In the past decade, artificial metalloenzymes (AMs) have emerged as an attractive alternative to the more traditional enzymes and homogeneous catalysts. Such hybrid catalysts result from the incorporation of an abiotic metal cofactor within a macromolecule (protein or oligonucleotide). Artificial metalloenzymes combine attractive features of both homogeneous catalysts and enzymes, including the possibility to genetically optimize the catalytic performance of new-to-nature organometallic reactions. Can artificial metalloenzymes become as catalytically efficient as naturally-evolved metalloenzymes, even in complex biological mixtures? Herein, we outline our efforts to address this challenge by localizing and evolving AMs within the periplasm of Escherichia coli. To achieve this objective, we will exploit AMs based on the biotin-streptavidin technology. Four subprojects have been tailored to address the challenges: i) knock-out deleterious components present in the periplasm; ii) improve the cofactor uptake through the outer-membrane; iii) engineer streptavidin to boost the AM’s performance; and iv) rely both on screening and selection strategies to evolve AMs in vivo. Relying on auxotrophs, we will demonstrate the potential of AMs to complement metabolic pathways. Only E. coli auxotrophs containing an evolved AM capable of producing the vital aminoacid-precursor will survive the stringent selection pressure. We have identified several selectable aminoacid precursors which can be produced by metathesis (indole, precursor of tryptophan), enone reduction (keto valine, precursor of valine) and allylic substitution (prephenate, precursor of tyrosine and phenylalanine). In a Darwinian evolution spirit, we anticipate that applying selection pressure will allow to evolve AMs to unprecedented catalytic performance.The main deliverable of the DrEAM is an engineered and evolvable E. coli strain capable of performing in vivo reaction cascades combining AMs and natural enzymes.