Unexpected Importance of Aromatic-Aliphatic and Aliphatic Side Chain-Backbone Interactions in the Stability of Amyloids.

The role of aromatic and nonaromatic amino acids in amyloid formation has been elucidated by calculating interaction energies between ß-sheets in amyloid model systems using density functional theory (B3LYP-D3/6-31G*). The model systems were based on experimental crystal structures of two types of amyloids: (1) with aromatic amino acids, and (2) without aromatic amino acids. Data show that these two types of amyloids have similar interaction energies, supporting experimental findings that aromatic amino acids are not essential for amyloid formation. However, different factors contribute to the stability of these two types of amyloids. In the former, the presence of aromatic amino acids significantly contributes to the strength of interactions between side chains; interactions between aromatic and aliphatic side chains are the strongest, followed by aromatic-aromatic interactions, while aliphatic-aliphatic interactions are the weakest. In the latter, that is, the amyloids without aromatic residues, stability is provided by interactions of aliphatic side chains with the backbone and, in some cases, by hydrogen bonds.

Stacking of metal chelates with benzene: can dispersion-corrected DFT be used to calculate organic-inorganic stacking?

CCSD(T)/CBS energies for stacking of nickel and copper chelates are calculated and used as benchmark data for evaluating the performance of dispersion-corrected density functionals for calculating the interaction energies. The best functionals for modeling the stacking of benzene with the nickel chelate are M06HF-D3 with the def2-TZVP basis set, and B3LYP-D3 with either def2-TZVP or aug-cc-pVDZ basis set, whereas for copper chelate the PBE0-D3 with def2-TZVP basis set yielded the best results. M06L-D3 with aug-cc-pVDZ gives satisfying results for both chelates. Most of the tested dispersion-corrected density functionals do not reproduce the benchmark data for stacking of benzene with both nickel (no unpaired electrons) and copper chelate (one unpaired electron), whereas a number of these functionals perform well for interactions of organic molecules.

Stacking of benzene with metal chelates: calculated CCSD(T)/CBS interaction energies and potential-energy curves.

Accurate values for the energies of stacking interactions of nickel- and copper-based six-membered chelate rings with benzene are calculated at the CCSD(T)/CBS level. The results show that calculations made at the ωB97xD/def2-TZVP level are in excellent agreement with CCSD(T)/CBS values. The energies of [Cu(C3H3O2)(HCO2)] and [Ni(C3H3O2)(HCO2)] chelates stacking with benzene are -6.39 and -4.77 kcal mol(-1), respectively. Understanding these interactions might be important for materials with properties that are dependent on stacking interactions.

What are the preferred horizontal displacements of aromatic-aromatic interactions in proteins? Comparison with the calculated benzene-benzene potential energy surface.

The data from protein structures from the Protein Data Bank and quantum chemical calculations indicate the importance of aromatic-aromatic interactions at large horizontal displacements (offsets). The protein stacking interactions of the phenylalanine residue show preference for large offsets (3.5-5.0 Å), while the calculations show substantially strong interactions, of about -2.0 kcal mol(-1).

Stacking interactions of Ni(acac) chelates with benzene: calculated interaction energies.

Piling 'em up: The stacking energy of the [Ni(acac)2]/benzene system is calculated at local CCSD(T) level and is in good agreement with the values obtained with the SCS-MP2 method. Energies calculated with several DFT-D methods are somewhat overestimated. The calculated stacking energy of the [Ni(acac)2]/benzene system is significantly stronger than that of the benzene dimer.

Parallel interactions at large horizontal displacement in pyridine-pyridine and benzene-pyridine dimers.

A study of crystal structures from the Cambridge Structural Database (CSD) and DFT calculations reveals that parallel pyridine-pyridine and benzene-pyridine interactions at large horizontal displacements (offsets) can be important, similar to parallel benzene-benzene interactions. In the crystal structures from the CSD preferred parallel pyridine-pyridine interactions were observed at a large horizontal displacement (4.0-6.0 Å) and not at an offset of 1.5 Å with the lowest calculated energy. The calculated interaction energies for pyridine-pyridine and benzene-pyridine dimers at a large offset (4.5 Å) are about 2.2 and 2.1 kcal mol(-1), respectively. Substantial attraction at large offset values is a consequence of the balance between repulsion and dispersion. That is, dispersion at large offsets is reduced, however, repulsion is also reduced at large offsets, resulting in attractive interactions.