Deep penetration by substrates through the size‐restricted channels of an apo‐ferritin cage results in size‐selective olefin hydrogenation at the Pd nanocluster core (see picture). The encapsulated zero‐valent cluster is synthesized in situ by chemical reduction of PdII ions in the apo‐ferritin cage.

In many subfields of chemistry and physics, numerous attempts have been made to accelerate scientific discovery using data-driven experimental design algorithms. Among them, Bayesian optimization has been proven to be an effective tool. A standard implementation (e.g., scikit-learn), however, can accommodate only small training data. We designed an efficient protocol for Bayesian optimization that employs Thompson sampling, random feature maps, one-rank Cholesky update and automatic hyperparameter tuning, and implemented it as an open-source python library called COMBO (COMmon Bayesian Optimization library). Promising results using COMBO to determine the atomic structure of a crystalline interface are presented. COMBO is available at https: //github.com/tsudalab/combo.

Construction of artificial metalloenzymes is one of the most important subjects in bioinorganic chemistry, because metalloenzymes catalyze chemical transformations with high selectivity and reactivity under mild conditions. [1][2][3][4] There are several reports on protein design: introduction of metal binding sites, [1,3,[5][6][7][8] design of substrate binding cavities, [9][10][11][12] chemical modification of prosthetic groups, [13][14][15][16] and covalent attachment of metal cofactors. [2,[17][18][19][20] In particular, the covalent modification of proteins is a powerful tool for the generation of new metalloenzymes, while the efficiency of the modification is very much dependent on the position and reactivity of the cysteinyl thiol functional group.[2] Herein, we describe a novel strategy for the preparation of artificial metalloenzymes by noncovalent insertion of metal-complex catalysts into protein cavities. The resulting semisynthetic metalloenzymes, apo-myoglobin (apo-Mb) reconstituted with Cr III Schiff base complexes, are able to catalyze enantioselective sulfoxidation.Manganese(iii) and chromium(iii) Schiff base complexes are known to be catalysts for various oxidations in organic solvents. [21,22] Jacobsen, [23] and Katsuki [24]

Polymerization reactions of phenylacetylene derivatives are promoted by rhodium complexes within the discrete space of apo-ferritin in aqueous media. The catalytic reaction provides polymers with restricted molecular weight and a narrow molecular weight distribution. These results suggest that protein nanocages have potential for use as various reaction spaces through immobilization of metal catalysts on the interior surfaces of the protein cages.

We report the preparation of organometallic Pd(allyl) dinuclear complexes in protein cages of apo-Fr by reactions with [Pd(allyl)Cl]2 (allyl = eta3-C3H5). One of the dinuclear complexes is converted to a trinuclear complex by replacing a Pd-coordinated His residue to an Ala residue. These results suggest that multinuclear metal complexes with various coordination structures could be prepared by the deletion or introduction of His, Cys, and Glu at appropriate positions on protein surface.

New methods for the synthesis of artificial metalloenzymes are important for the construction of novel biocatalysts and biomaterials. Recently, we reported new methodology for the synthesis of artificial metalloenzymes by reconstituting apo-myoglobin with metal complexes (Ohashi, M. et al., Angew Chem., Int. Ed. 2003, 42, 1005-1008). However, it has been difficult to improve their reactivity, since their crystal structures were not available. In this article, we report the crystal structures of M(III)(Schiff base).apo-A71GMbs (M = Cr and Mn). The structures suggest that the position of the metal complex in apo-Mb is regulated by (i) noncovalent interaction between the ligand and surrounding peptides and (ii) the ligation of the metal ion to proximal histidine (His93). In addition, it is proposed that specific interactions of Ile107 with 3- and 3'-substituent groups on the salen ligand control the location of the Schiff base ligand in the active site. On the basis of these results, we have successfully controlled the enantioselectivity in the sulfoxidation of thioanisole by changing the size of substituents at the 3 and 3' positions. This is the first example of an enantioselective enzymatic reaction regulated by the design of metal complex in the protein active site.

Protein scaffolds provide unique metal coordination environments that promote biomineralization processes. It is expected that protein scaffolds can be developed to prepare inorganic nanomaterials with important biomedical and material applications. Despite many promising applications, it remains challenging to elucidate the detailed mechanisms of formation of metal nanoparticles in protein environments. In the present work, we describe a crystalline protein cage constructed by crosslinking treatment of a single crystal of apo-ferritin for structural characterization of the formation of sub-nanocluster with reduction reaction. The crystal structure analysis shows the gradual movement of the Au ions towards the centre of the three-fold symmetric channels of the protein cage to form a sub-nanocluster with accompanying significant conformational changes of the amino-acid residues bound to Au ions during the process. These results contribute to our understanding of metal core formation as well as interactions of the metal core with the protein environment.

Accumulation of metal ions on protein surfaces is an important subject in the field of materials science because these processes are applicable to the preparation of bioinspired inorganic materials. While previous studies related to this subject have focused on the preparation of nanomaterials using protein scaffolds, the detailed processes of metal ion deposition and metal core formation on a protein surface require clarification. Elucidation of the coordination structures of multinuclear metal binding sites on proteins at an early stage as well as intermediate and fully occupied stages of the metal ion deposition will help us to understand the reaction mechanisms so that desirable inorganic materials can be prepared using protein scaffolds. In this Article, we report on the detailed processes of accumulation of Pd(II) ions demonstrated by a series of X-ray crystal structural analyses of apo-ferritin (apo-Fr), an iron storage protein, containing different amounts of Pd(II) ions in the protein cage. We have identified the specific binding sites of Pd(II) ions and analyzed the dynamic changes in the coordination structure by a combination of the crystal structures and ICP quantitative analyses of apo-Fr containing low, intermediate, and high content of Pd(II) ions. Our studies on Pd(II).apo-Frs provide intriguing implications for the preparation of many other inorganic materials using protein surfaces.