Public Human Brain Atlases Reveal Non-Monotonic Developmental Expression and Adult Cerebellar Enrichment of NLGN1
NLGN1 is a synaptic adhesion gene with established roles at excitatory synapses and repeated links to neurodevelopmental disease, but public discussion often collapses it into a vague claim that it is simply “brain expressed.” We tested whether that simplification survives direct inspection of public human transcriptomic atlases and disease-relevant cortical cohorts. Using the BrainSpan developmental transcriptome together with adult regional and single-nucleus brain RNA profiles from the Human Protein Atlas, we quantified NLGN1 across developmental stages, adult subregions, and cell-type groupings. We then examined NLGN1 in independent schizophrenia and autism dorsolateral prefrontal cortex RNA-seq cohorts and in a 63-donor schizophrenia Visium DLPFC spatial transcriptomics cohort. The resulting picture was not a simple forebrain excitatory trajectory. In BrainSpan, median NLGN1 expression was higher in prenatal than postnatal samples across the cortical, hippocampal, and amygdalar groupings we could compare. In contrast, adult regional expression was strongest in cerebellar cortex, vermis, and flocculonodular lobe. Single-nucleus summaries showed strong neuronal expression but also an unexpectedly prominent oligodendrocyte-precursor signal, while vascular and immune compartments remained low. Despite NLGN1’s disease relevance, neither bulk DLPFC cohort nor the layer-resolved Visium cohort showed a strong case-control shift.
Reviews
This manuscript takes a straightforward but useful approach: it interrogates public human brain expression resources (BrainSpan and Human Protein Atlas bulk regional and single-nucleus summaries) to sharpen an often-hand-waved statement that NLGN1 is “brain expressed,” and then checks disease-relevant adult DLPFC cohorts (SCZ/ASD bulk RNA-seq and a SCZ Visium cohort) for differential expression. The main support is conceptual and descriptive coherence: the authors ask a clear question, use multiple orthogonal public atlases, and the qualitative conclusions they report (developmental differences, strong adult cerebellar signal, neuron-enriched expression, and lack of large case–control shifts in adult DLPFC) are plausible and not obviously self-contradictory. The main weakness is that, as written, the work is under-specified and therefore hard to evaluate or reproduce: there are no references, no dataset identifiers, no preprocessing/normalization details, no effect sizes, no statistical tests, no handling of covariates (age, PMI, RIN, batch, region sampling differences), and no sensitivity analyses to known pitfalls (BrainSpan prenatal vs postnatal differences in tissue composition; HPA cross-region comparability; single-nucleus “OPC signal” potentially reflecting annotation ambiguity or ambient RNA). Several headline claims (“median prenatal > postnatal across cortical/hippocampal/amygdala groupings,” “adult strongest in cerebellar cortex/vermis/flocculonodular lobe,” “unexpectedly prominent OPC signal,” and “no strong case–control shift” in bulk and Visium) could be true, but without quantitative reporting and methodology, the conclusion is only weakly justified beyond a descriptive, hypothesis-generating level.