Abstract
Biological environments are typically referred to as "crowded" due to presence of distinct biomolecules, such as proteins, nucleic acids, metabolites at very high concentrations (300-400 g/L). Such high biomolecular concentrations restrict the accuracy of experimental and atomistic simulation studies, which do not take the interactions with these environments explicitly into account. Computer simulation approaches, such as rigid-body Brownian Dynamics and
Monte Carlo simulations have emerged as promising modeling tools to study impacts of macromolecular crowding at a microscopic level. In this thesis, we have described the impact of macromolecular crowding on protein-protein interactions, relative stabilizations and destabilizations of fexible polymer conformations in self crowding conditions and in presence of protein crowders. For this purpose, we employed a multiple conformation Monte Carlo approach previously introduced by Prytkova et al. [1] The accuracy of rigid-body simulations are impeded due to the missing description of intramolecular degrees of freedom or conformational exibilities of a molecule. In the mcMC method
conformational exibilities are incorporated by a discrete, finite-sized structural ensemble containing distinct conformations of a molecule, which are sampled from extensive MD simulation of the molecule of interest. The simulated molecules are then able to switch between conformations in this ensemble. We follow previous work by Prytkova et al. and incorporate conformational flexibilities in many-protein Monte Carlo simulations that essentially alleviates the structural bias, which is otherwise imposed due to rigid-description of a molecule. We have chosen solutions of Hen-Egg white lysozyme protein as a test system to validate and benchmark the MC simulation protocol. Distinct conformations of HEWL are sampled from extensive Replica exchange MD simulations of a single protein in water and employed in mcMC simulations for mimicking conformational flexibilities of HEWL protein molecule. We validate simulations of protein solutions by comparing with available experimental data and investigate the effects of: 1) Extent of REMD conformational sampling;
2) size of the conformational ensemble used to describe intramolecular
degrees of freedom; 3) empirical parameters in the protein-protein interaction potentials energies.
We have employed the mcMC method to study the influence of macromolecular crowding on the conformational equilibrium of 10-kDa polyethylene glycol (PEG) polymers in self-crowding conditions. We described the conformational flexibility of PEG polymers by a structural ensemble containing a finite number of PEG conformations of varying radii of gyration, describing a wide-size distribution of PEG structures from very compact to highly extended structures. We have shown that the impact of varying self-crowding conditions resulted from distinct intermolecular interactions on conformational free energy changes of PEG polymer structures can be qualitatively explained
by mcMC simulation method. Furthermore, we described the compensating effects of repulsive volume exclusion and non-specific attractive intermolecular interactions by modifying intermolecular interaction energies, which stabilizes compact and extended PEG conformations, respectively. We have further proposed
an analytical model derived from our simulation results and described
the relative stabilities of PEG conformations at distinct intermolecular interactions, as a function of radius of gyration and solvent accessible surface of area of PEG structures. The proposed model allows us to further extrapolate to situations that are not easily accessible by mcMC simulation method. We successfully described the impact of solvent accessible surfaces on non-specific attractive interactions, which counteracts repulsive excluded volume effect.
Following the modeling of self-crowding conditions for PEG, we further studied crowding effects on PEG polymer structures in presence of rigid protein crowder (HEWL and BSA) solutions. We have extended the mcMC simulation method to allow for simulations of a single
exible PEG conformation embedded in protein solutions. We have varied intermolecular interactions between the crowders and the PEG polymer to obtain distinct crowding conditions.
We have shown that non-specific attractive interactions between crowders and PEG counteract the repulsive excluded volume effect also in this class of systems, while the interactions between the crowder molecules play an equally important role. We have explored crowder size dependence of overall crowding intensity by using a small (HEWL) and large (BSA) protein crowder. We further defined a relative crowding intensity parameter to quantify crowding intensities obtained by varying intermolecular interactions.
