The story begins with a puzzle the researchers uncovered by dissecting the synovium, the soft tissue lining the joint capsule, at single-cell resolution. Using single-cell RNA sequencing of human osteoarthritic tissue, they identified seven major cell populations and found that two of them—fibroblast-like synoviocytes, the structural workhorses of the synovium, and infiltrating macrophages—stood out with dramatically elevated glycolytic activity. These cells cranked up expression of SLC2A1, the gene encoding the glucose transporter GLUT1, along with PFKFB3, a key enzyme that amplifies glycolytic flux. Immunofluorescence staining of human knee synovium and of mouse models confirmed that GLUT1 accumulated precisely in the activated, tumor-necrosis-factor-producing cells driving the inflammation. Pseudotime analysis suggested that high SLC2A1 expression marked a terminal, pathologically transformed cell state, and the GLUT1-high subsets carried strong inflammatory, fibrotic, and catabolic signatures.

But the metabolic story had a second, darker thread. Excessive glycolysis is tightly linked to oxidative stress, and the team found that the synovium’s selenium-dependent antioxidant system was in disarray. Mass spectrometry revealed reduced selenium levels in the synovial fluid of osteoarthritis patients, and single-cell analysis showed broad dysregulation of selenoprotein genes, including the antioxidant regulators TXNRD1 and SELENOH, across multiple cell populations. Strikingly, the GLUT1-high cells that were burning glucose fastest also showed the lowest selenoprotein expression and the highest oxidative stress scores, an inverse relationship confirmed by spatial analysis of tissue sections. The picture that emerged was a vicious cycle: glucose flooding in through GLUT1 feeds inflammatory signaling, inflammatory signaling suppresses the selenoprotein shield, and the resulting reactive oxygen species further entrench the inflammatory state.

Selenium’s role here is not incidental. The trace element is incorporated into selenoproteins as the amino acid selenocysteine, which sits at the catalytic heart of enzymes that neutralize peroxides and maintain cellular redox balance. Selenium deficiency is a recognized cause of joint disease in humans—Kashin-Beck disease, an endemic osteoarthropathy, is the classic clinical archetype—and prior work has shown that selenium supplementation protects cartilage. What remained unclear was how selenium status affected the inflamed synovium itself, and whether it could be exploited therapeutically. The researchers’ answer was to build a delivery vehicle that would exploit the very metabolic quirk that makes osteoarthritic synoviocytes pathological: their GLUT1-overexpressing, glucose-hungry surface.

The vehicle they engineered is elegantly simple in concept. Spherical selenium nanoparticles roughly 100 nanometers in diameter were synthesized by reducing sodium selenite with ascorbic acid in the presence of a PEGylated lipid, then decorated with mannose, a sugar that hijacks GLUT1-mediated uptake pathways. Transmission electron microscopy and dynamic light scattering confirmed uniform spheres with excellent colloidal stability across seven days at lysosomal, inflammatory, and physiological pH values. Fourier-transform infrared spectroscopy verified the carbohydrate coating through its characteristic O-H, C-H, and C-O stretching vibrations, and the zeta potential shifted from −2.7 to −1.2 millivolts after modification, a fingerprint of successful grafting. Cytotoxicity testing in synoviocytes and macrophages showed viability above 95 percent at concentrations up to 20 micromolar over 24 hours, setting a safe working dose well within international biocompatibility standards.

What happens to the nanoparticles once they enter a cell proved just as important as getting them there. Using high-performance liquid chromatography coupled with inductively coupled plasma mass spectrometry, the team tracked the intracellular fate of the selenium and found that the nanoparticles were rapidly metabolized into bioactive organic species, predominantly selenocystine and methylselenocysteine, the very precursors cells use to build selenoproteins. In other words, the particles do not merely dump an inert antioxidant into the cell; they feed the cell’s own selenoprotein assembly line. Quantitative proteomics of inflamed synoviocytes treated with the particles bore this out: of 8,868 proteins quantified, the antioxidant selenoprotein TXNRD1 more than doubled, and several other redox- and endoplasmic-reticulum-associated selenoproteins rose alongside it, while inflammatory mediators such as IL-6, CXCL1, CCL2, VCAM-1, and ICAM-1 fell.

The metabolic reprogramming was equally striking. Functional assays using Seahorse flux analysis showed that treated cells reduced their extracellular acidification rate, glucose uptake, and lactate production—hallmarks of glycolysis—while increasing their oxygen consumption rate, a sign that mitochondrial respiration was being restored. The particles suppressed the pro-glycolytic enzyme PFKFB3 and raised TXNIP, a known brake on glucose uptake, and the transcription factor ETS1, which correlates strongly with PFKFB3 in osteoarthritic tissue, declined in parallel. Mitochondrial membrane potential, which collapses under inflammatory oxidative assault, was stabilized as measured by JC-1 staining. In macrophages, the treatment pushed polarization away from the pro-inflammatory M1 phenotype, reducing expression of Il1b, Il6, Tnfa, Nos2, and Cd86, and shifting the population toward reparative M2-like cells in the mouse synovium.

Delivery was the make-or-break question for any intra-articular therapy, and here the mannose modification paid visible dividends. In a surgically induced mouse model of osteoarthritis, fluorescently labeled mannose-coated particles persisted in the knee joint for roughly fourteen days after a single injection, compared with about ten days for unmodified particles. Using genetically engineered reporter mice in which fibroblast-like synoviocytes glow green and myeloid cells glow red, the researchers showed that the modified particles accumulated preferentially inside both target cell types in the inflamed synovium. Systemic distribution was limited, with the liver as the main clearance organ, and histological, blood-count, and serum chemistry analyses found no signs of organ injury or toxicity after repeated dosing.

The functional payoff was measurable in the animals’ own behavior. Mice treated with the nanoparticles every two weeks for twelve weeks showed higher paw withdrawal thresholds on von Frey testing, meaning less mechanical hypersensitivity, along with longer rotarod latency and normalized gait parameters including stride length, swing time, and weight-bearing symmetry between the operated and healthy limbs. Expression of TRPA1, a ROS-sensing ion channel in sensory neurons and synoviocytes that has been implicated in osteoarthritic pain, was suppressed both in the dorsal root ganglia and in the joint lining. Histology told the same story from the inside: reduced synovial lining thickening and inflammatory infiltration, preserved proteoglycan content in the cartilage, lower OARSI degeneration scores, and diminished MMP13, the matrix-degrading enzyme that serves as a barometer of both cartilage catabolism and synovial inflammation.

The authors are careful about what their data do and do not prove. The restoration of TXNRD1 accompanied the anti-inflammatory effects but has not yet been shown to be causally required, and the ETS1–PFKFB3 relationship remains an association pending direct regulatory validation. Whether TRPA1 suppression is a direct effect of the selenium or a secondary consequence of a calmer joint environment is also unresolved. Even so, the study delivers a compelling proof of concept: a disease-modifying osteoarthritis therapy that does not simply block a single cytokine, but instead rebuilds the metabolic and redox architecture of the inflamed synovium from within. By turning the GLUT1-high phenotype of diseased synoviocytes from an accomplice of pathology into a delivery address, the work opens a genuinely new avenue for treating a condition that afflicts hundreds of millions of people worldwide and currently has no cure.

Subject of Research: Mannose-modified selenium nanoparticles that remodel glycolysis and redox homeostasis in inflamed synovium for osteoarthritis therapy

Article Title: Metabolic‐Redox Remodeling of Inflammatory Synovium by Mannose‐Modified Selenium Nanoparticles for Osteoarthritis Therapy

Article References: Zou, Y.-Q., Xu, C.-H., Li, B.-W., Chan, L., Xu, Y.-D., Su, H.-B., Peng, R., Luo, F.-J., Peng, Y.-C., Li, Z.-Y., Wong, V. K. W., Zheng, X.-F., Ma, X., Zha, Z.-G., Chen, T., & Zhang, H.-T. (2026). Metabolic‐Redox Remodeling of Inflammatory Synovium by Mannose‐Modified Selenium Nanoparticles for Osteoarthritis Therapy. Advanced Science, Article e78188. https://doi.org/10.1002/advs.78188

Image Credits: AI Generated

DOI: 10.1002/advs.78188

Keywords: osteoarthritis, selenium nanoparticles, mannose targeting, GLUT1, glycolysis, selenoproteins, synovial inflammation, macrophage polarization, redox homeostasis, PFKFB3, TRPA1, nanomedicine

The team began by isolating so-called red blood cell ghosts, the hemoglobin-free membrane remnants left behind after erythrocytes are lysed under controlled laboratory conditions. When human cardiac microvascular endothelial cells were exposed to these ghosts for just two hours, bulk RNA sequencing revealed 113 differentially expressed genes, the vast majority of them upregulated. Among the most enriched biological pathways were inflammatory response and blood vessel morphogenesis, with prominent induction of genes such as VEGF, TNFA, PTGS2 and NFKBIA. Bioinformatic mapping also linked the altered genes to disease pathways including ischemia, peripheral vascular disease, arteriosclerosis and diabetic retinopathy, hinting that the response is directly relevant to cardiovascular pathology.

Chromatin immunoprecipitation sequencing added an epigenetic dimension to the picture. The researchers detected 77 sites of altered histone 3 lysine 27 acetylation, a mark of active enhancers, with the overwhelming majority gaining acetylation after ghost treatment. Motif analysis of these newly opened genomic regions showed a striking predominance of binding sites for the activator protein-1, or AP-1, family of transcription factors, which accounted for 55 percent of the target sequences. Immunofluorescence and cellular fractionation confirmed that JUNB, a key AP-1 member, accumulated in the nuclei of treated endothelial cells, indicating that the ghost signal rewires the transcriptional landscape rather than merely nudging existing programs.

On the inflammatory side, the membrane fraction triggered robust activation of the transcription factor NFκB. Within one to two hours, endothelial cells showed increased phosphorylation of the NFκB p65 subunit at serine 536, followed by phosphorylation at serine 276, a site targeted by cyclic AMP-dependent protein kinase and critical for transcriptional activity. The inflammatory enzyme COX2 rose in parallel. Blocking NFκB activation with the IκBα inhibitor BAY11-7082 abolished both the COX2 increase and, crucially, the angiogenic response, establishing NFκB as the central gatekeeper of the ghost-induced program.

Functionally, the treated endothelial cells behaved like cells primed for vessel building. In spheroid angiogenesis assays, exposure to red blood cell ghosts increased the number of sprouts per spheroid and the total endothelial network length, an effect comparable to recombinant VEGF. The cells also proliferated more readily, as measured by Ki-67 staining, and migrated faster in scratch-wound assays. Notably, the hemoglobin-free membrane fraction alone was sufficient to drive these effects, while isolated heme had no impact, ruling out the iron-containing pigment as the responsible agent and pointing instead to components retained in the membrane itself.

Metabolic profiling revealed the engine behind this angiogenic conversion. Using Seahorse extracellular flux analysis, the researchers observed that ghost addition boosted endothelial glycolysis within minutes, with elevated proton efflux and extracellular acidification sustained for roughly five hours. Mass spectrometry-based metabolomics confirmed accumulation of glycolytic intermediates, activation of the pentose phosphate pathway, and increased acetyl-CoA and TCA cycle metabolites. Isotope tracing showed glucose-derived ribose feeding purine synthesis and glutamine carbon flowing into pyrimidine synthesis, a classic anabolic profile. The glycolytic regulator PFKFB3, previously established as a hub of vessel sprouting metabolism, rose significantly at the protein level, and blocking glycolysis with the glucose analog 2-deoxy-D-glucose or inhibiting glutaminolysis with BPTES abolished the sprouting response.

The hunt for the transcriptional conductor of this program led to NR4A1, an orphan nuclear receptor also known as NUR77, which was among the most strongly induced genes in both the RNA and ChIP sequencing datasets. NR4A1 protein, nearly undetectable in resting cells, surged after two hours of ghost exposure and accumulated in the nucleus in an NFκB-dependent manner. Silencing NR4A1 with siRNA or blocking it pharmacologically prevented the induction of JUNB, PFKFB3 and VEGF and significantly reduced sprout formation. Intriguingly, NR4A1 knockdown also increased basal NFκB phosphorylation, revealing a built-in negative feedback loop: the receptor dampens inflammation while simultaneously switching on the metabolic and angiogenic machinery.

The causal trigger turned out to be adenosine triphosphate stored within the erythrocyte membrane. Fluorescent ATP analog labeling confirmed ATP binding sites on red cell membranes, and extracellular ATP at micromolar concentrations reproduced the full ghost response, from NR4A1 induction to sprouting. Degrading extracellular ATP with apyrase blunted the angiogenic effect. Pharmacological and genetic experiments identified two purinergic receptors as essential: the fast-acting ionotropic P2X7 receptor, which drives the inflammatory NFκB activation, and the metabotropic P2Y11 receptor, which couples to adenylyl cyclase and cyclic AMP production. Real-time FRET imaging with a genetically encoded cAMP biosensor captured robust cytosolic cAMP surges in single endothelial cells exposed to either ATP or ghosts, and inhibiting membrane-bound adenylyl cyclase or preventing cAMP degradation with PDE4 inhibitors modulated the response accordingly.

The translational payoff came from patients with severe peripheral artery disease, a condition defined by defective vessel growth. Immunostaining of ischemic muscle from mice and patients confirmed that erythrocyte extravasation accompanies ischemia in vivo. When endothelial cells were treated with ghosts prepared from PAD patients, the inflammatory activation persisted, but the upregulation of PFKFB3, VEGF and the cAMP sensor EPAC1 was significantly blunted, and the sprouting response was markedly weaker than with ghosts from age- and sex-matched healthy controls. Remarkably, inhibiting phosphodiesterase 4 with rolipram to prevent cAMP degradation restored the patient ghost response to healthy levels, suggesting that the ATP–P2Y11–cAMP axis fails in disease and can be pharmacologically rescued.

The authors caution that their patient cohort was small and exploratory, that P2Y11 receptors are absent in mice, limiting animal validation, and that other red cell membrane factors may contribute alongside ATP. Even so, the study fundamentally repositions erythrolysis as an instructive event in vascular repair. The same mechanism that helps restore perfusion after hemorrhage could be exploited therapeutically to boost angiogenesis in ischemic limbs, or conversely suppressed in settings such as cancer or retinal disease, where erythrocyte leakage and pathological vessel growth feed each other. Targeting endothelial cAMP metabolism, the work suggests, may be a promising lever for tipping that balance in either direction.

Subject of Research: How lysed erythrocyte membranes and their ATP content metabolically prime endothelial cells for angiogenesis

Article Title: Lysed erythrocyte membranes metabolically prime endothelial cells for angiogenesis

Article References: Gogiraju, R., Moiko, K., Bochenek, M. L., Greulich, F., Witzler, C., Schmitz, W., Zifkos, K., Derieux, C., Ghasemi, I., Guliani, P., Sun, B., Espinola-Klein, C., Uhlenhaut, H. N., Bock, A., Ruf, W., Madhusudhan, T., Lurz, P., & Schäfer, K. (2026). Lysed erythrocyte membranes metabolically prime endothelial cells for angiogenesis. Angiogenesis, 29(4), Article 74. https://doi.org/10.1007/s10456-026-10096-y

Image Credits: AI Generated

DOI: 10.1007/s10456-026-10096-y

Keywords: angiogenesis, endothelial cells, erythrocytes, hemolysis, ATP signaling, NFκB, NR4A1, PFKFB3, cAMP, P2X7 receptor, P2Y11 receptor, peripheral artery disease

The researchers began from a metabolic observation that has been gaining traction across vascular biology. When blood vessels grow abnormally, the endothelial cells that line them shift their energy production toward glycolysis, the rapid, oxygen-independent breakdown of glucose. The enzyme PFKFB3, an allosteric activator of the glycolytic pathway, is central to this switch. Earlier studies showed that blocking PFKFB3 even partially reduces pathological vessel sprouting, and that endothelial cells lacking it cannot form vessels efficiently. What remained unclear was how PFKFB3 protein levels are controlled in the endothelial cells infiltrating endometriotic lesions.

To answer that question, the team examined vascular endothelial cells in ectopic endometrial tissue taken from both human patients and mouse models. In both species, they found that PFKFB3 was significantly upregulated, and that this elevation tracked closely with increased expression of YTHDF3, a member of the YTH domain family of proteins. These proteins are known as readers of N6-methyladenosine, or m6A, the most abundant internal chemical modification in eukaryotic messenger RNA. The m6A mark does not change the genetic code; instead, it influences the fate of each transcript, dictating whether it is translated into protein, degraded, or stored.

The causal importance of YTHDF3 became clear in genetic experiments. When the researchers specifically deleted the Ythdf3 gene in endothelial cells of mice, PFKFB3 protein levels dropped and angiogenesis within the ectopic endometrial lesions was suppressed. Notably, the manipulation targeted the translation step rather than the abundance of the mRNA itself. This distinction matters because it points to a regulatory layer that most existing therapies never touch: the efficiency with which a stable message is actually converted into functional protein.

Delving into the mechanism, the team focused on endometrial microvascular endothelial cells, where the PFKFB3 messenger RNA carried increased m6A modifications. The writer of those marks is METTL3, the catalytic core of the enzyme complex that installs m6A on transcripts. According to the study, METTL3 enhances m6A modification of the PFKFB3 mRNA, and YTHDF3 then recognizes these modified sites, promoting the translation of PFKFB3 and thereby raising glycolytic activity within the cells. The metabolic boost translates directly into cellular behavior: ovarian microvascular endothelial cells showed enhanced tube formation, migration and proliferation when the pathway was active.

The m6A system has been implicated in metabolism and angiogenesis before. Previous work showed that the modification regulates glycolysis in cancer cells through enzymes such as PDK4, and reviews have catalogued multiple RNA modifications shaping vessel growth. YTHDF3 itself has a colorful track record: it has been shown to promote the translation of m6A-enriched transcripts in breast cancer brain metastasis, to modulate antiviral interferon responses through FOXO3, and to assist translation by recruiting initiation factors. The new study extends this translation-enhancing role into a gynecological disease context, linking it to a hormone-driven pathology.

That hormonal link is perhaps the most clinically provocative element of the findings. Endometriosis is an estrogen-dependent disease, and the researchers found that estrogen upregulates both METTL3 and PFKFB3 through the estrogen receptor ERα. In other words, the very hormone that drives the growth of endometrial tissue also appears to arm the vessel-feeding machinery that sustains it, by amplifying the RNA-modification writer and the metabolic target it acts upon. This creates a coherent loop in which hormonal signaling, epitranscriptomic regulation and endothelial metabolism converge to support lesion establishment and expansion.

Angiogenesis has long been recognized as a requirement for endometriotic lesions to survive and grow after they are seeded within the pelvic cavity, and anti-angiogenic strategies have been proposed as adjunct therapies for years. The difficulty has been specificity: vessels in healing wounds and normal tissues also depend on glycolytic endothelial cells, so systemic blockade of PFKFB3 carries risks. The YTHDF3-m6A-PFKFB3 axis offers a more layered target. Interfering with the reader protein or with the estrogen-driven upregulation of METTL3 could, in principle, blunt pathological vessel growth in lesions while sparing some of the baseline vascular functions that depend on other regulators.

The study’s authors, led by Xiaosa Li, Jiale Wang and Liang Yuan as co-first authors under the direction of senior investigators including Xiaodong Fu, Huiping Lin and Junxiu Liu, argue that their findings establish the YTHDF3-m6A-PFKFB3 pathway as a critical driver of angiogenesis in endometriosis and suggest that targeting this pathway represents a promising therapeutic strategy. The work was supported by the National Natural Science Foundation of China and by a research fund from Guangzhou, and the authors declare no competing interests.

As with any mechanistic study, the road from molecular circuit to clinic is long. Animal models and cell culture systems capture only part of the human disease, and m6A readers act on many transcripts at once, so systemic interventions would need careful dosing and delivery strategies. Still, the study adds endometriosis to a growing list of conditions in which chemical tags on RNA, rather than changes in the genes themselves, determine whether disease-associated cells thrive. For millions of patients whose pain and fertility struggles have outpaced available treatments, the idea that a single reader protein controls the fuel supply of the vessels feeding their lesions is a compelling new lead, and one that researchers are likely to pursue with urgency.

Subject of Research: m6A reader YTHDF3 promotes angiogenesis in endometriosis by enhancing PFKFB3 translation

Article Title: YTHDF3 promotes angiogenesis in endometriosis by enhancing the translation efficiency of PFKFB3

Article References: Li, X., Wang, J., Yuan, L., Wang, Y., Wei, J., Li, P., Wang, R., Xu, X., Mai, Q., Liu, J., Lin, H., & Fu, X. (2026). YTHDF3 promotes angiogenesis in endometriosis by enhancing the translation efficiency of PFKFB3. Angiogenesis, 29(4), Article 70. https://doi.org/10.1007/s10456-026-10095-z

Image Credits: AI Generated

DOI: 10.1007/s10456-026-10095-z

Keywords: endometriosis, angiogenesis, YTHDF3, m6A, PFKFB3, METTL3, glycolysis, endothelial cells, estrogen, translation, PFKFB3 mRNA, vascular biology