Stochastic dynamics simulations have been performed to investigate macromolecular diffusion in the cytoplasm of the bacterium Escherichia coli. The cytoplasm was modeled as a polydisperse mixture of three different types of spherical particles representing ribosomes, proteins, and tRNA molecules at their physiological concentrations, The interactions between the particles consisted of short-range Lennard-Jones and long-range electrostatic terms, A number of properties, including the pair distributions functions, the static structure factor, and the long-time self-diffusion coefficients, were obtained for various model systems differing from each other by the size of the charges on the particles. The effect of macromolecular crowding on these properties is examined, and their implication for the dynamics of macromolecules within living cells is discussed.