The study began as a comparison of lifestyle and noncommunicable disease feature importance between Indian (n = 528) and Japanese (n = 70) university students, using cross-sectional data collected from June to August 2024. India and Japan present contrasting epidemiological and dietary profiles that made the comparison appealing. India faces a rapid nutrition transition, with sedentary behaviour independently associated with obesity risk among urban Indian adults, while Japan has undergone decades of continuous dietary westernization alongside a strong cultural link between lifestyle behaviour change and subjective well-being. Body mass index also carries different metabolic implications in South Asian populations than in Western reference populations, making BMI-based thresholds an imperfect, population-dependent proxy for adiposity rather than a universal standard.

But the preliminary analysis contained three specific, checkable pitfalls that the authors identified in their own work. First, the outcome of obesity, defined as BMI ≥ 25 kg/m², was tautologically predicted by including BMI itself as a model input. Second, model evaluation was performed without a held-out test set, meaning the models were evaluated on the same data they were trained on. Third, feature-importance comparisons ignored sampling variability from an eightfold difference in cohort size. The team addressed each problem using penalized logistic regression, Random Forest, and gradient-boosted trees with proper train/test splitting, held-out permutation importance, and bootstrap confidence intervals.

The most dramatic result came from quantifying the BMI-obesity circularity. With BMI included as a predictor, all three India-obesity models achieved perfect discrimination on the held-out test set, with an AUC of 1.000 for logistic regression, Random Forest, and XGBoost alike. BMI’s Gini importance was 0.91 with a bootstrap 95% confidence interval of [0.84, 0.92], and its held-out permutation importance was 0.50, meaning permuting BMI alone destroys half of the model’s discriminative AUC. This is definitional, not empirical: obesity is BMI ≥ 25 by construction. When BMI is removed from the predictor set, test AUC collapses to 0.568 for logistic regression, 0.526 for Random Forest, and 0.503 for XGBoost, indistinguishable from chance. Lifestyle features carry essentially no independent discriminative signal for obesity once the tautological predictor is removed.

For the remaining viable outcome, elevated self-rated stress, held-out predictive performance was modest at best. In India, test AUC was 0.67 for logistic regression, 0.67 for Random Forest, and 0.71 for XGBoost, with a bootstrap 95% confidence interval of [0.49, 0.72]. In Japan, the models did not clearly exceed chance, with AUCs of 0.57, 0.57, and 0.49, and a very wide bootstrap confidence interval of [0.29, 0.82] reflecting Japan’s small test set of only 14 observations. None of the AUC values in the 0.92 to 0.98 range obtained under the preliminary, non-held-out evaluation approach were reproduced once a genuine held-out test set was used.

The reanalysis also revealed a critical discrepancy in how feature importance was measured. Gini importance, computed on the training fit, reflects how much a variable reduced impurity during tree construction, while permutation importance on held-out data reflects how much actual predictive information the model loses without that variable. These two measures frequently disagreed. Most notably, sleep duration’s Gini importance in Japan was 0.22, superficially close to the 0.47 value initially reported as headline evidence for a Japanese psychosocial phenotype, but its held-out permutation importance was −0.004, indicating no measurable out-of-sample discriminative contribution at all. By contrast, BMI with a permutation importance of 0.067 and exercise frequency at 0.040 contributed more to real, held-out predictive performance than sleep duration did. The claim that sleep duration is the primary stress predictor in Japan is not supported once importance is assessed out-of-sample.

The study’s central contribution is a sample-size sensitivity analysis that directly tests whether the India-Japan feature-importance difference exceeds what unequal cohort size alone predicts. The team repeatedly subsampled the Indian cohort, without replacement, down to Japan’s analytic sample size of 70, refitting the Random Forest pipeline on each of 300 draws and recording the resulting distribution of feature importances. For every feature examined, including age, sex, BMI, sleep duration, exercise frequency, water intake, and all food-preference indicators except one near-zero-importance category, Japan’s actual observed Gini importance value fell inside the 95% range produced by this subsampling procedure alone. In other words, drawing a random subsample of 70 Indian participants and fitting the identical pipeline would routinely observe importance values as different from India’s full-sample estimate as Japan’s actual values are.

The reanalysis also uncovered a measurement-instrument mislabeling that the authors disclose transparently. Perceived stress was intended to be assessed using the Perceived Stress Scale (PSS-10), a validated 10-item instrument. But on auditing the raw survey exports, no PSS-10 items exist in either dataset. The actual instrument administered at both sites was a single self-rated item asking participants to rate their stress or anxiety level on a scale of 0 to 10, which was binarized at 5 or above to define elevated stress. A single-item stress rating has not been validated for cross-cultural measurement equivalence, and the authors treat all stress results as exploratory accordingly. Similarly, once the study’s own chronic-illness exclusion criterion was correctly applied, positive diabetes cases fell to zero in both cohorts, and Japan’s survey instrument never asked a hypertension-related question at all.

The authors emphasize that both cohorts are convenience samples recruited from a single institution in each country with substantially different response patterns. India retained 54.0% of initial respondents as age-eligible and not excluded, while Japan retained only 20.2%. Any apparent difference between them may reflect institutional, disciplinary, or self-selection factors as readily as national culture. The authors also note that they did not conduct formal cross-cultural measurement-invariance testing for any instrument used in the study, meaning that interpreting comparable constructs across sites remains a working assumption rather than a demonstrated fact. Self-reported height and weight typically underestimate true BMI, introducing further measurement uncertainty.

Despite the negative findings, the authors regard the analysis as a legitimate and useful contribution in its own right. The pattern illustrated here, including small paired cohorts, no held-out evaluation, circular outcome definitions, Gini importance treated as ground truth, and an oversampling step that was described but never actually implemented, is not unique to this dataset. The checks the authors describe, including quantifying circularity by ablating the offending predictor, computing held-out permutation importance alongside Gini importance, and running a sample-size-matched subsampling sensitivity analysis, are general-purpose and inexpensive to apply before any cross-cultural machine learning feature-importance claim is published. The authors suggest that future studies routinely report an explicit list of which outcome-defining variables are excluded from the predictor set, held-out permutation importance alongside any impurity-based importance measure, and a sample-size-matched subsampling sensitivity check before interpreting any between-population difference as substantive. A genuine test of cross-cultural machine learning transferability would train a model on one cohort and evaluate its discrimination and calibration on the other, directly measuring transfer performance rather than comparing independently fitted feature importances. With complete, runnable code accompanying the paper, every number in the study can be independently verified, and the authors hope this transparency will help digital health research avoid mistaking statistical artifacts for cultural insight.

Subject of Research: Do machine learning models really encode cross-cultural lifestyle-health differences? A leakage-free reanalysis of two university cohorts in India and Japan

Article Title: Do machine learning models really encode cross-cultural lifestyle-health differences? A leakage-free reanalysis of two university cohorts in India and Japan

Article References: Takaoka, E., Ruban, S., Mohammed Siraj, B., Vinay, V., Shanthini, H., & Dias, A. (2026). Do machine learning models really encode cross-cultural lifestyle-health differences? A leakage-free reanalysis of two university cohorts in India and Japan. Frontiers in Digital Health, 8, Article 1791066. https://doi.org/10.3389/fdgth.2026.1791066

Image Credits: AI Generated

DOI: 10.3389/fdgth.2026.1791066

Keywords: machine, learning, models, really, encode, cross-cultural, lifestyle-health, differences, leakage-free, reanalysis, university, cohorts

The logic behind long-duration trains is important for interpreting the present findings. A 500 ms train of biphasic pulses delivered at 333 Hz gives cortical neurons time to engage recurrent and downstream circuitry, including spinal and brainstem networks, in a way that a brief 50 ms train cannot. The resulting movements are not simply longer versions of twitch responses; they are coordinated, multi-joint actions that often converge toward stereotyped final postures regardless of where the limb began. This convergence property, observed in primates and rodents alike, suggests that the cortex encodes movement goals or trajectories rather than individual muscle activations. In the rat hindlimb, the identification of Advance and Elevation as distinct, reproducible movement classes fits this framework: each represents a coherent kinematic pattern that the cortex can generate as a unit.

Principal component analysis plays a central role in making such classifications objective rather than impressionistic. By reducing high-dimensional kinematic data, in this case the trajectories of markers placed on the iliac crest, hip, knee, ankle, metatarsals, and toes, into a small number of axes that capture the largest share of movement variance, PCA allows researchers to ask whether evoked movements cluster into discrete categories or form a continuum. The finding that overall movement amplitude and joint flexion patterns systematically separated Advance from Elevation movements indicates that these are genuinely different motor programs, not variations of a single movement produced at different intensities. This analytical approach mirrors strategies used in motor cortex recording studies, where low-dimensional neural subspaces have been shown to organize population activity during reaching, grasping, and locomotion.

The spatial segregation of the two movement types within the hindlimb cortex adds an organizational principle that was previously undocumented. Advance movements, which propel the limb forward, were evoked from anteromedial subregions, while Elevation movements, which draw the limb upward and backward, arose from posterolateral sites. This kind of within-limb functional parcellation echoes earlier observations in the forelimb, where long-duration stimulation revealed distinct zones for grasping, reaching, and retraction. It also resonates with the classical understanding that the hindlimb representation sits medially within the sensorimotor cortex of rodents, but it refines that picture by showing that the medial hindlimb territory is itself subdivided according to movement direction and purpose rather than according to individual joints or muscles.

Stimulating at a depth of approximately 1,550 micrometers targets layer V, the output layer of the neocortex containing pyramidal tract neurons and corticospinal cells that project directly to spinal circuitry. This targeting choice matters because it means the evoked movements arise from the same population of neurons that normally transmits voluntary motor commands. In rodents, the corticospinal tract is less dominant in motor control than in primates, with rubrospinal, reticulospinal, and other descending pathways sharing the burden, but layer V stimulation nonetheless engages the cortical output machinery in a physiologically meaningful way. The use of a fixed current intensity of 150 microamperes, established through pilot testing, helps ensure that movements evoked across different cortical sites reflect genuine differences in cortical organization rather than differences in stimulation effectiveness.

The dense mapping grid used in the study, with penetrations spaced roughly 354 micrometers apart across an exposed region extending from 3 mm anterior to 5 mm posterior to bregma, provides the spatial resolution needed to detect subregional boundaries. Earlier mapping studies often used coarser grids that could easily miss fine-grained subdivisions within a body part representation. By sampling at near-microcolumnar spacing and recording evoked movements with a high-speed camera synchronized to stimulation, the experimenters could build a detailed functional map in which the borders between Advance and Elevation territories emerged from the data rather than being imposed by prior assumptions.

Perhaps the most consequential experiment is the one conducted during active locomotion. Testing stimulation during treadmill walking at 23 centimeters per second addresses a question that static mapping cannot: does the cortex actually use these subregions to influence ongoing behavior? The result that phase-coherent stimulation, timed to the locomotor cycle, selectively biased hindlimb trajectories forward or backward in an amplitude-dependent manner demonstrates that these cortical sites can interact with the spinal locomotor pattern generator without abolishing it. This is consistent with a growing body of evidence that descending cortical commands are superimposed on, rather than replacing, brainstem and spinal rhythms underlying rhythmic movement. The amplitude dependence is particularly informative, suggesting a graded control mechanism in which stronger cortical drive produces larger trajectory deviations, a property that would be essential for fine motor adjustments in natural behavior.

This finding connects to work in non-human primates showing that the leg area of primary motor cortex maintains locomotor-related dynamics in one neural subspace while encoding volitional gait adjustments, such as obstacle avoidance, in another. The rat data suggest that an analogous division of labor may exist at the level of cortical subregions: one zone biased toward forward propulsion, another toward backward or elevating movements, each capable of modulating stepping without disrupting its rhythm. Such an arrangement would allow an animal to adjust its gait flexibly, for example to extend a stride, arrest forward motion, or lift a limb over an obstacle, while the basic locomotor program continues uninterrupted.

The ethological relevance of the two movement classes deserves consideration. Forward advancement of the hindlimb is fundamental to propulsion during walking and running, while elevation and backward drawing movements contribute to limb clearance during the swing phase, to postural adjustments, and to behaviors such as rearing or stepping over obstacles. That long-duration stimulation recovers precisely these behaviorally meaningful patterns, rather than arbitrary muscle combinations, supports the view that cortical motor maps are shaped by the movement repertoire of the animal. It also suggests that the segregation observed under anesthesia reflects a genuine functional architecture rather than an artifact of the stimulation protocol.

From a translational standpoint, the identification of discrete, directionally specific hindlimb cortical zones has direct implications for neuroprosthetics. Brain-machine interfaces and intracortical microstimulation-based therapies for spinal cord injury or stroke depend on knowing where in the cortex specific movements can be elicited or decoded. If forward and backward hindlimb commands originate from separable cortical territories, then stimulation or recording electrodes could be targeted to these zones to promote or suppress particular phases of gait in injured animals or patients. The demonstration that stimulation during locomotion produces graded, phase-appropriate modulation is an encouraging proof of principle for such interventions, since therapeutic stimulation would likewise need to integrate with residual locomotor rhythms rather than override them.

The choice of female Long-Evans rats in a moderate weight range provides a standardized preparation whose hindlimb kinematics can be measured reliably with external markers. Placing markers on the iliac crest in addition to limb joints allows pelvic motion to be quantified, which is important because hindlimb movement in quadrupeds involves trunk and pelvic contributions that are easily overlooked. Recording at 119 frames per second captures the temporal structure of movements lasting several hundred milliseconds, sufficient to resolve the trajectory shape that distinguishes an Advance from an Elevation movement.

It is worth noting what the anesthetized mapping paradigm can and cannot reveal. Anesthesia suppresses voluntary movement and alters cortical excitability, so movements evoked under ketamine and xylazine may not perfectly replicate those generated during wakefulness. The locomotion experiments partially bridge this gap by showing that the same cortical sites influence movement in the behaving animal, but the precise contribution of these subregions to naturally initiated, voluntary hindlimb movements remains an open question. Future studies combining chronic implants, cortical recording, and behavioral tasks could determine whether Advance and Elevation zones are differentially active during specific locomotor phases or during goal-directed hindlimb actions.

The study also fills a notable gap in the comparative literature. Because most long-duration stimulation work has emphasized the forelimb, particularly in primates where hand movements are behaviorally dominant, the hindlimb cortex has been comparatively neglected despite its obvious importance for locomotion, the most common voluntary movement in terrestrial animals. Establishing that the rat hindlimb cortex follows the same principle of complex, topographically organized movement representation strengthens the case that this organizational scheme is a general property of mammalian motor cortex rather than a specialization of forelimb or hand areas.

More broadly, the results contribute to an evolving conception of the motor cortex as a repository of movement primitives, reusable building blocks that combine to produce the rich variety of natural behavior. Under this view, somatotopy remains real but operates at the level of movements and their goals rather than individual muscles. The rat hindlimb cortex, with its anteromedial Advance zone and posterolateral Elevation zone, offers a compact and experimentally tractable model for studying how such primitives are laid down, how they interact with spinal locomotor circuitry, and how they might be harnessed to restore movement after neurological injury. The convergence of quantitative kinematics, systematic cortical mapping, and behavioral validation in a single preparation provides a methodological template that other laboratories studying motor cortex organization across species can readily adopt.

Subject of Research: Distinct hindlimb motor cortex subregions encode forward and backward hindlimb movements in rats

Article Title: Distinct hindlimb motor cortex subregions encode forward and backward hindlimb movements in rats

Article References: St-Arnault, J., Brown, A. R., Bonizzato, M., & Martinez, M. (2026). Distinct hindlimb motor cortex subregions encode forward and backward hindlimb movements in rats. Frontiers in Systems Neuroscience, 20, Article 1912148. https://doi.org/10.3389/fnsys.2026.1912148

Image Credits: AI Generated

DOI: 10.3389/fnsys.2026.1912148

Keywords: neuroscience, motor cortex, microstimulation, locomotion, neuroprosthetics, Distinct, hindlimb, motor, cortex, subregions, encode, forward

The work, published in BMC Biology by Mariachiara Grieco, Tommaso Becchi, Gabriele Boscagli and colleagues under the supervision of Matteo Cereda, spans the University of Milan, the Italian Institute for Genomic Medicine, IFOM ETS, the University of Turin and the IRCCS E. Medea institute. The central difficulty the team set out to address is that RNA-binding proteins rarely act through a single, cleanly defined binding site. Instead, they often recognize multivalent RNA motifs, combinations of short sequence elements arranged in specific positions relative to the splice sites that mark exon boundaries. Connecting such a motif to its candidate trans-acting protein from sequence information alone has remained stubbornly difficult, because many proteins recognize similar sequence patterns and because the functional consequence of binding depends heavily on where in the transcript the motif sits.

RNAmotifs 2.0 approaches the problem by coupling motif discovery with two independent streams of experimental evidence: in vivo binding data and measured splicing responses. The heart of the new framework is a module called MaRs, short for Motifs and Regulators of splicing, which integrates these data types into a single association score linking each multivalent RNA motif to a candidate RBP. The method does not treat all binding evidence equally. It learns position-specific binding and regulatory expectations, meaning it develops a statistical model of where a given protein tends to bind relative to exons and what effect that binding typically has on splicing outcomes. It also weights binding quality, so that stronger, more confident cross-linking signals contribute more to the association than weak or ambiguous ones.

Technically, the framework divides the region around each cassette exon, the type of alternative exon that is either included or skipped in the mature message, into three splice-proximal regulatory regions: the upstream intron, the exon body itself, and the downstream intron. Enrichment of a motif in these regions is captured through a combined regional enrichment measure, and the positional information is critical because a motif that enhances exon inclusion when placed downstream may repress inclusion when placed upstream. The association score that MaRs computes blends several components, including an enrichment score reflecting how strongly a motif is overrepresented near regulated exons, a binding score derived from enhanced cross-linking and immunoprecipitation, or eCLIP, experiments that map where proteins physically touch RNA inside living cells, and a measure of similarity between the predicted splicing response and the observed changes in percent spliced-in values following protein depletion.

A major methodological advance in version 2.0 is the optimization of the framework’s parameters on a per-protein basis. Rather than applying a single fixed set of thresholds and weights to every RBP, the authors use a guided hyperparameter search to tune the analysis for each protein individually. They compared a traditional grid search with Bayesian optimization, a strategy that intelligently samples the parameter space instead of exhaustively testing every combination, and found the Bayesian approach efficient for navigating the large parameter grid. To ensure that these optimized parameters generalize rather than simply overfit the training data, the team employed leave-one-RBP-out cross-validation, in which the framework is tuned without access to the protein being tested and then asked to identify that held-out protein from scratch.

The benchmark for this validation came from the ENCODE project, which has systematically depleted hundreds of RNA-binding proteins in cultured human cells using siRNA and measured the resulting changes in gene expression and splicing through RNA sequencing, alongside eCLIP maps of where each protein binds. Across this large collection of knockdown experiments, RNAmotifs 2.0 consistently prioritized the perturbed regulator as the top candidate for the motifs it discovered, with performance particularly strong when the splicing effects were large, as quantified by the absolute change in mean percent spliced-in upon protein depletion. The framework’s ability to recover the true regulator was evaluated using the area under the receiver operating characteristic curve, a standard measure of ranking quality, and the results were summarized across dozens of proteins in both HepG2 and K562 cell lines.

One practical concern with any method that integrates RNA-seq and eCLIP data is that these datasets are often generated in different cell types, and the authors explicitly tested the robustness of their approach to such mismatches. In a cross-cell-line transfer analysis, they paired knockdown exon sets from one cell line with binding maps from another, for example HepG2 knockdown data analyzed against K562 eCLIP profiles and vice versa. The framework retained substantial predictive power under these mismatched conditions, indicating that the learned positional and regulatory expectations capture biology that is at least partly shared across cellular contexts rather than being an artifact of perfectly matched datasets. An ablation analysis of the association-score components further revealed which ingredients of the score contribute most to performance, providing a transparent account of why the method works rather than presenting it as an opaque black box.

To demonstrate that the framework generalizes beyond the ENCODE benchmark, the team performed their own experimental validation in PC3 prostate cancer cells depleted of HNRNPK, a member of the heterogeneous nuclear ribonucleoprotein family with well-documented roles in RNA processing. After knocking down HNRNPK and profiling splicing changes with rMATS, a standard tool for detecting differential splicing from replicate RNA-seq data, they ran RNAmotifs 2.0 on the resulting set of regulated cassette exons. The framework identified HNRNPK as the splicing regulator, confirming that the method can recover the correct answer in a completely new biological setting with data generated independently of the training resources. This out-of-sample success is arguably the most convincing evidence that the learned motif-to-regulator associations reflect genuine molecular recognition rules.

Interpretability is a defining feature of the new framework and one that distinguishes it from purely predictive machine-learning approaches. Because the association score is built from named, quantifiable components, enrichment, binding quality, positional specificity and response similarity, researchers can inspect exactly why a particular protein was nominated as a candidate regulator of a particular set of exons. The authors also condensed the output into a compact visualization that summarizes the results for each RBP, making it feasible to survey many proteins and cell lines at a glance. Supplementary analyses addressed additional design choices, including the effect of motif width on RBP identification and the robustness of eCLIP-derived binding profiles, with the motif-width benchmark showing how this parameter influences both the identification of the correct protein and the number of enriched multivalent motifs detected.

The significance of this work extends well beyond method development. Alternative splicing is frequently misregulated in cancer and in a wide range of genetic diseases, and mutations in splice-regulatory motifs are increasingly recognized as drivers of pathology. A tool that can connect altered splicing patterns to the proteins responsible, using only sequence data, binding maps and expression measurements, gives researchers a systematic way to dissect these regulatory networks and to generate hypotheses about which factors might be therapeutically targeted. The framework is openly available, and the study’s authors, whose work was supported by AIRC and the Compagnia di San Paolo Foundation, have released extensive supplementary documentation covering every analytical decision. As eCLIP datasets continue to accumulate across tissues and disease states, frameworks like RNAmotifs 2.0 are poised to become standard instruments for translating the growing mountain of RNA data into concrete statements about which molecular players control which splicing decisions.

Subject of Research: A computational framework for linking multivalent RNA motifs to their candidate RNA-binding protein regulators of alternative splicing

Article Title: RNAmotifs 2.0: a framework for inferring multivalent RNA motifs and their candidate regulators of splicing

Article References: Grieco, M., Becchi, T., Boscagli, G., Priante, F., Caizzi, L., Pozzoli, U., & Cereda, M. (2026). RNAmotifs 2.0: a framework for inferring multivalent RNA motifs and their candidate regulators of splicing. BMC Biology. https://doi.org/10.1186/s12915-026-02745-x

Image Credits: AI Generated

DOI: 10.1186/s12915-026-02745-x

Keywords: alternative splicing, RNA-binding proteins, multivalent RNA motifs, RNAmotifs 2.0, eCLIP, RNA sequencing, splicing regulation, ENCODE, Bayesian optimization, HNRNPK, motif discovery, computational biology

The structures in question are known as artery tertiary lymphoid organs, or ATLOs. Unlike the classical lymph nodes and spleen, which develop during embryogenesis and serve the entire body, tertiary lymphoid organs arise postnatally in response to chronic inflammation at affected tissues. In atherosclerosis, they form in the adventitia, the outermost layer of the arterial wall, adjacent to plaques rich in immune cells. Previous work had established that these structures contain germinal centers, the specialized microenvironments where B cells proliferate, mutate their antibody genes, and undergo selection for increasingly potent antigen binding. What remained unknown was what these germinal centers were actually targeting, and whether the antibodies they produced mattered for disease.

To answer those questions, the research team, led by Zihao Zhang, Yue Zhang, Shu Ran, Ying Wang and colleagues, combined single-cell transcriptomic profiling with B cell receptor sequencing of germinal center B cells harvested from artery tertiary lymphoid organs. Single-cell transcriptomics allows researchers to capture the complete gene expression program of individual cells, revealing their identity, activation state, and developmental trajectory. B cell receptor sequencing, meanwhile, reads out the rearranged antibody genes carried by each B cell, providing a molecular fingerprint of the antigen-binding receptor that defines the cell’s specificity. By layering these two datasets on top of one another, the investigators could reconstruct, at unprecedented resolution, the population of B cells being educated within the arterial wall itself.

The analysis revealed a clonally expanded population of autoreactive B cells within the ATLO germinal centers, indicating that these cells had undergone the canonical process of somatic hypermutation and antigen-driven selection that normally refines antibody responses during infection or vaccination. From this population, the team cloned a monoclonal antibody, designated A6, and demonstrated that it binds with high affinity to histone 2B, or H2B, one of the core histone proteins around which DNA is wrapped in every cell nucleus. Histones are potent immunological stimuli, well known to activate innate immune pathways, but the identification of a germinal center-derived, high-affinity antibody specific for H2B within the artery wall was unexpected and mechanistically significant.

The pathogenic relevance of the finding was established through two complementary lines of experimentation. First, the researchers tested H2B vaccination: when animals were immunized with histone 2B, the resulting immune response markedly accelerated the progression of atherosclerosis, confirming that a targeted anti-H2B response is sufficient to worsen disease. Second, and more directly, the team performed adoptive transfer of the anti-H2B A6 monoclonal antibody itself. Simply delivering the purified antibody into circulation was enough to significantly speed plaque development, demonstrating that the antibody is not merely a marker of the disease process but an active driver of it.

These results carry substantial mechanistic weight because they close a causal loop. Earlier studies had documented that tertiary lymphoid organs appear in the arterial adventitia as atherosclerosis advances, and that their germinal centers become more prominent with disease severity. But correlation is not causation, and it remained possible that the local immune structures were protective, attempting to contain plaque inflammation, or merely incidental. By identifying a specific autoantibody-autoantigen pair encoded within these structures and showing that the antibody accelerates disease when transferred, the study establishes that ATLOs are not passive observers but productive factories of pathogenic humoral immunity.

The technical achievement underlying the discovery deserves emphasis. Germinal center B cells are notoriously difficult to study because they are short-lived, rapidly dividing, and spatially restricted within organized lymphoid tissue. Extracting them from the arterial wall, dissociating the tissue into single cells while preserving RNA quality, and obtaining paired heavy- and light-chain B cell receptor sequences from each cell requires meticulous single-cell handling. The successful cloning of a functional monoclonal antibody from this dataset, one that retains its specificity and pathogenic activity when expressed recombinantly, validates the entire pipeline and opens the door to systematically cataloging the full repertoire of antigens targeted within atherosclerotic ATLOs.

The identification of H2B as the target antigen also connects atherosclerosis to a broader theme in autoimmunity. Anti-histone antibodies are hallmarks of several systemic autoimmune diseases, most notably drug-induced lupus, where they bind nuclear components released by dying cells. In atherosclerotic plaques, cell death is abundant: foam cells undergo apoptosis, necrotic cores accumulate, and nuclear debris is released into the plaque microenvironment. It is plausible that this ongoing cell death provides a continuous supply of histone antigen, which drains to the adjacent adventitial lymphoid structures and drives the affinity maturation of autoreactive B cells. The new findings suggest that atherosclerosis may therefore share fundamental immunological machinery with classical autoimmune diseases, a concept that reframes how the field thinks about cardiovascular inflammation.

Therapeutically, the work suggests several possible avenues, though the researchers are careful to note that translation from experimental models to human patients will require further validation. If comparable anti-H2B responses are confirmed in human atherosclerotic arteries, the autoantibody could serve as a biomarker for identifying patients whose disease is driven by autoreactive immunity. More ambitiously, the pathways that organize and sustain artery tertiary lymphoid organs, including the lymphotoxin and chemokine networks that recruit and retain lymphocytes in the vessel wall, could become targets for interventions designed to quiet local antibody production without globally suppressing protective immunity. Given that atherosclerosis remains the leading cause of death worldwide despite aggressive lipid-lowering therapy, any mechanism that explains residual cardiovascular risk in patients is of enormous clinical interest.

The study also raises fundamental questions about how immune tolerance breaks down inside the artery. Germinal centers normally enforce stringent selection against self-reactive B cells, yet the ATLO germinal centers clearly nurtured a high-affinity autoreactive clone to maturity. Understanding why the arterial environment licenses this breach, whether through the character of the antigens presented, the cytokine milieu, or the absence of regulatory cell networks found in conventional lymphoid tissue, may reveal general principles applicable to other chronic inflammatory diseases in which tertiary lymphoid structures arise, from rheumatoid arthritis to organ transplant rejection. For now, the demonstration that a single cloned antibody targeting a nuclear protein can measurably accelerate the arterial disease that kills more people than any other stands as a vivid reminder that the immune system, in the wrong context, can be as dangerous as any traditional risk factor.

The adventitia, where these lymphoid structures take root, has long been treated as a largely passive scaffold of collagen, vasa vasorum, and resident fibroblasts. The demonstration that it can host affinity-matured humoral responses repositions this outer arterial layer as an immunologically active compartment. It also helps explain earlier observations that immune-cell clusters in the adventitia expand in parallel with plaque burden and that lymphatic vessels draining the arterial wall become more prominent as disease advances, providing a plausible route by which plaque-derived antigens reach the local immune structures.

Histone 2B belongs to a family of proteins that are normally sequestered inside the nucleus, where they package DNA into nucleosomes. When cells die, particularly through the inflammatory forms of cell death common in plaques, histones can escape into the extracellular space, where they act as damage-associated molecular patterns that stimulate innate receptors. The new work shows that this same nuclear material can also serve as a target for affinity-matured adaptive responses, linking innate danger signaling and autoreactive antibody production within a single tissue site.

The adoptive transfer result is especially informative, because it separates the antibody itself from the cells that produce it. Accelerated plaque development following antibody delivery alone indicates that circulating anti-H2B immunoglobulin is sufficient to promote disease, pointing future mechanistic studies toward how the antibody engages antigens within the vessel wall and what effector pathways it recruits once bound.

Subject of Research: Artery tertiary lymphoid organs encode a pathogenic high-affinity autoantibody−autoantigen pair in atherosclerosis

Article Title: Artery tertiary lymphoid organs encode a pathogenic high-affinity autoantibody−autoantigen pair in atherosclerosis

Article References: Zhang, C., Zhang, X., Ran, Y., Wang, Z., Li, L., Wang, S., Zheng, J., Zhang, Y., Sun, T., Li, Y., Lu, S., Hong, M., Ma, Z., Steffens, S., Hristov, M., Blanchet, X., Zhu, J., Dou, X., Deng, X., … Yin, C. (2026). Artery tertiary lymphoid organs encode a pathogenic high-affinity autoantibody−autoantigen pair in atherosclerosis. Nature Cardiovascular Research, 5(9), 869-890. https://doi.org/10.1038/s44161-026-00864-w

Image Credits: AI Generated

DOI: 10.1038/s44161-026-00864-w

Keywords: Artery, tertiary, lymphoid, organs, encode, pathogenic, high-affinity, autoantibody, autoantigen, pair, atherosclerosis, scientific research