Biology subjects

Subramanian, S.

Publications and source records attributed to Subramanian, S..

At least 19 recordsLinked to original sources

bioRxiv2026

Stemness overcomes IFNγ-induced antitumour exhaustion in human macrophages

Macrophages hold great promise for antitumour therapies (1-4), particularly owing to their ability to infiltrate solid tumours (5). Yet therapeutic success has been limited (6,7), because macrophages exhibit rapid exhaustion of antitumorigenic activity in the tumour (5,8,9). Here, we induced a stem-like self-renewal programme in human induced pluripotent stem cell (iPSC)-derived macrophages by deleting the self-renewal restricting transcription factors MAFB and MAF (MAF-DKO). Human MAF-DKO macrophages showed activated stem cell signatures, increased clonogenicity and proliferative potential. Increased self-renewal capacity was accompanied by enhanced mitochondrial fitness and invigorated energy metabolism. Importantly, these stemness features prevented exhaustion, cellular aging-associated parameters and loss of energy efficiency that are typically associated with proinflammatory macrophage activation. In addition, activated MAF-DKO macrophages showed a broad loss of IFN{gamma}-triggered immunosuppressive pathways (10,11), induced a super-activated state with robust cytotoxic activity and resisted exhaustion upon tumour cell exposure. Consequently, activated MAF-DKO macrophages showed intratumoral persistence with a stable antitumoral phenotype in vivo, conferred robust reduction of tumour burden and extended survival of tumour-bearing mice. Together, this demonstrates that the induction of stemness in human macrophages can overcome critical roadblocks of antitumour therapy by conferring resistance to activation-triggered exhaustion of antitumour activity in the TME.

immunology↗

bioRxiv2026

Relatedness of Gardnerella prophages and their endolysins

The Gardnerella genus includes a complex sub-taxonomy of bacterial species and subgroups which have been implicated in bacterial vaginosis. We analyzed G. leopoldii, G. swidsinski, G. piotii, and G. vaginalis genomes available in public databases and predicted proviral sequences using Cenote-Taker3. We clustered 266 prophage sequences into 140 viral genera level groupings that encode four distinct gene clusters and 19 different lysin proteins. Analysis of the bacterial and prophage sequence phylogenies reveals that while there were no specific prophage sequence lineages that infected each bacterial host, there is clustering of the prophage sequences by bacterial species within the Gardnerella genus.

microbiology↗

bioRxiv2026

A rational design strategy and validation for protease-resistant fusion-inhibitor antiviral peptides

Peptide-based fusion inhibitors are promising pharmaceuticals in the fight against enveloped viruses relying on membrane fusion for host infection. However, peptide therapeutic applications have long been hindered by their poor stability in vivo. Here, we discovered that peptide inhibitors with the wildtype sequence of the heptad repeat 2 (HR2) domain of the SARS-CoV-2 spike protein are efficiently cleaved by Transmembrane Protease, Serine 2 (TMPRSS2), a key protease involved in the SARS-CoV-2 virus-cell fusion pathway. We then identified the corresponding cleavage sites and designed three protease-resistant peptides using ranking based on deep mutational scanning and natural occurrence. The three candidates all exhibit inhibitory activity in a cell-cell fusion assay. A high-resolution cryo-EM structure of the top candidate, HR2-NHN, bound to its HR1 target reveals the molecular basis for its potent activity. The top candidate of the cell-based screening assay significantly improved efficacy relative to the wildtype peptide when administered 12 h before infection in both an authentic virus-cell infection assay and a mouse assay. More broadly, our results suggest that the design strategies for protease-resistant peptides could be applied to a broad spectrum of other enveloped viruses and pave the way for the development of safe, prophylactic antivirals that can be administered before exposure.

microbiology↗

bioRxiv2026

Towards a Physiological Scaling Law: Model Quality vs. Cohort Size for Stochastic Sequence Data

Scaling laws help determine the optimal data size for training large models but are established in domains where the target is deterministic. Physiological signals are different: heartbeat sequences are stochastic, so part of the error is irreducible even with large amounts of data. Metrics such as MAE do not account for non-deterministic behavior, and therefore assessing scaling requires evaluating distributional calibration (measuring how well predicted probability densities capture true conditional characteristics). We formulate a scaling law metric(n) = E + A n- and evaluate it with five metrics: accuracy (MAE, RMSE), distributional calibration (KS distance, goodness-of-fit), and training objective (negative log loss) using a neural temporal point process trained on a cohort of four-ECG datasets. The law fits all five metrics. While point accuracy is near saturation at n = 183, KS distance and goodness-of-fit improve by 6% and 12% respectively when extrapolated to 10,000 subjects, showing that scaling decisions in stochastic domains must be guided by distributional calibration rather than point accuracy.

physiology↗

bioRxiv2026

Lognormal Neural Point Process Models for Interpretable Heartbeat Dynamics

Neural temporal point processes (NTPPs) are powerful tools for modeling sequences of timestamped events with statistical temporal structure. Density-based NTPPs, in particular, are an interesting opportunity to merge the universal function approximation capability of neural networks with a defined statistical model in a way that has many potential applications. We demonstrate one such application to heartbeat dynamics, a physiologic point process. We specifically apply a lognormal mixture NTPP to compute instantaneous estimates of the mean and standard deviation of beat-to-beat intervals. We compare our results to the state of art (Barbieri et al.) point process model for heartbeat dynamics, which uses a more physiologically rigorous inverse Gaussian model. We find that the NTPP model maintains reasonable accuracy while improving upon robustness to noise.

physiology↗

bioRxiv2026

A Neurotensin Brake on Exploratory Drive under Persistent Threat

Avoidance behavior is an adaptive response that delays exploration to promote survival. Avoidance is enhanced by psychological stress and is a hallmark of many neuropsychiatric disorders. Neural circuits that control avoidance by integrating stressful stimuli and modulating exploratory behavior remain underexplored. Elucidating the functional dynamics of this highly conserved phenomenon and the underlying neural mechanism of avoidance is an important open-ended question, with relevance to understanding both innate behaviors and neuropsychiatric disorders. Using predator odor as an innate, chronic stressor to increase avoidance behaviors in mice, we identified a neural population in the lateral septum (LS) that integrates threat information and modulates latency to explore. Calcium recordings in freely exploring mice combined with activity-based transcriptomics revealed that predator-responsive LS neurons are GABAergic and express neurotensin (LSNT). Further, single nuclei RNA-seq analysis revealed that among predator-responsive neurons, NT-enriched inhibitory clusters are predominant. Chronic activation of LSNT neurons induces avoidance behaviors in the absence of predator odor, while synaptic silencing of this population abrogates predator-enhanced avoidance. Using transgenic mouse models to indelibly tag predator-responsive neurons, we defined the downstream circuit that connects the encoding of predator odor information to the lateral hypothalamus. Projection-specific activation of LSNT[->]LHA neurons recapitulate stress-induced avoidance behaviors in mice. Finally, we showed that deletion of neurotensin from LS neurons prevented the effects of predator odor on exploration. Together, these findings offer a genetically-and projection-defined, top-down circuit linking the limbic neurotensinergic system to chronic psychological stress and avoidance behaviors in mice.

neuroscience↗

bioRxiv2026

Antibody-drug conjugate combination therapy targeting LGR5 and MET with different payloads enhances efficacy in preclinical colorectal cancer models

Leucine-rich repeat-containing G protein-coupled receptor 5 (LGR5) is a marker of cancer stem-like cells frequently upregulated in colorectal cancer (CRC) with lower expression in normal tissues, making it a favorable target for antibody-drug conjugates (ADCs). ADCs combine antibody specificity with potent cytotoxic payloads to enhance efficacy while minimizing systemic toxicity. LGR5-targeting ADCs incorporating different payloads demonstrate strong initial tumor inhibition, yet tumors eventually recur due in part to LGR5 downregulation, necessitating more effective strategies to prevent relapse. We show treatment with chemotherapies or an LGR5-targeting ADC coupled to a topoisomerase 1 inhibitor payload (8E11-CPT2) reduces LGR5 levels and increases MET receptor expression and/or activation in CRC cells, supporting a therapeutic approach to target LGR5 and MET simultaneously. Accordingly, we engineered a MET-targeting ADC (ABT-700-SG3199) via site-specific conjugation of the anti-MET antibody telisotuzumab (ABT-700) with the DNA-crosslinking pyrrolobenzodiazepine (PBD) dimer SG3199. ABT-700-SG3199 exhibited superior potency and efficacy in CRC models compared to the clinical-stage MET-targeting ADCs ABBV-399 and ABBV-400, which use the same antibody backbone conjugated to different payloads. Treatment with ABT-700-SG3199 increased LGR5 expression, and the combination of ABT-700-SG3199 with 8E11-CPT2 enhanced CRC cell-killing efficacy, reinforcing the rationale for a dual-targeting approach. In patient-derived xenografts, combined administration of 8E11-CPT2 and ABT-700-SG3199 markedly delayed tumor relapse and prolonged survival compared with single-agent treatment. Taken together, these findings reveal a reciprocal regulation between MET and LGR5 in response to LGR5- or MET-targeting ADCs in CRC models and support a LGR5/MET dual-targeting therapeutic strategy to enhance efficacy and potentially overcome resistance and relapse.

cancer biology↗

bioRxiv2026

Stroke-induced lipocalin-2-expressing red pulp macrophages reprogram peripheral immunity

Acute ischemic stroke (AIS) induces profound systemic immune alterations that contribute to infection susceptibility. Here, we identify lipocalin-2 (LCN-2) as a rapidly induced and conserved regulator of stroke-associated immunosuppression. Using 3-MACE-Seq, cytokine profiling, and immunofluorescence in C57BL/6J mice subjected to transient middle cerebral artery occlusion (tMCAO), we show that LCN-2 is strongly upregulated in splenic red pulp macrophages (RPMs) within 24 hours and again 7 days post-tMCAO. LCN-2-expressing RPMs form immunological synapses with CD3+ T cells, thereby impacting T cell trafficking. Recombinant LCN-2 directly reprogrammed T cells and monocytes toward hyporesponsive, tolerogenic phenotypes by suppressing inflammatory cytokines, impairing chemotaxis, enhancing phagocytosis, and uncoupling oxidative burst. Human spleens likewise displayed LCN-2-expressing CD68+ RPMs, and LCN-2 preconditioning of monocytes reproduced reduced HLA-DR, CD80, CD206, and ROS with increased uptake of E. coli bioparticles. These findings identify LCN-2 signaling as a central orchestrator of stroke-induced peripheral immunoreprogramming and a potential therapeutic target to mitigate post-stroke immunodepression. SummaryAcute ischemic stroke induces LCN-2 in splenic red pulp macrophages, which reprogram T cells and monocytes toward tolerogenic, hyporesponsive states. Mouse and human data identify LCN-2 as a driver of peripheral immunodepression and a potential target to reduce infection risk. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=143 SRC="FIGDIR/small/733904v1_ufig1.gif" ALT="Figure 1"> View larger version (47K): org.highwire.dtl.DTLVardef@fb0caforg.highwire.dtl.DTLVardef@1fdcd41org.highwire.dtl.DTLVardef@1ae99b0org.highwire.dtl.DTLVardef@1ec0837_HPS_FORMAT_FIGEXP M_FIG Graphical abstract C_FIG

immunology↗

bioRxiv2026

Characterization of Electrophysiological and Transcriptomic Alterations in Patient-Derived Neurons from CHAMP1 Syndrome

Mutations in chromosome alignment maintaining phosphoprotein 1 (CHAMP1) have been linked to neurodevelopmental disorders characterized by intellectual disability, developmental delay, and autism spectrum disorder. However, the cellular and electrophysiological mechanisms by which CHAMP1 mutations disrupt human neuronal development remain poorly understood. In the present study, we used patient-derived induced pluripotent stem cells (iPSCs) carrying two pathogenic CHAMP1 mutations and generated neural progenitor cells (NPCs) and excitatory neurons to investigate the effects of each mutation on neuronal maturation and function, DNA repair, and gene expression. Proliferative capacity declines with CHAMP1 dosage, while DNA repair dysfunction is allele-specific. Whole-cell patch-clamp electrophysiology revealed that CHAMP1 mutant neurons exhibit significant alterations in intrinsic membrane properties during early developmental stages, including depolarized resting membrane potential, reduced action potential firing, and impaired waveform kinetics. These functional deficits were accompanied by reduced sodium and potassium current densities, suggesting impaired ion channel accumulation during neuronal maturation. Furthermore, recordings of spontaneous excitatory postsynaptic currents indicated altered synaptic activity and reduced proportions of synaptically active neurons. Morphological analyses showed that CHAMP1-deficient neurons exhibit impaired neurite outgrowth and branching, supporting a defect in neuronal maturation. Single-nucleus transcriptomic profiling further revealed delayed developmental trajectories and mutation-specific dysregulation of synaptic gene programs enriched for autism, ADHD, and epilepsy risk genes. Together, these findings demonstrate that CHAMP1 mutations disrupt multiple aspects of neuronal development, including homologous recombination (HR) dysfunction in NPCs, membrane excitability, ion channel function, and synaptic connectivity. Our results provide insights into the neurobiological consequences of CHAMP1 mutations and establish patient-derived neurons as a platform to investigate cellular pathophysiology and potential therapeutic strategies for CHAMP1-associated neurodevelopmental disorders.

neuroscience↗

bioRxiv2026

Towards Passive Dietary Monitoring: Dilated CNN-Based Meal Detection Using Ambulatory High-Resolution Electrogastrography

BackgroundObjective tools for longitudinal dietary monitoring, despite its importance in the health triad of diet, sleep, and exercise, remain limited. Widely available ambulatory options include continuous glucose monitoring, which is a delayed response, and manual logging, which is rife with human error. Clinical tools such as gastric emptying scintigraphy are impractical for everyday use. High-resolution electrogastrography (HR-EGG) offers an alternative by treating gastric myoelectric activity as a biomarker of digestive state. However, its utility for ambulatory meal detection remains unclear. We hypothesize that HR-EGG and accelerometry together encode distinct postprandial gastric signatures to enable automated meal detection, and that postural context represents a relevant source of variation in signal detectability. MethodsHR-EGG and accelerometry data were collected from seven healthy adults across sixteen 150-minute meal sessions under IRB-approved protocol. Each session included a 30-minute fasted baseline, consumption of a standardized meal, and 90-minute postprandial period of sitting, walking, and lying in a randomized order. Features of the gastric slow wave, including raw and normalized bandpower, phase gradient directionality (PGD), wave direction, and wave speed, were extracted alongside triaxial accelerometer magnitude. A dilated one-dimensional convolutional network (1D CNN) was trained to classify meal consumption at five-minute resolution using leave-one-subject-out cross-validation. Postural effects on gastric myoelectric metrics were assessed using the Friedman test. ResultsThe model achieved a mean AUROC of 0.925 (95% CI: [0.857, 0.993]) and mean AUPRC of 0.824 (95% CI: [0.668, 0.980]; null model: 0.20). Feature ablation showed PGD as the most informative input ({Delta}AUPRC = -0.188), with wave propagation speed the least informative ({Delta}AUPRC = -0.105). Walking produced the highest signal-to-noise ratio (9.94 dB), lying had the most stable gastric rhythm (89.1% normogastric), and sitting demonstrated the greatest frequency instability (dominant frequency standard deviation = 0.825 cpm). ConclusionA dilated 1D CNN applied to spatiotemporal HR-EGG features enables temporally aware passive meal detection across ambulatory contexts. This study framework addresses a gap between clinical need for objective dietary monitoring and the limitations of current detection methods.

bioengineering↗

bioRxiv2026

Evaluating Long-Range Temporal Structure in Foundation Model-Based Forecasts of Heartbeat Dynamics

We examine the long-range temporal structure of forecasts produced by Time-Series Foundation Models (TSFMs) on heartbeat dynamics using the MIT-BIH Normal Sinus Rhythm Database (NSRDB). Our findings indicate that these models do not adequately capture long-range dependencies, as reflected in growing errors in RR-interval predictions over longer forecast horizons. Code is available at https://github.com/SubramanianLab/ecg-tsfm-benchmark.

bioinformatics↗

bioRxiv2026

Redesign of energetically frustrated regions rescues function in defective T4 clamp loaders

DNA polymerase clamp loaders are AAA+ ATPases that load sliding clamps on DNA for high- speed replication. Using a platform for high-throughput mutagenesis of replication proteins in T4 bacteriophage, we carried out saturation mutagenesis of the AAA+ ATPase module of the T4 clamp loader bearing a mutation, Gln 118{lozenge}Asn (Q118N), that reduces fitness. We identified residues for which different mutations improve the fitness of the Q118N variant but are neutral in the wild-type background. These conditionally neutral "rescue hotspots" overlap with those identified earlier in another defective variant (D110C). These rescue hotspots localize to regions where the sequence is not optimal for the structure, as determined by energetic frustration analysis. We designed new sequences for three of these regions, using the protein-design algorithm ProteinMPNN. In two helical regions, several designed sequences increased the fitness of both wild-type and mutant proteins, likely due to enhanced stability. An inter-domain hinge in AAA+ module changes conformation during activation, and designs for the hinge lead to loss of fitness in the wild-type background. However, when using the active conformation as the template, designs for the hinge increase the fitness of defective variants. In contrast designs templated on the inactive conformation led to loss of fitness, suggesting that a proper conformational balance is crucial. Thus, adaptive capacity in the clamp loader resides in a network of conditionally neutral sites that enable functional tuning through shifts in stability and conformational equilibria.

biochemistry↗

bioRxiv2026

Screening metatranscriptomes for ultrastable RNA secondary structures reveals hidden bacteriophages and novel capsid nanomaterials

Metatranscriptomics has transformed our view of RNA bacteriophage diversity, revealing vast numbers of single-stranded RNA (ssRNA) phages whose protein capsids can be engineered for biotechnology applications. However, many ssRNA phages remain hidden from current detection methods, which require protein-level similarity to known phages. Here we show that RNA structure provides an additional signal for the detection of ssRNA phages in metatranscriptomes, including hidden phages missed by prior protein-based methods. By computationally folding each contig and screening for exceptionally stable RNA secondary structures, we find evidence of thousands of previously unrecognized phages encoding novel coat proteins. We express a library of 12,000 such coat proteins in E. coli and find that most assemble into nuclease-resistant capsids. We determine the 3D structure of one such capsid by cryo-electron microscopy and demonstrate that it can be disassembled and reassembled in vitro to package heterologous RNA--a key step toward repurposing these particles as RNA delivery vehicles. We compile the newly discovered ssRNA phages with previously known ones into a database that contains sequence and structural information for over 460,000 unique RNA molecules and over 100,000 distinct coat proteins, providing a comprehensive resource for microbiology and nanomaterials research.

biochemistry↗

bioRxiv2026

Structural analyses of Trichomonas vaginalis pyrophosphate-dependent phosphofructokinase (TvPPi-PFK)

Trichomonas vaginalis causes trichomoniasis, the most common non-viral sexually transmitted disease in humans. T. vaginalis pyrophosphate-dependent phosphofructokinase (TvPPi-PFK) is a putative target for rational, structure-based drug discovery, given its absence in mammals and its importance for parasite survival. TvPPi-PFK is a cytosolic enzyme that catalyzes the phosphorylation of fructose-6-phosphate using pyrophosphate (PPi) as the phosphoryl donor. This reversible reaction, catalyzed by TvPPi-PFK, is the first committed step in glycolysis. Its reverse reaction is vital for gluconeogenesis in T. vaginalis. The purification, crystallization, structure determination, and crystal structures of TvPPi-PFK are reported. TvPPi-PFK is the first reported eukaryotic PPi-PFK structure. TvPPi-PFK retains the overall PPi-PFK topology observed in bacterial PPi-PFK including conserved motifs essential for pyrophosphate binding and PPi-PFK catalytic activity. In addition to the catalytic PPi-PFK binding sites, TvPPi-PFK has two additional ligand binding sites. The first binds AMP usurped during protein production and helps stabilize the TvPPi-PFK tetramer. A second ligand binding site was observed in proximity to the AMP-binding site and accommodates sugar phosphates soaked into preformed crystals. This sugar phosphates binding site is distinct from the TvPPi-PFK active site that binds fructose-6-phosphate. Future mutagenesis and activity studies are planned to determine the relevance of both sites. SynopsisThe production, crystallization, and crystal structures of a pyrophosphate-dependent phosphofructokinase from Trichomonas vaginalis (TvPPi-PFK) are reported. TvPPi-PFK has a prototypical PPi-PFK active site as well as unexpected AMP and sugar-phosphate binding sites at the dimer interface.

biochemistry↗

bioRxiv2026

Gene network centrality affects parallel evolution and local adaptation in wild yeast

The predictability of evolution remains a fundamental question in biology. While parallel evolution has been observed across taxa, we lack a mechanistic framework for predicting when evolution will be repeatable versus contingent. Here, we test whether gene network architecture predicts evolutionary repeatability and local (mal)adaptation in natural populations. We sequenced 38 wild Hanseniaspora uvarum yeast strains from two replicated apple varieties across four Connecticut orchards, yielding 29,489 high-quality SNPs. Population genomic analyses suggested that genetic differentiation was structured primarily by orchard environment. Reciprocal transplant experiments across all orchard-variety combinations demonstrated local (mal)adaptation at multiple ecological scales, with strongest (mal)adaptive responses at the orchard-variety level. By mapping H. uvarum genes to the comprehensive Saccharomyces cerevisiae genetic interaction network, we found that network centrality affects evolutionary outcomes. Highly connected genes occupying bow-tie network positions evolved in parallel across replicated apple varieties and central genes harbored alleles with adaptive fitness effects. In contrast, peripheral genes with fewer interactions facilitated rapid, non-parallel evolution to orchard-variety interactions and were associated with local maladaptation. Analysis of 150-gene sets evolving to apple variety within each orchard revealed greater-than-random similarity at genetic, functional, and network levels, with network-level predictability significantly exceeding gene-level predictability. These results suggest that gene network architecture could provide a mechanistic foundation for determining which evolutionary changes will be repeatable, potentially enabling evolutionary predictions even when specific genes cannot be identified.

evolutionary biology↗

bioRxiv2026

Automated extraction and optimization of protein purification protocols using multi-agent large language models

Recent advances in Large Language Models (LLMs) present new opportunities for automating critical bottlenecks in scientific workflows such as literature reviews or protocol design. One such bottleneck is the purification of recombinant proteins, a vital aspect of biomedical research that frequently fails. To improve success rates, researchers must manually define optimal large-scale purification conditions and establish robust rescue protocols for proteins with low stability or solubility - a time-intensive process. To address this gap, we introduce a multi-agent LLM system that automates the creation and optimization of protein purification protocols to facilitate the production of high-concentration, high-purity protein samples. Our application streamlines the labor-intensive manual process of sequence similarity searches, literature reviews, and protocol comparison. Operating in a tool-like constrained workflow, the system identifies analogous proteins, leverages specialized LLM agents to extract successful purification methodologies from primary source literature, and cross-references them against failed protocols to generate optimization recommendations. Evaluation on a select number of targets demonstrated high accuracy in protocol extraction and the generation of scientifically sound, expert-validated optimization recommendations. While this system reduces complex analysis time from hours to minutes, we identify the lack of programmatic open access to literature, specifically primary citations in the Protein Data Bank, as a fundamental limitation to LLM agent-based scientific workflows. Ultimately, this system demonstrates the feasibility of using LLM agents to streamline wet-lab workflows while preserving methodological transparency and reproducibility.

bioinformatics↗

bioRxiv2026

Discrete Transcriptional States Define Biphasic Immune Response and Dynamic CMS Transitions in Colorectal Cancer

BackgroundSequential alterations in APC, KRAS, TP53, and SMAD4 have been proposed as a framework for colorectal cancer progression. Human colorectal cancer datasets have not revealed the biological transitions associated with these mutations. When examining a cohort of TCGA-colorectal tumors grouped as AK (APC/KRAS), AKP (APC/KRAS/TP53), and AKPS (APC/KRAS/TP53/SMAD4), we observed no significant differences in immune-cell composition, four previously defined Consensus Molecular Subtypes (CMS1/2/3/4), or transcriptomic clustering between these genomic groups. Therefore, these canonical alterations do not sufficiently characterize the known properties of metastatic progression in human colorectal cancer. MethodsTo overcome these limitations, we developed a genetically defined, organoid-based, orthotopic mouse model whereby mouse colon organoids modeling sequential APC, KRAS, TP53, and SMAD4 alterations were orthotopically injected into the colon. This was followed by RNA-sequence processing, normalization with DESeq2, differential expression, pathway enrichment, and immune/stromal inference. Gene co-expression modules were identified from variance-stabilized mouse expression data, mapped to 1:1 human orthologs, and summarized as eigengenes. A multinomial logistic regression model trained on mouse eigengenes was applied to TCGA-COAD human tumors to assign them to mouse-informed transcriptomic states (AK-like, AKP-like, AKPS-like), which were then used for downstream visualization and comparative analyses. ResultsWhole-transcriptome analysis revealed discrete transcriptional states and immune-cell differences between the organoid AK/AKP/AKPS groups. Early TP53 loss led to strong activation of immune pathways, accompanied by increased infiltration of NK and T cells. As tumors progressed with SMAD4 loss and metastasis, this immune activity collapsed, giving rise to broad immune suppression. CMS classifications also shifted, with AK tumors resembling epithelial CMS2, AKP tumors displaying immune-rich CMS1 features, and AKPS and metastatic lesions adopting mesenchymal CMS4 characteristics. We then applied a progression-based transcriptomic classifier to 460 human colorectal tumors. This reclassification revealed conserved immune remodeling, CMS transitions, pathway-level differences, and significant differences in patient survival. ConclusionWe show that organoid-derived progression profiles reveal hidden evolutionary structure in human colorectal cancer and provide a transcriptional framework for interpreting metastatic potential and clinical outcomes.

cancer biology↗

bioRxiv2026

Structural Basis of Glycolytic Control in Trypanosoma cruzi: Insights from Enolase and PGI

Trypanosoma cruzi, the etiological agent of Chagas disease, depends on glycolysis for ATP production, rendering its glycolytic enzymes attractive targets for therapeutic development. Here, we report the high-resolution crystal structures of two essential glycolytic enzymes, glucose-6-phosphate isomerase (Tc PGI, 1.8 [A]) and enolase (Tc enolase, 2.4 [A]) and provide structural and computational analyses to support structure-based drug design. Tc PGI adopts a dimeric {beta} sandwich fold and features a parasite-specific 53-residue N-terminal extension and a unique C-terminal hook region which both distinguish it from its human ortholog. Tc enolase exhibits the conserved (/{beta}) 8 TIM barrel fold but harbors minor distinct structural deviations, including an extended 17 helix and a structured 1 region, which differentiate it from human isoforms. Both enzymes exhibited high thermal stability, consistent with adaptation to the parasites complex life cycle. Structure-based virtual screening using a scaffold with known multi-target potential identified distinct high-affinity inhibitors for each enzyme. Molecular dynamics simulations further confirmed stable enzyme-inhibitor interactions and favorable binding energetics. Collectively, these findings reveal structural signatures unique to T. cruzi glycolytic enzymes and lay the groundwork for the development of selective antiparasitic therapeutics.

biochemistry↗