INTERMOLECULAR INTERACTION PREDICTION OF DRUG MIXTURE: COMPUTATIONAL STUDY OF SALICYLIC ACID AND PHENYLACETIC ACID COCRYSTAL MODEL

• Main Author: Perdana Kusuma, Aris
• Format: Dissertations
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/74733
• Institution: Institut Teknologi Bandung

Pharmaceutical interactions can affect dosage forms. These interactions can lead to precipitation, eutectic formation, and degradation. Such interactions occur due to changes in the physicochemical properties of the drug. One type of this interaction is the formation of cocrystals. Cocrystals are multicomponent crystals composed of two or more different molecules connected by non-covalent bonds and forming a solid system at room temperature. The packing of the newly formed cocrystal is different from that of the single compound. The formation of cocrystals can be observed through laboratory tests with polarizing microscope instruments, DSC and XRD. However, the COVID-19 pandemic has changed some perspectives regarding non-essential experimental studies, making it difficult to carry out laboratory work. Therefore, computational prediction is expected to bridge this gap. With computational prediction, the negative effects of mixing two drugs that have the potential to form cocrystals can be anticipated earlier. In addition, the selection of cocrystal coformer screening can still be carried out despite the limitations of laboratory work. Computational predictions can be applied to various target of cocrystal parts, one of which is the synthon part. Regarding synthons, one of the most interesting ones is the carboxylic acid-pyridine heterosynthon (CPHS). CPHS is one of the strongest types of synthon and is found in many cocrystal structures. This synthon is composed of a hydrogen atom of the carboxylic group bonded to the nitrogen atom of the pyridine group. Due to its frequent occurrence, this synthon may act as a precursor for cocrystal formation. The synthon may have existed since the beginning of the cocrystal formation process and will continue to exist until the final crystal packing is formed. Although there have been many studies regarding synthons, there is no consensus on which factors play the most significant role. Related to this, the computation needs to be carried out at the dimer conformation level. In mixing two molecules, there are several existing and non-existing/putative conformations. The existing conformation is a conformation that is actually formed in cocrystal packing, one of which contains CPHS. Meanwhile, putative conformation is a conformation that is not formed in cocrystal packing. iv The purpose of this study was to determine the synthon formation factors in cocrystals. In addition, this research is aimed at obtaining a computational method that can predict the occurrence of a cocrystal. In this study, the cocrystal models of salicylic acid (SAC) - nicotinamide (NIC) {SACNIC)}, salicylic acid - isonicotinamide (INA) {SACINA}, and phenylacetic acid (PYC) - nicotinamide {PYCNIC} were used. All three contain small molecules that have hydrogen bond donors and acceptors. The SACNIC, SACINA, and PYCNIC have CPHS. Computational analysis was performed using density functional theory (DFT) at the B3LYP-D3BJ and WB97M-D3BJ levels with a basis set of 6-311G (d,p) in the existing and putative conformations. This theoretical level and basis set can be used for conformational analysis of dimers in crystals and cocrystals. The descriptors are geometric shape, total energy, interaction energy, single hydrogen bond energy, gap of highest occupied molecular orbital (HOMO) - lowest unoccupied molecular orbital (LUMO), Laplacian bond order (LBO), and natural bond orbital (NBO). Next, computational techniques along with appropriate descriptors will be applied to several existing cocrystals to determine the extent of the method's predictive ability. The computational results show that the conformation with CPHS appears to have an advantage in the single hydrogen bond energy descriptor, compared to other conformations, both existing and putative. Given that this type of synthon is almost always present in cocrystal pairs that have carboxylic groups and pyridine groups, this energy descriptor can be used as one of the predictive parameters for the occurrence of cocrystals containing CPHS.