Androgens amplify SARS-CoV-2 inflammation while estrogen restrains it

Biological sex has long been recognized as a determinant of how severely coronavirus disease 2019 progresses, with male patients consistently showing higher rates of hypoxemic respiratory failure, hyperinflammation, and mortality. Low testosterone concentrations in hospitalized men correlate with elevated pro-inflammatory cytokines, particularly interleukin-6, and progressive testosterone decline accompanied by rising interleukin-10 levels tracks with increasing disease severity and death risk. Elevated estradiol-to-testosterone ratios, conversely, are associated with hyperinflammatory states and higher hospital mortality. Yet the molecular machinery through which sex steroids regulate immune and endothelial responses to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has remained only partially characterized. A new study published in Frontiers in Cellular and Infection Microbiology by Sandra Bižić-Radulović of the University Clinical Center of Serbia and colleagues now provides detailed mechanistic evidence that androgenic and estrogenic signaling exert opposing, cell-type-specific effects on virus-induced inflammatory pathways.

The research team hypothesized that β-estradiol and dihydrotestosterone (DHT), the most potent androgen, would differentially modulate inflammatory signaling and monocyte–endothelial interactions during SARS-CoV-2 exposure. To test this, they exposed HULEC-5a lung microvascular endothelial cells, THP-1 monocytic leukemia cells, and primary monocytes and neutrophils isolated from patients with confirmed COVID-19 to in vitro-amplified Wuhan-1 SARS-CoV-2 at a multiplicity of infection of 0.001 for 24 and 48 hours. All procedures involving live virus were conducted in a certified Biosafety Level 3 laboratory in accordance with institutional biosafety regulations and World Health Organization recommendations. Cells were treated with DHT, β-estradiol, or the androgen receptor antagonist flutamide at nanomolar to micromolar concentrations before viral exposure. Receptor gene expression was measured by RT-qPCR, inflammatory signaling by Western blot against phosphorylated and total NFκB, STAT3, and p38 MAPK proteins, and monocyte transendothelial migration by Boyden chamber assay. Patient samples were collected between June 2020 and April 2022 at the Clinic for Infectious and Tropical Diseases of the University Clinical Center of Serbia.

The first key finding concerns how SARS-CoV-2 itself reshapes the expression of sex steroid receptors in a cell-specific manner. The virus reduced expression of estrogen receptor 1 (ESR1) in endothelial cells after 24 hours (p<0.05), while simultaneously increasing ESR1 expression in neutrophils (p<0.05) and monocytes (p<0.01) from patients with COVID-19. Androgen receptor (AR) expression was inversely regulated in neutrophils and monocytes: the virus reduced AR in neutrophils after 24 hours (p<0.001) but increased it in monocytes (p<0.05). These differential patterns suggest that SARS-CoV-2 actively reprograms the hormonal responsiveness of both vascular and immune compartments. The authors propose that suppression of ESR1 in endothelial cells may contribute to endothelial dysfunction, vascular inflammation, and thrombotic complications of severe disease, whereas increased ESR1 in circulating innate immune cells may represent a compensatory immunoregulatory response to systemic inflammation.

In THP-1 monocytic cells, SARS-CoV-2 generally activated inflammatory signaling pathways including NFκB, STAT3, and p38 MAPK, with activation reaching up to 1.9-fold above baseline. The NFκB response was further augmented by DHT (44.6%) and attenuated by flutamide (44.1%), indicating that androgen receptor signaling amplifies a central mediator of virus-induced cytokine production. STAT3 signaling was also activated by the virus (p<0.01), and both AR activation and AR blockade were associated with sustained STAT3 phosphorylation at 48 hours, though with different temporal dynamics: the DHT-associated increase at 48 hours was 3.4-fold lower than at 24 hours, whereas the flutamide-associated level remained unchanged between the two time points. Prolonged p38 MAPK activation was maintained by a low dose of DHT (10 nM) and reduced by low-dose flutamide (100 nM), but not by higher flutamide concentrations, suggesting dose-dependent and pathway-specific modulation of inflammatory cascades by androgen-responsive mechanisms.

In HULEC-5a lung endothelial cells of human male origin, the picture was notably different. SARS-CoV-2 reduced STAT3 activation at 24 hours (p<0.05), an effect not further modified by DHT or flutamide. After 48 hours, DHT alone reduced STAT3 activation even in the absence of virus (p<0.01), indicating a direct effect of androgen signaling on baseline STAT3 activity in endothelial cells. In SARS-CoV-2-exposed cells, DHT twice reduced virus-induced activation of the p38 MAPK pathway (p<0.05). These results suggest that androgen signaling can partially suppress inflammatory signaling in endothelial cells both at baseline and during viral exposure, a finding that contrasts with the amplifying effects observed in monocytic cells and underscores the highly cell-specific nature of sex steroid immunoregulation during SARS-CoV-2 infection.

Primary immune cells from patients with COVID-19 revealed further complexity. In neutrophils, β-estradiol inhibited NFκB signaling 2.2-fold, whereas DHT stimulated it 4.9-fold. SARS-CoV-2 reduced NFκB activation in neutrophils from female patients after 24 hours (p<0.001), and β-estradiol further reduced NFκB activation in a dose-dependent manner irrespective of viral exposure (p<0.01). In contrast, DHT increased NFκB activation after 48 hours independently of SARS-CoV-2. Simultaneously, the virus strongly induced p38 MAPK activation in neutrophils (>10-fold) after 24 hours (p<0.001). The authors note that persistent p38 MAPK activation may sustain neutrophil activation despite partial NFκB inhibition, contributing to oxidative stress, neutrophil extracellular trap formation, and tissue damage associated with severe disease. They also highlight that G protein-coupled estrogen receptor 1 (GPER1), a membrane-associated estrogen receptor expressed in human monocytes and neutrophils, may regulate neutrophil inflammatory responses through cAMP/PKA/CREB and MAPK signaling pathways.

In monocytes from patients with COVID-19, DHT at 1 nM enhanced NFκB activation in the presence of SARS-CoV-2 after 24 hours (p<0.05), while β-estradiol at 100 nM increased p38 MAPK activation (p<0.01) and at the lowest concentration tested (1 nM) increased p38 MAPK activation irrespective of viral exposure (p<0.05). Comparison of exposure durations showed stronger SARS-CoV-2-induced activation of NFκB and p38 MAPK after 24 hours than after 48 hours (p<0.05), indicating a transient but pronounced activation of these pathways during early viral exposure. Notably, SARS-CoV-2 activation of inflammatory signaling deteriorates over time in monocytes from patients with COVID-19 (p<0.05), suggesting that androgen-mediated amplification of NFκB and MAPK cascades may contribute to exaggerated innate immune activation and heightened inflammatory responses in male patients with severe disease.

Perhaps the most functionally significant results emerged from the Boyden chamber migration assay. DHT significantly increased trans-endothelial migration of monocytes from patients with COVID-19 (p<0.001) and promoted detachment of SARS-CoV-2-exposed HULEC-5a endothelial cells (p<0.01). Monocyte migration through SARS-CoV-2-exposed endothelial layers was predominantly enhanced by DHT and, to a lesser extent, by β-estradiol and flutamide (p<0.001). After correction for endothelial background fluorescence and monocyte-only controls, increased migration remained evident, particularly in DHT-treated conditions. These findings suggest that androgen signaling promotes endothelial dysfunction and facilitates monocyte transmigration during SARS-CoV-2 infection, key events in COVID-19-associated immunothrombosis and vascular injury. Enhanced leukocyte trafficking across the endothelial barrier may promote tissue infiltration by inflammatory cells and aggravate local cytokine production.

The authors acknowledge limitations, including that sex-stratified analyses were not uniformly performed across all experiments involving primary immune cells because of limited availability of freshly isolated samples, and that the use of HULEC-5a cells derived from a single male donor restricts assessment of sex-dependent differences in endothelial responses. Nevertheless, the findings elucidate molecular mechanisms driving male-biased COVID-19 severity and underscore the potential of targeting androgen-responsive pathways as a therapeutic strategy to mitigate severe viral inflammation and endothelial dysfunction. Beyond SARS-CoV-2, the differential regulation of NFκB, STAT3, and p38 MAPK pathways observed here may reflect fundamental mechanisms of sex hormone-mediated immunoregulation that extend to other immune-mediated and vascular diseases in which sex steroids contribute to disease susceptibility and progression.

The study’s experimental design merits closer attention because it bridges two levels of analysis that are often treated separately in COVID-19 research: controlled cell-line experiments and primary cells drawn from acutely ill patients. By using the same Wuhan-1 viral stock, verified by sequencing and titrated by TCID50 and plaque assay, for every condition, the investigators minimized viral preparation as a confounding variable. The low multiplicity of infection of 0.001 is notable, as it means relatively few cells received productive viral input, yet robust signaling changes were still detectable. This suggests that even limited viral exposure can trigger substantial inflammatory pathway activation, consistent with the idea that a small number of infected cells can amplify responses through paracrine signaling, although the study did not directly test such amplification mechanisms.

The choice of signaling readouts also connects the findings to established COVID-19 biology. NFκB is the canonical transcription factor driving transcription of tumor necrosis factor, interleukin-6, and interleukin-1 family cytokines, all of which are elevated in severe disease and correlate with the testosterone deficits described in the clinical literature. STAT3 links cytokine signaling, particularly through the interleukin-6 receptor JAK-STAT axis, to cell survival and inflammatory differentiation programs in monocytes and neutrophils. p38 MAPK is a stress-activated kinase implicated in cytokine translation, oxidative burst, and neutrophil extracellular trap formation. The observation that SARS-CoV-2 activated all three pathways in monocytic cells, and that androgen receptor agonism and antagonism shifted the magnitude of these responses in opposite directions, provides a mechanistic scaffold for interpreting why male patients with lower testosterone might nonetheless experience androgen-driven hyperinflammation, since DHT effects were observed at concentrations within the physiological range.

The dose-dependence of the androgen effects deserves particular emphasis. In the THP-1 experiments, low-dose DHT at 10 nanomolar sustained p38 MAPK activation while low-dose flutamide at 100 nanomolar reduced it, but higher flutamide concentrations did not reproduce this effect. Such non-monotonic dose relationships are a recognized feature of sex steroid biology, where receptor occupancy, receptor isoform expression, and co-regulator availability can produce qualitatively different outcomes at different ligand concentrations. For translational purposes, this complexity cautions against simple assumptions that androgen blockade will uniformly suppress inflammatory signaling across all cell types and doses, and it highlights the need for careful pharmacological characterization before any therapeutic repurposing of androgen receptor antagonists in viral disease.

The cell-type specificity of the findings is equally important. DHT amplified NFκB signaling in monocytic cells and neutrophils, yet it twice reduced virus-induced p38 MAPK activation in endothelial cells and lowered baseline STAT3 activity after 48 hours. This divergence means that androgen signaling cannot be characterized as simply pro- or anti-inflammatory; rather, its net effect on systemic inflammation depends on the relative contribution of vascular versus immune compartments to a given clinical phenotype. In severe COVID-19, where both endothelial injury and myeloid cell hyperactivation drive pathology, the opposing effects could partially offset each other or, alternatively, the monocyte- and neutrophil-directed amplification may dominate because these cells produce the circulating cytokines and tissue-damaging mediators observed in patients.

The transendothelial migration results add a functional dimension that goes beyond signaling biochemistry. Monocyte recruitment across the endothelial barrier is a hallmark of vascular inflammation and contributes directly to immunothrombosis, the process by which leukocyte activation, tissue factor expression, and platelet engagement generate thrombi in the pulmonary and systemic microvasculature. The finding that DHT promoted monocyte migration with high statistical significance, and that it also induced detachment of virus-exposed endothelial cells, links androgen signaling to two mechanistically connected events: weakening of the endothelial monolayer and enhanced leukocyte trafficking through it. Detachment of endothelial cells, if it occurs in vivo, could expose subendothelial matrix and tissue factor, providing a procoagulant surface that complements the inflammatory effects of infiltrating monocytes.

The differential regulation of estrogen receptor expression by the virus itself adds another layer of interpretation. Reduced ESR1 in endothelial cells could diminish the protective, anti-inflammatory tone that estrogen signaling normally provides to the vasculature, including nitric oxide-mediated vasodilation and suppression of adhesion molecule expression. Increased ESR1 in neutrophils and monocytes, conversely, might represent a compensatory attempt by innate immune cells to enhance estrogen-dependent immunoregulation during systemic inflammation. Whether this compensatory upregulation is functionally effective in patients, or is overwhelmed by the concurrent androgen-driven amplification of NFκB and MAPK signaling, remains an open question that would require measuring receptor protein levels and downstream estrogen-responsive gene expression in patient cells.

The temporal dimension of the findings, with stronger pathway activation at 24 hours than at 48 hours in patient monocytes and progressive deterioration of inflammatory signaling over the course of disease, aligns with clinical observations that COVID-19 inflammation evolves in phases. Early innate immune activation, followed by sustained or dysregulated myeloid responses, characterizes progression toward severe disease. The observation that signaling responses in patient monocytes worsened over time suggests that longitudinal sampling, rather than single time-point measurements, may be necessary to capture the full trajectory of sex steroid modulation in COVID-19, and it raises the possibility that therapeutic timing, whether targeting androgen signaling early or late in disease, could substantially influence outcomes.

Finally, the study’s reliance on the Wuhan-1 variant, collected during the first years of the pandemic, invites consideration of how these mechanisms might apply to later variants with altered inflammatory profiles. The core signaling pathways investigated here, NFκB, STAT3, and p38 MAPK, are activated by multiple pattern recognition receptors and cytokine circuits, so the sex steroid modulation described is unlikely to be variant-specific in principle. However, the magnitude of viral replication, the efficiency of cell entry, and the specific cytokine milieu induced by different variants could shift the balance between androgen-amplified and estrogen-attenuated responses. Extending this experimental framework to contemporary variants, and to cells from vaccinated individuals with pre-existing adaptive immunity, would help determine whether the androgen-responsive mechanisms identified here remain relevant targets for mitigating severe viral inflammation and endothelial dysfunction in current clinical practice.

Subject of Research: Sex steroids differentially modulate SARS-CoV-2-induced inflammatory responses in endothelial and immune cells

Article Title: Sex steroids differentially modulate SARS-CoV-2-induced inflammatory responses in endothelial and immune cells

Article References: Bižić-Radulović, S., Subotički, T., Kovačević, A., Živković, E., Popović, O., Mitrović-Ajtić, O., Diklić, M., Jovanović, S., Čokić, V. P., & Beleslin-Čokić, B. (2026). Sex steroids differentially modulate SARS-CoV-2-induced inflammatory responses in endothelial and immune cells. Frontiers in Cellular and Infection Microbiology, 16, Article 1976945. https://doi.org/10.3389/fcimb.2026.1976945

Image Credits: AI Generated

DOI: 10.3389/fcimb.2026.1976945

Keywords: steroids, differentially, modulate, SARS-CoV-2-induced, inflammatory, responses, endothelial, immune, cells, scientific research