De novo protein design-From new structures to programmable functions.

Methods from artificial intelligence (AI) trained on large datasets of sequences and structures can now "write" proteins with new shapes and molecular functions de novo, without starting from proteins found in nature. In this Perspective, I will discuss the state of the field of de novo protein design at the juncture of physics-based modeling approaches and AI. New protein folds and higher-order assemblies can be designed with considerable experimental success rates, and difficult problems requiring tunable control over protein conformations and precise shape complementarity for molecular recognition are coming into reach. Emerging approaches incorporate engineering principles-tunability, controllability, and modularity-into the design process from the beginning. Exciting frontiers lie in deconstructing cellular functions with de novo proteins and, conversely, constructing synthetic cellular signaling from the ground up. As methods improve, many more challenges are unsolved.

Modulation of FGF pathway signaling and vascular differentiation using designed oligomeric assemblies.

Many growth factors and cytokines signal by binding to the extracellular domains of their receptors and driving association and transphosphorylation of the receptor intracellular tyrosine kinase domains, initiating downstream signaling cascades. To enable systematic exploration of how receptor valency and geometry affect signaling outcomes, we designed cyclic homo-oligomers with up to 8 subunits using repeat protein building blocks that can be modularly extended. By incorporating a de novo-designed fibroblast growth factor receptor (FGFR)-binding module into these scaffolds, we generated a series of synthetic signaling ligands that exhibit potent valency- and geometry-dependent Ca2+ release and mitogen-activated protein kinase (MAPK) pathway activation. The high specificity of the designed agonists reveals distinct roles for two FGFR splice variants in driving arterial endothelium and perivascular cell fates during early vascular development. Our designed modular assemblies should be broadly useful for unraveling the complexities of signaling in key developmental transitions and for developing future therapeutic applications.

De novo design of modular protein hydrogels with programmable intra- and extracellular viscoelasticity.

Relating the macroscopic properties of protein-based materials to their underlying component microstructure is an outstanding challenge. Here, we exploit computational design to specify the size, flexibility, and valency of de novo protein building blocks, as well as the interaction dynamics between them, to investigate how molecular parameters govern the macroscopic viscoelasticity of the resultant protein hydrogels. We construct gel systems from pairs of symmetric protein homo-oligomers, each comprising 2, 5, 24, or 120 individual protein components, that are crosslinked either physically or covalently into idealized step-growth biopolymer networks. Through rheological assessment, we find that the covalent linkage of multifunctional precursors yields hydrogels whose viscoelasticity depends on the crosslink length between the constituent building blocks. In contrast, reversibly crosslinking the homo-oligomeric components with a computationally designed heterodimer results in viscoelastic biomaterials exhibiting fluid-like properties under rest and low shear, but solid-like behavior at higher frequencies. Exploiting the unique genetic encodability of these materials, we demonstrate the assembly of protein networks within living mammalian cells and show via fluorescence recovery after photobleaching (FRAP) that mechanical properties can be tuned intracellularly in a manner similar to formulations formed extracellularly. We anticipate that the ability to modularly construct and systematically program the viscoelastic properties of designer protein-based materials could have broad utility in biomedicine, with applications in tissue engineering, therapeutic delivery, and synthetic biology.

In silico evolution of autoinhibitory domains for a PD-L1 antagonist using deep learning models.

There has been considerable progress in the development of computational methods for designing protein-protein interactions, but engineering high-affinity binders without extensive screening and maturation remains challenging. Here, we test a protein design pipeline that uses iterative rounds of deep learning (DL)-based structure prediction (AlphaFold2) and sequence optimization (ProteinMPNN) to design autoinhibitory domains (AiDs) for a PD-L1 antagonist. With the goal of creating an anticancer agent that is inactive until reaching the tumor environment, we sought to create autoinhibited (or masked) forms of the PD-L1 antagonist that can be unmasked by tumor-enriched proteases. Twenty-three de novo designed AiDs, varying in length and topology, were fused to the antagonist with a protease-sensitive linker, and binding to PD-L1 was measured with and without protease treatment. Nine of the fusion proteins demonstrated conditional binding to PD-L1, and the top-performing AiDs were selected for further characterization as single-domain proteins. Without any experimental affinity maturation, four of the AiDs bind to the PD-L1 antagonist with equilibrium dissociation constants (KDs) below 150 nM, with the lowest KD equal to 0.9 nM. Our study demonstrates that DL-based protein modeling can be used to rapidly generate high-affinity protein binders.

De novo protein fold design through sequence-independent fragment assembly simulations.

De novo protein design generally consists of two steps, including structure and sequence design. Many protein design studies have focused on sequence design with scaffolds adapted from native structures in the PDB, which renders novel areas of protein structure and function space unexplored. We developed FoldDesign to create novel protein folds from specific secondary structure (SS) assignments through sequence-independent replica-exchange Monte Carlo (REMC) simulations. The method was tested on 354 non-redundant topologies, where FoldDesign consistently created stable structural folds, while recapitulating on average 87.7% of the SS elements. Meanwhile, the FoldDesign scaffolds had well-formed structures with buried residues and solvent-exposed areas closely matching their native counterparts. Despite the high fidelity to the input SS restraints and local structural characteristics of native proteins, a large portion of the designed scaffolds possessed global folds completely different from natural proteins in the PDB, highlighting the ability of FoldDesign to explore novel areas of protein fold space. Detailed data analyses revealed that the major contributions to the successful structure design lay in the optimal energy force field, which contains a balanced set of SS packing terms, and REMC simulations, which were coupled with multiple auxiliary movements to efficiently search the conformational space. Additionally, the ability to recognize and assemble uncommon super-SS geometries, rather than the unique arrangement of common SS motifs, was the key to generating novel folds. These results demonstrate a strong potential to explore both structural and functional spaces through computational design simulations that natural proteins have not reached through evolution.

Tpgen: a language model for stable protein design with a specific topology structure.

BACKGROUND: Natural proteins occupy a small portion of the protein sequence space, whereas artificial proteins can explore a wider range of possibilities within the sequence space. However, specific requirements may not be met when generating sequences blindly. Research indicates that small proteins have notable advantages, including high stability, accurate resolution prediction, and facile specificity modification. RESULTS: This study involves the construction of a neural network model named TopoProGenerator(TPGen) using a transformer decoder. The model is trained with sequences consisting of a maximum of 65 amino acids. The training process of TopoProGenerator incorporates reinforcement learning and adversarial learning, for fine-tuning. Additionally, it encompasses a stability predictive model trained with a dataset comprising over 200,000 sequences. The results demonstrate that TopoProGenerator is capable of designing stable small protein sequences with specified topology structures. CONCLUSION: TPGen has the ability to generate protein sequences that fold into the specified topology, and the pretraining and fine-tuning methods proposed in this study can serve as a framework for designing various types of proteins.

Alternative target recognition elements for chimeric antigen receptor (CAR) T cells: beyond standard antibody fragments.

BACKGROUND AIMS: Chimeric antigen receptor T (CAR-T) cells are a remarkably efficacious, highly promising and rapidly evolving strategy in the field of immuno-oncology. The precision of these targeted cellular therapies is driven by the specificity of the antigen recognition element (the "binder") encoded in the CAR. This binder redirects these immune effector cells precisely toward a defined antigen on the surface of cancer cells, leading to T-cell receptor-independent tumor lysis. Currently, for tumor targeting most CAR-T cells are designed using single-chain variable fragments (scFvs) derived from murine or human immunoglobulins. However, there are several emerging alternative binder modalities that are finding increasing utility for improved CAR function beyond scFvs. METHODS: Here we review the most recent developments in the use of non-canonical protein binding domains in CAR design, including nanobodies, DARPins, natural ligands, and de novo-designed protein elements. RESULTS: Overall, we describe how new protein binder formats, with their unique structural properties and mechanisms of action, may possess key advantages over traditional scFv CAR designs. CONCLUSIONS: These alternative binder designs may contribute to enhanced CAR-T therapeutic options and, ultimately, improved outcomes for cancer patients.

Transplanted Murine Tumours SPECT Imaging with 99mTc Delivered with an Artificial Recombinant Protein.

99mTc is a well-known radionuclide that is widely used and readily available for SPECT/CT (Single-Photon Emission Computed Tomography) diagnosis. However, commercial isotope carriers are not specific enough to tumours, rapidly clear from the bloodstream, and are not safe. To overcome these limitations, we suggest immunologically compatible recombinant proteins containing a combination of metal binding sites as 99mTc chelators and several different tumour-specific ligands for early detection of tumours. E1b protein containing metal-binding centres and tumour-specific ligands targeting integrin αvß3 and nucleolin, as well as a short Cys-rich sequence, was artificially constructed. It was produced in E. coli, purified by metal-chelate chromatography, and used to obtain a complex with 99mTc. This was administered intravenously to healthy Balb/C mice at an activity dose of about 80 MBq per mouse, and the biodistribution was studied by SPECT/CT for 24 h. Free sodium 99mTc-pertechnetate at the same dose was used as a reference. The selectivity of 99mTc-E1b and the kinetics of isotope retention in tumours were then investigated in experiments in C57Bl/6 and Balb/C mice with subcutaneously transplanted lung carcinoma (LLC) or mammary adenocarcinoma (Ca755, EMT6, or 4T1). The radionuclide distribution ratio in tumour and adjacent normal tissue (T/N) steadily increased over 24 h, reaching 15.7 ± 4.2 for EMT6, 16.5 ± 3.8 for Ca755, 6.7 ± 4.2 for LLC, and 7.5 ± 3.1 for 4T1.

Revolutionizing Molecular Design for Innovative Therapeutic Applications through Artificial Intelligence.

The field of computational protein engineering has been transformed by recent advancements in machine learning, artificial intelligence, and molecular modeling, enabling the design of proteins with unprecedented precision and functionality. Computational methods now play a crucial role in enhancing the stability, activity, and specificity of proteins for diverse applications in biotechnology and medicine. Techniques such as deep learning, reinforcement learning, and transfer learning have dramatically improved protein structure prediction, optimization of binding affinities, and enzyme design. These innovations have streamlined the process of protein engineering by allowing the rapid generation of targeted libraries, reducing experimental sampling, and enabling the rational design of proteins with tailored properties. Furthermore, the integration of computational approaches with high-throughput experimental techniques has facilitated the development of multifunctional proteins and novel therapeutics. However, challenges remain in bridging the gap between computational predictions and experimental validation and in addressing ethical concerns related to AI-driven protein design. This review provides a comprehensive overview of the current state and future directions of computational methods in protein engineering, emphasizing their transformative potential in creating next-generation biologics and advancing synthetic biology.

Modular repeat protein sculpting using rigid helical junctions.

The ability to precisely design large proteins with diverse shapes would enable applications ranging from the design of protein binders that wrap around their target to the positioning of multiple functional sites in specified orientations. We describe a protein backbone design method for generating a wide range of rigid fusions between helix-containing proteins and use it to design 75,000 structurally unique junctions between monomeric and homo-oligomeric de novo designed and ankyrin repeat proteins (RPs). Of the junction designs that were experimentally characterized, 82% have circular dichroism and solution small-angle X-ray scattering profiles consistent with the design models and are stable at 95 °C. Crystal structures of four designed junctions were in close agreement with the design models with rmsds ranging from 0.9 to 1.6 Å. Electron microscopic images of extended tetrameric structures and ∼10-nm-diameter "L" and "V" shapes generated using the junctions are close to the design models, demonstrating the control the rigid junctions provide for protein shape sculpting over multiple nanometer length scales.

Robust folding of a de novo designed ideal protein even with most of the core mutated to valine.

Protein design provides a stringent test for our understanding of protein folding. We previously described principles for designing ideal protein structures stabilized by consistent local and nonlocal interactions, based on a set of rules relating local backbone structures to tertiary packing motifs. The principles have made possible the design of protein structures having various topologies with high thermal stability. Whereas nonlocal interactions such as tight hydrophobic core packing have traditionally been considered to be crucial for protein folding and stability, the rules proposed by our previous studies suggest the importance of local backbone structures to protein folding. In this study, we investigated the robustness of folding of de novo designed proteins to the reduction of the hydrophobic core, by extensive mutation of large hydrophobic residues (Leu, Ile) to smaller ones (Val) for one of the designs. Surprisingly, even after 10 Leu and Ile residues were mutated to Val, this mutant with the core mostly filled with Val was found to not be in a molten globule state and fold into the same backbone structure as the original design, with high stability. These results indicate the importance of local backbone structures to the folding ability and high thermal stability of designed proteins and suggest a method for engineering thermally stabilized natural proteins.

A de novo peroxidase is also a promiscuous yet stereoselective carbene transferase.

By constructing an in vivo-assembled, catalytically proficient peroxidase, C45, we have recently demonstrated the catalytic potential of simple, de novo-designed heme proteins. Here, we show that C45's enzymatic activity extends to the efficient and stereoselective intermolecular transfer of carbenes to olefins, heterocycles, aldehydes, and amines. Not only is this a report of carbene transferase activity in a completely de novo protein, but also of enzyme-catalyzed ring expansion of aromatic heterocycles via carbene transfer by any enzyme.

Recent advances in de novo protein design: Principles, methods, and applications.

The computational de novo protein design is increasingly applied to address a number of key challenges in biomedicine and biological engineering. Successes in expanding applications are driven by advances in design principles and methods over several decades. Here, we review recent innovations in major aspects of the de novo protein design and include how these advances were informed by principles of protein architecture and interactions derived from the wealth of structures in the Protein Data Bank. We describe developments in de novo generation of designable backbone structures, optimization of sequences, design scoring functions, and the design of the function. The advances not only highlight design goals reachable now but also point to the challenges and opportunities for the future of the field.

De Novo Peptide and Protein Design Using Generative Adversarial Networks: An Update.

Nowadays, machine learning and deep learning approaches are widely utilized for generative chemistry and computer-aided drug design and discovery such as de novo peptide and protein design, where target-specific peptide-based/protein-based therapeutics have been suggested to cause fewer adverse effects than the traditional small-molecule drugs. In light of current advancements in deep learning techniques, generative adversarial network (GAN) algorithms are being leveraged to a wide variety of applications in the process of generative chemistry and computer-aided drug design and discovery. In this review, we focus on the up-to-date developments for de novo peptide and protein design research using GAN algorithms in the interdisciplinary fields of generative chemistry, machine learning, deep learning, and computer-aided drug design and discovery. First, we present various studies that investigate GAN algorithms to fulfill the task of de novo peptide and protein design in the drug development pipeline. In addition, we summarize the drawbacks with respect to the previous studies in de novo peptide and protein design using GAN algorithms. Finally, we depict a discussion of open challenges and emerging problems for future research.

Perturbing the energy landscape for improved packing during computational protein design.

The FastDesign protocol in the molecular modeling program Rosetta iterates between sequence optimization and structure refinement to stabilize de novo designed protein structures and complexes. FastDesign has been used previously to design novel protein folds and assemblies with important applications in research and medicine. To promote sampling of alternative conformations and sequences, FastDesign includes stages where the energy landscape is smoothened by reducing repulsive forces. Here, we discover that this process disfavors larger amino acids in the protein core because the protein compresses in the early stages of refinement. By testing alternative ramping strategies for the repulsive weight, we arrive at a scheme that produces lower energy designs with more native-like sequence composition in the protein core. We further validate the protocol by designing and experimentally characterizing over 4000 proteins and show that the new protocol produces higher stability proteins.

Simple yet functional phosphate-loop proteins.

Abundant and essential motifs, such as phosphate-binding loops (P-loops), are presumed to be the seeds of modern enzymes. The Walker-A P-loop is absolutely essential in modern NTPase enzymes, in mediating binding, and transfer of the terminal phosphate groups of NTPs. However, NTPase function depends on many additional active-site residues placed throughout the protein's scaffold. Can motifs such as P-loops confer function in a simpler context? We applied a phylogenetic analysis that yielded a sequence logo of the putative ancestral Walker-A P-loop element: a ß-strand connected to an α-helix via the P-loop. Computational design incorporated this element into de novo designed ß-α repeat proteins with relatively few sequence modifications. We obtained soluble, stable proteins that unlike modern P-loop NTPases bound ATP in a magnesium-independent manner. Foremost, these simple P-loop proteins avidly bound polynucleotides, RNA, and single-strand DNA, and mutations in the P-loop's key residues abolished binding. Binding appears to be facilitated by the structural plasticity of these proteins, including quaternary structure polymorphism that promotes a combined action of multiple P-loops. Accordingly, oligomerization enabled a 55-aa protein carrying a single P-loop to confer avid polynucleotide binding. Overall, our results show that the P-loop Walker-A motif can be implemented in small and simple ß-α repeat proteins, primarily as a polynucleotide binding motif.

Designing protein structures and complexes with the molecular modeling program Rosetta.

Proteins perform an amazingly diverse set of functions in all aspects of life. Critical to the function of many proteins are the highly specific three-dimensional structures they adopt. For this reason, there is strong interest in learning how to rationally design proteins that adopt user-defined structures. Over the last 25 years, there has been significant progress in the field of computational protein design as rotamer-based sequence optimization protocols have enabled accurate design of protein tertiary and quaternary structure. In this award article, I will summarize how the molecular modeling program Rosetta is used to design new protein structures and describe how we have taken advantage of this capability to create proteins that have important applications in research and medicine. I will highlight three protein design stories: the use of protein interface design to create therapeutic bispecific antibodies, the engineering of light-inducible proteins that can be used to recruit proteins to specific locations in the cell, and the de novo design of new protein structures from pieces of naturally occurring proteins.

Function2Form Bridge-Toward synthetic protein holistic performance prediction.

Protein engineering and synthetic biology stand to benefit immensely from recent advances in silico tools for structural and functional analyses of proteins. In the context of designing novel proteins, current in silico tools inform the user on individual parameters of a query protein, with output scores/metrics unique to each parameter. In reality, proteins feature multiple "parts"/functions and modification of a protein aimed at altering a given part, typically has collateral impact on other protein parts. A system for prediction of the combined effect of design parameters on the overall performance of the final protein does not exist. Function2Form Bridge (F2F-Bridge) attempts to address this by combining the scores of different design parameters pertaining to the protein being analyzed into a single easily interpreted output describing overall performance. The strategy comprises of (a) a mathematical strategy combining data from a myriad of in silico tools into an OP-score (a singular score informing on a user-defined overall performance) and (b) the F2F Plot, a graphical means of informing the wetlab biologist holistically on designed construct suitability in the context of multiple parameters, highlighting scope for improvement. F2F predictive output was compared with wetlab data from a range of synthetic proteins designed, built, and tested for this study. Statistical/machine learning approaches for predicting overall performance, for use alongside the F2F plot, were also examined. Comparisons between wetlab performance and F2F predictions demonstrated close and reliable correlations. This user-friendly strategy represents a pivotal enabler in increasing the accessibility of synthetic protein building and de novo protein design.

Chemoselective N-H insertion catalyzed by a de novo carbene transferase.

The ability to perform organic reactions with chemoselectivity is of critical importance in synthesis. Recently, we reported that a de novo carbene transferase, a tetra-α-helical c-type heme-containing protein, C45, is proficient at N-H insertion reactions, proceeding via the intermolecular transfer of a metallocarbenoid intermediate into the N-H σ-bond to form a new N-C σ-bond. Here we demonstrate that C45 can also catalyse N-H insertion reactions chemoselectively, even when the substrate contains an unprotected hydroxyl group.

A protein constructed de novo enables cell growth by altering gene regulation.

Recent advances in protein design rely on rational and computational approaches to create novel sequences that fold and function. In contrast, natural systems selected functional proteins without any design a priori. In an attempt to mimic nature, we used large libraries of novel sequences and selected for functional proteins that rescue Escherichia coli cells in which a conditionally essential gene has been deleted. In this way, the de novo protein SynSerB3 was selected as a rescuer of cells in which serB, which encodes phosphoserine phosphatase, an enzyme essential for serine biosynthesis, was deleted. However, SynSerB3 does not rescue the deleted activity by catalyzing hydrolysis of phosphoserine. Instead, SynSerB3 up-regulates hisB, a gene encoding histidinol phosphate phosphatase. This endogenous E. coli phosphatase has promiscuous activity that, when overexpressed, compensates for the deletion of phosphoserine phosphatase. Thus, the de novo protein SynSerB3 rescues the deletion of serB by altering the natural regulation of the His operon.