Improving protein expression, stability, and function with ProteinMPNN

Abstract

Natural proteins are highly optimized for function, but are often difficult to produce at a scale suitable for biotechnological applications due to poor expression in heterologous systems, limited solubility, and sensitivity to temperature. Thus, a general method that improves the physical properties of native proteins while maintaining function could have wide utility for protein-based technologies. Here we show that the deep neural network ProteinMPNN together with evolutionary and structural information provides a route to increasing protein expression, stability, and function. For both myoglobin and tobacco etch virus (TEV) protease, we generated designs with improved expression, elevated melting temperatures, and improved function. For TEV protease, we identified multiple designs with improved catalytic activity as compared to the parent sequence and previously reported TEV variants. Our approach should be broadly useful for improving the expression, stability, and function of biotechnologically important proteins.

Reengineering of a flavin‐binding fluorescent protein using ProteinMPNN

“…In a paper submitted after this one, Sumida and colleagues demonstrate that ProteinMPNN can be used to reengineer another ligand‐binding colored protein, human myoglobin, with a comparable success rate (Sumida et al, 2024). There, a more conservative cutoff of 7 Å for fixing the ligand‐proximal amino acids during reengineering was chosen, but given the larger size of myoglobin, resulting designs had 41%–55% sequence identity with the most similar protein in the UniRef100 database (Sumida et al, 2024). Several examples where protein‐ or peptide‐binding proteins (TEV protease, ubiquitin, ghrelin receptor) were reengineered using ProteinMPNN similarly display high success rates (de Haas et al, 2023; Goverde et al, 2023; Sumida et al, 2024).…”

“…There, a more conservative cutoff of 7 Å for fixing the ligand‐proximal amino acids during reengineering was chosen, but given the larger size of myoglobin, resulting designs had 41%–55% sequence identity with the most similar protein in the UniRef100 database (Sumida et al, 2024). Several examples where protein‐ or peptide‐binding proteins (TEV protease, ubiquitin, ghrelin receptor) were reengineered using ProteinMPNN similarly display high success rates (de Haas et al, 2023; Goverde et al, 2023; Sumida et al, 2024). Finally, new methods called LigandMPNN and CARBonAra were recently described that explicitly model non‐protein components, but their codes are not yet readily available (Dauparas et al, 2023; Krapp et al, 2023; Krishna et al, 2023).…”

Opportunities and Challenges for Machine Learning-Assisted Enzyme Engineering

“…Proteins that express well in a host organism for evolution are also preferred. Generative models have the potential to address this need for enzymes that are better starting points than natural enzymes: for example, ProteinMPNN was able to design wet-lab validated enzymes with higher expression and thermostability . With proper labels about enzyme activity on different substrates, generative design models could be conditioned to generate enzymes with several of these desirable attributes.…”

Thermal Stabilization of a Bacterial Zn(II)-Dependent Phospholipase C through Consensus Sequence Design

“…3−5 The sequences of myoglobin and TEV protease were recently redesigned using RoseTTAFold 6 and ProteinMPNN, 3 resulting in proteins that are substantially more stable and at the same time improving the expression and retaining or improving the activity. 7 The use of a consensus sequence was suggested initially by Steipe et al 8 and subsequently tested on several different proteins. 4,9−11 This strategy takes advantage of the information contained in multiple sequence alignments of homologous proteins.…”

“…Over the last decades, several strategies for the stabilization of protein folds were proposed, ranging from structure-based rational mutagenesis to random mutagenesis and selection, and more recently, machine learning approaches. − The sequences of myoglobin and TEV protease were recently redesigned using RoseTTAFold and ProteinMPNN, resulting in proteins that are substantially more stable and at the same time improving the expression and retaining or improving the activity …”