From X-ray crystallographic structure to intrinsic thermodynamics of protein–ligand binding using carbonic anhydrase isozymes as a model system
Vaida Paketuryte-Latve, Alexey Smirnov, Elena Manakova, Lina Baranauskiene, Vytautas Petrauskas, Asta Zubriene, Jurgita Matuliene, Virginija Dudutiene, Edita Capkauskaite, Audrius Zaksauskas, Janis Leitans, Saulius Grazulis, Kaspars
Tars and Daumantas Matulis [1]
Molecular Tour
Rational design of novel pharmaceutical molecules requires deep understanding of how small molecules recognize and bind to the disease target protein molecule. Unfortunately, the mechanism of recognition is still quite poorly understood. Experimentally, the binding reaction can be described structurally and energetically. The X-ray crystallographic structure of the drug candidate molecule bound to the protein molecule provides detailed information on the structural arrangement and the interface between the compound and the protein. However, the structural information does not yield a clue on the strength and affinity, thus the energy of the interaction. Full energetic description should include the measurement of the thermodynamic binding parameters, including the Gibbs energy, enthalpy, entropy, heat capacity, compressibility, and several others and the kinetic parameters, especially the dissociation rate constant or residence time. When the reaction is fully characterized both structurally and energetically, we can begin studying the principles of structure – thermodynamics correlations.
However, both the structural and energetic characterization may succumb to numerous pitfalls. The structure of the protein – ligand complex may be determined by numerous structural methods, such as X-ray crystallography, NMR, and Cryo-EM. The conditions of these experiments may vary a lot, the temperature may be 100K, much lower than the physiological temperature. Dynamic information may be more accessible in NMR than in other methods. Limited resolution may also be insufficient to determine exact positions of all atoms including hydrogens. There are always limitations of what the structural picture can provide. On the other hand, energetic description is also highly complex. There are numerous techniques to characterize the affinity and other thermodynamic parameters of binding. Each technique has limitations as described in this manuscript. However, here we especially emphasize the intrinsic energetics of binding. Numerous protein – ligand binding reactions are linked to protonation reactions that contribute significant energies and without the dissection of such energies it is impossible to draw proper structure – thermodynamics correlations.
In this manuscript we illustrate how detailed structural and thermodynamic characterization, especially taking into account the intrinsic parameters, helped to design high-affinity and high-specificity chemical compounds that would bind carbonic anhydrase IX (CAIX), a protein that is highly overexpressed in most solid hypoxic tumors. The protein participates in the acidification of the tumor microenvironment, helps promote invasion and metastasis processes in cancer. Thus, a possible anticancer strategy could involve inhibition of the protein by inhibitors that would not bind to any other proteins and thus not cause possible toxic side effects. We synthesized over 1000 molecules and demonstrated chemical structure features of a compound to exhibit high affinity for CAIX and low affinity for remaining 11 catalytically active vital human carbonic anhydrase isozymes. The compounds were arranged into a database (plbd.org) to help researchers apply AI approaches and study the structure – thermodynamics correlations for rational drug design.
References
- ↑ Paketurytė-Latvė V, Smirnov A, Manakova E, Baranauskiene L, Petrauskas V, Zubrienė A, Matulienė J, Dudutienė V, Čapkauskaitė E, Zakšauskas A, Leitans J, Gražulis S, Tars K, Matulis D. From X-ray crystallographic structure to intrinsic thermodynamics of protein-ligand binding using carbonic anhydrase isozymes as a model system. IUCrJ. 2024 Jul 1;11(Pt 4):556-569. PMID:38856178 doi:10.1107/S2052252524004627
