BioMed X Institute

The human intestine plays a pivotal role in nutrient absorption and immune system regulation. Along the longitudinal axis, cell-type composition changes to meet the varying functional requirements. Therefore, our protocol focuses on the processing of the whole human intestine to facilitate the analysis of region-specific characteristics such as tissue architecture and changes in cell populations. We describe how to generate a biobank that can be used to isolate specific immune cell subtypes, generate organoid lines, and establish autologous immune cell-organoid co-cultures. Key features • Dissection and tissue analysis of whole human intestines. • Cryopreservation for biobank generation. • Optimized protocols for the isolation of epithelial and immune cells. • Autologous co-culture of organoids and lamina propria–derived immune cells.

The Korean rhinoceros beetle, Trypoxylus dichotomus , an edible insect in the Republic of Korea, is an important insect for industrial farming, with various uses ranging from medical purposes to pets. However, there have been difficulties in the production and management of this species over the past 10 years as a result of nudivirus. This virus belongs to the family Nudiviridae, and infection with this virus proves fatal for rhinoceros beetles. The goal of this work was to develop an antibody for fundamental research on virus diagnosis. The ODV‐E66 gene encoding an envelope protein was selected for antibody production. The peptide region was derived from in silico analysis considering the conserved motif region and secondary structure of the peptide. In addition, the final peptide immunogen was validated by analyzing the physicochemical properties of the peptide to increase the rate of antibody formation. The production of polyclonal antibodies was induced by injecting peptide immunogens into rabbits. Antibody activity was evaluated using the enzyme‐linked immunosorbent assay (ELISA) method. As a result, we confirmed that antibody formation was possible with nudivirus mimotopes, and we also confirmed that the antigen–antibody reaction differs depending on the peptide. These results suggest a new perspective for diagnosing insect viruses and contribute to the broader knowledge of peptide immunogens.

Repair of double-stranded breaks generated by CRISPR/Cas9 is highly dependent on the flanking DNA sequence. To learn about interactions between DNA repair and target sequence, we measure frequencies of over 236,000 distinct Cas9-generated mutational outcomes at over 2800 synthetic target sequences in 18 DNA repair deficient mouse embryonic stem cells lines. We classify the outcomes in an unbiased way, finding a specialised role for Prkdc (DNA-PKcs protein) and Polm in creating 1 bp insertions matching the nucleotide on the protospacer-adjacent motif side of the break, a variable involvement of Nbn and Polq in the creation of different deletion outcomes, and uni-directional deletions dependent on both end-protection and end-resection. Using our dataset, we build predictive models of the mutagenic outcomes of Cas9 scission that outperform the current standards. This work improves our understanding of DNA repair gene function, and provides avenues for more precise modulation of Cas9-generated mutations.

Real-time monitoring of therapeutic drugs is crucial for treatment management and pharmacokinetic studies. We present the optimization and affinity tuning of split-aptamer sandwich assay for real-time monitoring of the narrow...

Severe acute respiratory syndrome coronavirus 2 mRNA vaccination has reduced effectiveness in certain immunocompromised individuals. However, the cellular mechanisms underlying these defects, as well as the contribution of disease-induced cellular abnormalities, remain largely unexplored. In this study, we conducted a comprehensive serological and cellular analysis of patients with autoimmune systemic lupus erythematosus (SLE) who received the Wuhan-Hu-1 monovalent mRNA coronavirus disease 2019 vaccine. Our findings revealed that patients with SLE exhibited reduced avidity of anti-receptor-binding domain antibodies, leading to decreased neutralization potency and breadth. We also observed a sustained anti-spike response in IgD⁻CD27⁻ ‘double-negative (DN)’ DN2/DN3 B cell populations persisting during memory responses and with greater representation in the SLE cohort. Additionally, patients with SLE displayed compromised anti-spike T cell immunity. Notably, low vaccine efficacy strongly correlated with higher values of a newly developed extrafollicular B and T cell score, supporting the importance of distinct B cell endotypes. Finally, we found that anti-BAFF blockade through belimumab treatment was associated with poor vaccine immunogenicity due to inhibition of naive B cell priming and an unexpected impact on circulating T follicular helper cells.

In the evolving landscape of biomedical research, the convergence of molecular biology and translational medicine has ushered in a new era of pharmaceutical innovation. This paradigm shift, characterized by significant advances in targeted therapies and gene editing, emphasizes the critical role of integrating academic research – and academic researchers – within industry settings. Contemporary innovation models are moving beyond traditional, corporation-centered frameworks, adopting more open, collaborative approaches. Here, we discuss the challenges and solutions brought about by this new direction in pharma innovation and describe the BioMed X innovation model, a unique open innovation approach that has been growing continuously over the past ten years.

A biochemical pathway consists of a series of interconnected biochemical reactions to accomplish specific life activities. The participating reactants and resultant products of a pathway, including gene fragments, proteins, and small molecules, coalesce to form a complex reaction network. Biochemical pathways play a critical role in the biochemical domain as they can reveal the flow of biochemical reactions in living organisms, making them essential for understanding life processes. Existing studies of biochemical pathway networks are mainly based on experimentation and pathway database analysis methods, which are plagued by substantial cost constraints. Inspired by the success of representation learning approaches in biomedicine, we develop the biochemical pathway prediction (BPP) platform, which is an automatic BPP platform to identify potential links or attributes within biochemical pathway networks. Our BPP platform incorporates a variety of representation learning models, including the latest hypergraph neural networks technology to model biochemical reactions in pathways. In particular, BPP contains the latest biochemical pathway-based datasets and enables the prediction of potential participants or products of biochemical reactions in biochemical pathways. Additionally, BPP is equipped with an SHAP explainer to explain the predicted results and to calculate the contributions of each participating element. We conduct extensive experiments on our collected biochemical pathway dataset to benchmark the effectiveness of all models available on BPP. Furthermore, our detailed case studies based on the chronological pattern of our dataset demonstrate the effectiveness of our platform. Our BPP web portal, source code and datasets are freely accessible at https://github.com/Glasgow-AI4BioMed/BPP.

Cardiac function relies on the autonomous molecular contraction mechanisms in the ventricular wall. Contraction is driven by ordered motor proteins acting in parallel to generate a macroscopic force. The averaged structure can be investigated by diffraction from model tissues such as trabecular and papillary cardiac muscle using collimated synchrotron beams, offering high resolution in reciprocal space. In the ventricular wall, however, the muscle tissue is compartmentalized into smaller branched cardiomyocytes, with a higher degree of disorder. We show that X-ray diffraction is now also capable of resolving the structural organization of actomyosin in single isolated cardiomyocytes of the ventricular wall. In addition to the hexagonal arrangement of thick and thin filaments, the diffraction signal of the hydrated and fixated cardiomyocytes was sufficient to reveal the myosin motor repeat (M3), the troponin complex repeat (Tn), and the sarcomere length. The sarcomere length signal comprised up to 13 diffraction orders, which were used to compute the sarcomere density profile based on Fourier synthesis. The Tn and M3 spacings were found in the same range as previously reported for other muscle types. The approach opens up a pathway to record the structural dynamics of living cells during the contraction cycle, toward a more complete understanding of cardiac muscle function.

Originally marketed as a safer tool for cigarette smoking cessation, electronic cigarettes (e-cigarettes) and vaping devices have now been associated with e-cigarette/vaping associated lung injury (EVALI). Given recent EVALI cases despite the absence of diacetyl and vitamin E acetate, it is reasonable to pursue an alternative cause of EVALI. We hypothesized that fragments of metal from the e-cigarette coil are aerosolized when heated and inhaled by the user, causing an accumulation of metals, such as titanium, in the lungs. Inhalation exposure to titanium causes pulmonary inflammation and may lead to lung fibrosis. We hypothesized that pulmonary function testing with spirometry and lung volumes is not sensitive enough to detect EVALI, and that DLCO should be included in evaluations. Within six weeks of discharge from the hospital, we conducted spirometry, lung volumes, and DLCO, according to ATS guidelines. Results were compared to each patient’s predicted value. Subjects underwent lung biopsy when clinically indicated. Lung tissue from one patient was examined for metals, including titanium, zinc, calcium, iron, chromium, and copper at Brookhaven National Laboratory’s National Synchrotron Light Source II Beam ID-5. Titanium, chromium, zinc, calcium, iron, and copper among other metals, were detected in the lung tissue of a subject with EVALI. This suggests the titanium from e-cigarette coils was burned and aerosolized prior to inhalation. Thus, we present a previously undescribed mechanism for EVALI. A similar finding of inhalational lung injury was observed in patients diagnosed with Iraq Afghanistan War Lung Injury (IAW-LI), who were exposed to burn pit plumes and found to have oxidized titanium in lung tissue. Subjects who vaped more frequently and for a longer duration had a lower DLCO, increased necessity of undergoing a lung biopsy, and a longer duration of hospitalization. Spirometry and lung volumes were not sensitive enough to detect EVALI. Vaping is not safe and may result in respiratory failure and hospitalization. To date, the mechanism of lung injury associated with EVALI is incompletely understood. However, we have detected an association between EVALI and deposition of metals in the lungs. The severity of EVALI correlates with more frequent and longer duration of vaping and lower DLCO. We believe serial DLCO testing to determine if these changes are acute or chronic in nature. Considering the rapid increase in popularity of vaping, especially among adolescents with developing lungs, there is notable concern for a future epidemic of vaping-related lung injury.

Organ-on-chip technology is a powerful tool for in vitro modeling. Combining it with organoids overcomes lumen inaccessibility while preserving cellular diversity and function of the intestinal epithelium. Here, we present a protocol for generating and analyzing organ-on-chips using human and mouse intestinal organoids. This protocol covers organoid line establishment, single-cell dissociation, chip preparation, and seeding. It outlines procedures for permeability assays, RNA isolation, staining, and imaging. Additionally, we describe independent stimulation and sampling of the apical and basal side.

In this study, we optimized the dissociation of synovial tissue biopsies for single-cell omics studies and created a single-cell atlas of human synovium in inflammatory arthritis. The optimized protocol allowed consistent isolation of highly viable cells from tiny fresh synovial biopsies, minimizing the synovial biopsy drop-out rate. The synovium scRNA-seq atlas contained over 100,000 unsorted synovial cells from 25 synovial tissues affected by inflammatory arthritis, including 16 structural, 11 lymphoid, and 15 myeloid cell clusters. This synovial cell map expanded the diversity of synovial cell types/states, detected synovial neutrophils, and broadened synovial endothelial cell classification. We revealed tissue-resident macrophage subsets with proposed matrix-sensing (FOLR2+COLEC12high) and iron-recycling (LYVE1+SLC40A1+) activities and identified fibroblast subsets with proposed functions in cartilage breakdown (SOD2highSAA1+SAA2+SDC4+) and extracellular matrix remodeling (SERPINE1+COL5A3+LOXL2+). Our study offers an efficient synovium dissociation method and a reference scRNA-seq resource, that advances the current understanding of synovial cell heterogeneity in inflammatory arthritis.

The identification of patient-derived, tumor-reactive T cell receptors (TCRs) as a basis for personalized transgenic T cell therapies remains a time- and cost-intensive endeavor. Current approaches to identify tumor-reactive TCRs analyze tumor mutations to predict T cell activating (neo)antigens and use these to either enrich tumor infiltrating lymphocyte (TIL) cultures or validate individual TCRs for transgenic autologous therapies. Here we combined high-throughput TCR cloning and reactivity validation to train predicTCR, a machine learning classifier that identifies individual tumor-reactive TILs in an antigen-agnostic manner based on single-TIL RNA sequencing. PredicTCR identifies tumor-reactive TCRs in TILs from diverse cancers better than previous gene set enrichment-based approaches, increasing specificity and sensitivity (geometric mean) from 0.38 to 0.74. By predicting tumor-reactive TCRs in a matter of days, TCR clonotypes can be prioritized to accelerate the manufacture of personalized T cell therapies.

Human intestinal epithelial cells are the interface between luminal content and basally residing immune cells. They form a tight monolayer that constantly secretes mucus creating a multilayered protective barrier. Alterations in this barrier can lead to increased permeability which is common in systemic lupus erythematosus (SLE) patients. However, it remains unexplored how the barrier is affected. Here, we present an in vitro model specifically designed to examine the effects of SLE on epithelial cells. We utilize human colon organoids that are stimulated with serum from SLE patients. Combining transcriptomic with functional analyses revealed that SLE serum induced an expression profile marked by a reduction of goblet cell markers and changed mucus composition. In addition, organoids exhibited imbalanced cellular composition along with enhanced permeability, altered mitochondrial function, and an interferon gene signature. Similarly, transcriptomic analysis of SLE colon biopsies revealed a downregulation of secretory markers. Our work uncovers a crucial connection between SLE and intestinal homeostasis that might be promoted in vivo through the blood, offering insights into the causal connection of barrier dysfunction and autoimmune diseases.

H3K27M, a driver mutation with T and B cell neoepitope characteristics, defines an aggressive subtype of diffuse glioma with poor survival. We functionally dissect the immune response of one patient treated with an H3K27M peptide vaccine who subsequently entered complete remission. The vaccine robustly expanded class II human leukocyte antigen (HLA)–restricted peripheral H3K27M-specific T cells. Using functional assays, we characterized 34 clonally unique H3K27M-reactive T cell receptors and identified critical, conserved motifs in their complementarity-determining region 3 regions. Using detailed HLA mapping, we further demonstrate that diverse HLA-DQ and HLA-DR alleles present immunogenic H3K27M epitopes. Furthermore, we identified and profiled H3K27M-reactive B cell receptors from activated B cells in the cerebrospinal fluid. Our results uncover the breadth of the adaptive immune response against a shared clonal neoantigen across multiple HLA allelotypes and support the use of class II–restricted peptide vaccines to stimulate tumor-specific T and B cells harboring receptors with therapeutic potential.

Background A recurrent point mutation at position 27 in the histone-3 gene (H3K27M) defines a distinct subtype of highly aggressive diffuse midline gliomas (DMG) characterized by high mortality and morbidity rates. Despite the high clinical need and several clinical trials focusing on novel treatments, effective treatments remain limited. A vaccine targeting the neoepitope H3K27M has been shown to induce a mutation-specific CD4⁺ T-cell response and to control H3K27M-mutated syngeneic tumors in an MHC humanized mouse model. Methods We have developed a protocol to expand neoantigen-reactive T cells from patients with H3K27M-mutated diffuse midline gliomas not eligible for the currently ongoing Phase 1 clinical trial (ClinicalTrials.gov Identifier: NCT04808245) and vaccinated with a mutant long peptide vaccine (H3-vac) to recover T-cell receptors (TCR) using combined single cell RNA and VDJ sequencing. We established a pipeline to clone TCRs from expanded populations of T cells and test their neoepitope specificity. Applying a novel DNA assembly process, we significantly decreased costs, allowing us to test more than 200 TCRs, derived from three different patients, for neoepitope specific reactivity using a co-culture assay with immortalized patient-derived B cell lines. Results Employing a panel of different healthy-donor-derived immortalized B cell lines and by establishing CRISPR/Cas9-mediated HLA knock-out lines we determined the HLA-restrictions of all identified H3.3K27M reactive TCRs. We demonstrated that the H3K27M peptide vaccine induces an immune response across diverse HLA alleleotypes: the deconvoluted HLA restrictions of cognate CD4+ TCRs demonstrate restrictions to members of multiple class II HLA lociAdditionally, we confirmed the presence of vaccine-induced mutation-specific CD4⁺ TCRs in a cerebrospinal fluid (CSF) sample from a vaccinated patient. The functionality of the CD4+ T cell clones identified in this study is further underlined by their central memory phenotype and the prominent cluster of activated B cells expressing H3K27M-specific BCRs in the CSF. We validated these B cells to encode H3K27M-specific antibodies. Cloning and testing two B cell receptors (BCRs) recovered from single B cells revealed mutation-specific binding of H3K27M peptide as well as full length protein. Conclusions Taken together, our data show that a neoepitope vaccine targeting H3K27M not only induces a CD4⁺ specific T-cell response across various HLA alleles but also induces a mutation-specific B-cell response with development of H3K27M-targeting antibodies.

Background Rheumatoid arthritis (RA) has a variable impact on different synovial joints, with inflammation being more commonly observed in some joints than others. Emerging evidence suggests that the anatomical variation in pathophysiology could result in differential responses to treatments across the joints, both within and between modes of action. This analysis aimed to characterize joint-specific responses to tofacitinib and methotrexate monotherapy in patients with RA. Methods This was a post hoc analysis of data from the phase III trial ORAL Start (NCT01039688), in methotrexate-naïve patients with RA. A paired joint pathology score (PJPS), derived from bilateral tender/swollen joint counts, was calculated. The percentage change from baseline in PJPS (%∆PJPS) and treatment-specific responses (tofacitinib 5 and 10 mg twice daily [BID] vs methotrexate; tofacitinib 5 vs 10 mg BID) for each patient joint pair, except for those with baseline/post-baseline PJPS = 0, were calculated at month 3, month 6, and month 12. Radiographic progression was similarly assessed using the Modified Total Sharp Score at month 6 and month 12. Results In methotrexate-naïve patients, differences in %∆PJPS demonstrated greater responses with tofacitinib vs methotrexate in most joint locations. Lesser responses with tofacitinib vs methotrexate were observed in most joints of the feet, particularly at month 12. Despite this, radiographic progression at month 12 was significantly worse in the foot (and metacarpophalangeal) joints of patients receiving methotrexate vs tofacitinib. Conclusion We observed variation in joint-specific responses with tofacitinib and methotrexate monotherapy. Despite a proximal–distal efficacy gradient, with better clinical responses in the feet, patients receiving methotrexate monotherapy demonstrated more radiographic progression in the foot joints compared with those receiving tofacitinib. These findings suggest that body site- and therapy-specific characteristics may interact to produce differential treatment responses. Trial registration ClinicalTrials.gov, NCT01039688.

BACKGROUND Clonal and mutually exclusive substitution of lysine 27 to methionine in histone H3-3A (H3K27M) defines an aggressive subtype of diffuse glioma. These diffuse midline gliomas (DMG) are aggressive, incurable primary brain tumors that occur predominantly in a midline location in children and young adults. Surgical treatment options remain limited, response to chemoradiation is poor and palliative radiotherapy remains the only intervention with proven benefit, resulting in a median overall survival between 10 and 15 months after initial diagnosis. Previous studies have shown that a H3K27M-specific long peptide vaccine (H3K27M-vac) induces mutation-specific immune responses capable of controlling H3K27M+ tumors in a major histocompatibility complex (MHC)-humanized mouse model. MATERIAL AND METHODS Here we describe the first-in-human treatment with H3K27M-vac of eight adult patients with recurrent H3K27M+ DMG. Induction of mutation-specific immune responses was determined in peripheral blood using Interferon-γ ELISpot assays. Class II HLA-restricted peripheral H3K27M-specific T cells were expanded in vitro to characterize and track unique H3K27M-reactive T cell responses in blood and cerebrospinal fluid (CSF). RESULTS Repeat vaccinations with H3K27M-vac were safe and induced mutation-specific immune responses in five of eight patients after a median of two vaccinations. Median progression free survival after start of vaccination was 6.2 months and median overall survival was 12.8 months. In vitro restimulation of peripheral CD4+ and CD8+ T cells with H3 mutant peptide or wild type peptide revealed that H3K27M specific immune responses are CD4-mediated and can be suppressed by blocking antibodies against MHC II, but not MHC I. Furthermore, proximity ligation assay of formalin-fixed paraffin-embedded primary tumor tissue revealed that H3K27M neoepitope co-localizes with HLA class II-DR. One patient showed radiographic pseudoprogression (PsPD) according to iRANO criteria within six weeks after first detection of mutation-specific peripheral immune response. Following PsPD, three out of the top ten vaccine-induced, K27M-expanded CD4+ TCR clonotypes from peripheral blood were detectable in the CSF of this patient who went into sustained complete remission for > 31 months. CONCLUSION H3K27M-vac is safe and induces H3K27M-specific CD4 T cell responses in patients with H3K27M+ DMG. An ongoing multicenter, phase 1 clinical trial for adult patients with newly diagnosed H3K27M+ DMG will assess the safety and immunogenicity of H3K27M-vac in combination with atezolizumab in standard-of-care radiotherapy (NCT04808245).

Generative modeling and representation learning of tandem mass spectrometry data aim to learn an interpretable and instrument-agnostic digital representation of metabolites directly from MS/MS spectra. Interpretable and instrument-agnostic digital representations would facilitate comparisons of MS/MS spectra between instrument vendors and enable better and more accurate queries of large MS/MS spectra databases for metabolite identification. In this study, we apply generative modeling and representation learning using variational autoencoders to understand the extent to which tandem mass spectra can be disentangled into their factors of generation (e.g., collision energy, ionization mode, instrument type, etc.) with minimal prior knowledge of the factors. We find that variational autoencoders can disentangle tandem mass spectra data with the proper choice of hyperparameters into meaningful latent representations aligned with known factors of variation. We develop a two-step approach to facilitate the selection of models that are disentangled, which could be applied to other complex and high-dimensional data sets.