Despite successes in replicating the primary–secondary–tertiary structure hierarchy of protein, it remains elusive to synthetically materialize protein functions that are deeply rooted in their chemical, structural and dynamic heterogeneities1–12. We propose that for polymers with backbone chemistries different from that of proteins, programming spatial and temporal projections of sidechains at the segmental level can be effective in replicating protein behaviours13,14; and leveraging the rotational freedom of polymer can mitigate deficiencies in monomeric sequence specificity and achieve behaviour uniformity at the ensemble level2,3,15–20. Here, guided by the active site analysis of about 1,300 metalloproteins, we design random heteropolymers (RHPs) as enzyme mimics based on one-pot synthesis. We introduce key monomers as the equivalents of the functional residues of protein and statistically modulate the chemical characteristics of key monomer-containing segments, such as segmental hydrophobicity21. The resultant RHPs form pseudo-active sites that provide key monomers with protein-like microenvironments, co-localize substrates with catalytic or cofactor-binding sidechains and catalyse reactions such as oxidation and cyclization of citronellal with isopulegol/menthoglycol selectivity. This RHP design led to enzyme-like materials that can retain catalytic activity under non-biological conditions, are compatible with scalable processing and have expanded substrate scope, including environmentally long-lasting antibiotic tetracycline22.
