Localized scleroderma, also called morphea, remains one of the most stubborn fibrotic skin diseases. It causes induration of the dermal and subcutaneous layers, and in severe cases leads to ulcers, disfiguring lesions, joint deformity, restricted movement, and even elevated fracture risk later in life. Current treatment relies mainly on immunosuppressants and glucocorticoids, with emerging options such as tocilizumab and JAK inhibitors, but no available therapy can reverse established skin fibrosis. That therapeutic gap drove the Sichuan team to look deeper into the molecular machinery that keeps fibroblasts, the collagen-producing cells of the dermis, locked in an activated, scar-forming state.

Their starting point was mitochondria. Previous work has established that mitochondrial dysfunction is a core driver of fibroblast activation in scleroderma, and that oxidative stress, a consequence of damaged mitochondria, plays an important role in the disease. Mitochondria are the main source of reactive oxygen species inside cells, and when their calcium handling goes awry, these reactive molecules accumulate and push cells toward senescence, a state of permanent growth arrest. Senescent cells then secrete a cocktail of inflammatory signals known as the senescence-associated secretory phenotype, which promotes fibroblasts and macrophages to adopt a fibrotic phenotype and fuels the scarring process.

Central to mitochondrial calcium control is a structure called the mitochondria-associated endoplasmic reticulum membrane, or MAM. At these contact sites, a channel complex composed of the proteins IP3R1, GRP75, and VDAC1 ferries calcium ions from the endoplasmic reticulum into mitochondria. When this flux becomes excessive, mitochondrial calcium overload damages the organelles, collapses their membrane potential, and elevates production of mitochondrial reactive oxygen species. Using transmission electron microscopy on skin samples from patients treated at West China Hospital between 2024 and 2025, the researchers found that the spatial proximity between the endoplasmic reticulum and mitochondria was significantly reduced in morphea lesions compared with healthy control tissue, providing the first preliminary evidence that MAM homeostasis is disturbed in this disease.

To find a drug that could restore that balance, the team turned to parthenolide, which has documented anti-inflammatory, antioxidative, and antifibrotic effects in liver, lung, kidney, and peritoneal fibrosis models. In human dermal fibroblasts stimulated with TGF-β, the master cytokine of fibrosis, parthenolide treatment reduced the number of β-galactosidase-positive senescent cells, increased proliferation as measured by EdU incorporation, and downregulated the senescence markers p21 and p16 along with the inflammatory cytokines IL-6, IL-1β, and TNF-α. At the highest concentration tested, 5 micromolar over 48 hours, the compound also lowered mRNA and protein levels of α-smooth muscle actin and type I collagen, the hallmarks of myofibroblast activation. Notably, the effects were concentration-dependent, with anti-aging effects appearing at medium and high doses and antifibrotic effects at high doses. When the researchers cleared senescent cells using the dasatinib and quercetin combination, fibroblast activation also eased, supporting the idea that parthenolide relieves fibrosis at least in part by modulating senescence.

The mechanistic hunt then moved to the MAM itself. Confocal microscopy showed that parthenolide diminished the colocalization between mitochondria and the endoplasmic reticulum and increased the inter-organelle distance in activated fibroblasts, while flow cytometry revealed restored mitochondrial membrane potential and reduced reactive oxygen species. RNA sequencing identified 1,111 genes differentially expressed after TGF-β stimulation and 290 after parthenolide treatment, with 161 genes shared between the two sets. Enrichment analyses pointed to wound healing, collagen-containing extracellular matrix, and endoplasmic reticulum lumen pathways. To narrow the field, the researchers performed inverse virtual screening against roughly 2,600 protein structures from the Protein Data Bank, ranking candidates by binding free energy, and intersected the top 100 hits with their differentially expressed genes. Two genes emerged: PI16 and Ero1L.

Ero1L, an oxidoreductase enzyme that lives predominantly at the MAM and responds to hypoxia and oxidative stress, became the prime suspect. Molecular docking predicted that parthenolide binds Ero1L with a binding energy of −7.3 kilocalories per mole, and surface plasmon resonance on a Biacore instrument confirmed a direct physical interaction. Docking simulations highlighted two key residues, ARG-449 and PHE-62, and site-directed mutagenesis showed that converting ARG-449 to alanine abolished the binding affinity, while the PHE-62 mutation had no significant effect, confirming the specificity of the interaction. Parthenolide also reduced Ero1L mRNA and protein expression in TGF-β-treated cells. Single-cell RNA sequencing of a public dataset covering fourteen localized scleroderma patients and fourteen healthy controls, encompassing 32,154 cells clustered into fifteen cell types, showed that Ero1L is expressed at higher levels in activated fibroblasts than in other cell subtypes, and immunohistochemistry revealed elevated Ero1L in patient skin biopsies that correlated positively with both senescence markers and the modified Localized Scleroderma Skin Severity Index.

Causality tests sealed the case. When the researchers overexpressed Ero1L in fibroblasts, the protective effects of parthenolide were reversed: senescence and fibrosis markers climbed back up, mitochondrial reactive oxygen species rose, membrane potential fell, and the compound’s suppression of ER-mitochondrial colocalization was blunted. Conversely, knocking Ero1L down with siRNA mimicked the drug’s benefits, reducing fibrotic and senescent markers, restoring mitochondrial health, increasing the average ER-mitochondria distance on electron micrographs, and weakening the interactions within the IP3R1-GRP75-VDAC1 calcium channel complex. Parthenolide treatment similarly downregulated the three channel proteins and eased mitochondrial calcium overload, suggesting that Ero1L acts upstream of calcium flux at the MAM. The authors propose that Ero1L may also promote fibrosis indirectly through redox regulation, since the enzyme reoxidizes protein disulfide isomerase to maintain the disulfide bonds needed for type I collagen folding and secretion.

Because parthenolide’s poor water solubility and low bioavailability have hindered clinical use, the team engineered a delivery system combining PLGA nanoparticles loaded with the drug and a thermoresponsive F127 poloxamer hydrogel. Electron microscopy showed spherical nanoparticles, and rheological testing confirmed the hydrogel’s injectability and temperature-dependent gel transition at body temperature. High-performance liquid chromatography demonstrated a slower, sustained release without an initial burst phase, and live-dead staining and CCK-8 assays confirmed good biocompatibility. In a bleomycin-induced mouse model of localized scleroderma, weekly injections of the formulation, designated F127@PLGA@PTL at 70 milligrams per kilogram, significantly attenuated dermal thickness, collagen deposition, and myofibroblast counts, reduced Ero1L protein expression, and showed anti-aging effects, while multiplex immunofluorescence indicated the treatment specifically quieted activated fibroblasts without measurable effects on vascular endothelial cells, keratinocytes, or Th2 lymphocytes.

The study is the first to implicate Ero1L in the pathogenesis of localized scleroderma and to demonstrate an anti-aging effect of parthenolide in skin fibrosis, positioning the ER-mitochondrial interface as a therapeutic target for hardening skin diseases. The authors caution that their findings rest on cell culture and mouse models and require validation in human subjects, and they note that parthenolide can behave differently in other contexts, even inducing mitochondrial dysfunction in cancer cells, so its biological effects appear cell-specific and shaped by metabolic state. Still, by connecting a humble herbal compound, a calcium-handling hub, and an aging program into one mechanistic chain, the work opens a concrete path toward therapies that could do what current medicine cannot: soften fibrotic skin by rejuvenating the cells that build it.

Subject of Research: Parthenolide targeting of Ero1L to regulate mitochondria-associated ER membranes in localized scleroderma fibrosis and cellular senescence

Article Title: Targeting Ero1L by Parthenolide Alleviates Cellular Senescence and Fibrosis of Localized Scleroderma by Regulating Mitochondria‐Associated Endoplasmic Reticulum Membranes Stabilization

Article References: Wang, F., Xing, R., Xie, M., Qin, Y., Zhang, Y., Cao, D., He, G., & Lyu, X. (2026). Targeting Ero1L by Parthenolide Alleviates Cellular Senescence and Fibrosis of Localized Scleroderma by Regulating Mitochondria‐Associated Endoplasmic Reticulum Membranes Stabilization. Aging Cell, 25(10), Article e70738. https://doi.org/10.1111/acel.70738

Image Credits: AI Generated

DOI: 10.1111/acel.70738

Keywords: localized scleroderma, parthenolide, Ero1L, mitochondria-associated ER membranes, cellular senescence, fibrosis, calcium flux, dermal fibroblasts, PLGA nanoparticles, F127 hydrogel, oxidative stress, TGF-beta

The compound at the center of the research is puerarin, an isoflavone extracted from the root of the kudzu vine, a plant long used in traditional Chinese medicine. Puerarin has already attracted attention for its antioxidant and anti-inflammatory properties, and previous studies have linked it to protection of heart, brain, and liver tissue. The problem is that puerarin is a notoriously difficult drug to deliver. It dissolves poorly in water, degrades quickly in the body, and is metabolized so rapidly that very little of an oral dose ever reaches target tissue. These pharmacokinetic weaknesses have kept it out of mainstream clinical use despite two decades of promising laboratory data.

To overcome that barrier, a team of researchers from Saudi Arabia and Egypt, led by Shaza Alyamani of Batterjee Medical College and Gehad Elshopakey of Mansoura University, encapsulated puerarin inside nanoparticles made of poly(lactic-co-glycolic acid), or PLGA, a biodegradable polymer already approved by regulators for use in drug delivery and medical devices. The resulting particles were remarkably uniform: roughly 69 nanometers in diameter on average, with individual particles ranging between 46 and 65 nanometers, a narrow size distribution, and a strongly negative surface charge that keeps them from clumping together. Using a modified nanoprecipitation and solvent evaporation technique, the team achieved an encapsulation efficiency of about 84 percent and a drug loading of 8.5 percent, meaning the tiny particles carry a substantial cargo of the active compound.

Laboratory release tests revealed why the formulation matters so much. Free puerarin dumped nearly all of its payload within 24 hours in a burst that would be wasted in the bloodstream. The PLGA-encapsulated version released only about 8 percent initially and then trickled out steadily, reaching roughly 52 percent at 12 hours and about 91 percent over 72 hours. That sustained profile means the drug stays in circulation longer, and Fourier transform infrared spectroscopy confirmed that puerarin and the polymer interact only through non-covalent forces, preserving the chemical integrity of both components.

With the formulation validated, the researchers turned to an animal model of arsenic poisoning. Sixty male Sprague-Dawley rats were randomly divided into six groups. One group served as untreated controls, two groups received free puerarin or the nanoparticle formulation alone, one group received arsenic at 10 milligrams per kilogram of body weight orally for 14 days to induce subacute liver injury, and two groups received arsenic alongside either free puerarin or the nanoformulation at an equivalent dose of 100 milligrams per kilogram. The study was designed with a formal power analysis and followed international ethical standards for animal research, including OECD guidelines and ARRIVE recommendations.

The arsenic-only animals showed all the classic signatures of a poisoned liver. Serum levels of the liver enzymes ALT, AST, and ALP surged, indicating that hepatocyte membranes had ruptured and leaked their contents into the blood. Albumin and total protein dropped, reflecting a failing synthetic capacity, and the lipid profile deteriorated in a pattern consistent with mitochondrial dysfunction. At the molecular level, the picture was even more striking. Arsenic exposure flooded liver cells with reactive oxygen species, oxidized DNA, as measured by the marker 8-OHdG, and drove iron accumulation. Crucially, the rats’ livers showed a coordinated collapse of the antioxidant machinery that normally guards against a recently recognized form of cell death called ferroptosis, an iron-dependent process in which lipid membranes are destroyed by runaway peroxidation. The ferroptosis-suppressing proteins GPX4 and SLC7A11 were sharply downregulated, while the pro-ferroptotic enzymes ACSL4 and transferrin receptor 1 climbed, alongside mounting levels of lipid peroxidation products such as MDA and 4-HNE.

Both forms of puerarin pushed back against this molecular storm, but the nanoparticle version did so more decisively. PU-PLGA treatment restored the SLC7A11/GPX4 axis, boosted nuclear levels of the master antioxidant regulator NRF2, slashed hepatic iron, and normalized markers of lipid peroxidation to near-control values. It also quieted the inflammatory cascade, suppressing NF-κB signaling and lowering the cytokines TNF-α, IL-6, and IL-1β, along with the enzyme COX-2. Fibrotic signaling, marked by TGF-β and collagen type I, retreated as well. The nanoformulation outperformed free puerarin on nearly every measure, and the differences between the two treatments were statistically significant for most biomarkers, including NRF2, GPX4, ACSL4, TNF-α, and the fibrosis markers.

Perhaps most striking were the cell-death pathways. Arsenic tipped the apoptotic balance toward death by raising pro-apoptotic Bax and caspase-3 while depleting the survival proteins Bcl-2 and phosphorylated AKT. PU-PLGA reversed this imbalance, restoring anti-apoptotic signaling more effectively than the crude compound. The benefits were visible under the microscope, too. Light microscopy revealed that arsenic-treated livers suffered dilated sinusoids, hepatocyte vacuolation, and necrotic nuclei, while transmission electron microscopy showed swollen mitochondria with shattered cristae and dilated endoplasmic reticulum. In the nanoparticle-treated animals, hepatic architecture and mitochondrial ultrastructure were largely preserved, and immunohistochemistry confirmed robust restoration of GPX4 in hepatocytes alongside suppression of ACSL4 staining.

The authors are appropriately careful about interpretation. They note that the ferroptosis findings rely on ELISA-based biomarkers rather than direct functional validation with lipid-reactive-oxygen-species probes or ferroptosis inhibitors such as ferrostatin-1, so the data support ferroptosis-associated pathway modulation rather than definitive proof of ferroptotic cell death being blocked. Similarly, pharmacokinetic measurements of puerarin levels in blood and liver were not performed, so the superior performance of the nanoformulation, while plausibly attributable to sustained release and improved bioavailability, cannot yet be pinned to a specific mechanism. The study also used only male rats over a short two-week window, leaving questions of sex differences and long-term fibrosis progression open for future work.

Even with those caveats, the study is a compelling demonstration of how nanotechnology can rescue a pharmacologically gifted but delivery-challenged natural product. By wrapping puerarin in biodegradable PLGA shells, the researchers achieved simultaneous suppression of oxidative stress, ferroptosis-related signaling, inflammation, apoptosis, and early fibrotic remodeling in a living model of arsenic hepatotoxicity. For the millions of people worldwide whose drinking water carries dangerous levels of arsenic, the work offers a proof of concept that a kudzu-derived isoflavone, delivered at the nanoscale, could one day become a genuinely protective therapy. Translational validation, safety profiling, and human pharmacokinetic studies remain the necessary next steps, but the groundwork is now firmly in place.

Subject of Research: Hepatoprotective effects of PLGA-loaded puerarin nanoparticles against arsenic-induced liver injury through modulation of ferroptosis and oxidative stress

Article Title: PLGA-loaded puerarin attenuates arsenic-induced liver injury through modulation of ferroptosis, oxidative stress, inflammation, and apoptotic signaling

Article References: Alyamani, S. A., Hifni, B. A., Alqahtani, N. S., Elmorsy, E. M., Al-Ghafari, A. B., Al Doghaither, H. A., Chatha, W. A., & Elshopakey, G. E. (2026). PLGA-loaded puerarin attenuates arsenic-induced liver injury through modulation of ferroptosis, oxidative stress, inflammation, and apoptotic signaling. BMC Pharmacology and Toxicology, 27(1), Article 129. https://doi.org/10.1186/s40360-026-01221-0

Image Credits: AI Generated

DOI: 10.1186/s40360-026-01221-0

Keywords: puerarin, PLGA nanoparticles, arsenic, hepatotoxicity, ferroptosis, oxidative stress, Nrf2, GPX4, liver injury, nanomedicine, inflammation, apoptosis

The biological inspiration for DPN comes from erythrocytes. Native red blood cells evade immune clearance through two complementary traits: surface protein expression and geometric deformability. While several research groups have pursued the first strategy by cloaking synthetic particles with membranes harvested directly from red blood cells, the Decuzzi group focused on the second, purely mechanical property. Their reasoning draws on well-established hemodynamic principles. In flowing blood, particles with a high aspect ratio and flat profile experience asymmetric hydrodynamic forces that push them laterally toward the vessel wall, a phenomenon known as margination. Unlike spheres, which tend to remain suspended in the central stream, discoidal particles drift into the red-cell-depleted plasma layer adjacent to the endothelium, positioning themselves for adhesion or drug release at pathological sites. Computational fluid dynamics studies have confirmed that margination efficiency improves with increasing particle size and with departure from spherical symmetry, because the reduced rotational freedom of flat particles near a surface prolongs their residence time at the vessel wall.

What distinguishes DPN from the vast body of PLGA-based nanoparticle research is not the choice of materials but the precision of their fabrication. Rather than relying on thermodynamic self-assembly, which produces predominantly spherical particles with limited independent control over geometry, DPN are manufactured through a top-down, template-assisted strategy. Particle diameter, height, aspect ratio, and mechanical stiffness can each be tuned independently by modifying the template geometry or the polymer paste composition, without altering any other property. The structural matrix is formed by the physical entanglement and crosslinking of PLGA and PEG chains, yielding a hydrogel-like composite that degrades through the well-characterized hydrolytic breakdown of PLGA into lactic and glycolic acid, metabolites with an extensively documented safety profile in humans and a place in several FDA-approved drug delivery products.

Perhaps the most conceptually novel aspect of the platform is the treatment of mechanical stiffness as an independently tunable design variable. By systematically varying the ratio of PLGA to PEG while holding all other parameters constant, the researchers produced DPN variants spanning from soft constructs with a Young’s modulus of approximately 1.3 kilopascals to rigid ones at around 15 kilopascals, all sharing the same 1,000 by 400 nanometer disc geometry and a surface zeta potential of roughly minus 14 millivolts. The biological consequences are substantial. Mechanistic studies with liquid-filled silica nanocapsules of calibrated stiffness have shown that softer particles resist macrophage internalization far more effectively than rigid counterparts, because the energetic cost of membrane wrapping around a deformable object exceeds that of engulfing a geometrically stable one. Quantitatively, macrophage uptake of soft nanocapsules has been found to be reduced by approximately threefold relative to stiff equivalents of comparable size and surface chemistry. Within the DPN framework, a systematic experimental study spanning circular, quadrangular, and elliptical geometries across stiffness values from roughly 100 kilopascals to 10 megapascals confirmed that reduced Young’s modulus consistently and substantially lowered phagocytic uptake regardless of particle shape. Crucially, this immune evasion is achieved through purely synthetic means, by tuning crosslink density, offering reproducibility and scalability advantages that become critical as the platform moves toward clinical development.

The pharmacokinetic payoff of combining discoidal geometry with mechanical softness has been striking in preclinical models. In tumor-bearing mice, including brain and skin cancer xenografts, intravenously administered soft DPN maintained detectable blood concentrations for approximately 24 hours post-injection and achieved tumor tissue accumulation of up to 20% of the injected dose per gram of tumor. That figure stands in dramatic contrast to the field-wide average of below 1%, and it is accomplished passively, without active targeting ligands, relying entirely on the physical properties of the construct. The mechanistic basis lies in the tumor microvasculature itself: irregular caliber, high tortuosity, and sluggish perfusion create flow conditions under which flat, deformable particles are more readily arrested and retained at the vascular wall than spherical ones.

Translating these advantages into effective chemotherapy required solving a genuine formulation problem. The soft, spongy polymer matrix that minimizes macrophage uptake retains small molecules of 500 to 1,000 daltons poorly, a limitation the authors addressed through three distinct engineering strategies. The first, a multi-passage approach, applies multiple polymer deposition layers without altering the PLGA-to-PEG ratio, achieving a twofold improvement in docetaxel loading efficiency. The resulting docetaxel-loaded DPN showed significantly higher in vivo efficacy than free intravenous docetaxel in an orthotopic murine model of triple-negative breast cancer, demonstrating therapeutic benefit at a low dose of 3 milligrams per kilogram. The second strategy involved synthesizing a series of docetaxel prodrugs conjugated to PEG chains of varying length, informed by coarse-grained molecular dynamics simulations of drug release from PLGA matrices. While an oleic acid conjugate showed excessively slow release, intermediate hydrophilic PEG derivatives performed well, and the lead candidate, PEG550-docetaxel-DPN, significantly slowed disease progression in an orthotopic glioblastoma model compared with temozolomide and untreated controls, with only 40% of treated mice succumbing before day 40 versus 80% of temozolomide-treated animals.

The third strategy pushed the platform further with micro-Combinatorial Hydrogel Particles, or µCGP, two-micrometer discoidal hydrogels built from a hydrolytically labile PEG network that encapsulates smaller nanoparticles of roughly 200 nanometers along with molecules, exosomes, or antibodies. Fabricated without organic solvents, these hierarchical particles combine the vascular margination and long circulation of DPN with the deep tissue penetration of small nanoparticles. Loaded with docetaxel-carrying polymeric nanoparticles and tested in a late-stage lung metastasis model of triple-negative breast cancer, µCGP showed preferential accumulation within the lung capillary network, preserved alveolar architecture in treated animals, and enabled approximately 50% of mice to survive four months post-treatment, whereas all control groups died within three months. Imaging confirmed substantial lung accumulation and markedly enhanced penetration of the released nanoparticles into metastatic nodules, supporting the authors’ argument that the enhanced permeability and retention effect is not a dominant mechanism in metastatic disease and that engineered delivery platforms must compensate.

Beyond oncology, the vascular confinement of DPN has enabled a compelling application in thrombolytic therapy. Tissue plasminogen activator, marketed as Alteplase, is the standard treatment for acute ischemic stroke, but systemically administered tPA can cross the disrupted blood-brain barrier and exacerbate neurological injury. By covalently conjugating tPA onto the DPN surface through EDC/NHS bioconjugation, the researchers confined the drug within the vascular compartment. In a mesenteric thrombosis model, tPA-DPN recanalized 90% of occluded vessels and reduced clot size by 50% within 35 minutes, compared with 40% recanalization and 20% clot reduction for free tPA. In a severe middle cerebral artery occlusion model of ischemic stroke, mice treated with tPA-DPN showed survival rates and behavioral scores comparable to saline controls, while free tPA-treated animals displayed severely impaired survival and neurological deficits, with histology and MRI confirming smaller infarcts and reduced blood-brain barrier leakage. Notably, enhanced thrombolytic efficacy was achieved with only 1 milligram per kilogram of tPA in mice, roughly one-tenth of the conventional murine equivalent of the clinical human dose. The team has further decorated tPA-DPN with fucoidan, a sulfated polysaccharide with strong affinity for P-selectin expressed on activated platelets and endothelial cells, and preliminary experiments showed stable binding of these functionalized particles to artificial clots where conventional tPA-DPN failed to remain attached.

Significant hurdles remain before clinical translation. The template-assisted fabrication process, though precise, involves sequential operations that must be scaled while preserving uniform cavity filling and consistent particle recovery, and the authors note that variations in polymer-paste rheology, solvent evaporation, and drying conditions could introduce batch-to-batch variability. Robust quality control, suitable sterilization methods, long-term storage conditions, and comprehensive characterization of biodistribution, immunogenicity, and toxicity will all be required, and multi-compartment platforms such as µCGP may be classified as combination products under FDA and EMA frameworks, complicating their regulatory pathway. The authors also point to computational and machine-learning approaches as a means to reduce the experimental design space and accelerate optimization of particle size, shape, and deformability. Still, the evidence assembled in this review presents a persuasive argument that geometry, mechanics, and surface functionality, engineered together rather than separately, can deliver what surface chemistry alone has not: a nanocarrier whose performance advantages are structurally built in rather than bolted on.

Subject of Research: Biomimetic deformable discoidal polymeric nanoconstructs for targeted drug delivery in cancer and thrombotic disease

Article Title: Deformable discoidal polymeric nanoconstructs: Design principles, therapeutic applications, and translational perspectives

Article References: Spanò, R., Palomba, R., Felici, A., Decuzzi, P., & Palange, A. L. (2026). Deformable discoidal polymeric nanoconstructs: Design principles, therapeutic applications, and translational perspectives. Biomedical Microdevices, 28(4), Article 69. https://doi.org/10.1007/s10544-026-00854-6

Image Credits: AI Generated

DOI: 10.1007/s10544-026-00854-6

Keywords: nanomedicine, drug delivery, PLGA nanoparticles, discoidal nanoconstructs, cancer nanotechnology, triple-negative breast cancer, glioblastoma, ischemic stroke, thrombolytic therapy, tissue plasminogen activator, particle mechanics, margination