The study, published in BMC Psychiatry, enrolled 66 patients with systemic lupus erythematosus and 33 healthy volunteers, all recruited from Taizhou People’s Hospital. Participants underwent neuropsychological assessment and resting-state functional magnetic resonance imaging, a technique that records brain activity while subjects lie still and think of nothing in particular. Rather than treating this spontaneous activity as a static signal, the team focused on a measure called dynamic regional homogeneity, or dReHo. Regional homogeneity reflects how synchronized the activity of neighboring voxels is at any given moment; high homogeneity suggests that a local patch of cortex is firing in a coordinated fashion, while low homogeneity suggests disarray. By computing this metric across sliding time windows, the researchers could track how that local coordination fluctuates over the course of a scan, capturing a dimension of brain function that conventional static analyses miss entirely.

The results were striking. Seven brain regions showed abnormal dynamic regional homogeneity in the lupus patients compared with healthy controls: the left gyrus rectus, the left and right middle temporal gyri, the left and right middle occipital gyri, the left precuneus, and the right calcarine fissure. What unites most of these regions is their role in vision. The calcarine fissure is the cortical landing strip for signals arriving from the eyes, the primary visual processing hub of the brain, while the middle occipital and middle temporal gyri form successive stages of the visual hierarchy, extracting motion, form, and spatial relationships from the raw retinal input. The precuneus, tucked into the medial parietal lobe, weaves visual and memory signals into self-referential thought and social cognition, and the gyrus rectus on the underside of the frontal lobe contributes to emotional and social processing. Functional annotation of the affected regions confirmed this impression, linking the abnormality map primarily to social cognition and visual processing.

Identifying where the brain behaves abnormally is only half the story. The deeper question is why those particular regions, and not others, are vulnerable in lupus. To address this, the team turned to an increasingly popular strategy known as imaging transcriptomics. The Allen Human Brain Atlas provides gene expression data from postmortem tissue sampled across the entire cerebral cortex, and researchers can ask whether the spatial fingerprint of an imaging abnormality, region by region, tracks the spatial fingerprint of any particular set of genes. If brain areas with high expression of a certain gene family are precisely the areas showing the greatest functional disruption, that gene family becomes a plausible molecular suspect. It is an inferential approach, and the authors are careful to describe their findings as hypothesis-generating, but it offers something that pure imaging studies cannot: a bridge from circuits to molecules.

The transcriptomic analysis pointed in a coherent direction. Genes whose cortical expression patterns matched the dReHo alteration map were significantly enriched for functions related to chemical synaptic transmission, the fundamental mechanism by which neurons communicate across the tiny gaps between them. The map also aligned with markers of excitatory neurons, the workhorse cells that drive cortical activity, and with genes preferentially expressed in the visual cortex. Two specific excitatory neuron subclasses stood out: Ex3-Granule-L4 cells, which populate the middle layer of the cortex where sensory input first arrives, and Ex8-Corticothalamic-Projecting-L6 cells, which send feedback from deep cortical layers back down to the thalamus, the brain’s relay station. In other words, the regions most affected in lupus are those built around a dense machinery of excitatory synaptic transmission in the sensory processing stream, suggesting that the disease may be disrupting the very synapses that carry visual and social information through the cortex.

The team went further, correlating the dReHo abnormality map with molecular maps of neurotransmitter systems, each derived from independent imaging and pharmacological data. Eight systems showed meaningful spatial correspondence: the dopamine transporter DAT, the serotonin receptors 5-HT2A and 5-HT6, the inhibitory transmitter GABA, the kappa-opioid receptor, the muscarinic acetylcholine receptor M1, the synaptic vesicle glycoprotein SV2A, and histone deacetylase, or HDAC. Each of these carries its own implications. Serotonin and dopamine are the classic targets of psychiatric pharmacology, and their involvement hints at why lupus patients experience depression, anxiety, and psychosis. GABA is the brain’s principal brake, and a relative loss of inhibitory control could explain the local desynchronization captured by the dReHo metric. SV2A marks the density of synaptic terminals themselves, while HDAC sits at the interface between gene regulation and inflammation, a pathway of particular interest in autoimmune disease.

For clinicians, the study offers a potential explanation for a frustrating clinical reality. Many neuropsychiatric lupus symptoms, from subtle cognitive slowing to visual disturbances and difficulty reading social cues, have historically been attributed to vascular injury, antibody-mediated inflammation, or the psychological burden of chronic illness, yet objective biomarkers have remained scarce. The finding that dynamic local coordination is disrupted specifically in visual and social cognition networks suggests that some of these symptoms may have a measurable functional signature, one that could in principle be detected with a standard resting-state MRI scanner already installed in most hospitals. The involvement of the visual cortex is especially provocative, because visual complaints in lupus are often dismissed as peripheral, confined to the eye itself, when the new data suggest that the brain’s central visual machinery may be quietly reorganizing.

The authors are appropriately measured about what their results can and cannot claim. The sample, while respectable for a study of this kind, comprised 66 patients and 33 controls from a single center, and the imaging-transcriptomic method is correlational by design: it identifies spatial associations between gene expression and functional disruption but cannot prove that any particular gene or neurotransmitter causes the abnormalities. The dReHo alterations could reflect downstream consequences of inflammation, vascular compromise, medication effects, or the immune assault itself, and disentangling these mechanisms will require longitudinal studies that follow patients from disease onset through treatment. Still, the framework is powerful. By anchoring a functional imaging phenotype to synaptic transmission genes, specific excitatory neuron classes, and eight distinct neurotransmitter systems, the study generates a ranked list of molecular targets that pharmacologists and immunologists can now interrogate.

What makes the work emblematic of a broader shift in neuroscience is its methodological synthesis. A decade ago, a rheumatology question would rarely have been answered with a brain atlas of gene expression; today, the tools exist to overlay a patient group’s functional abnormality map onto the transcriptomic landscape of the cortex and read out candidate biology in a single analysis. The Taizhou team’s integration of dynamic fMRI metrics, the Allen Human Brain Atlas, neurotransmitter receptor maps, and single-cell-derived neuron subclass data represents exactly the kind of multi-scale approach that is reshaping how researchers study neuropsychiatric disease. For the estimated millions of people worldwide living with lupus, and particularly for the substantial fraction who experience neurological or psychiatric complications, the study offers something intangible but important: evidence that the fog of neuropsychiatric lupus has structure, that it follows the brain’s molecular geography, and that with the right combination of imaging and genomics, that geography can finally be mapped.

Subject of Research: Dynamic brain activity changes and their transcriptomic correlates in neuropsychiatric systemic lupus erythematosus

Article Title: Dynamic regional homogeneity changes and their genetic correlates in neuropsychiatric systemic lupus erythematosus: a combined transcriptome-neuroimaging study

Article References: Pu, X., Li, Y., Sun, Z., Liu, Y., Ji, W., Yin, X., Wu, Q., Dong, L., Zou, H., Zhang, J., & Xia, J. (2026). Dynamic regional homogeneity changes and their genetic correlates in neuropsychiatric systemic lupus erythematosus: a combined transcriptome-neuroimaging study. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08653-2

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08653-2

Keywords: neuropsychiatric lupus, systemic lupus erythematosus, resting-state fMRI, dynamic regional homogeneity, imaging transcriptomics, Allen Human Brain Atlas, visual cortex, synaptic transmission, neurotransmitters, excitatory neurons, autoimmune disease, brain mapping