Balaz Lab

Research Overview

The overall, long-term goal is to contribute to the understanding of processes affecting the fates and effects of drugs in the body, in terms of structures and properties of drugs and body components, by a concerted deployment of bench-top and computational techniques. The approaches can be divided into two main streams: Subcellular Pharmacokinetics and Drug-Receptor Interactions, which are integrated into Quantitative Structure-Time-Activity Relationships. QSTARs are comprehensive descriptions of pharmacokinetics and pharmacodynamics for a set of drugs. The approaches represent a basis for rational drug discovery and development. The streams are broadly described below. Individual research themes are listed here:

Drug-phospholipid Interactions Slow Inhibition Kinetics of Metzincins
Subcellular Pharmacokinetics Receptor-based Affinity Predictions
Similarity of Metzincins QM/MM Prediction of Metzincin Affinities
Prediction of Distribution Volume Multi-species, Multi-mode Binding


Subcellular Pharmacokinetics aims at a description of the rate and extent of drug disposition in the body. Most drugs cross the majority of membranes in the body by passive diffusion through phospholipid bilayers. This simple process is still not understood to a level allowing accurate and reliable structure-based predictions.           

Trans-bilayer transport and accumulation are studied in experimental systems like immobilized phospholipid monolayers, liposomes, cells, and tissues using isothermal titration calorimetry, fluorescence spectroscopy, confocal fluorescence microscopy, and conventional analytical techniques (HPLC, GC-MS). The analysis is using drug solvation data in the surrogate phase for headgroups, the hydrated di-acetyl phosphatidylcholine. 

The results are analyzed by fitting conceptual models to the experimental data. The obtained transport (micro)parameters are related to drug and bilayer structures and properties utilizing computational techniques like molecular dynamics at the atomic and coarse-grain levels. Finally, the data characterizing one monolayer or bilayer are used to simulate drug distribution in more complex systems (cells, tissues, etc.).

The bilayer-based models describe well passive trans-cellular absorption and, after integration with protein binding, also distribution. This framework can be extended by adding excretion and metabolism, The resulting models of Subcellular Pharmacokinetics are comprehensive kinetic descriptions of drug fates in the body.

Prediction of Distribution Volume is based on the in vitro measurement and prediction of binding to the preponderant body constituents: phospholipids, triglycerides, albumin, extracellular matrix, and actin. Binding to proteins cannot be approximated by physicochemical properties of chemicals and 3D-QSAR models are necessary.

Drug-Receptor Interactions are the key steps in drug action and need to be estimated as precisely as possible. We use and develop both ligand-based and structure-based computational approaches.

The lab pioneered the conceptual use of multiple bound ligand orientations or conformations (binding modes) in the receptor site modeling. The classical approaches suffer from the need to define one active mode, in which all studied molecules are to be aligned. Multi-mode approach allows the user to input several modes and the procedure objectively selects the best mode(s). Our approach provides a straightforward treatment of multiple drug species, such as protomers and tautomers, interacting with the receptor. The multi-species, multi-mode approach is being implemented in Comparative Molecular Field Analysis (CoMFA) that represents the most frequently used ligand-based (3D-QSAR) method.

The energies of binding to receptors of known structures are computationally estimated using molecular dynamics-based methods.  We improved the Linear Response method by replacing the ensemble averages of van der Waals and electrostatic interactions by the QM/MM energy of time-averaged structures from the simulations.

The drug-receptor interactions are also studied experimentally, using state-of-the-art spectroscopic and calorimetric techniques. We are focusing on matrix metalloproteinases (MMPs) and other metzincins because they are implicated in diseases like cancer metastasis, arthritis, and neurological disorders. The inhibitors, which are used to treat aberrant activities of these enzymes, frequently need up to several hours to exert full, steady-state effects. This period can be prohibitively long in vivo. We analyze the slow metzincin inhibition and strive to identify the rate-limiting step.