Characterization of Guest DNA Transport and Adsorption within Host Porous Protein Crystals.

Nucleic acid transport through protein-based pores is a well-characterized phenomenon due in part to advancements in nanopore sequencing. A less studied area is nucleic acid transport through extended protein-based channels, where the additional surface area and increased contact time allow for the study of prolonged binding interactions. Porous protein crystals composed of "CJ", a putative polyisoprenoid-binding protein from , represent a favorable, highly ordered material for studying DNA transport and binding/unbinding along protein-based channels. These crystals adopt a hexagonal prism habit and contain a densely packed hexagonal array of 13 nm diameter axial nanopores that run from the top to the bottom of the crystal. After cross-linking, the crystals are easily manipulated for experimentation. An adsorption isotherm between host crystals and guest double-stranded 8 base pair DNA (8mer) revealed a high equilibrium adsorption constant of 206 ± 30 L/g. Fluorescence confocal microscopy tracked the loading of guest DNA into host crystals predominately along the major axial crystal nanopores. Four different computational models based on the finite volume (FV) method were assessed to model the transport process for guest 8mer and 15mer dsDNA loading into empty host crystals in terms of fundamental parameters, such as the intrapore diffusion constant. Fitting the models to the data revealed that the most basic FV model sufficed to describe the observed loading behavior, characterized by a single effective diffusion coefficient. Leveraging Fick's first law, we more directly fit a numerical range for the observed intrapore diffusion coefficient as a function of time, position within the crystal, and relative guest concentration. This new transport analysis strategy was applied to both out-of-equilibrium loading and fluorescence recovery after photobleaching (FRAP) experiments. The intrapore diffusion constants are comparable between 8mer and 15mer dsDNA and were found to be 2 orders of magnitude faster for DNA loading into empty crystals than that observed in FRAP experiments, which averaged (10 ± 4) × 10 cm/s.