Biology subjects

Jenkins, M. R.

Publications and source records attributed to Jenkins, M. R..

6 recordsLinked to original sources

bioRxiv2026

CXCL9-Dependent Immune Networks of Spontaneous Tertiary Lymphoid Structures in a Murine Subcutaneous Tumor Model

Recent discoveries have identified human tumor tertiary lymphoid structures (TLS) as organized clusters of innate and adaptive immune cells and stromal cells. Cancer-associated TLS can promote tumor clearance and enhance responses to immune checkpoint blockade (ICB) immunotherapy. However, the absence of TLS in conventional mouse tumor models, particularly subcutaneous transplant models, limits experimental manipulation and study of TLS development. By tuning tumor antigen expression, we have generated a subcutaneous tumor model that naturally forms TLS within the immune-hot, vasculature-rich peritumoral adipose tissue. These TLS were identified as clusters of B and T cells in close proximity to high endothelial venules. TLS organization was underpinned by spatially compartmentalized chemokine expression, with CCL21 produced in the T cell zone of perivascular TLS and CXCL13 present in the B cell region. Single cell transcriptomic sequencing showed that the immune composition in the TLS forming peritumoral tissue resembled human cancer TLS signatures. Upon anti-PD-L1 ICB treatment, TLS underwent expansion, which was associated with increased interferon responses in macrophages and coincided with tumor clearance. We further identified the interferon-inducible chemokine CXCL9 as a key regulator of the TLS microenvironment. CXCL9-expressing macrophages populated TLS, interacting with B cells following ICB treatment. CXCL9 deficiency reduced B cell recruitment to peritumoral tissue and TLS formation. Further T and B cell receptor sequencing indicated a CXCL9-dependent promotion of antigen-specific responses within the tumor and TLS niche. Combined, these findings establish a tractable subcutaneous model of naturally arising tumor TLS and define the cellular and molecular networks that regulate their formation and function. This provides a platform to dissect how TLS develop and test strategies that promote TLS formation to enhance cancer therapy.

immunology↗

bioRxiv2026

Active Learning Enables Efficient Directed Evolution of a Far-Red Fluorescent Protein with Minimal Experimental Data

Fluorescent proteins are fundamental tools for cellular imaging. Most fluorescent proteins in routine use, including GFP, are derived from the jellyfish Aequorea victoria and emit blue-green light, which is strongly absorbed and scattered by tissue, limiting imaging depth. Far-red and near-infrared fluorescent proteins, engineered from bacteriophytochromes, address this limitation because far-red light penetrates tissue considerably further. However, these proteins are typically much dimmer than their A. victoria -derived counterparts. Improving brightness by conventional directed evolution requires screening large random mutant libraries, a process that is slow, labor-intensive, and often impractical outside specialized laboratories. We utilized an active-learning-guided directed evolution workflow that identified improved variants from substantially less data than conventional screening. Each round coupled automated, miniaturized cell-free protein expression directly from a DNA template without cloning or cell culture, with a machine-learning model retrained on cumulative sequence-function data to nominate the most informative variants for the next round. Applied to miRFP670nano3, this workflow screened 120 variants across successive rounds and identified twelve with improved brightness, the best four-fold brighter in bacterial systems. However, these gains did not translate when the variants were evaluated in mammalian cells, indicating that performance can be strongly dependent on cellular context. Retrospective simulation across benchmark datasets from ProteinGym showed that performing more experimental batches with fewer samples per batch consistently accelerated convergence to high-fitness sequences. Incorporating protein-language-model derived zero-shot fitness priors also accelerated convergence, but only in proportion to how well each prior score correlated with the true fitness landscape. Together, these findings established generalizable design rules, favoring smaller acquisition batches and confidence-weighted priors, for engineering proteins from minimal experimental data. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=55 SRC="FIGDIR/small/744534v1_ufig1.gif" ALT="Figure 1"> View larger version (11K): org.highwire.dtl.DTLVardef@14cd238org.highwire.dtl.DTLVardef@7d8334org.highwire.dtl.DTLVardef@30e9f2org.highwire.dtl.DTLVardef@14f3a0c_HPS_FORMAT_FIGEXP M_FIG C_FIG

synthetic biology↗

bioRxiv2025

Age-dependent tumor-immune interactions underlie immunotherapy response in pediatric cancer

Pediatric cancers originate in rapidly growing tissues within the context of a developing host. However, the interactions between cancer cells and the developing immune system are incompletely understood. Here, we established a suite of pediatric syngeneic mouse cancer models across diverse anatomical sites and compared their tumor immune microenvironment with that in adult mice. Tumors in pediatric mice exhibited significantly accelerated growth and diminished leukocyte infiltration, dominated by naive-like PD-1low/CD8+ T cells, and proliferative MHCIIlow/PD-L1hi/CD86low macrophages. Tumor-infiltrating leukocytes in pediatric mice were enriched for MYC targets, which was also observed in pediatric patient samples. Furthermore, pediatric mice displayed poor responses to anti-PD-1/PD-L1 or bispecific T cell engager antibodies, which could be reversed by inducing a proinflammatory microenvironment via MYC inhibition or inducing macrophage polarization to an MHCIIhi phenotype. These findings underscore the significant influence of young age on cancer immune responses and reveal potential new therapeutic opportunities for pediatric cancers. HIGHLIGHTSO_LIAllograft tumors exhibit markedly accelerated growth in pediatric hosts compared to adults. C_LIO_LITumors growing in pediatric mice have reduced leukocyte infiltration, dominated by naive-like PD-1low/CD8+ T cells, and MHCIIlow/M2-like macrophages. C_LIO_LIEnrichment of MYC target genes is observed in pediatric mouse tumors and confirmed in primary patient tumor samples. C_LIO_LIPediatric mice display reduced response to anti-PD-1/PD-L1 and BiTE immunotherapy, which can be reversed by remodeling the TIME, using either MYC inhibition or macrophage polarization. C_LI

cancer biology↗

bioRxiv2024

TIRE-seq: an Integrated Sample Extraction and Transcriptomics Workflow for High Throughput Perturbation Studies

RNA sequencing (RNA-seq) is widely used in biomedical research, advancing our understanding of gene expression across biological systems. Traditional methods require upstream RNA extraction from biological inputs, adding time and expense to workflows. We developed TIRE-seq (Turbocapture Integrated RNA Expression Sequencing) to address these challenges. TIRE-seq integrates mRNA purification directly into library preparation, eliminating the need for a separate extraction step. This streamlined approach reduces turnaround time, minimizes sample loss, and improves data quality. A comparative study with the widely used Prime-seq protocol demonstrates TIRE-seqs superior sequencing efficiency with crude cell lysates as inputs. TIRE-seqs utility was demonstrated across three biological applications. It captured transcriptional changes in stimulated human T cells, revealing activation-associated gene expression profiles. It also identified key genes driving murine dendritic cell differentiation, providing insights into lineage commitment. Lastly, TIRE-seq analyzed the dose-response and time-course effects of temozolomide on patient-derived neurospheres, identifying differentially expressed genes and enriched pathways linked to the drugs mechanism of action. With its simplified workflow and high sequencing efficiency, TIRE-seq offers a cost-effective solution for large-scale gene expression studies across diverse biological systems.

genomics↗

bioRxiv2020

De novo designed receptor transmembrane domains enhance CAR-T cell cytotoxicity and attenuate cytokine release

De novo designed receptor transmembrane domains (TMDs) present opportunities for precise control of cellular receptor functions. We developed a de novo design strategy for generating programmed membrane proteins (proMPs): single-pass -helical TMDs that self-assemble through computationally defined and crystallographically validated interfaces. We used these proMPs to program specific oligomeric interactions into a chimeric antigen receptor (CAR) and found that both in vitro CAR T cell cytokine release and in vivo antitumor activity scaled linearly with the oligomeric state encoded by the receptor TMD, from monomers up to tetramers. All programmed CARs (proCARs) stimulated substantially lower T cell cytokine release relative to the commonly used CD28 TMD, which we show elevated cytokine release through lateral recruitment of the endogenous T cell costimulatory receptor CD28. Precise design using orthogonal and modular TMDs thus provides a new way to program receptor structure and predictably tune activity for basic or applied synthetic biology.

synthetic biology↗

bioRxiv2019

Digitally Deconstructing Leaves in 3D Using X-ray Microcomputed Tomography and Machine Learning

Premise of the studyX-ray microcomputed tomography (microCT) can be used to measure 3D leaf internal anatomy, providing a holistic view of tissue organisation. Previously, the substantial time needed for segmenting multiple tissues limited this technique to small datasets, restricting its utility for phenotyping experiments and limiting our confidence in the conclusion of these studies due to low replication numbers. Methods and ResultsWe present a Python codebase for random-forest machine learning segmentation and 3D leaf anatomical trait quantification which dramatically reduces the time required to process single leaf microCT scans into detailed segmentations. By training the model on each scan using 6 hand segmented image slices out of >1500 in the full leaf scan, it achieves >90% accuracy in background and tissue segmentation. ConclusionOverall, this 3D segmentation and quantification pipeline can reduce one of the major barriers to using microCT imaging in high-throughput plant phenotyping.

plant biology↗