The chemistry of highly evolved protein-based compartments has inspired the design of new catalytically active materials that self-assemble from biological components. A frontier of this biodesign is the potential to contribute new catalytic systems for the production of sustainable fuels, such as hydrogen. Here, we show the encapsulation and protection of an active hydrogen-producing and oxygen-tolerant [NiFe]-hydrogenase, sequestered within the capsid of the bacteriophage P22 through directed self-assembly. We co-opted Escherichia coli for biomolecular synthesis and assembly of this nanomaterial by expressing and maturing the EcHyd-1 hydrogenase prior to expression of the P22 coat protein, which subsequently self assembles. By probing the infrared spectroscopic signatures and catalytic activity of the engineered material, we demonstrate that the capsid provides stability and protection to the hydrogenase cargo. These results illustrate how combining biological function with directed supramolecular self-assembly can be used to create new materials for sustainable catalysis.

Nucleic acids are the natural cargo of viruses and key determinants that affect viral shell stability. In some cases the genome structurally reinforces the shell, whereas in others genome packaging causes internal pressure that can induce destabilization. Although it is possible to pack heterologous cargoes inside virus-derived shells, little is known about the physical determinants of these artificial nanocontainers’ stability. Atomic force and three-dimensional cryo-electron microscopy provided mechanical and structural information about the physical mechanisms of viral cage stabilization beyond the mere presence/absence of cargos. We analyzed the effects of cargo–shell and cargo–cargo interactions on shell stability after encapsulating two types of proteinaceous payloads. While bound cargo to the inner capsid surface mechanically reinforced the capsid in a structural manner, unbound cargo diffusing freely within the shell cavity pressurized the cages up to ~30 atm due to steric effects. Strong cargo–cargo coupling reduces the resilience of these nanocompartments in ~20% when bound to the shell. Understanding the stability of artificially loaded nanocages will help to design more robust and durable molecular nanocontainers.

The time-dependent localization of the metal-to-ligand charge transfer (MLCT) excited states of ruthenium(II) complexes containing 2,2'-bipyridine (bpy) and 1,10-phenanthroline (phen) ligands was studied by femtosecond transient absorption spectroscopy. Time-resolved anisotropy measurements indicate that the excited state hops randomly among the three ligands of each complex by subpicosecond interligand electron transfer (ILET). Although the bpy- and phen-localized (3)MLCT states have similar energies and steady-state emission spectra, pronounced differences in their excited-state absorption spectra make it possible to observe changes in excited state populations using magic angle transient absorption measurements. Analysis of the magic angle signals shows that the excited electron is equally likely to be found on any of the three ligands approximately 1 ps after excitation, but this statistical distribution subsequently evolves to a Boltzmann distribution with a time constant of approximately 10 ps. The apparent contradiction between ultrafast ILET revealed by time-dependent anisotropy measurements and the slower ILET seen in magic angle measurements on the tens of picoseconds time scale is explained by a model in which the underlying rates depend dynamically on excess vibrational energy. The insight that ILET can occur over multiple time scales reconciles contradictory literature observations and may lead to improved photosensitizer performance.

Viral protein cages, with their regular and programmable architectures, are excellent platforms for the development of functional nanomaterials. The ability to transform a virus into a material with intended structure and function relies on the existence of a well-understood model system, a noninfectious virus-like particle (VLP) counterpart. Here, we study the factors important to the ability of P22 VLP to retain or release various protein cargo molecules depending on the nature of the cargo, the capsid morphology, and the environmental conditions. Because the interaction between the internalized scaffold protein (SP) and the capsid coat protein (CP) is noncovalent, we have studied the efficiency with which a range of SP variants can dissociate from the interior of different P22 VLP morphologies and exit by traversing the porous capsid. Understanding the types of cargos that are either retained or released from the P22 VLP will aid in the rational design of functional nanomaterials.