Biology subjects

Upadhyay, V.

Publications and source records attributed to Upadhyay, V..

At least 19 recordsLinked to original sources

bioRxiv2026

An AI-powered cloud biofoundry for autonomous biological research

Autonomous experimentation, where an automated system designs, performs, analyzes, and iteratively improves experiments with minimal human intervention represents a transformative scientific research paradigm. However, it remains hampered by the disconnect between computational design and experimental execution. Here we present iCloudBiofoundry, a cloud-accessible, scalable, and self-evolving platform that bridges this divide. A scientific multi-agent system, iBioGenie, plans research workflows by converting natural language research objectives into testable hypotheses and experimentally grounded protocols using literature, databases, and computational tools. A physical AI layer, BiofoundryAI, translates them into executable workflows that run autonomously on distributed robotic biofoundries, and collects the data for iterative design. We demonstrate the end-to-end capabilities of iCloudBiofoundry across three representative case studies spanning enzyme retrieval, protein engineering, and metabolic engineering. By integrating agentic AI with physical AI, iCloudBiofoundry provides a cloud accessible platform that unifies computational design and robotic execution, enabling autonomous biological research.

synthetic biology↗

bioRxiv2026

The ModelSEED Biochemistry Database, 2026 update: grading multi-source thermodynamics

The ModelSEED Biochemistry Database (https://modelseed.org) supplies foundational mass- and charge-balanced reaction networks for metabolic reconstructions. Here we present an update on the biochemistry where the database has been expanded to [~] 46,000 compounds, and [~] 56,000 reactions, featuring [~] 37,000 metabolic structures. We have expanded our approach for handling thermodynamic data, enabling multiple sources of data to be derived, integrated, and presented to the wider research community. We now publish predictions of pKa, reaction energy (and respective uncertainties), and estimates of reaction direction from multiple independent sources. Each reaction is graded gold, silver or bronze according to the strength of the evidence behind it, so that users can weigh its reliability directly. We also release reaction directions predicted by an ensemble of large language models. This multi-source approach exposes agreements and discrepancies between sources for [~] 33,000 reactions. Finally, to ensure data integrity, a new conflict-resolution pipeline reconciles structures across sources, documenting input from curators. Our work is publicly available at https://github.com/ModelSEED/ModelSEEDDatabase.

biochemistry↗

bioRxiv2026

Female-enriched Eggerthella lenta drives neuroinflammation and IFN-γ via host receptor TLR2

Women are at increased risk of autoimmune diseases, including multiple sclerosis (MS); however, the degree to which sex differences in the gut microbiota impact autoimmunity remains largely unexplored. Our 27-cohort meta-analysis revealed 60 sex-associated gut bacterial species. Leveraging an independent clinical cohort, we demonstrate that female-enriched species significantly associate with MS status and clinical disability (EDSS). Top female-enriched species Eggerthella lenta drove disease in the experimental autoimmune encephalomyelitis (EAE) MS model, consistent with brain and gut lamina propria T cell infiltration and MS-associated T helper (Th) signatures. E. lenta induced intestinal Th1 and Th17 in healthy mice, independent of bacterial viability. Mechanistically, we demonstrate that TLR2 directly drives E. lenta-induced IFN-{gamma} production in Th cells and is necessary for exacerbation of EAE. Together, we identify a causal host-microbe axis contributing to sex differences in autoimmunity and provide a framework for evaluating sex as a biological variable in human microbiome research. HIGHLIGHTSO_LI27-cohort meta-analysis identifies a robust sex-signature in human gut microbiota. C_LIO_LIFemale-enriched species are associated with MS risk and severity. C_LIO_LIFemale-enriched Eggerthella lenta exacerbates the EAE model. C_LIO_LIE. lenta impacts neuroinflammation via toll-like receptor 2. C_LI

microbiology↗

bioRxiv2025

Shifts in the human gut microbiome during cancer chemotherapy are diet-dependent

Numerous studies have implicated both dietary intake and the human gut microbiome in colorectal cancer (CRC) treatment outcomes. However, little is known about how patients adjust their dietary intake during cancer chemotherapy or if these dietary changes contribute to treatment-associated alterations in the gut microbiome. We performed paired longitudinal diet and microbiome analysis during CRC treatment with oral fluoropyrimidines (NCT04054908) and validated key associations using cell culture assays. At each timepoint, diet was measured by averaging up to 3 consecutive days of 24-hour dietary records (NCI ASA24; 35 patients; 2.5{+/-}1.1 timepoints per patient), while microbiome composition was measured from stool (16S rRNA gene sequencing, metagenomics, qPCR; 40 patients; 5.6{+/-}1.4 timepoints per patient). Diet quality significantly decreased during chemotherapy. Carbohydrate and refined grain intake increased, accompanied by decreased consumption of fats, nuts and seeds, and fat-soluble micronutrients. Multiple individual dietary components were strongly linked to the gut microbiome. Decreases in theobromine intake correlated with decreases in overall microbial diversity and more gastrointestinal toxicities. Diet shifts partly explained changes in bacterial abundance during chemotherapy, including more severe depletion of Faecalibacterium prausnitzii in patients with decreased vitamin K1 intake. Changes in diet were correlated with multiple bacterial gene families involved in micronutrient metabolism and drug sensitivity. Increased copper intake was linked to decreased Fusobacterium nucleatum in patients and inhibited F. nucleatum in cell assays. Together, these data suggest that changes in diet during chemotherapy contribute to changes in gut bacterial diversity, taxonomic composition, and gene abundance. Our approach may generalize to other cancer therapies and emphasizes the need for collecting more robust dietary data in clinical microbiome studies.

microbiology↗

bioRxiv2025

MechFind: A computational framework for de novo prediction of enzyme mechanisms

Despite the importance of understanding the step-by-step mechanism of enzymatically catalyzed reactions, fewer than one thousand cataloged mechanistic annotations can be found in the open literature and databases. Herein, we introduce MechFind, a computational tool that generates elementally and charge-balanced putative enzyme mechanisms. Unlike previous methods that require structural data or user-supplied active site residues, MechFind uses only the overall reaction stoichiometry as input, abstracting individual reaction steps as the gain or loss of chemical moieties. An optimization framework then identifies the top ten most parsimonious (i.e., fewest steps) mechanistic descriptions for the overall transformation, which are then re-ranked based on their mechanistic similarity to a database of known, curated mechanisms. MechFind recovered the correct mechanism for 72% of the Mechanism and Catalytic Site Atlas (M-CSA) training dataset within the top ten predictions and was independently validated on six enzyme mechanisms absent from the training set. When deployed at scale on 28,412 reactions from the Rhea database, MechFind identified a plausible mechanism for 64% of all entries, generating over 18,000 novel mechanistic hypotheses expanding significantly upon the current state of the art in prediction (i.e., a 20-fold increase). By proposing detailed reaction mechanisms MechFind provides information that can be leveraged by modern ML-based de novo protein design tools, offering a resource for improving functional annotation and accelerating the engineering of novel biocatalysts. All codes, curated datasets, and results are available at https://github.com/maranasgroup/MechFind.

biochemistry↗

bioRxiv2025

α7 nicotinic acetylcholine receptors regulate radial glia fate in the developing human cortex

Prenatal nicotine exposure impairs fetal cortical grey matter volume, but the precise cellular mechanisms remain poorly understood. This study elucidates the role of nicotinic acetylcholine receptors (nAChRs) in progenitor cells and radial glia (RG) during human cortical development. We identify two nAChR subunits--CHRNA7 and the human-specific CHRFAM7A--expressed in SOX2+ progenitors and neurons, with CHRFAM7A particularly enriched along RG endfeet. nAChR activation in organotypic slices and dissociated cultures increases RG proliferation while decreasing neuronal differentiation, whereas nAChR knockdown reduces RG and increases neurons. Single-cell RNA sequencing reveals that nicotine exposure downregulates key genes in excitatory neurons (ENs), with CHRNA7 or CHRFAM7A selectively modulating these changes, suggesting an evolutionary divergence in regulatory pathways. Furthermore, we identify YAP1 as a critical downstream effector of nAChR signaling, and inhibiting YAP1 reverses nicotine-induced phenotypic alterations in oRG cells, highlighting its role in nicotine-induced neurodevelopmental pathophysiology.

neuroscience↗

bioRxiv2025

Gut bacterial lactate stimulates lung epithelial mitochondria and exacerbates acute lung injury

Acute respiratory distress syndrome (ARDS) is an often fatal critical illness where lung epithelial injury leads to intrapulmonary fluid accumulation. ARDS became widespread during the COVID-19 pandemic, motivating a renewed effort to understand the complex etiology of this disease. Rigorous prior work has implicated lung endothelial and epithelial injury in response to an insult such as bacterial infection; however, the impact of microorganisms found in other organs on ARDS remains unclear. Here, we use a combination of gnotobiotic mice, cell culture experiments, and re-analyses of a large metabolomics dataset from ARDS patients to reveal that gut bacteria impact lung cellular respiration by releasing metabolites that alter mitochondrial activity in lung epithelium. Colonization of germ-free mice with a complex gut microbiota stimulated lung mitochondrial gene expression. A single human gut bacterial species, Bifidobacterium adolescentis, was sufficient to replicate this effect, leading to a significant increase in mitochondrial membrane potential in lung epithelial cells. We then used genome sequencing and mass spectrometry to confirm that B. adolescentis produces L-lactate, which was sufficient to increase mitochondrial activity in lung epithelial cells. Finally, we found that serum lactate was significantly associated with disease severity in patients with ARDS from the Early Assessment of Renal and Lung Injury (EARLI) cohort. Together, these results emphasize the importance of more broadly characterizing the microbial etiology of ARDS and other lung diseases given the ability of gut bacterial metabolites to remotely control lung cellular respiration. Our discovery of a single bacteria-metabolite pair provides a proof-of-concept for systematically testing other microbial metabolites and a mechanistic biomarker that could be pursued in future clinical studies. Furthermore, our work adds to the growing literature linking the microbiome to mitochondrial function, raising intriguing questions as to the bidirectional communication between our endo- and ecto-symbionts.

molecular biology↗

bioRxiv2024

novoStoic2.0: An integrated framework for pathway synthesis, thermodynamic evaluation, and enzyme selection

Computational pathway design and retro-biosynthetic approaches can facilitate the development of innovative biochemical production routes, biodegradation strategies, and the funneling of multiple precursors into a single bioproduct. However, effective pathway design necessitates a comprehensive understanding of biochemistries, enzyme activities, and thermodynamic feasibility. Herein, we introduce novoStoic2.0, an integrated platform that combines tools for estimating overall stoichiometry, designing de novo synthesis pathways, assessing thermodynamic feasibility, and selecting enzymes. novoStoic2.0 offers a unified web-based interface as a part of the AlphaSynthesis platform (http://novostoic.platform.moleculemaker.org/) tailored for the synthesis of thermodynamically viable pathways as well as the selection of enzymes for re-engineering required for novel reaction steps. We exemplify the utility of the platform to identify novel pathways for hydroxytyrosol synthesis, which are shorter than the known pathways and require reduced cofactor usage. In summary, novoStoic2.0 aims to streamline the process of pathway design contributing to the development of sustainable biotechnological solutions.

synthetic biology↗

bioRxiv2024

Expansion of a bacterial operon during cancer treatment ameliorates drug toxicity

Dose-limiting toxicities remain a major barrier to drug development and therapy, revealing the limited predictive power of human genetics. Herein, we demonstrate the utility of a more comprehensive approach to studying drug toxicity through longitudinal study of the human gut microbiome during colorectal cancer (CRC) treatment (NCT04054908) coupled to cell culture and mouse experiments. 16S rRNA gene sequencing revealed significant shifts in gut microbial community structure during oral fluoropyrimidine treatment across multiple patient cohorts, in mouse small and large intestinal contents, and in patient-derived ex vivo communities. Metagenomic sequencing revealed marked shifts in pyrimidine-related gene abundance during oral fluoropyrimidine treatment, including enrichment of the preTA operon, which is sufficient for the inactivation of active metabolite 5-fluorouracil (5-FU). preTA+ bacteria depleted 5-FU in gut microbiota grown ex vivo and the mouse distal gut. Germ-free and antibiotic-treated mice experienced increased fluoropyrimidine toxicity, which was rescued by colonization with the mouse gut microbiota, preTA+ E. coli, or preTA-high CRC patient stool. Finally, preTA abundance was negatively associated with fluoropyrimidine toxicity in patients. Together, these data support a causal, clinically relevant interaction between a human gut bacterial operon and the dose-limiting side effects of cancer treatment. Our approach is generalizable to other drugs, including cancer immunotherapies, and provides valuable insights into host-microbiome interactions in the context of disease. One Sentence Summary: Gut microbial enzymes can be used to predict and prevent anticancer drug toxicity.

systems biology↗

bioRxiv2023

A diet-dependent host metabolite shapes the gut microbiota to protect from autoimmunity

Diet can protect from autoimmune disease; however, whether diet acts via the host and/or microbiome remains unclear. Here, we use a ketogenic diet (KD) as a model to dissect these complex interactions. A KD rescued the experimental autoimmune encephalomyelitis (EAE) mouse model of multiple sclerosis in a microbiota-dependent fashion. Dietary supplementation with a single KD-dependent host metabolite ({beta}-hydroxybutyrate, {beta}HB) rescued EAE whereas transgenic mice unable to produce {beta}HB in the intestine developed more severe disease. Transplantation of the {beta}HB-shaped gut microbiota was protective. Lactobacillus sequence variants were associated with decreased T helper 17 (Th17) cell activation in vitro. Finally, we isolated a L. murinus strain that protected from EAE, which was phenocopied by the Lactobacillus metabolite indole lactic acid. Thus, diet alters the immunomodulatory potential of the gut microbiota by shifting host metabolism, emphasizing the utility of taking a more integrative approach to study diet-host-microbiome interactions.

microbiology↗

bioRxiv2023

Cell type specificity of mosaic chromosome 1q gain resolved by snRNA-seq in a case of epilepsy with hyaline protoplasmic astrocytopathy

STRUCTURED ABSTRACTO_ST_ABSIntroductionC_ST_ABSMosaic gain of chromosome 1q (chr1q) has been associated with malformation of cortical development (MCD) and epilepsy. Hyaline protoplasmic astrocytopathy (HPA) is a rare neuropathological finding seen in cases of epilepsy with MCD. The cell-type specificity of mosaic chr1q gain in the brain and the molecular signatures of HPA are unknown. MethodsWe present a child with pharmacoresistant epilepsy who underwent epileptic focus resections at age 3 and 5 years and was found to have mosaic chr1q gain and HPA. We performed single-nuclei RNA-sequencing (snRNA-seq) of brain tissue from the second resection. ResultssnRNA-seq showed increased expression of chr1q genes specifically in subsets of neurons and astrocytes. Differentially expressed genes associated with inferred chr1q gain included AKT3 and genes associated with cell adhesion or migration. A subpopulation of astrocytes demonstrated marked enrichment for synapse-associated transcripts, possibly linked to the astrocytic inclusions observed in HPA. DiscussionsnRNA-seq may be used to infer the cell type-specificity of mosaic chromosomal copy number changes and identify associated gene expression alterations, which in the case of chr1q gain may involve aberrations in cell migration. Future studies using spatial profiling could yield further insights on the molecular signatures of HPA.

neuroscience↗

bioRxiv2023

Exploring the interaction of inhibitor YM-1 with Plasmodium falciparum HSP70s through in silico methods

YM-1 belongs to the rhodacyanine class of the 70kDa heat shock protein (Hsp70) inhibitors. It antagonizes nucleotide exchange factor (NEF) BAG3 to block the critical ATPase cycle of HSP70 chaperones thereby disabling them. The anti-cancer potential of this small molecule derivative of MKT-077 is attributed to its improved solubility and stability over its parent. In the current study, we have used bioinformatics tools including molecular docking to study the binding strength of YM-1 for Plasmodium falciparum (Pf) HSP70s. Of the four PfHSP70 homologs, the mitochondrial counterpart was predicted to bind most efficiently with YM-1 at a previously reported inhibitor binding site. The interaction of YM-1 with mitochondrial PfHSP70 was significantly better than with any of the other Pf homologs. Further analysis of this PfHSP70-YM-1 complex helped us to predict not just the types of interactions, but also specific residues of the protein involved in inhibitor engagement. Information reported in this article is important from the standpoint of design and development of newer rhodacyanine based drugs against malaria.

bioinformatics↗

bioRxiv2023

Analyzing interaction of rhodacyanine inhibitor MKT-077 with Plasmodium falciparum HSP70 nucleotide binding domains

MKT-077 and its derivatives are rhodacyanine inhibitors that hold potential in treatment of cancer, neurodegenerative diseases and malaria. These allosteric drugs act by inhibiting the ATPase action of heat shock proteins of 70kDa (HSP70). MKT-077 accumulates in the mitochondria and displays differential activity against HSP70 homologs. The four Plasmodium falciparum HSP70s (PfHSP70) are present at various subcellular locations to perform distinct functions. In the present study, we have used bioinformatics tools to understand interactions between MKT-077 and PfHSP70s. Nature of the identified interactions is primarily hydrophobic with different PfHSP70 homologs showing variable propensity to bind MKT-077. Our molecular docking approach has helped us to predict the binding pocket and specific residues on PfHSP70s that are involved in interacting with MKT-077. Information derived from this article may form the foundation for design and development of MKT-077 based drugs against malaria.

bioinformatics↗

bioRxiv2023

Understanding the structural basis for differential binding of lapachol with PfHSP70s

The 70 kDa heat shock proteins from Plasmodium falciparum (PfHSP70) are an important family of proteins that may be exploited for antimalarial design. Plant derived inhibitor lapachol is reported to efficiently inhibit the exported PfHSP70-x while having poor activity against the parasite PfHSP70-1. In the present study, we have used in silico tools including molecular docking to understand the molecular basis for the above differences in lapachol inhibition activity against PfHSP70s. We found a significant gap in the binding energies and hence affinity of PfHSP70-1 and PfHSP70-x for lapachol with the latter having a better binding propensity. Our data highlight notable differences in the type of interactions between the two complexes. Detailed molecular analysis of the complexes has helped us to predict specific amino acid residues from both these PfHSP70 homologs that may be involved in lapachol binding. The above information may be utilized for design of PfHSP70 inhibitors with antimalarial potential.

bioinformatics↗

bioRxiv2023

Biophysical evolution of the receptor binding domains of SARS-CoVs

With hundreds of coronaviruses (CoVs) identified in bats that are capable of infecting humans, it is important to understand how CoVs that affected the human population have evolved. Seven known coronaviruses have infected humans, of which three CoVs caused severe disease with high mortality rates: SARS-CoV emerged in 2002, MERS-CoV in 2012, and SARS-CoV-2 in 2019. Both SARS-CoV and SARS-CoV-2 belong to the same family, follow the same receptor pathway, and use their receptor binding domain (RBD) of spike protein to bind to the ACE2 receptor on the human epithelial cell surface. The sequence of the two RBDs is divergent, especially in the receptor binding motif (RBM) that directly interacts with ACE2. We probed the biophysical differences between the two RBDs in terms of their structure, stability, aggregation, and function. Since RBD is being explored as an antigen in protein subunit vaccines against CoVs, determining these biophysical properties will also aid in developing stable protein subunit vaccines. Our results show that despite RBDs having a similar three-dimensional structure, they differ in their thermodynamic stability. RBD of SARS-CoV-2 is significantly less stable than that of SARS-CoV. Correspondingly, SARS-CoV-2 RBD shows a higher aggregation propensity. Regarding binding to ACE2, less stable SARS-CoV-2 RBD binds with a higher affinity than more stable SARS-CoV RBD. In addition, SARS-CoV-2 RBD is more homogenous in terms of its binding stoichiometry towards ACE2, compared to SARS-CoV RBD. These results indicate that SARS-CoV-2 RBD differs from SARS-CoV RBD in terms of its stability, aggregation, and function, possibly originating from the diverse RBMs. Higher aggregation propensity and decreased stability of SARS-CoV-2 RBD warrants further optimization of protein subunit vaccines that use RBD as an antigen either by inserting stabilizing mutations or formulation screening. Statement of SignificanceThis study holds significant relevance in the context of the COVID-19 pandemic and the broader understanding of coronaviruses. A comparison of the receptor binding domains (RBDs) of SARS-CoV and SARS-CoV-2 reveals significant differences in their structure, stability, aggregation, and function. Despite divergent sequences, the RBDs share a similar fold and ACE2 receptor binding capability, likely through convergent evolution. These findings are crucial for understanding coronavirus evolution, interactions with human receptors, and the spillover of coronaviruses from animals to humans. The study also has implications for vaccine design strategies for SARS-CoVs, where the RBD is used as an antigen in protein subunit vaccines. By anticipating future outbreaks and enhancing our understanding of zoonotic spillover, this research contributes to safeguarding human health.

biophysics↗

bioRxiv2023

PhIP-Seq uncovers novel autoantibodies and unique endotypes in interstitial lung disease

Interstitial lung diseases (ILDs) are a heterogeneous group of disorders that can develop in patients with connective tissue diseases (CTD). Establishing autoimmunity in ILD impacts prognosis and treatment. ILD patients are screened for autoimmunity by assaying for anti-nuclear autoantibodies, rheumatoid factors and other non-specific tests. However, this approach has not been rigorously validated and may miss autoimmunity that manifests as autoantibodies to tissue antigens not previously defined in ILD. Here, we use Phage Immunoprecipitation-Sequencing (PhIP-Seq) to conduct a large, multi-center unbiased autoantibody discovery screen of ILD patients and controls. PhIP-Seq identified 17 novel autoreactive targets, and machine learning classifiers derived from these targets discriminated ILD serum from controls. Among these 17 candidates, we validated Cadherin Related Family Member 5 (CDHR5) as an autoantigen and found CDHR5 autoantibodies in patients with rheumatologic disorders and importantly, subjects not previously diagnosed with autoimmunity. Lung tissue of CDHR5 autoreactive patients showed transcriptional profiles consistent with activation of NF{kappa}B signaling and upregulation of chitotriosidase (CHIT1), a molecular pathway linked to fibrosis. Our study shows PhIP-Seq uncovers novel autoantibodies in ILD patients not revealed by standard clinical tests. Furthermore, CDHR5 autoantibodies may define a novel molecular endotype of ILD characterized by inflammation and fibrosis.

molecular biology↗

bioRxiv2022

Mild SARS-CoV-2 infection results in long-lasting microbiota instability

Viruses targeting mammalian cells can indirectly alter the gut microbiota, potentially compounding their phenotypic effects. Multiple studies have observed a disrupted gut microbiota in severe cases of SARS-CoV-2 infection that require hospitalization. Yet, despite demographic shifts in disease severity resulting in a large and continuing burden of non-hospitalized infections, we still know very little about the impact of mild SARS-CoV-2 infection on the gut microbiota in the outpatient setting. To address this knowledge gap, we longitudinally sampled 14 SARS-CoV-2 positive subjects who remained outpatient and 4 household controls. SARS-CoV-2 cases exhibited a significantly less stable gut microbiota relative to controls, as long as 154 days after their positive test. These results were confirmed and extended in the K18-hACE2 mouse model, which is susceptible to SARS-CoV-2 infection. All of the tested SARS-CoV-2 variants significantly disrupted the mouse gut microbiota, including USA-WA1/2020 (the original variant detected in the United States), Delta, and Omicron. Surprisingly, despite the fact that the Omicron variant caused the least severe symptoms in mice, it destabilized the gut microbiota and led to a significant depletion in Akkermansia muciniphila. Furthermore, exposure of wild-type C57BL/6J mice to SARS-CoV-2 disrupted the gut microbiota in the absence of severe lung pathology. IMPORTANCETaken together, our results demonstrate that even mild cases of SARS-CoV-2 can disrupt gut microbial ecology. Our findings in non-hospitalized individuals are consistent with studies of hospitalized patients, in that reproducible shifts in gut microbial taxonomic abundance in response to SARS-CoV-2 have been difficult to identify. Instead, we report a long-lasting instability in the gut microbiota. Surprisingly, our mouse experiments revealed an impact of the Omicron variant, despite producing the least severe symptoms in genetically susceptible mice, suggesting that despite the continued evolution of SARS-CoV-2 it has retained its ability to perturb the intestinal mucosa. These results will hopefully renew efforts to study the mechanisms through which Omicron and future SARS-CoV-2 variants alter gastrointestinal physiology, while also considering the potentially broad consequences of SARS-CoV-2-induced microbiota instability for host health and disease.

microbiology↗

bioRxiv2022

Biophysical fitness landscape of the SARS-CoV-2 Delta variant receptor binding domain

Among the five known SARS-CoV-2 variants of concern, Delta is the most virulent leading to severe symptoms and increased number of deaths. Our study seeks to examine how the biophysical parameters of the Delta variant correlate to the clinical observations. Receptor binding domain (RBD) is the first point of contact with the human host cells and is the immunodominant form of the spike protein. Delta variant RBD contains two novel mutations L452R and T478K. We examined the effect of single mutations as well as the double mutation on RBD expression in human Expi293 cells, RBD stability using urea and thermal denaturation, and RBD binding to angiotensin converting enzyme 2 (ACE2) receptor and to neutralizing antibodies using isothermal titration calorimetry. Delta variant RBD showed significantly higher expression compared to the wild-type RBD, and the increased expression is due to L452R mutation. Despite their non-conservative nature, none of the mutations significantly affected RBD structure and stability. All mutants showed similar binding affinity to ACE2 and to Class 1 antibodies (CC12.1 and LY-CoV016) as that of the wild-type. Delta double mutant L452R/T478K showed no binding to Class 2 antibodies (P2B-2F6 and LY-CoV555) and a hundred-fold weaker binding to a Class 3 antibody (REGN10987), and the decreased antibody binding is determined by the L452R mutation. These results indicate that the immune escape from neutralizing antibodies, rather than receptor binding, is the main biophysical parameter determining the fitness landscape of the Delta variant RBD and is determined by the L452R mutation.

biophysics↗