Sampling the Small Molecule Universe with NMR, WebCocon & ConArch+

Results shown

A molecular formula, as obtained by MS or elemental analysis, normally comes with a huge number of possible constitutions, configurations and conformations, a universe of possibilities. Nuclear Magnetic Resonance is one of the methods commonly used in order to reduce this universe to a single solution, or eventually a small set of solutions.

In a first step chemical shifts (mainly Carbon and Proton) and different correlation experiments (COSY, HSQC, 13C-HMBC, 1,1-ADEQUATE, 15N-HMBC, among others) are used to obtain connectivity information. As correlation data sets might be incomplete or just not enough to define all connectivities of all atoms, various constitutions might be compatible with the NMR correlation data, and various assignments might be observed for a single constitution. This first step is carried out by CoCon [1,2], which outputs all possible constitutions with all possible assignments for a given correlation data set. WebCoCon uses different approaches to reduce the set of possible solutions created by CoCon.

  1. A simple Molecular Dynamics (MD) is used to create 3D structures. The total energy is used for the ranking of the solution within the list. Sometimes, specially with complex ring systems, this approach fails. But, it is the computationally cheapest method [2].
  2. A simple Distance Geometry (DG) is used to create 3D structures. DG is much more robust against complex molecular structures and generally provides better results, at slightly higher computational cost. The total energy is used for the ranking of the solutions.
  3. A Simple DG with NOE restraints is used to create improved 3D structures. DG uses very coarsely defined distances for the structure calculation, at essentially no additional cost. The obtained solutions compare much better to reference X-Ray structures, different from the previous solutions [3].
  4. A simple ORCA [4] 13C chemical shift calculation. The calculation of chemical shifts based on previously defined conformations seems very promising, but the structures generated by WebCocon without NOEs are not good enough for high quality results.

The results obtained by DG with NOEs indicate that using more experimental data from NMR can improve the selection process for the correct constitution and assignment, and lead to better solutions overall. Thus, ConArch+ [5] was interfaced with CoCon, as it is a DG package that not only handles NOEs, but also J-couplings, RDC, RCSA and other anisotropic data. Initially, only NOEs and RDCs are supported. When tested with Brucine, one (of two) assignment of the correct constitution was clearly favored in the results, when compared to the other two possible constitutions (with two assignments each) that CoCon found with theoretical NMR correlation data. Additionally, the 13C chemical shift calculations performed with ORCA also improved in quality for the correct assignment, showing considerably lower deviations from experimental data when compared to the results obtained with the other methods.

We are now working on obtaining other examples for this application. Unfortunately the publication of such complete NMR data sets is not very frequent.

Literature cited

The following literature references are used for this poster:

  1. (a) Lindel, T.; Junker, J.; Köck, M. J. Mol. Model. 1997, 3, 364–368; (b) Lindel, T.; Junker, J.; Köck, M. Eur. J. Org. Chem. 1999, 573–577; (c) Köck, M.; Junker, J.; Maier, W.; Will, M.; Lindel, T. Eur. J. Org. Chem. 1999, 579–586; (d) Junker, J.; Maier, W.; Lindel, T.; Köck, M. Org. Lett. 1999, 1, 737–740; (e) Köck, M.; Junker, J.; Lindel, T. Org. Lett. 1999, 1, 2041–2044.
  2. Junker, J., J. Cheminform. 2011, 3: 31.
  3. Köck, M.; Lindel, T.; Junker, J., Molecules 2021, 26: 4846.
  4. (a) Neese, F. The ORCA program system, Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2: 73–78; (b) Neese, F. Software update: the ORCA program system, version 4.0, Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2017, 8: e1327
  5. (a) Immel, S.; Köck, M., Reggelin, M.; Chem. Eur. J. 2018, 24, 13918-13930; (b) Immel, S.; Köck, M., Reggelin, M.; Chirality 2019, 31, 384-400; (c) Köck, M., Reggelin, M., Immel, S.; Marine Drugs 2020, 18; (d) Immel, S.; Köck, M., Reggelin, M.; J. Nat. Prod. 2022, 85, 1837-1849.
The Poster

This was upgraded to a talk, so there is no poster ;-)