The study, led by Zhilan Li of Xiangya Hospital at Central South University in collaboration with Jena R. Wirth, C. Alex. Colvert, Rachael J. Werner, and Melissa A. Cunningham of the Medical University of South Carolina, set out to address a persistent gap in lupus research. Most work on estrogen signaling in autoimmunity has focused on lymphocytes, the adaptive immune cells that produce antibodies. Far less is known about how estrogen receptor variants behave in innate immune cells, the first responders of the immune system, or in structural cells such as those lining the kidney’s filtering units, which also participate in inflammatory signaling during lupus nephritis. Understanding these cells matters because innate immune activation, driven in part by immune complexes containing nucleic acids sensed through Toll-like receptor 7, is central to the pathology of systemic lupus erythematosus.

To dissect the contributions of the two receptor variants, the team used a transfection strategy in which cultured cells were given plasmids encoding either ERα66, ERα46, both variants in equal proportion, or an empty control vector. The full-length ERα66 contains all of the functional domains of the receptor, including the two activation function regions that recruit co-regulators to hormone-responsive genes. ERα46, by contrast, lacks the amino-terminal AF-1 domain entirely, which means it cannot engage the same set of transcriptional co-activators and may instead act as a modulator or even a dominant inhibitor of ERα66-driven transcription. Because both proteins are encoded by the same gene and differ only through alternative splicing, any divergent effects between them can be attributed to the presence or absence of specific protein domains rather than to different genes altogether.

The first cellular model was HMC3, a human microglial cell line derived from a male donor. Microglia are the resident immune cells of the central nervous system, and they are believed to play a critical role in neuropsychiatric or central nervous system lupus, one of the most devastating manifestations of the disease. When the researchers overexpressed ERα46 in these cells, they observed a clear promotion of cell proliferation, measured by EdU incorporation assays that detect cells in the act of DNA synthesis, alongside CCK-8 assays that track metabolic viability. Notably, both receptor variants ultimately decreased the levels of inflammatory cytokines, but the mechanism was sobering rather than protective: the apparent anti-inflammatory effect stemmed from reduced cell viability, meaning fewer surviving cells were available to churn out inflammatory mediators. This distinction between a genuine immunomodulatory effect and one that simply reflects cell loss is a technical point with major interpretive consequences for the field.

The picture shifted substantially when the team moved to a second myeloid model, the murine macrophage line RAW 264.7, also male-derived. Here, overexpressing ERα46 promoted cell viability but did not spur proliferation, a subtle but important dissociation between survival and division. More strikingly, when the macrophages were stimulated with a Toll-like receptor 7 agonist, both ERα66 and ERα46 promoted the expression of inflammatory cytokines. TLR7 is a sensor of single-stranded RNA, and its overactivity is tightly linked to lupus pathogenesis, since self-RNA contained in immune complexes can chronically activate this pathway in susceptible individuals. The finding that both receptor variants amplified TLR7-driven inflammation in macrophages directly contradicts the anti-inflammatory picture seen in microglia and underscores that estrogen receptor biology cannot be generalized across the innate immune compartment.

Stimulation experiments extended beyond TLR7. The researchers challenged the transfected cells with interferon-alpha, interferon-gamma, and lipopolysaccharide, three inflammatory triggers that represent distinct arms of innate immune activation: type I interferon signaling, which is a hallmark of lupus; type II interferon signaling; and bacterial endotoxin sensing through TLR4. Cytokine output was quantified at both the protein level, by enzyme-linked immunosorbent assay, and the messenger RNA level, by quantitative reverse transcription polymerase chain reaction, providing complementary confirmation of the transcriptional and secretory responses. The convergent message across these assays was that the two receptor variants are not interchangeable, and that their influence on inflammatory gene expression is highly dependent on which stimulus is applied and which cell is listening.

The most provocative results emerged from human renal mesangial cells, primary cells obtained from both male and female donors. Mesangial cells occupy a unique position in lupus nephritis. Although they are structural components of the glomerulus, the kidney’s blood-filtering unit, they display innate immune-like activities during injury, secreting cytokines and chemokines that recruit and amplify inflammatory attack on the kidney tissue. In mesangial cells from male donors, both ERα variants inhibited proliferation but had only a minimal effect on the inflammatory phenotype. In mesangial cells from female donors, however, ERα66 and ERα46 modulated the expression of multiple inflammatory cytokines, hinting at genuinely sex-specific functions of the receptor variants in the kidney. The authors are careful to note that these observations require validation with additional donors, since primary human cells carry inherent inter-individual variability, but the pattern is consistent with the broader hypothesis that estrogen signaling contributes to the female predominance of lupus.

Taken together, the findings dismantle the simple narrative in which estrogen is uniformly bad, or uniformly good, for people with lupus. Instead, the two major splice variants of estrogen receptor alpha act as context-dependent regulators whose effects on proliferation, survival, and cytokine production flip sign depending on the cellular setting. In microglia, the shorter variant encourages division; in macrophages, both variants fan the flames of TLR7-driven inflammation; in male mesangial cells, both variants restrain growth while leaving inflammation largely untouched; and in female mesangial cells, both variants reshape the cytokine landscape. For clinicians and drug developers, this heterogeneity is both a warning and an opportunity. Therapies that broadly block estrogen receptor signaling could produce very different outcomes in the brain, the blood, and the kidney, while strategies that selectively target ERα46, or exploit its missing AF-1 domain, might one day allow tissue-specific tuning of the immune response.

The study was supported by the National Institutes of Health through the National Institute of Arthritis and Musculoskeletal and Skin Diseases, with additional support from the China Scholarship Council, and the authors acknowledge Professor Geoffrey Greene of the University of Chicago for providing the human ERα plasmids. While the work remains at the in vitro stage, and the authors themselves caution that the sex-associated differences in mesangial cells await confirmation across larger donor panels, the research provides a mechanistic framework for a question that has shadowed lupus medicine for generations: why a disease of dysregulated immunity strikes women so disproportionately. The answer, it now appears, is written not just in the hormone itself but in the alternative splice forms of its receptor, and in the distinct ways that microglia, macrophages, and mesangial cells interpret the same molecular message.

Subject of Research: Differential effects of estrogen receptor alpha splice variants ERα66 and ERα46 on proliferation and inflammatory responses in in vitro models of systemic lupus erythematosus inflammation.

Article Title: Estrogen receptor alpha variants differentially influence proliferation and inflammatory responses in in vitro models of lupus inflammation

Article References: Li, Z., Wirth, J. R., Colvert, C. A., Werner, R. J., & Cunningham, M. A. (2026). Estrogen receptor alpha variants differentially influence proliferation and inflammatory responses in in vitro models of lupus inflammation. Biology of Sex Differences. https://doi.org/10.1186/s13293-026-00990-x

Image Credits: AI Generated

DOI: 10.1186/s13293-026-00990-x

Keywords: estrogen receptor alpha, ERα46, lupus, systemic lupus erythematosus, autoimmune disease, microglia, macrophages, mesangial cells, TLR7, inflammatory cytokines, sex differences, splice variants

NF1 is caused by pathogenic variants in the NF1 gene, an unusually large gene located on chromosome 17 that carries the instructions for neurofibromin, a protein widely described as a tumor suppressor. Neurofibromin’s principal job is to act as a brake on cellular growth signaling, specifically by accelerating the conversion of the RAS protein from its active to its inactive state, thereby dampening the mitogen-activated protein kinase, or MAPK, pathway. When that brake fails, cells in nerve sheaths, skin pigment systems, bone, and brain can proliferate abnormally, producing the diverse constellation of features clinicians recognize: café-au-lait macules, freckling in skin folds, Lisch nodules in the iris, skeletal abnormalities, learning difficulties, and an elevated risk of benign and malignant tumors.

The Shanghai cohort offered an unusually rich window into that complexity. Every child in the study underwent genetic testing, and researchers painstakingly curated both the genetic and phenotypic data, using SPSS, GraphPad Prism, and Python for statistical analysis and generating growth curves and variant distribution maps from the integrated dataset. What emerged was a picture of remarkable heterogeneity: patients exhibited a broad range of multisystem manifestations, with skeletal abnormalities and developmental or intellectual impairments standing out as particularly common and as features showing notable differences between the sexes.

Among the most consequential findings is the study’s contribution to understanding growth in children with NF1. Males in the cohort had significantly lower height standard deviation scores than females, and when the team constructed P50, or median, growth curves for their patients, those curves sat consistently below the reference values for healthy children of the same population. This is not a trivial statistical curiosity. Height assessment is one of the most routine measurements in pediatrics, and for children with NF1 it has long been a source of uncertainty: is a child simply short for their family, or is the disorder itself reshaping their growth trajectory?

The new sex-specific growth curves directly address that uncertainty. By incorporating longitudinal growth data from childhood into the analysis, the researchers built a precise reference standard tailored to children with NF1, giving clinicians a benchmark against which growth abnormalities can be evaluated with far more confidence than was previously possible. Earlier studies have established that height impairment in NF1 is characterized by decreased pubertal growth velocity in both sexes, and growth hormone deficiency has been documented in some patients even without suprasellar brain lesions, but the field has lacked contemporary, population-matched charts of the kind this cohort now provides. The implications run in both directions: clinicians will be better able to identify children whose growth deviates even from the NF1-specific curve, signaling a possible endocrine complication that warrants investigation, while also sparing families unnecessary anxiety when a child is tracking appropriately along the disorder’s own, slightly lower, trajectory.

On the genetic side, the study’s findings illuminate how different classes of NF1 variants translate into different clinical realities. Truncating variants, which introduce premature stop signals and typically render the neurofibromin protein nonfunctional, were the most common mutation type in the cohort, accounting for 60.8 percent of cases. But the correlations the researchers uncovered went beyond simple frequency counts. Patients harboring single amino acid variations, in which one building block of the protein is swapped for another without truncating it, exhibited a significantly higher prevalence of short stature than patients with truncating variants. This counterintuitive result, in which a seemingly milder molecular defect produces a more pronounced growth phenotype, echoes prior reports of specific missense mutations affecting the p.Arg1809 residue being associated with Noonan syndrome features, including short stature and pulmonic stenosis, and suggests that altered, rather than simply absent, neurofibromin function may perturb growth signaling in distinctive ways.

Splice variants, which disrupt the precise editing of the gene’s messenger RNA, told a different story. In this cohort, splice variants were associated with unidentified bright objects, known as UBOs, the mysterious hyperintense spots that appear on brain MRI scans of many children with NF1 and have been linked in prior neuroimaging natural history studies to clinical features, and with abnormalities of the optic nerve, the structure most vulnerable to optic pathway gliomas, low-grade tumors that are among the most feared complications of pediatric NF1. If confirmed in larger and more diverse populations, such associations could eventually guide surveillance strategies, allowing clinicians to prioritize MRI screening or ophthalmologic monitoring for children whose genetic test results fall into higher-risk variant classes.

The cohort’s genetic architecture also yielded fundamental insights into how NF1 enters a family in the first place. Among the 229 patients for whom the parental origin of the variant was known, a striking 77.29 percent carried de novo variants, meaning the mutation arose spontaneously in the child rather than being inherited from a parent. This high proportion of fresh mutations is consistent with NF1’s status as one of the classic de novo mutation disorders in human genetics, and it carries real counseling weight: parents of a child with a confirmed de novo variant face a low recurrence risk in future pregnancies, whereas an inherited variant implies a one-in-two risk for each subsequent child. The study also expanded the known variant landscape, identifying 37 novel NF1 variants not previously cataloged, adding to the growing global database that underpins accurate molecular diagnosis. The processed genotype and phenotype data, with patient information anonymized, have been made publicly available on GitHub, reflecting a growing commitment to open data in rare disease research, although raw sequencing data remain restricted by institutional and ethical privacy protections.

The broader significance of the work lies in how it reframes NF1 as a disorder whose clinical course can, at least in part, be read from its genetic code. Revised international diagnostic criteria, published in 2021, now allow a molecular diagnosis of NF1 even in young children who have not yet developed enough clinical features to meet the classic criteria, making genetic testing an increasingly front-line tool. Studies like this one, conducted over more than a decade at a single center with consistent methodology, provide exactly the kind of large, internally coherent evidence base that such testing depends on. They also lay groundwork for the next therapeutic chapter: with MEK inhibitors already transforming the treatment landscape for inoperable NF1-related plexiform neurofibromas, a deeper understanding of how specific variant classes drive specific downstream pathways may eventually inform which patients benefit most from which targeted agents.

Challenges remain, as the researchers themselves acknowledge through the careful framing of their conclusions. Single-center cohorts, however large, reflect one population and one clinical referral pattern, and genotype-phenotype correlations in NF1 are probabilistic rather than deterministic; even within a single variant class, two children can follow dramatically different courses. Modifier genes, stochastic developmental events, and environmental factors all likely contribute to the variability that makes NF1 so difficult to predict. Yet the Shanghai study moves the field measurably forward. It documents, in 301 children followed across 12 years, the breadth of the disorder’s multisystem phenotype; it quantifies sex-related differences in skeletal and neurodevelopmental features; it delivers the first sex-specific growth references built from contemporary childhood data; and it ties variant classes to growth failure, brain imaging findings, and optic nerve abnormalities with statistical rigor. For the families of the estimated one in every 2,500 to 3,000 children born with NF1 worldwide, and for the clinicians who care for them, the message is one of cautious optimism: the era of precisely mapping this unpredictable disease, and of using that map to anticipate and treat its complications, is gathering speed.

Subject of Research: Genotype-phenotype correlations and growth characteristics in pediatric neurofibromatosis type 1

Article Title: Phenotypic and genotypic characteristics of pediatric patients with neurofibromatosis type 1: a 12-year single-center cohort study

Article References: Li, Z.-Y., Li, X., Wen, T., Feng, B.-Y., Song, Y., Hu, J.-Y., Yu, T.-T., Yao, R.-E., & Wang, X.-M. (2026). Phenotypic and genotypic characteristics of pediatric patients with neurofibromatosis type 1: a 12-year single-center cohort study. World Journal of Pediatrics. https://doi.org/10.1007/s12519-026-01092-9

Image Credits: AI Generated

DOI: 10.1007/s12519-026-01092-9

Keywords: neurofibromatosis type 1, NF1 gene, genotype-phenotype correlation, pediatric genetics, short stature, growth charts, de novo variants, truncating variants, splice variants, optic pathway glioma, neurofibromin, MAPK pathway