Influence ofpH on two-dimensional crystals of streptavidin

We observe two-dimensional (2D) protein crystals on lipid monolayers to study the interaction of protein molecules confined to a plane. This geometry bears close resemblance to actual structures in biological systems, as well as to fledgling protein-based devices envisioned for engineering applications. In addition, 2D protein crystals have direct utility for academic studies in structural biology, biophysics, and materials science. For these reasons, we seek to determine the important factors and principles governing the growth of these arrays.; Experimentally, we confine protein molecules to a plane, and permit them to self-organize. Under proper conditions, proteins organize into one of several arrays exactly one molecule thick (2D crystals) with very high degrees of positional and orientational order. We observe these crystals on a macroscopic level in situ using fluorescence microscopy, and also observe the crystals on a molecular level using electron microscopy.; The bacterial protein Streptavidin forms 2D crystals on a fluid lipid monolayer derivitized with the vitamin biotin. We use this system to investigate the effects of changing subphase pH on 2D crystals. Changing pH alters the protein-protein and protein-solvent interactions. We observe a variety of different pH-dependent macroscopic crystal shapes which correspond to different molecular arrangements of the protein. These different structures appear in quite well-defined pH regimes, and most of them may be reversibly transformed into each other through appropriate gradual pH changes.; These observations can be reconciled by a hypothesis of a free-energy surface that changes upon changing the pH of the system. Using molecular models, we observe that the different crystal lattices introduce different sets of contacts between neighboring molecules with pH-sensitive amino acid residues.; We complement these experiments with calculations of meanfield free energies of the observed lattices. Our model considers protein-protein interactions in manner motivated by protein folding simulations and considers protein-solvent interactions based on polymer physics. For both sets of interactions, all parameters are based on experiment. Using this model, we find agreement with experimental observations of microscopic crystal packing and macroscopic crystal shape for pH {dollar}>{dollar} 5, and the energy differences between the lattices are very small.