Deep Mutational Scanning Research Articles

High throughput profiling identified PA-L106R amino acid substitution in A(H1N1)pdm09 influenza virus that confers reduced susceptibility to baloxavir in vitro

Baloxavir acid (BXA) is a pan-influenza antiviral that targets the cap-dependent endonuclease of the polymerase acidic (PA) protein required for viral mRNA synthesis. To gain a comprehensive understanding on the molecular changes associated with reduced susceptibility to BXA and their fitness profile, we performed a deep mutational scanning at the PA endonuclease domain of an A (H1N1)pdm09 virus. The recombinant virus libraries were serially passaged in vitro under increasing concentrations of BXA followed by next-generation sequencing to monitor PA amino acid substitutions with increased detection frequencies. Enriched PA amino acid changes were each introduced into a recombinant A (H1N1)pdm09 virus to validate their effect on BXA susceptibility and viral replication fitness in vitro. The I38 T/M substitutions known to confer reduced susceptibility to BXA were invariably detected from recombinant virus libraries within 5 serial passages. In addition, we identified a novel L106R substitution that emerged in the third passage and conferred greater than 10-fold reduced susceptibility to BXA. PA-L106 is highly conserved among seasonal influenza A and B viruses. Compared to the wild-type virus, the L106R substitution resulted in reduced polymerase activity and a minor reduction of the peak viral load, suggesting the amino acid change may result in moderate fitness loss. Our results support the use of deep mutational scanning as a practical tool to elucidate genotype-phenotype relationships, including mapping amino acid substitutions with reduced susceptibility to antivirals.

Deep mutational scanning reveals sequence to function constraints for SWEET family transporters.

Protein science is entering a transformative phase enabled by deep mutational scans that provide an unbiased view of the residue level interactions that mediate function. However, it has yet to be extensively used to characterize the mutational and evolutionary landscapes of plant proteins. Here, we apply the method to explore sequence-function relationships within the sugar transporter AtSWEET13. DMS results describe how mutational interrogation throughout different regions of the protein affects AtSWEET13 abundance and transport function. Our results identify novel transport-enhancing mutations that are validated using the FRET sensor assays. Extending DMS results to phylogenetic analyses reveal the role of transmembrane helix 4 (TM4) which makes the SWEET family transporters distinct from prokaryotic SemiSWEETs. We show that transmembrane helix 4 is intolerant to motif swapping with other clade-specific SWEET TM4 compositions, despite accommodating single point-mutations towards aromatic and charged polar amino acids. We further show that the transfer learning approaches based on physics and ML based In silico variant prediction tools have limited utility for engineering plant proteins as they were unable to reproduce our experimental results. We conclude that DMS can produce datasets which, when combined with the right predictive computational frameworks, can direct plant engineering efforts through derivative phenotype selection and evolutionary insights.

Unsupervised modeling of mutational landscapes of adeno-associated viruses viability

Adeno-associated viruses 2 (AAV2) are minute viruses renowned for their capacity to infect human cells and akin organisms. They have recently emerged as prominent candidates in the field of gene therapy, primarily attributed to their inherent non-pathogenic nature in humans and the safety associated with their manipulation. The efficacy of AAV2 as gene therapy vectors hinges on their ability to infiltrate host cells, a phenomenon reliant on their competence to construct a capsid capable of breaching the nucleus of the target cell. To enhance their infection potential, researchers have extensively scrutinized various combinatorial libraries by introducing mutations into the capsid, aiming to boost their effectiveness. The emergence of high-throughput experimental techniques, like deep mutational scanning (DMS), has made it feasible to experimentally assess the fitness of these libraries for their intended purpose. Notably, machine learning is starting to demonstrate its potential in addressing predictions within the mutational landscape from sequence data. In this context, we introduce a biophysically-inspired model designed to predict the viability of genetic variants in DMS experiments. This model is tailored to a specific segment of the CAP region within AAV2’s capsid protein. To evaluate its effectiveness, we conduct model training with diverse datasets, each tailored to explore different aspects of the mutational landscape influenced by the selection process. Our assessment of the biophysical model centers on two primary objectives: (i) providing quantitative forecasts for the log-selectivity of variants and (ii) deploying it as a binary classifier to categorize sequences into viable and non-viable classes.

Enhancing efficiency of protein language models with minimal wet-lab data through few-shot learning

Accurately modeling the protein fitness landscapes holds great importance for protein engineering. Pre-trained protein language models have achieved state-of-the-art performance in predicting protein fitness without wet-lab experimental data, but their accuracy and interpretability remain limited. On the other hand, traditional supervised deep learning models require abundant labeled training examples for performance improvements, posing a practical barrier. In this work, we introduce FSFP, a training strategy that can effectively optimize protein language models under extreme data scarcity for fitness prediction. By combining meta-transfer learning, learning to rank, and parameter-efficient fine-tuning, FSFP can significantly boost the performance of various protein language models using merely tens of labeled single-site mutants from the target protein. In silico benchmarks across 87 deep mutational scanning datasets demonstrate FSFP’s superiority over both unsupervised and supervised baselines. Furthermore, we successfully apply FSFP to engineer the Phi29 DNA polymerase through wet-lab experiments, achieving a 25% increase in the positive rate. These results underscore the potential of our approach in aiding AI-guided protein engineering.

Deep Mutational Scanning in Disease-related Genes with Saturation Mutagenesis-Reinforced Functional Assays (SMuRF).

Interpretation of disease-causing genetic variants remains a challenge in human genetics. Current costs and complexity of deep mutational scanning methods hamper crowd-sourcing approaches toward genome-wide resolution of variants in disease-related genes. Our framework, Saturation Mutagenesis-Reinforced Functional assays (SMuRF), addresses these issues by offering simple and cost-effective saturation mutagenesis, as well as streamlining functional assays to enhance the interpretation of unresolved variants. Applying SMuRF to neuromuscular disease genes FKRP and LARGE1, we generated functional scores for all possible coding single nucleotide variants, which aid in resolving clinically reported variants of uncertain significance. SMuRF also demonstrates utility in predicting disease severity, resolving critical structural regions, and providing training datasets for the development of computational predictors. Our approach opens new directions for enabling variant-to-function insights for disease genes in a manner that is broadly useful for crowd-sourcing implementation across standard research laboratories.

Natural variation in neuraminidase activity influences the evolutionary potential of the seasonal H1N1 lineage hemagglutinin.

The antigenic evolution of the influenza A virus hemagglutinin (HA) gene poses a major challenge for the development of vaccines capable of eliciting long-term protection. Prior efforts to understand the mechanisms that govern viral antigenic evolution mainly focus on HA in isolation, ignoring the fact that HA must act in concert with the viral neuraminidase (NA) during replication and spread. Numerous studies have demonstrated that the degree to which the receptor-binding avidity of HA and receptor-cleaving activity of NA are balanced with each other influences overall viral fitness. We recently showed that changes in NA activity can significantly alter the mutational fitness landscape of HA in the context of a lab-adapted virus strain. Here, we test whether natural variation in relative NA activity can influence the evolutionary potential of HA in the context of the seasonal H1N1 lineage (pdmH1N1) that has circulated in humans since the 2009 pandemic. We observed substantial variation in the relative activities of NA proteins encoded by a panel of H1N1 vaccine strains isolated between 2009 and 2019. We comprehensively assessed the effect of NA background on the HA mutational fitness landscape in the circulating pdmH1N1 lineage using deep mutational scanning and observed pronounced epistasis between NA and residues in or near the receptor-binding site of HA. To determine whether NA variation could influence the antigenic evolution of HA, we performed neutralizing antibody selection experiments using a panel of monoclonal antibodies targeting different HA epitopes. We found that the specific antibody escape profiles of HA were highly contingent upon NA background. Overall, our results indicate that natural variation in NA activity plays a significant role in governing the evolutionary potential of HA in the currently circulating pdmH1N1 lineage.

Deep mutational scanning quantifies DNA binding and predicts clinical outcomes of PAX6 variants.

Nonsense and missense mutations in the transcription factor PAX6 cause a wide range of eye development defects, including aniridia, microphthalmia and coloboma. To understand how changes of PAX6:DNA binding cause these phenotypes, we combined saturation mutagenesis of the paired domain of PAX6 with a yeast one-hybrid (Y1H) assay in which expression of a PAX6-GAL4 fusion gene drives antibiotic resistance. We quantified binding of more than 2700 single amino-acid variants to two DNA sequence elements. Mutations in DNA-facing residues of the N-terminal subdomain and linker region were most detrimental, as were mutations to prolines and to negatively charged residues. Many variants caused sequence-specific molecular gain-of-function effects, including variants in position 71 that increased binding to the LE9 enhancer but decreased binding to a SELEX-derived binding site. In the absence of antibiotic selection, variants that retained DNA binding slowed yeast growth, likely because such variants perturbed the yeast transcriptome. Benchmarking against known patient variants and applying ACMG/AMP guidelines to variant classification, we obtained supporting-to-moderate evidence that 977 variants are likely pathogenic and 1306 are likely benign. Our analysis shows that most pathogenic mutations in the paired domain of PAX6 can be explained simply by the effects of these mutations on PAX6:DNA association, and establishes Y1H as a generalisable assay for the interpretation of variant effects in transcription factors.

A parameterized two-domain thermodynamic model explains diverse mutational effects on protein allostery

New experimental findings continue to challenge our understanding of protein allostery. Recent deep mutational scanning study showed that allosteric hotspots in the tetracycline repressor (TetR) and its homologous transcriptional factors are broadly distributed rather than spanning well-defined structural pathways as often assumed. Moreover, hotspot mutation-induced allostery loss was rescued by distributed additional mutations in a degenerate fashion. Here, we develop a two-domain thermodynamic model for TetR, which readily rationalizes these intriguing observations. The model accurately captures the in vivo activities of various mutants with changes in physically transparent parameters, allowing the data-based quantification of mutational effects using statistical inference. Our analysis reveals the intrinsic connection of intra- and inter-domain properties for allosteric regulation and illustrate epistatic interactions that are consistent with structural features of the protein. The insights gained from this study into the nature of two-domain allostery are expected to have broader implications for other multi-domain allosteric proteins.

A parameterized two-domain thermodynamic model explains diverse mutational effects on protein allostery.

New experimental findings continue to challenge our understanding of protein allostery. Recent deep mutational scanning study showed that allosteric hotspots in the tetracycline repressor (TetR) and its homologous transcriptional factors are broadly distributed rather than spanning well-defined structural pathways as often assumed. Moreover, hotspot mutation-induced allostery loss was rescued by distributed additional mutations in a degenerate fashion. Here, we develop a two-domain thermodynamic model for TetR, which readily rationalizes these intriguing observations. The model accurately captures the in vivo activities of various mutants with changes in physically transparent parameters, allowing the data-based quantification of mutational effects using statistical inference. Our analysis reveals the intrinsic connection of intra- and inter-domain properties for allosteric regulation and illustrate epistatic interactions that are consistent with structural features of the protein. The insights gained from this study into the nature of two-domain allostery are expected to have broader implications for other multi-domain allosteric proteins.

High-throughput assays to assess variant effects on disease.

Interpreting the wealth of rare genetic variants discovered in population-scale sequencing efforts and deciphering their associations with human health and disease present a critical challenge due to the lack of sufficient clinical case reports. One promising avenue to overcome this problem is deep mutational scanning (DMS), a method of introducing and evaluating large-scale genetic variants in model cell lines. DMS allows unbiased investigation of variants, including those that are not found in clinical reports, thus improving rare disease diagnostics. Currently, the main obstacle limiting the full potential of DMS is the availability of functional assays that are specific to disease mechanisms. Thus, we explore high-throughput functional methodologies suitable to examine broad disease mechanisms. We specifically focus on methods that do not require robotics or automation but instead use well-designed molecular tools to transform biological mechanisms into easily detectable signals, such as cell survival rate, fluorescence or drug resistance. Here, we aim to bridge the gap between disease-relevant assays and their integration into the DMS framework.

Rosace: a robust deep mutational scanning analysis framework employing position and mean-variance shrinkage

Deep mutational scanning (DMS) measures the effects of thousands of genetic variants in a protein simultaneously. The small sample size renders classical statistical methods ineffective. For example, p-values cannot be correctly calibrated when treating variants independently. We propose Rosace, a Bayesian framework for analyzing growth-based DMS data. Rosace leverages amino acid position information to increase power and control the false discovery rate by sharing information across parameters via shrinkage. We also developed Rosette for simulating the distributional properties of DMS. We show that Rosace is robust to the violation of model assumptions and is more powerful than existing tools.

Molecular basis for antibody recognition of multiple drug–peptide/MHC complexes

The HapImmuneTM platform exploits covalent inhibitors as haptens for creating major histocompatibility complex (MHC)-presented tumor-specific neoantigens by design, combining targeted therapies with immunotherapy for the treatment of drug-resistant cancers. A HapImmune antibody, R023, recognizes multiple sotorasib-conjugated KRAS(G12C) peptides presented by different human leukocyte antigens (HLAs). This high specificity to sotorasib, coupled with broad HLA-binding capability, enables such antibodies, when reformatted as T cell engagers, to potently and selectively kill sotorasib-resistant KRAS(G12C) cancer cells expressing different HLAs upon sotorasib treatment. The loosening of HLA restriction could increase the patient population that can benefit from this therapeutic approach. To understand the molecular basis for its unconventional binding capability, we used single-particle cryogenic electron microscopy to determine the structures of R023 bound to multiple sotorasib-peptide conjugates presented by different HLAs. R023 forms a pocket for sotorasib between the VH and VL domains, binds HLAs in an unconventional, angled way, with VL making most contacts with them, and makes few contacts with the peptide moieties. This binding mode enables the antibody to accommodate different hapten-peptide conjugates and to adjust its conformation to different HLAs presenting hapten-peptides. Deep mutational scanning validated the structures and revealed distinct levels of mutation tolerance by sotorasib- and HLA-binding residues. Together, our structural information and sequence landscape analysis reveal key features for achieving MHC-restricted recognition of multiple hapten-peptide antigens, which will inform the development of next-generation therapeutic antibodies.

Exploring the ability of the MD+FoldX method to predict SARS-CoV-2 antibody escape mutations using large-scale data.

Antibody escape mutations pose a significant challenge to the effectiveness of vaccines and antibody-based therapies. The ability to predict these escape mutations with computer simulations would allow us to detect threats early and develop effective countermeasures, but a lack of large-scale experimental data has hampered the validation of these calculations. In this study, we evaluate the ability of the MD+FoldX molecular modeling method to predict escape mutations by leveraging a large deep mutational scanning dataset, focusing on the SARS-CoV-2 receptor binding domain. Our results show a positive correlation between predicted and experimental data, indicating that mutations with reduced predicted binding affinity correlate moderately with higher experimental escape fractions. We also demonstrate that better performance can be achieved using affinity cutoffs tailored to distinct antibody-antigen interactions rather than a one-size-fits-all approach. We find that 70% of the systems surpass the 50% precision mark, and demonstrate success in identifying mutations present in significant variants of concern and variants of interest. Despite promising results for some systems, our study highlights the challenges in comparing predicted and experimental values. It also emphasizes the need for new binding affinity methods with improved accuracy that are fast enough to estimate hundreds to thousands of antibody-antigen binding affinities.

Proscan: a structure-based proline design web server.

The ability to control protein conformations and dynamics through structure-based design has been useful in various scenarios, including engineering of viral antigens for vaccines. One effective design strategy is the substitution of residues to proline amino acids, which due to its unique cyclic side chain can favor and rigidify key backbone conformations. To provide the community with a means to readily identify and explore proline designs for target proteins of interest, we developed the Proscan web server. Proscan provides assessment of backbone angles, energetic and deep learning-based favorability scores, and other parameters for proline substitutions at each position of an input structure, along with interactive visualization of backbone angles and candidate substitution sites on structures. It identifies known favorable proline substitutions for viral antigens, and was benchmarked against datasets of proline substitution stability effects from deep mutational scanning and thermodynamic measurements. This tool can enable researchers to identify and prioritize designs for prospective vaccine antigen targets, or other designs to favor stability of key protein conformations. Proscan is available at: https://proscan.ibbr.umd.edu.

Epistasis facilitates functional evolution in an ancient transcription factor.

A protein's genetic architecture - the set of causal rules by which its sequence produces its functions - also determines its possible evolutionary trajectories. Prior research has proposed that the genetic architecture of proteins is very complex, with pervasive epistatic interactions that constrain evolution and make function difficult to predict from sequence. Most of this work has analyzed only the direct paths between two proteins of interest - excluding the vast majority of possible genotypes and evolutionary trajectories - and has considered only a single protein function, leaving unaddressed the genetic architecture of functional specificity and its impact on the evolution of new functions. Here, we develop a new method based on ordinal logistic regression to directly characterize the global genetic determinants of multiple protein functions from 20-state combinatorial deep mutational scanning (DMS) experiments. We use it to dissect the genetic architecture and evolution of a transcription factor's specificity for DNA, using data from a combinatorial DMS of an ancient steroid hormone receptor's capacity to activate transcription from two biologically relevant DNA elements. We show that the genetic architecture of DNA recognition consists of a dense set of main and pairwise effects that involve virtually every possible amino acid state in the protein-DNA interface, but higher-order epistasis plays only a tiny role. Pairwise interactions enlarge the set of functional sequences and are the primary determinants of specificity for different DNA elements. They also massively expand the number of opportunities for single-residue mutations to switch specificity from one DNA target to another. By bringing variants with different functions close together in sequence space, pairwise epistasis therefore facilitates rather than constrains the evolution of new functions.

Epistasis facilitates functional evolution in an ancient transcription factor

A protein’s genetic architecture – the set of causal rules by which its sequence produces its functions – also determines its possible evolutionary trajectories. Prior research has proposed that the genetic architecture of proteins is very complex, with pervasive epistatic interactions that constrain evolution and make function difficult to predict from sequence. Most of this work has analyzed only the direct paths between two proteins of interest – excluding the vast majority of possible genotypes and evolutionary trajectories – and has considered only a single protein function, leaving unaddressed the genetic architecture of functional specificity and its impact on the evolution of new functions. Here, we develop a new method based on ordinal logistic regression to directly characterize the global genetic determinants of multiple protein functions from 20-state combinatorial deep mutational scanning (DMS) experiments. We use it to dissect the genetic architecture and evolution of a transcription factor’s specificity for DNA, using data from a combinatorial DMS of an ancient steroid hormone receptor’s capacity to activate transcription from two biologically relevant DNA elements. We show that the genetic architecture of DNA recognition consists of a dense set of main and pairwise effects that involve virtually every possible amino acid state in the protein-DNA interface, but higher-order epistasis plays only a tiny role. Pairwise interactions enlarge the set of functional sequences and are the primary determinants of specificity for different DNA elements. They also massively expand the number of opportunities for single-residue mutations to switch specificity from one DNA target to another. By bringing variants with different functions close together in sequence space, pairwise epistasis therefore facilitates rather than constrains the evolution of new functions.

Structural basis of broad SARS-CoV-2 cross-neutralization by affinity-matured public antibodies

Descendants of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Omicron variant now account for almost all SARS-CoV-2 infections. The Omicron variant and its sublineages have spike glycoproteins that are highly diverged from the pandemic founder and first-generation vaccine strain, resulting in significant evasion from monoclonal antibody therapeutics and vaccines. Understanding how commonly elicited antibodies can broaden to cross-neutralize escape variants is crucial. We isolate IGHV3-53, using "public" monoclonal antibodies (mAbs) from an individual 7months post infection with the ancestral virus and identify antibodies that exhibit potent and broad cross-neutralization, extending to the BA.1, BA.2, and BA.4/BA.5 sublineages of Omicron. Deep mutational scanning reveals these mAbs' high resistance to viral escape. Structural analysis via cryoelectron microscopy of a representative broadly neutralizing antibody, CAB-A17, in complex with the Omicron BA.1 spike highlights the structural underpinnings of this broad neutralization. By reintroducing somatic hypermutations into a germline-reverted CAB-A17, we delineate the role of affinity maturation in the development of cross-neutralization by a public class of antibodies.

Functional and antigenic characterization of SARS-CoV-2 spike fusion peptide by deep mutational scanning

The fusion peptide of SARS-CoV-2 spike protein is functionally important for membrane fusion during virus entry and is part of a broadly neutralizing epitope. However, sequence determinants at the fusion peptide and its adjacent regions for pathogenicity and antigenicity remain elusive. In this study, we perform a series of deep mutational scanning (DMS) experiments on an S2 region spanning the fusion peptide of authentic SARS-CoV-2 in different cell lines and in the presence of broadly neutralizing antibodies. We identify mutations at residue 813 of the spike protein that reduced TMPRSS2-mediated entry with decreased virulence. In addition, we show that an F823Y mutation, present in bat betacoronavirus HKU9 spike protein, confers resistance to broadly neutralizing antibodies. Our findings provide mechanistic insights into SARS-CoV-2 pathogenicity and also highlight a potential challenge in developing broadly protective S2-based coronavirus vaccines.

Deep mutational scanning reveals a correlation between degradation and toxicity of thousands of aspartoacylase variants

Unstable proteins are prone to form non-native interactions with other proteins and thereby may become toxic. To mitigate this, destabilized proteins are targeted by the protein quality control network. Here we present systematic studies of the cytosolic aspartoacylase, ASPA, where variants are linked to Canavan disease, a lethal neurological disorder. We determine the abundance of 6152 of the 6260 ( ~ 98%) possible single amino acid substitutions and nonsense ASPA variants in human cells. Most low abundance variants are degraded through the ubiquitin-proteasome pathway and become toxic upon prolonged expression. The data correlates with predicted changes in thermodynamic stability, evolutionary conservation, and separate disease-linked variants from benign variants. Mapping of degradation signals (degrons) shows that these are often buried and the C-terminal region functions as a degron. The data can be used to interpret Canavan disease variants and provide insight into the relationship between protein stability, degradation and cell fitness.

Computationally restoring the potency of a clinical antibody against Omicron.

The COVID-19 pandemic underscored the promise of monoclonal antibody-based prophylactic and therapeutic drugs1-3 and revealed how quickly viral escape can curtail effective options4,5. When the SARS-CoV-2 Omicron variant emerged in 2021, many antibody drug products lost potency, including Evusheld and its constituent, cilgavimab4-6. Cilgavimab, like its progenitor COV2-2130, is a class 3 antibody that is compatible with other antibodies in combination4 and is challenging to replace with existing approaches. Rapidly modifying such high-value antibodies to restore efficacy against emerging variants is a compelling mitigation strategy. We sought to redesign and renew the efficacy of COV2-2130 against Omicron BA.1 and BA.1.1 strains while maintaining efficacy against the dominant Delta variant. Here we show that our computationally redesigned antibody, 2130-1-0114-112, achieves this objective, simultaneously increases neutralization potency against Delta and subsequent variants of concern, and provides protection in vivo against the strains tested: WA1/2020, BA.1.1 and BA.5. Deep mutational scanning of tens of thousands of pseudovirus variants reveals that 2130-1-0114-112 improves broad potency without increasing escape liabilities. Our results suggest that computational approaches can optimize an antibody to target multiple escape variants, while simultaneously enriching potency. Our computational approach does not require experimental iterations or pre-existing binding data, thus enabling rapid response strategies to address escape variants or lessen escape vulnerabilities.