**4. Conclusions**

Fluorine is a common substituent in medicinal chemistry and is found in the structure of several currently available blockbuster drugs. This element influences many pharmacokinetic and pharmacodynamic properties of drugs, but its role in stabilizing ligand– biomolecule systems still remains unclear. In this study, we performed a statistical and theoretical analysis of HBs with fluorine found in the PDB database, focusing on the different HB donors (hydroxyl, amine, and methyl groups). The energy range of distinct HBs (i.e., <sup>F</sup>··· H–O, F··· H–N+, F··· H–N, F··· H–C) and optimal ranges of geometric parameters of HBs with fluorine were determined based on the selected PDB complexes.

The results of the analyses showed significant differences in the interaction of fluorine attached to an aliphatic carbon (Fal) and fluorine attached to an aromatic carbon (Far). The F··· H–O HBs with Far are more frequently formed with SER and THR, while those with Fal are formed by all amino acids with a polar hydroxyl group. Typically, F··· H–N HBs are formed with amino acids that have an amino group in their side chain (ARG, LYS, ASN, GLU). Hydrophobic amino acids most often form F··· H–C interactions, which suggests that fluorine prefers a hydrophobic environment in biological systems. It is worth noting that due to the three free electron pairs of fluorine, HBs are only influenced by the donor– acceptor distance and not by the angles. Although the three free electron pairs occupy the entire space around fluorine, F··· H–X HBs exhibit the characteristics of HBs, with exceeded standard angles. However, no significant differences were noted in the energies of HBs with fluorine depending on the donor type, which indicates that fluorine acts as a weak HB acceptor for all types of atoms. The optimal ranges of geometric parameters for HBs with fluorine were found to be 150◦–120◦ and 2.9–3.6 Å. For F··· N+ interactions, an HB distance shorter than 2.8 Å showed a destabilizing character in almost 70% of the cases.

It must be emphasized that all the analyzed crystal structures may not be crystallized at the lowest free energy form, and hence the observed interactions might not be in optimal geometries [47]. However, the results suggest that HBs with fluorine are forced to form, due to the stronger ligand–receptor neighboring interactions, which make fluorine the "donor's last resort" [48]. This is in line with Margareth Etter's rule that stronger HBs form first, and weaker donors and acceptors interact afterward [47]. All these findings suggest that fluorine does not form strong, stabilizing intermolecular interactions, and thus it seems that indirect influence of this element (electrostatic, inductive, and resonance effects) has a greater impact on the biological activity of compounds than his influences on the pharmacodynamics. The results of this study may contribute to a thorough understanding of hydrogen bonding with fluorine in biological systems which may serve to improve the tools currently available for the rational design of new fluorinated drugs.

**Supplementary Materials:** The following supporting information can be downloaded, Figure S1: Points of geometrical parameters of hydrogen bonds for which the energy value was calculated; Figure S2: Box plots showing the distribution of the different energy components: kinetic (red), electrostatic (cyan), Coulomb (blue), XC (olive), dispersion (magenta) energy for HB donors and fluorine attached to (A) an aliphatic fragment and (B) an aromatic ring. The energy values were calculated for specific geometric parameters using ETS-NOCV approach. Table S1: Correlation coefficient values calculated between results from every method; Table S2: Pearson test values calculated between results from every method.

**Author Contributions:** W.P.: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Writing—original draft, Visualization, Project administration; R.K. (Rafał Kurczab): Conceptualization, Resources, Writing—original draft, Supervision, Project administration; R.K. (Rafał Kafel): Software, Visualization; A.J.B.: Investigation, Writing—original draft, Funding acquisition. All authors have read and agreed to the published version of the manuscript.

**Funding:** The authors acknowledge the financial support from the National Science Centre, Poland (grant no. 2019/35/N/NZ7/04312) and the statutory funding from the Maj Institute of Pharmacology, Polish Academy of Sciences, Poland.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data were obtained from the PDB repository and the DrugCentral 2021 database (accessed 30 April 2021).

**Acknowledgments:** Numerical simulations were performed by PLGrid Infrastructure (Prometheus, ACC Cyfronet, AGH). WP acknowledges the support of InterDokMed project no. POWR.03.02.00-00- I013/16. The authors thank Justyna Kalinowska-Tłu´scik (Jagiellonian University), Mariusz Mitoraj (Jagiellonian University), and Filip Sagan (Jagiellonian University) for providing technical and substantive assistance in research.

**Conflicts of Interest:** The authors declare no conflict of interest.

**Sample Availability:** Samples of the compounds are not available from the authors.
