53, 95-98

[7] C. D. Schwieters, J. J. Kuszewski, G.M. Clore, Progr. NMR Spectroscopy 2006, 48, 47-62

[8] R. Berger, C. Fischer, M. Klessinger, J. Phys. Chem. A 1998, 102, 7157-7167