Advanced nanoformulations of NUAK1 regulate NLRP3 inflammasome for preeclampsia management in mice Authors: Peiyue Jiang, Xue Ying, Zhi Li, Weidong Fei, Yao Yao, Jia Xu, Hetong Li, Lujiao Chen, Qiong Luo Published: 2026 ABSTRACT AbstractPreeclampsia (PE) is a severe pregnancy complication driven by placental dysfunction and oxidative stress, which currently lacks effective targeted therapies. Here, we evaluated the therapeutic potential of folic acid-modified lipid nanoparticles (NPs) co-loaded with polydopamine and NUAK Family Kinase 1 (NUAK1), termed FLN@(PDA + NUAK1) NPs, for mitigating PE. In vitro, these NPs effectively reduced reactive oxygen species levels and suppressed activation of the NLR Family Pyrin Domain Containing 3 (NLRP3) inflammasome, while promoting autophagic activity and inhibiting apoptosis under oxidative stress. In vivo analysis using an L-NAME-induced PE mouse model further confirmed that FLN@(PDA + NUAK1) NPs improved placental structure and fetal growth parameters. This study highlights the dual functionality of NPs in activating protective autophagy and suppressing inflammasome overactivation, offering a promising strategy for the treatment of PE.Subject terms: Molecular biology, Biotechnology, Diseases Introduction Preeclampsia (PE) is a serious complication of pregnancy-induced hypertension, primarily characterized by hypertension and proteinuria1. It typically arises after 20 weeks of gestation and, in severe cases, may progress to eclampsia, posing substantial risks to both maternal and fetal health2. According to the World Health Organization, PE and eclampsia are among the leading causes of maternal mortality and adverse fetal outcomes3. Globally, ~2–8% of pregnancies are affected by PE, with a disproportionate burden in developing countries1,4. PE can result in multi-organ dysfunction, including renal impairment, hepatic abnormalities, and neurological complications, and may progress to HELLP syndrome (hemolysis, elevated liver enzymes, and low platelet count)5,6. Therefore, early diagnosis and effective treatment are crucial. Despite extensive research, the etiology of PE remains incompletely understood, with genetic, immune, and vascular factors implicated7,8. Oxidative stress and inflammation are critical in PE development9–11. Oxidative stress results from an imbalance between free radicals and antioxidants, leading to excess production of reactive oxygen species (ROS)12–14. ROS can damage cellular membranes, proteins, and DNA, ultimately impairing cellular function15,16. In placentas affected by preeclampsia, oxidative stress levels are markedly elevated, which can lead to damage to placental cells and their macromolecular structures. This, in turn, induces placental dysfunction and exacerbates pathological conditions in both the mother and the fetus17,18. Furthermore, inflammation also plays a pivotal role in PE progression19. The abnormal expression of inflammatory mediators, such as cytokines and chemokines, disrupts placental vascular function, reducing placental perfusion and impairing fetal development20,21. Nanotechnology holds great promise in biomedical applications due to its unique small size, surface, and quantum effects22. Engineered nanoparticles (NPs) can serve as targeted drug-delivery systems, enhancing therapeutic efficacy and minimizing off-target effects23. Functionalization of NP surfaces can improve disease-specific targeting, increase local drug concentrations, and reduce toxicity to healthy tissues24,25. Nanotechnology has shown promising translational potential in various diseases, including cancer26,27. Polydopamine (PDA), a biopolymer formed by oxidative self-polymerization of dopamine (DA), was first introduced by the Messersmith group in 200728. Subsequently, PDA coatings have been widely applied across various fields, including the surface functionalization of nanomaterials, drug delivery, and multiple biomedical applications29. Enriched with catechol groups, PDA exhibits excellent adhesive properties, and its surface charge tends to become negatively charged in response to environmental pH changes, thereby contributing to enhanced dispersibility and stability of PDA-coated nanomaterials30,31. In addition, catechol moieties serve as effective chelators of multivalent metal ions (e.g., iron, zinc, copper, cerium, and gadolinium), enabling the coordination-based adsorption of therapeutic agents such as drugs32 and DNA33. PDA’s biocompatibility and low cytotoxicity support its broad application in biomedicine34, while its antioxidant properties help protect biomolecules from oxidative damage29. Folic acid (FA) is vital for fetal development, especially in preventing neural tube defects and other congenital abnormalities. During pregnancy, maternal demand for FA increases 5–10-fold to ensure sufficient fetal supply. The placenta mediates this transfer via specific folate transporters: folate receptors, proton-coupled folate transporters, and reduced folate carriers35–37. Studies have reported that FA regulates trophoblast function via the mTOR signaling pathway, thereby influencing fetal growth. FA deficiency suppresses the activity of mTORC1 and mTORC2, reduces amino acid transport capacity, and consequently impairs fetal nutrient supply38. In addition, FA-modified NPs have been shown to precisely target triple-negative breast cancer cells, enhance intratumoral drug accumulation, improve antitumor efficacy, reduce cytotoxicity to normal cells, and prolong survival in tumor-bearing mice, demonstrating favorable biosafety and offering a promising therapeutic strategy for breast cancer39. Based on this, FA-modified NPs have been employed for targeted delivery to the placenta. NUAK Family Kinase 1 (NUAK1) is a protein kinase involved in cellular energy metabolism and stress response regulation. Recent studies suggest that NUAK1 is implicated in the pathogenesis of various diseases, including cancers, metabolic disorders, and inflammatory conditions40–42. The NLR Family Pyrin Domain Containing 3 (NLRP3) inflammasome is a multi-protein complex that senses pathological stimuli and mediates inflammation through the release of pro-inflammatory cytokines43,44. In the context of PE, NUAK1 and NLRP3 may influence placental structure and function by modulating autophagy and intracellular ROS levels. Our study delves into the effects of FA-modified lipid NPs co-loaded with polydopamine and NUAK1 (FLN@(PDA + NUAK1) NPs) on placental function in PE. Specifically, we examine their influence on autophagic flux, NLRP3 inflammasome activity, and ROS production in placental cells by modulating oxidative stress and mitochondrial function. Using both in vitro trophoblast cell models and an in vivo PE mouse model, we aim to demonstrate the potential of these NPs to enhance placental function and alleviate or prevent PE symptoms. Ultimately, our findings will contribute to a deeper understanding of PE pathophysiology and offer a scientific rationale for the development of novel therapeutic strategies to improve maternal and fetal health. Results Construction of ROS-responsive dopamine-NUAK1 protein kinase NPs PE is a pregnancy-associated hypertensive disorder with an incompletely understood etiology. Increasing evidence highlights the central role of oxidative stress in its pathophysiology21,45. In PE, abnormal placental development and dysfunction result in impaired placental perfusion and excessive production of ROS46. These ROS not only damage placental tissues but also trigger systemic inflammatory responses, vascular endothelial injury, and hypertension in the mother46,47. Therefore, reducing or neutralizing ROS is considered a critical therapeutic strategy for the treatment and prevention of PE. To address this, we developed a ROS-responsive NP system designed to selectively release therapeutic agents in oxidative stress conditions, thereby targeting and repairing damaged tissues to restore maternal and fetal health. We first synthesized DSeMSN and conducted comprehensive structural and chemical characterization. FTIR spectroscopy confirmed the presence of diselenide bonds, revealing characteristic Se–Se absorption peaks alongside vibrational signatures for silanol (Si–OH) and carbon–carbon (C–C) bonds (Supplementary fig. 1A). BET analysis showed a high specific surface area (564.3 m2/g), pore volume (0.31 cm3/g), and an average pore diameter of ~9.23 nm, conducive to effective protein encapsulation (Supplementary Fig. 1B). Furthermore, XPS analysis demonstrated changes in the chemical state of selenium in aqueous and H2O2 environments. Upon H2O2 exposure, cleavage of diselenide bonds was observed, with a notable increase in Se binding energy from 56 to 60 eV, confirming the ROS-responsiveness of DSeMSN (Supplementary Fig. 1C–E). Next, PDA and NUAK1 protein kinase were loaded into DSeMSN, followed by lipid coating with DSPE-PEG, PC, and CHO using an extrusion method to yield FLN@(PDA + NUAK1) NPs (Fig. 1A). TEM imaging revealed that the FLN@(PDA + NUAK1) NPs possessed a uniform spherical morphology. Particle size increased following PDA and NUAK1 encapsulation, with NUAK1-containing NPs significantly larger than those loaded with PDA alone (Fig. 1B). DLS analysis indicated that FLN@(PDA + NUAK1) NPs had an average hydrodynamic diameter of 160.23 ± 15.52 nm, a narrow size distribution, and a PDI of 0.19, suggesting good monodispersity (Fig. 1B, C). Zeta potential measurements showed a surface charge of −22.21 ± 1.31 mV (Fig. 1D), indicating good colloidal stability and biocompatibility under physiological conditions. In addition, stability tests were performed on the various NP formulations, and the results showed that all NPs remained stable over a 72-h period, with no significant changes in particle size or PDI (Supplementary Fig. 2A–C). Long-term stability was confirmed by monitoring particle size and PDI in PBS at 37 °C for 35 days, showing no substantial variation throughout the period (Fig. 1E, Supplementary Fig. 2D–F). HPLC analysis revealed a NUAK1 EE of 78.24 ± 4.18% and a drug loading content of 14.36 ± 1.27% (Fig. 1F). To assess ROS-responsiveness, drug release profiles were evaluated under normal and oxidative conditions. In the presence of 100 μM H2O2, diselenide bond cleavage triggered liposome disintegration and a marked increase in drug release, confirming the system’s ROS-sensitive behavior (Fig. 1G). Preparation and characterization of FLN@(PDA + NUAK1) NPs. A Schematic illustration of the preparation of FLN@(PDA + NUAK1) NPs (Created in BioRender. Jiang, P. (2026) https://BioRender.com/m2zwiu4). B TEM image showing the morphology, bar=50 nm. C Hydrodynamic diameter of the nanoparticles measured by DLS. D Zeta potential distribution graph of the nanoparticles. E Short-term stability test results of FLN@(PDA + NUAK1) NPs. F Standard curve for the release of the FITC-labeled model drug determined by HPLC. G Release profiles under various conditions using FITC as a model drug. Data are presented as mean ± SD (n = 6 independent biological replicates). NUAK1 NUAK Family Kinase 1, PDA polydopamine, PDI polydispersity index, PBS phosphate-buffered saline, h hour. These results successfully demonstrate the construction of NUAK1 protein kinase-loaded NPs with effective ROS-responsiveness, verifying their efficacy and stability regarding size distribution, surface charge, drug release, and long-term stability. This lays a solid foundation for subsequent in vitro and in vivo studies. Targeting efficiency and ROS responsiveness of FLN@(PDA + NUAK1) NPs in placental cells After constructing the FLN@(PDA + NUAK1) NPs, we further evaluated their placental cell-targeting uptake behavior and release characteristics under oxidative stress conditions. Because subsequent experiments required good cellular compatibility, we first assessed the in vitro cytotoxicity of the NPs. Macrophages are abundant in the endometrium and play key roles in implantation and placental development48. Cytotoxicity assays were conducted using placental trophoblast cells (HTR8/SVneo) and macrophages (RAW264.7). Cells were treated with increasing concentrations of FLN@(PDA + NUAK1) NPs. Live/dead staining confirmed that NPs exhibited no significant cytotoxicity up to a concentration of 100 μg/mL (Supplementary Fig. 3A–E). The cellular uptake of FITC-labeled FLN@(PDA + NUAK1) in HTR8/SVneo cells was visualized by confocal microscopy (Supplementary Fig. 4A). After 6 h of incubation, fluorescence intensity in cells treated with FLN@(PDA + NUAK1) was markedly higher than in those treated with unmodified LN@(PDA + NUAK1), confirming the targeting enhancement conferred by FA modification (Supplementary Fig. 4B). Quantitative analysis revealed that uptake of FLN@(PDA + NUAK1) reached ~50%, which was ~3-fold higher than that of the unmodified NPs (Supplementary Fig. 4C), demonstrating that FA significantly improves NP uptake efficiency in trophoblast cells. To mimic the oxidative microenvironment of preeclamptic placentas, HTR8/SVneo cells were treated with 100 μM H2O2, and the ROS-responsive release behavior of NPs was assessed. Under high oxidative stress, fluorescence spectroscopy revealed a significant increase in fluorescence intensity, indicating efficient release of NUAK1 protein kinase from NPs (Supplementary Fig. 4D). Quantitative data showed that the release at 2 h post-H2O2 exposure reached ~50%, which was 2.5 times higher than the control group without ROS stimulation. This enhanced release is attributed to the cleavage of disulfide bonds in NPs under oxidative stress, facilitating rapid release of the active payload (Supplementary Fig. 4E), thus confirming the ROS-responsive nature of the delivery system. Further treatment of HTR8/SVneo cells with different endocytic inhibitors showed that, compared with the group without inhibitor treatment, Dynasore, CPZ, and EIPA each reduced the cellular uptake of FLN@(PDA + NUAK1) NPs to varying extents. In addition, pre-incubation with excess free FA markedly decreased nanoparticle uptake, suggesting that folate receptor-mediated uptake is involved in the endocytic process of these NPs (Supplementary Fig. 4F and G). Collectively, these results confirm that FA-modified FLN@(PDA + NUAK1) NPs exhibit efficient targeting and internalization in placental trophoblast cells, coupled with ROS-triggered drug release under oxidative stress conditions. The involvement of clathrin- and caveolin-mediated endocytosis pathways in NP uptake further supports their mechanistic basis and therapeutic potential for the treatment of PE. Protective effects of FLN@(PDA + NUAK1) NPs against oxidative stress in placental cells Having confirmed the efficient targeting and ROS-responsiveness of FLN@(PDA + NUAK1) NPs in placental cells, we further evaluated their protective role against oxidative stress-induced cellular damage. A series of in vitro experiments was conducted to investigate NPs’ ability to reduce ROS levels, enhance antioxidant enzyme expression, and inhibit apoptosis in placental trophoblast cells (Fig. 2A). Impact of FLN@(PDA + NUAK1) NPs on oxidative stress in HTR8/SVneo cells. HTR8/SVneo cells were treated with 100 μM H₂O₂ for 24 h, then co-incubated with the synthesized nanoparticles for another 24 h. A Schematic diagram of the experimental procedure for protecting HTR8/SVneo cells from oxidative stress damage with FLN@(PDA + NUAK1) NPs (Created in BioRender. Jiang, P. (2026) https://BioRender.com/1sg6pq4). B ROS levels in HTR8/SVneo cells detected using the DCFH-DA fluorescent probe, bar=25 μm. C Average fluorescence intensity of ROS in HTR8/SVneo cells. D Cell viability of HTR8/SVneo cells measured by CCK-8 assay. E Western Blot analysis of SOD1, GPx4, and CAT protein expression in HTR8/SVneo cells. F Relative quantification of SOD1, GPx4, and CAT protein expression. G RT-qPCR analysis of mRNA expression of BCL2, Bak, and Bax, and Cytochrome c in HTR8/SVneo cells. H Schematic diagram of the mechanism by which FLN@(PDA + NUAK1) NPs protect HTR8/SVneo cells (Created in BioRender. Jiang, P. (2026) https://BioRender.com/bty96px). I TUNEL staining to detect apoptosis in HTR8/SVneo cells, bar = 50 μm. J Fluorescence intensity statistics of TUNEL staining. Data are presented as mean ± SD (n = 6 independent biological replicates). Data were analyzed using one-way ANOVA followed by Tukey’s multiple comparisons test (two-sided); For (C) ROS intensity: Significant differences were found between groups (Residual degrees of freedom = 25). Adjusted p < 0.0001 for Control vs. H₂O₂, H₂O₂ vs. FLN@PDA, H₂O₂ vs. FLN@NUAK1, and H₂O₂ vs. FLN@(PDA + NUAK1). The difference between Control and FLN@(PDA + NUAK1) was not significant (p = 0.997). For (D) Cell viability: The analysis showed significant differences between groups (residual degrees of freedom = 25). Adjusted p < 0.0001 for comparisons of H₂O₂ vs. Control, H₂O₂ vs. FLN@PDA, and H₂O₂ vs. FLN@NUAK1. The difference between H₂O₂ + FLN@PDA and H₂O₂ + FLN@NUAK1 was significant (adjusted p = 0.0026), as was the difference between H₂O₂ + FLN@NUAK1 and H₂O₂ + FLN@(PDA + NUAK1) (adjusted p = 0.015). For (F) Protein expression: Data were analyzed using two-way ANOVA followed by Tukey’s multiple comparisons test (residual degrees of freedom = 75). For SOD1, GPx4, and CAT, comparisons between H₂O₂ and all treatment groups (FLN@PDA, FLN@NUAK1, FLN@(PDA + NUAK1)) showed significant differences with adjusted p < 0.0001. For (G) mRNA expression: Data were analyzed using two-way ANOVA followed by Tukey’s multiple comparisons test (residual degrees of freedom = 100). Comparisons between H₂O₂ and all treatment groups were significant for all genes (adjusted p < 0.0001). For (J) apoptosis rate: Significant differences were found (F(4, 25) = 352.6, P < 0.0001). Adjusted p < 0.0001 for H₂O₂ vs. all other groups. Significant differences were also found for H₂O₂ + FLN@PDA vs. H₂O₂ + FLN@NUAK1 (p = 0.0012) and H₂O₂ + FLN@NUAK1 vs. H₂O₂ + FLN@(PDA + NUAK1) (p = 0.0065). * indicates p < 0.05, ** p < 0.01, *** p < 0.001 vs. control; # indicates p < 0.05, ## p < 0.01, ### p < 0.001 vs. H₂O₂ group; & indicates p < 0.05, & & p < 0.01 vs. H₂O₂ + FLN@PDA group; % indicates p < 0.05 vs. H₂O₂ + FLN@NUAK1 group. Samples derive from the same experiment, and the gels/blots were processed in parallel. High-resolution versions of the figures are available at 10.6084/m9.figshare.30919301. NUAK1 NUAK Family Kinase 1, PDA polydopamine, ROS reactive oxygen species, DCFH-DA 2′,7′-dichlorodihydrofluorescein diacetate, SOD1 superoxide dismutase 1, GPx4 glutathione peroxidase 4, CAT catalase, TUNEL terminal deoxynucleotidyl transferase dUTP nick end labeling. HTR8/SVneo cells were treated with 100 μM H₂O₂ for 24 h to mimic the oxidative stress conditions present in preeclamptic placental tissue. Subsequently, cells were co-incubated with FLN@PDA, FLN@NUAK1, or FLN@(PDA + NUAK1) NPs for another 24 h. As expected, ROS levels were low in the untreated control group but markedly increased following H₂O₂ exposure. Notably, treatment with any of the NP formulations significantly reduced intracellular ROS, with the FLN@(PDA + NUAK1) group showing the greatest reduction, indicating a significant effect in mitigating oxidative stress (Fig. 2B and C). CCK-8 assay demonstrated that H₂O₂ exposure significantly decreased trophoblast viability, while NP treatment markedly improved it. Among the treatment groups, FLN@(PDA + NUAK1) demonstrated the highest level of protection (Fig. 2D), further supporting its efficacy in mitigating oxidative damage. Western blot analysis revealed that antioxidant enzymes, SOD1, GPx4, and CAT, were significantly upregulated in cells treated with FLN@(PDA + NUAK1) compared to both the H₂O₂ group and the individual FLN@PDA and FLN@NUAK1 groups (Fig. 2E, F). These results confirm that the combined NP formulation enhances antioxidant defenses more effectively than the single-component systems. In addition, RT-qPCR demonstrated that FLN@(PDA + NUAK1) treatment significantly downregulated Bak, Bax, and Cytochrome c expression while upregulating Bcl2, with greater modulation than observed in the FLN@PDA or FLN@NUAK1 groups (Fig. 2G). Apoptosis was further evaluated using TUNEL staining. H₂O₂ exposure significantly increased the proportion of apoptotic cells, whereas all NP treatments reduced apoptosis, with the FLN@(PDA + NUAK1) group showing the most substantial reduction (Fig. 2I, J). These findings demonstrate that FLN@(PDA + NUAK1) NPs confer robust protection to placental trophoblast cells by reducing oxidative stress, enhancing the expression of antioxidant enzymes, and preventing apoptosis (Fig. 2H). Mitigating oxidative stress-induced mitochondrial dysfunction with FLN@(PDA + NUAK1) NPs While FLN@(PDA + NUAK1) NPs have demonstrated potent effects in reducing ROS levels and inhibiting apoptosis, the health and survival of placental cells also critically depend on mitochondrial integrity and function. As the primary source of cellular energy and a central regulator of apoptosis and oxidative homeostasis, mitochondrial protection is essential for maintaining trophoblast viability. Therefore, we further investigated the effects of FLN@(PDA + NUAK1) NPs on mitochondrial function under oxidative stress to elucidate their protective mechanisms in placental cells. HTR8/SVneo cells were pre-treated with 100 μM H₂O₂ for 24 h to simulate the oxidative environment of PE, followed by co-incubation with FLN@PDA, FLN@NUAK1, or FLN@(PDA + NUAK1) NPs for an additional 24 h. Key indicators of mitochondrial health, including MMP, ATP production, mitochondrial ROS levels, and mitochondrial abundance and morphology, were systematically assessed (Fig. 3A). Impact of FLN@(PDA + NUAK1) NPs on mitochondrial function in HTR8/SVneo cells. HTR8/SVneo cells were treated with 100 μM H₂O₂ for 24 h, followed by 24-h co-incubation with each synthesized nanoparticle formulation. A Schematic diagram of the experimental procedure for regulating mitochondrial function in HTR8/SVneo cells under oxidative stress with FLN@(PDA + NUAK1) NPs (Created in BioRender. Jiang, P. (2026) https://BioRender.com/7j4on3y). B MMP in HTR8/SVneo cells assessed using JC-1 staining, bar = 25 μm. C Statistical graph of the red/green fluorescence ratio of mitochondrial membrane potential in HTR8/SVneo cells. D Measurement of ATP levels in HTR8/SVneo cells. E, F Western Blot analysis and relative quantification of PGC1α and TFAM protein expression in HTR8/SVneo cells. G Detection of mitochondrial ROS generation in HTR8/SVneo cells using MitoSOX red fluorescent probe, bar = 25 μm. H Statistical graph of mitochondrial ROS levels in HTR8/SVneo cells. I Statistical graph of mitochondrial count in HTR8/SVneo cells marked with MitoTracker green fluorescent probe. J TEM observation of ultrastructural mitochondrial changes in HTR8/SVneo cells; red arrows indicate mitochondrial structures, bar = 500 nm. K Schematic diagram of the mechanism by which FLN@(PDA + NUAK1) NPs protect mitochondrial function in HTR8/SVneo cells (created in BioRender. Jiang, P. (2026) https://BioRender.com/vt11k2n). Data are presented as mean ± SD (n = 6 independent biological replicates). Data were analyzed using one-way ANOVA (for C, D, H, I) or two-way ANOVA (for F) followed by Tukey’s multiple comparisons test (two-sided). C MMP ratio: Significant differences were found (F(4, 25) = 59.46, p < 0.0001). Comparisons of Control vs. H₂O₂ and H₂O₂ vs. all treatment groups yielded adjusted P < 0.0001. The difference between Control and FLN@(PDA + NUAK1) was significant (p = 0.0002). D ATP content: Significant differences were found (F(4, 25) = 23.31, p < 0.0001). Adjusted p < 0.0001 for Control vs. H₂O₂ and H₂O₂ vs. all treatment groups. No significant difference was observed between Control and FLN@(PDA + NUAK1) (P = 0.78). The difference between FLN@PDA and FLN@NUAK1 was significant (p = 0.0064). F Protein expression: Significant differences were found for PGC1α and TFAM (df = 50). Comparisons of Control vs. H₂O₂ and H₂O₂ vs. all treatment groups were highly significant (adjusted p < 0.0001). Significant differences were also observed between H₂O₂ + FLN@NUAK1 and H₂O₂ + FLN@(PDA + NUAK1) for PGC1α (p = 0.0007) and TFAM (p < 0.0001). H MitoSOX levels: Significant differences were found (F(4, 25) = 33.81, p < 0.0001). Adjusted p < 0.0001 for Control vs. H₂O₂ and H₂O₂ vs. all treatment groups. Control vs. FLN@(PDA + NUAK1) was p = 0.039. H₂O₂ + FLN@PDA vs. H₂O₂ + FLN@NUAK1 was p = 0.0055. I Mitochondria number: Significant differences were found (F(4, 25) = 56.03, p < 0.0001). Adjusted p < 0.0001 for Control vs. H₂O₂, H₂O₂ vs. all treatment groups, and comparisons between all treatment groups (e.g., PDA vs. NUAK1, NUAK1 vs. Dual). Control vs. FLN@(PDA + NUAK1) was p = 0.0001. *** Indicates p < 0.001 compared to the control group, # indicates p < 0.05 compared to the H2O2-treated group, ##p < 0.01, ###p < 0.001. High-resolution versions of the figures are available at 10.6084/m9.figshare.30919301. NUAK1 NUAK Family Kinase 1, PDA polydopamine, ROS reactive oxygen species, MMP mitochondrial membrane potential. MMP was evaluated using JC-1 staining. As expected, H₂O₂ treatment markedly decreased MMP, indicated by a shift from red to green fluorescence. In contrast, FLN@(PDA + NUAK1) treatment significantly preserved MMP, as evidenced by increased red and decreased green fluorescence (Fig. 3B, C). Consistently, intracellular ATP levels measured using an ATP assay kit were significantly elevated in cells treated with FLN@(PDA + NUAK1) (Fig. 3D), further confirming enhanced mitochondrial energy metabolism. Western blot analysis was conducted to examine the expression of mitochondrial biogenesis markers PGC1α and TFAM. Both proteins were significantly downregulated by H₂O₂ exposure, whereas FLN@(PDA + NUAK1) treatment restored their expression levels to near-control values (Fig. 3E, F). Additionally, mitochondrial-specific ROS production was assessed using the MitoSOX Red probe. H₂O₂ treatment dramatically elevated mitochondrial ROS, while treatment with FLN@NUAK1 and especially FLN@(PDA + NUAK1) significantly reduced mitochondrial ROS accumulation (Fig. 3G, H). MitoTracker Green staining further revealed that H₂O₂ exposure reduced mitochondrial number and caused morphological abnormalities, whereas FLN@(PDA + NUAK1) treatment restored mitochondrial quantity and preserved normal morphology (Fig. 3G, I). Furthermore, TEM analysis provided ultrastructural evidence of mitochondrial integrity. H₂O₂-treated cells displayed mitochondrial swelling, cristae disruption, and membrane damage, while FLN@(PDA + NUAK1) treatment preserved mitochondrial structure and cristae organization (Fig. 3J). Together, these results demonstrate that FLN@(PDA + NUAK1) NPs effectively mitigate mitochondrial dysfunction in HTR8/SVneo cells exposed to oxidative stress. NPs preserve MMP, reduce mitochondrial ROS production, enhance mitochondrial biogenesis, and maintain mitochondrial morphology and ultrastructure (Fig. 3K). Regulation of autophagy levels and its protective role in placental cells Mitophagy, a selective form of autophagy responsible for mitochondrial quality control, plays a pivotal role in maintaining cellular homeostasis by eliminating damaged mitochondria under oxidative stress or providing nutrients and energy during metabolic stress49. To further elucidate the protective mechanisms of FLN@(PDA + NUAK1) NPs, we investigated their regulatory effects on autophagy in HTR8/SVneo placental trophoblast cells. Western blot analysis was first performed to evaluate the expression of autophagy-related proteins LC3, Beclin1, ATG5, and ATG7. Exposure to H₂O₂ markedly increased the LC3II/LC3I ratio while significantly decreasing the expression of Beclin1, ATG5, and ATG7, indicating impaired autophagic flux. In contrast, treatment with FLN@(PDA + NUAK1) NPs significantly reduced the LC3II/LC3I ratio and restored the expression of Beclin1, ATG5, and ATG7, suggesting a reactivation of autophagic processes. Notably, the addition of the autophagy inhibitor 3-MA reversed these effects, leading to reduced expression of autophagy markers and a decreased LC3II/LC3I ratio, confirming suppression of autophagy (Fig. 4A, B). Regulation of autophagy in placental trophoblast cells by FLN@(PDA + NUAK1) NPs. A, B Western Blot analysis of autophagy-related protein expression in HTR8/SVneo cells. Immunofluorescence detection of Beclin 1 levels (C) and corresponding quantitative graph (D), bar = 25 μm. Immunofluorescence detection of p62 levels (E) and corresponding quantitative graph (F), bar = 25 μm. G, H Confocal microscopy images showing the number of autophagosomes and autolysosomes in mRFP-GFP-LC3 adenovirus-transfected cells, bar = 15 μm, with quantitative analysis of autophagosomes and autolysosomes (H). I, J TEM analysis of autophagosomes in HTR8/SVneo cells, with red arrows indicating autophagic vesicles, bar = 1 μm, and quantitative graph of autophagosome numbers (J). K, L Lysosomal activity measured using Magic Red cathepsin B staining and confocal microscopy, with red fluorescence intensity positively correlated with lysosomal activity, bar = 15 μm. M Schematic diagram showing the mechanism by which FLN@(PDA + NUAK1) NPs restore autophagic flux and protect placental cells in PE (Created in BioRender. Jiang, P. (2026) https://BioRender.com/1ab3dqg). Data are presented as mean ± SD (n = 6 independent biological replicates). Data were analyzed using one-way ANOVA (for D, F, J, L) or two-way ANOVA (for B) followed by Tukey’s multiple comparisons test (two-sided). B Protein expression: Significant differences were observed for all autophagy markers (df = 120). Comparisons of Control vs. H₂O₂ (G1) and H₂O₂ vs. FLN@(PDA + NUAK1) (G4) were highly significant (adjusted p < 0.0001). The increase in autophagy markers induced by Dual NPs (G4) was significantly inhibited by 3-MA (G5) (adjusted p < 0.0001 for all proteins). D Beclin 1 positive cells: Significant differences were found (F(4, 30) = 37.16, p < 0.0001). Comparisons of G0 vs. G1 and G1 vs. G4 yielded adjusted p < 0.0001. The difference between G4 and G5 was significant (adjusted p < 0.0001). F p62 positive cells: Significant differences were found (F(4, 30) = 22.89, p < 0.0001). G1 significantly increased p62 levels vs. G0 (adjusted p < 0.0001), while G4 reduced them (adjusted p < 0.0001). This reduction was blocked in G5 (adjusted p < 0.0001 vs. G4). J Autophagosomes (TEM): Significant differences were found (F(4, 30) = 41.85, p < 0.0001). G4 treatment significantly increased autophagosome numbers compared to G1 (adjusted p < 0.0001). This effect was reversed in G5 (adjusted p = 0.0003 vs. G4). L Cathepsin B activity: Significant differences were found (F(4, 30) = 55.27, p < 0.0001). Lysosomal activity was restored in G4 (adjusted p < 0.0001 vs. G1) and significantly inhibited in G5 (adjusted p < 0.0001 vs. G4). No significant difference was found between G0 and G4 (p = 0.13). *** Indicates p < 0.001 compared to the control group, # indicates p < 0.05 compared to the H₂O₂-treated group, ## p < 0.01, ### p < 0.001, & & & indicates p < 0.001 compared to the FLN@(PDA + NUAK1) NPs treatment group. High-resolution versions of the figures are available at 10.6084/m9.figshare.30919301. NUAK1 NUAK Family Kinase 1, PDA polydopamine. Immunofluorescence staining further supported these findings. H₂O₂ exposure led to diminished Beclin-1 expression and accumulation of p62, consistent with blocked autophagic flux. Conversely, FLN@(PDA + NUAK1) treatment upregulated Beclin1 and reduced p62 levels, indicating effective restoration of autophagic activity. Co-treatment with 3-MA again suppressed Beclin1 expression and increased p62 accumulation, confirming inhibition of the autophagy pathway (Fig. 4C–F). To visualize autophagosome dynamics, we employed a GFP-mRFP-LC3 dual-fluorescence assay. H₂O₂-treated cells exhibited a notable elevation in green puncta (autophagosomes) and a reduction in red puncta (autolysosomes), indicating a block in autophagosome–lysosome fusion and impaired autophagic degradation. FLN@(PDA + NUAK1) treatment decreased green puncta and increased red puncta, reflecting enhanced autophagosome maturation and restored autophagic flux. In contrast, 3-MA treatment inhibited early autophagosome formation, reducing puncta overall and disrupting autophagosome turnover (Fig. 4G, H). Ultrastructural analysis via TEM revealed notable differences across treatment groups. In the H₂O₂ group, numerous double-membrane autophagosomes containing partially degraded organelle fragments were observed, indicating defective autophagic clearance. FLN@(PDA + NUAK1) treatment reduced autophagosome accumulation and preserved organelle integrity, suggesting efficient clearance of damaged mitochondria and restoration of autophagic flux. In the 3-MA combination group, autophagosome numbers were again reduced, yet an abundance of morphologically abnormal mitochondria remained undegraded, confirming autophagic blockade (Fig. 4I, J). Lysosomal function was evaluated using LysoTracker and Magic Red cathepsin B co-staining. H₂O₂ treatment significantly diminished lysosomal activity, while FLN@(PDA + NUAK1) restored lysosomal function. Co-treatment with 3-MA again reduced lysosomal activity, further supporting the inhibition of autophagic flux (Fig. 4K, L). Collectively, these results demonstrate that FLN@(PDA + NUAK1) NPs alleviate H₂O₂-induced autophagic dysfunction by restoring autophagic flux and lysosomal activity, thereby protecting placental trophoblast cells from oxidative damage (Fig. 4M). FLN@(PDA + NUAK1) NPs inhibit NLRP3 inflammasome-mediated pyroptosis and inflammatory responses through autophagy activation Under oxidative stress, activation of the NLRP3 inflammasome promotes cellular pyroptosis and drives the maturation and release of pro-inflammatory cytokines50. Impaired mitophagy leads to the accumulation of damaged mitochondria, which further amplify NLRP3 activation and exacerbate cell injury and inflammatory responses51,52. In this study, we investigated whether FLN@(PDA + NUAK1) NPs, by restoring autophagic flux under oxidative stress, could suppress NLRP3 inflammasome activation and consequently inhibit pyroptosis and inflammatory cytokine release (Fig. 5A). Regulation of pyroptosis and inflammatory responses in oxidatively stressed HTR8/SVneo cells by FLN@(PDA + NUAK1) NPs. A Schematic diagram of the experimental procedure for investigating the regulatory effects of FLN@(PDA + NUAK1) NPs on pyroptosis and inflammatory responses in HTR8/SVneo cells under oxidative stress (created in BioRender. Jiang, P. (2026) https://BioRender.com/rgbqmqo). B Western Blot analysis of NLRP3 inflammasome-related proteins (NLRP3, ASC, caspase1), pyroptosis-related proteins (caspase4, caspase5, GSDMD), and GSDMD cleavage product GSDMD-N. Relative quantification of NLRP3, ASC, caspase1 (C), and caspase4, caspase5, GSDMD-N (D) protein expression. E, F Immunofluorescence analysis of GSDMD localization and expression in HTR8/SVneo cells, bar = 25 μm, with a quantitative graph of GSDMD-N fluorescence intensity (F). G, H Brightfield microscopy showing morphological changes in trophoblast cells, with a quantitative graph of pyroptotic cell percentage (H), bar = 50/15 μm. I ELISA analysis of the secretion levels of inflammatory cytokines IL18, TNFα, IL6, and IL1β in the cell culture supernatant. J Schematic diagram illustrating the mechanism by which nanoparticles inhibit pyroptosis and inflammatory cytokine secretion by activating autophagic flux in HTR8/SVneo cells (created in BioRender. Jiang, P. (2026) https://BioRender.com/76stypg). Data are presented as mean ± SD (n = 6 independent biological replicates). Data were analyzed using one-way ANOVA (for F, H, I) or two-way ANOVA (for C, D) followed by Tukey’s multiple comparisons test. C Inflammasome proteins: Significant differences were observed for all markers (df = 100). H₂O₂ (G1) treatment significantly upregulated NLRP3, Caspase1 p20, and ASC, and downregulated Pro-caspase1 compared to Control (G0) (adjusted p < 0.0001). Dual NPs (G4) significantly reversed these changes compared to G1 (adjusted p < 0.0001 for all proteins). D Pyroptosis proteins: Significant differences were observed for Caspase4, Caspase5, and GSDMD-NT/GSDMD (df = 75). H₂O₂ (G1) treatment significantly increased the expression of these proteins compared to Control (adjusted p < 0.0001). Dual NPs (G4) significantly reduced their expression compared to G1 (adjusted p < 0.0001). Notably, GSDMD-NT/GSDMD levels in the G4 group returned to baseline levels (p = 0.24 vs. Control). F GSDMD intensity: Significant differences were found (F(4, 25) = 42.40, p < 0.0001). G1 increased GSDMD levels (adjusted p < 0.0001 vs. G0), while G4 significantly reduced them (adjusted p < 0.0001 vs. G1). The difference between G0 and G4 was significant (p = 0.045). H Pyroptotic cells: Significant differences were found (F(4, 25) = 34.81, p < 0.0001). The percentage of pyroptotic cells was significantly increased in G1 (adjusted p < 0.0001) and restored to normal levels in G4 (adjusted p < 0.0001 vs. G1; p = 0.87 vs. G0). I Cytokines (ELISA): Significant differences were observed (df = 25). Levels of all cytokines were elevated in G1 (adjusted p < 0.0001) and significantly reduced in G4 (adjusted p < 0.0001). Comparisons between G0 and G4 showed specific P values: p = 0.027 (IL-18), p = 0.015 (TNFα), p = 0.0004 (IL-6), and p = 0.062 (IL-1β, not significant). ***Indicates p < 0.001 compared to the control group, # indicates p < 0.05 compared to the H2O2-treated group, ##p < 0.01, ###p < 0.001. Samples derive from the same experiment and the gels/blots were processed in parallel. High-resolution versions of the figures are available at 10.6084/m9.figshare.30919301. NUAK1 NUAK family kinase 1, PDA polydopamine, ELISA enzyme-linked immunosorbent assay, IL interleukin. To evaluate inflammasome activation, Western blot analysis was conducted to examine the expression of NLRP3, ASC, and caspase1 in HTR8/SVneo cells. H₂O₂ treatment significantly increased the levels of these proteins, confirming NLRP3 inflammasome activation. However, treatment with FLN@(PDA + NUAK1) NPs markedly reduced the expression of NLRP3, ASC, and caspase1, suggesting that activation of autophagy effectively restrained inflammasome hyperactivation (Fig. 5B, C). Caspase1 activation facilitates cleavage of gasdermin D (GSDMD), generating its pore-forming N-terminal fragment (GSDMD-N), which integrates into the plasma membrane, disrupts membrane integrity, and triggers pyroptotic cell death53,54. We next assessed the expression of pyroptosis-related proteins—caspase4/5, full-length GSDMD, and cleaved GSDMD-N. Western blot results showed that H₂O₂ treatment significantly upregulated caspase4/5 and GSDMD-N levels, indicating active pyroptosis. In contrast, FLN@(PDA + NUAK1) treatment significantly suppressed the expression of these pyroptosis markers (Fig. 5B, D). Immunofluorescence staining revealed that the expression and cellular localization of GSDMD were substantially reduced in cells treated with FLN@(PDA + NUAK1), further supporting the inhibition of pyroptosis (Fig. 5E, F). Morphological observations under brightfield microscopy showed that H₂O₂-treated cells exhibited hallmark features of pyroptosis, including swelling, balloon-like membrane structures, and abrupt cell rupture. These changes were reversed by NP treatment, resulting in reduced pyroptotic cell proportions (Fig. 5G, H). Given the association between pyroptosis and inflammation, we quantified the secretion of pro-inflammatory cytokines (IL18, TNFα, IL6, and IL1β) in the cell culture supernatant via ELISA. H₂O₂ exposure significantly elevated all four cytokines, while treatment with FLN@(PDA + NUAK1) NPs significantly suppressed their levels, indicating effective inhibition of inflammasome-mediated cytokine maturation and release (Fig. 5I). In summary, FLN@(PDA + NUAK1) NPs restore autophagic flux under oxidative stress conditions, thereby restraining NLRP3 inflammasome activation and preventing pyroptosis and excessive inflammatory responses (Fig. 5J). This mechanism underpins the substantial protective effects of FLN@(PDA + NUAK1) in placental trophoblast cells and provides a strong experimental foundation for the therapeutic potential of these NPs in oxidative stress-related pathologies such as PE. In vitro studies demonstrate NPs promote placental trophoblast cell invasion and migration, restoring placental function Oxidative stress is known to impair mitochondrial function and damage placental trophoblast cells (HTR8/SVneo), contributing to reduced energy metabolism and exacerbating pyroptosis and inflammation via NLRP3 inflammasome activation55,56. These pathological events also negatively affect the invasive and migratory capacities of trophoblast cells—functions critical for normal placental development. To elucidate the role of FLN@(PDA + NUAK1) NPs in mitigating such dysfunction, we investigated their effects on trophoblast cell (HTR8/SVneo cells) migration and invasion, as well as associated molecular pathways, under oxidative stress conditions (Fig. 6A). Effect of nanoparticles on migration, invasion, and placental function markers in HTR8/SVneo cells. A Schematic diagram of the experimental procedure for assessing the effect of nanoparticles on cell behavior in HTR8/SVneo cells under oxidative stress (created in BioRender. Jiang, P. (2026) https://BioRender.com/n3eqom4). B, C Wound assay to evaluate the migration ability of HTR8/SVneo cells, bar = 100 μm, with a quantitative analysis of the placental trophoblast cells wound percentage (C). D, E Transwell invasion assay to assess the invasion capability of HTR8/SVneo cells, bar = 50 μm, with quantitative analysis of the number of invading cells (E). F, G Western Blot analysis of proteins related to invasion and migration (MMP2, MMP9, E-cadherin, ICAM1), with quantitative analysis of relative protein expression levels (G). H–K ELISA analysis of the secretion levels of placental function markers (ET1, sFLT1, PLGF, VEGF) in the cell culture supernatant. Data are presented as mean ± SD (n = 6 independent biological replicates). Data were analyzed using one-way ANOVA (for C, E, H–K) or two-way ANOVA (for G) followed by Tukey’s multiple comparisons test. For (C): Significant differences were found between the model group (G1) and all other groups (adjusted p < 0.0001). Comparisons between treatment groups (G2 vs G3, G3 vs. G4) were significant (adjusted p < 0.0001). No significant difference was observed between Control (G0) and the combined treatment group (G4) (p = 0.85); For (E): The model group (G1) showed significant differences compared to all other groups (adjusted p < 0.0001). The combined treatment group (G4) was significantly different from the single treatment groups G2 (p < 0.0001) and G3 (p = 0.0003). There was no significant difference between G2 and G3 (p = 0.97); For (G) Protein expression (MMP2, MMP9, E-cadherin, ICAM1): The model group showed significant differences compared to all other groups for all proteins (adjusted p < 0.0001, df=100). The combined treatment (Dual) showed significantly better efficacy than single treatments (PDA or NUAK1) for MMP2, E-cadherin, and ICAM1 (adjusted p < 0.0001). However, for MMP9, there was no significant difference between the NUAK1 group and the Dual group (adjusted p = 0.91). Significant differences were also observed between the PDA and NUAK1 groups for E-cadherin (p = 0.0074) and MMP2 (p = 0.0001); For (H–K): The model group showed significant differences compared to the control group in all panels (adjusted p < 0.0001, df = 25). Comparisons between the two single-treatment groups (PDA vs. NUAK1) showed no significant differences in any of these panels (p > 0.09). The combined treatment (Dual) was significantly more effective than single treatments in all cases (adjusted p < 0.0001). Notably, for I–K, the combined treatment restored values to levels statistically indistinguishable from the Control group ((I): p = 0.11; (J): p = 0.14; (K): p = 0.12). However, for (H), although improved, the combined treatment group remained significantly different from the Control group (p < 0.0001). *** Indicates p < 0.001 compared to the control group, # indicates p < 0.05 compared to the H2O2-treated group, ##p < 0.01, ###p < 0.001. Samples derive from the same experiment and that gels/blots were processed in parallel. High-resolution versions of the figures are available at 10.6084/m9.figshare.30919301. NUAK1 NUAK Family Kinase 1, PDA polydopamine, ELISA enzyme-linked immunosorbent assay. Cell migration was first assessed using a wound healing (scratch) assay. Treatment with 100 μM H₂O₂ significantly impaired HTR8/SVneo cell migration. However, following treatment with FLN@(PDA + NUAK1) NPs, HTR8/SVneo cell migration was significantly enhanced compared to the H₂O₂-only group, demonstrating NPs’ ability to rescue oxidative stress-impaired migration (Fig. 6B, C). Next, we assessed the effects of the NPs on the invasive capacity of HTR8/SVneo cells using a Transwell invasion assay. In the H₂O₂-treated group, the number of HTR8/SVneo cells that invaded the lower side of the Transwell membrane was markedly reduced, indicating that oxidative stress significantly impaired trophoblast cell invasion. In contrast, treatment with FLN@(PDA + NUAK1) NPs significantly increased the number of invading cells, further demonstrating the ability of the NPs to restore and enhance trophoblast invasive capacity (Fig. 6D, E). To investigate the underlying molecular mechanisms, we examined the expression of migration- and invasion-associated proteins by Western blot. H₂O₂ exposure significantly downregulated matrix metalloproteinases MMP2 and MMP9, as well as E-cadherin, while upregulating ICAM1, reflecting a suppression of migration/invasion and enhanced HTR8/SVneo cell adhesion. Treatment with FLN@(PDA + NUAK1) restored the expression of MMP2, MMP9, and E-cadherin while decreasing ICAM1 expression, indicating that NPs promote HTR8/SVneo cell motility and invasiveness through modulation of these key molecules (Fig. 6F, G). To assess placental functional restoration, ELISA was performed to quantify the secretion of ET1, sFLT1, PLGF, and VEGF—critical markers of placental function. H₂O₂ treatment significantly elevated ET1 and sFLT1, while reducing PLGF and VEGF, consistent with a dysfunctional trophoblast phenotype. Treatment with FLN@(PDA + NUAK1) reversed these alterations, reducing ET1 and sFLT1 levels and restoring PLGF and VEGF secretion (Fig. 6H–K), indicating that NPs contribute to restoring placental trophoblast (HTR8/SVneo cells) function. In summary, FLN@(PDA + NUAK1) NPs effectively reverse oxidative stress-induced impairment of trophoblast (HTR8/SVneo cells) function by promoting HTR8/SVneo cell migration and invasion, and restoring the expression of key placental markers. These findings provide compelling experimental evidence supporting the therapeutic potential of FLN@(PDA + NUAK1) NPs in placental disorders such as PE. In vivo studies demonstrate FLN@(PDA + NUAK1) NPs effectively improve placental function in a PE mouse model To evaluate the therapeutic efficacy of FLN@(PDA + NUAK1) NPs in vivo, we first assessed their biosafety in healthy mice. After 10 consecutive days of intravenous administration, histological analysis via H&E staining and biochemical assessment of liver and kidney function showed no significant toxicity (Supplementary Fig. 5). Subsequently, the in vivo targeting ability of FLN@(PDA + NUAK1) was evaluated in healthy pregnant mice. On E15.5, following intravenous injection of fluorescently labeled FLN@(PDA + NUAK1), biodistribution was assessed through in vivo imaging and ex vivo imaging of major organs (Supplementary Fig. 6A). Real-time fluorescence imaging at 30 min post-injection revealed predominant accumulation in the liver, indicating early hepatic uptake. However, at 12 h post-injection, the signal in the abdominal region—particularly the placenta—increased markedly, demonstrating delayed but robust placental targeting (Supplementary Fig. 6B). Further ex vivo imaging showed that FLN@(PDA + NUAK1) accumulation in the placenta was notably higher than that in other control groups, while fluorescence signals in the liver, spleen, and other organs were relatively low, confirming the targeted distribution of FLN@(PDA + NUAK1) in vivo (Supplementary Fig. 6C and D). Quantitative analysis revealed that fluorescence intensity in the placenta was approximately four times higher than in the control group and exceeded that in all other major organs, validating NPs’ excellent placental targeting and retention capabilities (Supplementary Fig. 6E). To evaluate therapeutic efficacy, a PE mouse model was established via daily subcutaneous injections of L-NAME starting on E10.5. FLN@(PDA + NUAK1) was administered concurrently, and mice were sacrificed on E17.5 for analysis (Fig. 7A). Effects of FLN@(PDA + NUAK1) NPs on the L-NAME-induced PE mouse model. A Schematic diagram of the animal experiment process and timeline for FLN@(PDA + NUAK1) NPs intervention (created in BioRender. Jiang, P. (2026) https://BioRender.com/3k8muwa). B, C Detection of SBP and urine protein levels. B Shows SBP changes at different time points, and C presents the quantitative analysis of urine protein content. D–F Evaluation of fetal development. D Shows images of fetuses in the mouse uterus. E Provides quantitative analysis of litter size and embryo resorption rate, and F displays representative images of fetuses and placentas, along with quantitative analyses of fetal and placental weights. G H&E staining of placental sections illustrating overall structural organization; the boundary between the labyrinth and junctional zones is marked by black lines. Scale bar = 1 mm. H Quantitative analysis of the proportions of the labyrinth and junctional zones. I H&E staining showing inflammatory cell infiltration in the decidual and labyrinth layers across treatment groups, bar = 100 μm. J ELISA analysis of the expression levels of placental inflammatory factors IL18, TNFα, IL6, and IL1β. K ELISA analysis of placental function markers ET1, sFLT1, PLGF, and VEGF in mouse serum. Data are presented as mean ± SD (n = 10 biologically independent animals per group). For placental and fetal indicators, values were averaged within each litter prior to statistical analysis. Statistical significance was determined using one-way ANOVA followed by Dunnett’s post-hoc test for multiple comparisons (two-sided). For (B) SBP: Significant differences were found (df = 108). At E12.5, the model group (T1) showed significantly higher SBP than Control (T0) (p < 0.0001). While the PDA group (T2) showed no significant difference compared to T1 (p = 0.50), the NUAK1 (T3) and Dual groups (T4, T5) significantly reduced SBP (p < 0.0001). Comparison between T2 and T3 showed NUAK1 was more effective (p = 0.0011). At E16.5, T1 remained significantly higher than T0 (p < 0.0001). All treatment groups significantly reduced SBP compared to T1 (p = 0.021 for T2; p < 0.0001 for T3, T4, T5). NUAK1 (T3) remained significantly more effective than PDA (T2) (p = 0.0012). For (C) Urine protein: Data were analyzed using one-way ANOVA followed by Tukey’s multiple comparisons test (df = 54). The model group (T1) showed significantly higher urine protein levels compared to the Control group (T0) (adjusted p < 0.0001). All treatment groups (T2–T5) significantly reduced urine protein levels compared to the model group (adjusted p < 0.0001). Comparisons between treatments revealed a hierarchy of efficacy: the NUAK1 group (T3) was significantly more effective than the PDA group (T2) (P < 0.0001), and both Dual treatment groups (T4, T5) were significantly more effective than the NUAK1 group (T3) (p < 0.0001). There was no significant difference between the low-dose (T4) and high-dose (T5) Dual treatment groups (p = 0.49). For Pups/Litter, the PDA group (T2) showed no significant improvement compared to the model group (p = 0.25), whereas the NUAK1 (T3) and Dual treatment groups (T4, T5) significantly increased litter size (p < 0.001). For Resorption Rate, all treatment groups (T2–T5) significantly reduced the rate compared to the model group (p < 0.001). Comparisons between treatments showed that the High-dose Dual group (T5) was significantly more effective than the NUAK1 group (T3) for both indicators (Litter size: p = 0.02; Resorption: p = 0.004). There were no significant differences between the Low (T4) and High (T5) dose Dual groups for either indicator (p > 0.5). For (F) Fetal and placental weights: Data were analyzed using one-way ANOVA followed by Tukey’s multiple comparisons test (df = 54). The model group (T1) showed significantly lower fetal and placental weights compared to the Control group (T0) (adjusted p < 0.0001). All treatment groups (T2–T5) significantly increased weights compared to the model group (adjusted p < 0.005). Comparisons between treatment groups revealed a significant dose-dependent effect: the High-dose Dual group (T5) resulted in significantly higher weights than the Low-dose Dual group (T4) for both the fetus (P < 0.0001) and the placenta (p = 0.002). Notably, the High-dose Dual treatment (T5) restored both fetal and placental weights to levels statistically indistinguishable from the Control group (Fetus: p = 0.053; Placenta: p = 0.94). For (H) Labyrinth/Junction zone ratio: Data were analyzed using one-way ANOVA followed by Tukey’s multiple comparisons test (df = 54). The model group (T1) showed a significantly reduced Labyrinth/Junction zone ratio compared to the Control group (T0) (adjusted p < 0.0001). All treatment groups (T2–T5) significantly increased the ratio compared to the model group (adjusted p < 0.0001). Comparisons between treatments showed that the NUAK1 group (T3) was significantly more effective than the PDA group (T2) (p < 0.0001), and both Dual treatment groups (T4, T5) were significantly more effective than the NUAK1 group (T3) (T4 vs. T3: p = 0.02; T5 vs. T3: P = 0.0002). Notably, both the Low-dose (T4) and High-dose (T5) Dual treatment groups restored the ratio to levels statistically indistinguishable from the Control group (T4 vs. T0: p = 0.88; T5 vs. T0: p > 0.99). For (J) Placental inflammatory factors (IL18, TNFα, IL6, IL1β): Data were analyzed using one-way ANOVA followed by Tukey’s multiple comparisons test (df = 54). The model group (T1) showed significantly elevated levels of all inflammatory factors compared to the Control group (T0) (adjusted p < 0.0001). All treatment groups (T2–T5) significantly reduced these levels compared to the model group (adjusted p < 0.0001). Comparisons between single treatments (PDA vs. NUAK1) showed no significant differences for IL18, IL6, and IL1β (p > 0.09), whereas for TNFα, the NUAK1 group was significantly more effective than the PDA group (p < 0.0001). The High-dose Dual treatment (T5) demonstrated the strongest anti-inflammatory effect. Notably, for TNFα and IL1β, the High-dose Dual treatment restored levels to values statistically indistinguishable from the Control group (TNFα: P = 0.68; IL1β: p = 0.52). For IL18 and IL6, although significantly improved, the levels remained higher than those of the Control group (p < 0.02). For (K) Serum placental function markers (ET1, sFLT1, PLGF, VEGF): Data were analyzed using one-way ANOVA followed by Tukey’s multiple comparisons test (df=54). The model group (T1) showed significantly impaired levels of all markers compared to the Control group (T0) (elevated ET1 and sFLT1; reduced PLGF and VEGF; adjusted p < 0.0001). All treatment groups (T2–T5) significantly improved these markers compared to the model group (adjusted p < 0.0001). Comparisons between treatment groups showed a clear dose-dependent effect: the High-dose Dual group (T5) was significantly more effective than the Low-dose Dual group (T4) for all markers (p < 0.0001 for sFLT1, PLGF, and VEGF; p = 0.03 for ET1). Notably, the High-dose Dual treatment (T5) restored levels of ET1 and VEGF to values statistically indistinguishable from the Control group (ET1: p = 0.42; VEGF: p = 0.42). For sFLT1 and PLGF, although significantly improved, the levels in the T5 group remained significantly different from the Control group (p < 0.001). *** Indicates p < 0.001 compared to the T0 group, # indicates p < 0.05 compared to the T1 group, ##p < 0.01, ###p < 0.001. High-resolution versions of the figures are available at 10.6084/m9.figshare.30919301. NUAK1 NUAK Family Kinase 1, PDA polydopamine, PDI polydispersity index, PBS phosphate-buffered saline, IL Interleukin. We further examined the effects of different doses of NPs on blood pressure and urinary protein levels in L-NAME-induced PE mice. In the L-NAME-treated group, SBP was significantly elevated, accompanied by a marked increase in urinary protein levels, indicating the development of PE-like symptoms. In contrast, mice treated with FLN@(PDA + NUAK1) NPs exhibited a significant reduction in SBP and a notable decrease in urinary protein levels, suggesting that the NPs effectively alleviated hypertension and renal dysfunction associated with PE. Moreover, the high-dose group demonstrated a more pronounced therapeutic effect than the low-dose group (Fig. 7B, C). Next, we evaluated the effects of the NPs on fetal development and placental function. In the L-NAME-treated group, fetal development was impaired, as evidenced by a reduced number of fetuses per litter and significantly lower fetal weight. In contrast, treatment with FLN@(PDA + NUAK1) NPs resulted in a significant increase in litter size and restoration of fetal weight, indicating that the NPs support normal fetal development. Furthermore, placental weight was markedly improved following NP treatment, providing additional evidence for their restorative effects on placental function. Notably, the high-dose group demonstrated a more pronounced improvement than the low-dose group (Fig. 7D–F). Histological examination of placental tissues revealed that the proportions of the labyrinth and junctional zones were significantly reduced in L-NAME-treated preeclamptic placentas, whereas treatment with FLN@(PDA + NUAK1) NPs markedly restored these proportions, with the high-dose group showing a more pronounced effect than the low-dose group (Fig. 7G, H). In the PE placentas, the abundance of trophoblast-specific protein α-positive (TPBPA⁺) invasive trophoblast giant cells (TGCs) in the junctional zone, monocarboxylate transporter 1-positive (MCT1⁺) syncytiotrophoblast layer I (SynTI) cells, MCT4⁺ syncytiotrophoblast layer II (SynTII) cells in the labyrinth zone, and laminin-positive fetal-derived vasculature was consistently reduced. Treatment with FLN@(PDA + NUAK1) NPs significantly restored the levels of these trophoblast markers, again with greater improvement in the high-dose group (Supplementary Fig. 7). H&E staining further demonstrated marked structural disorganization in placentas from L-NAME-treated mice, characterized by abnormal arrangement of trophoblast and decidual cells and pronounced inflammatory cell infiltration. In contrast, placental architecture was substantially improved following FLN@(PDA + NUAK1) NPs treatment, with a notable reduction in inflammatory cells in both the decidual and labyrinth layers and an overall restoration of normal placental structure. These improvements were more prominent in the high-dose group compared with the low-dose group (Fig. 7I). Finally, we assessed the expression levels of placental inflammatory factors and functional markers using ELISA. In L-NAME-treated preeclamptic mice, ET1 and sFlt1 levels were significantly elevated, whereas PLGF and VEGF levels were markedly reduced, reflecting impaired placental function. In contrast, treatment with FLN@(PDA + NUAK1) NPs led to a significant decrease in ET1 and sFlt1 levels and a notable restoration of PLGF and VEGF expression to levels comparable to those in healthy mice. Again, the high-dose group showed more pronounced improvements than the low-dose group (Fig. 7J, K). In conclusion, FLN@(PDA + NUAK1) NPs significantly ameliorated PE symptoms in vivo by lowering blood pressure and proteinuria, improving placental structure and trophoblast function, and normalizing expression of placental biomarkers. These findings provide compelling evidence for the therapeutic potential of FLN@(PDA + NUAK1) in the treatment of PE and support further investigation in clinical models. FLN@(PDA + NUAK1) NPs reduce oxidative stress and improve placental function in PE In this study, we systematically evaluated the therapeutic efficacy and underlying mechanisms of FLN@(PDA + NUAK1) NPs in an L-NAME-induced PE mouse model, focusing on their roles in alleviating oxidative stress, promoting autophagic flux, and inhibiting NLRP3 inflammasome-mediated pyroptosis to preserve placental function (Fig. 8A). Protective effects of FLN@(PDA + NUAK1) NPs in the L-NAME-induced PE mouse model. A Schematic diagram of the experimental procedure for evaluating the protective effects of FLN@(PDA + NUAK1) NPs on oxidative stress and placental function in the PE mouse model (Created in BioRender. Jiang, P. (2026) https://BioRender.com/7lxom13). B, C DCFH-DA staining to detect ROS levels in placental tissue, bar = 25 μm. D, E 4-HNE staining to assess lipid peroxidation levels, bar = 25 μm. F, G 8-OHdG immunohistochemical staining to detect DNA oxidative damage, bar = 100 μm. H, I Western Blot analysis of NLRP3, Pro-caspase 1, GSDMD, GSDMD-NT, LC3I/II, and p62 expression, with relative quantification of protein expression (I). J, K p62 immunofluorescence staining to show autophagic flux blockage, bar = 25 μm. L, M TUNEL staining to detect apoptotic cells in placental tissue, bar = 25 μm. N TEM to observe ultrastructural changes in mitochondria in placental cells, bar = 1 μm. Data are presented as mean ± SD (n = 10 biologically independent animals per group). Data were analyzed using one-way ANOVA (for C, E, G, K, M) or two-way ANOVA (for I) followed by Tukey’s multiple comparisons test. C ROS levels (DCFH-DA): Significant differences were found (F(5, 54) = 56.78, p < 0.0001). The model group (T1) showed significantly increased ROS levels compared to Control (T0) (adjusted p < 0.0001). All treatment groups significantly reduced ROS levels compared to T1 (adjusted p < 0.0001). The High-dose Dual group (T5) was significantly more effective than the Low-dose group (T4) (p = 0.0004), although ROS levels in T5 remained higher than in T0 (p < 0.0001). No significant difference was observed between PDA (T2) and NUAK1 (T3) groups (p = 0.91). E Lipid peroxidation (4-HNE): Significant differences were observed (F(5, 54) = 48.99, p < 0.0001). T1 showed significantly elevated 4-HNE levels vs. T0 (adjusted p < 0.0001). T5 treatment significantly reduced these levels compared to T1 (adjusted p < 0.0001) and was significantly more effective than T4 (p = 0.017). Levels in T5 remained significantly higher than in T0 (p < 0.0001). T2 and T3 showed no significant difference (p = 0.13). G DNA damage (8-OHdG): Significant differences were found (F(5, 54) = 42.47, p < 0.0001). T1 significantly increased DNA damage vs. T0 (adjusted p < 0.0001). T5 treatment reduced damage most effectively compared to T1 (adjusted p < 0.0001), showing a significant improvement over T4 (p = 0.005). However, levels in T5 remained higher than in T0 (p < 0.0001). No significant difference was found between T2 and T3 (p > 0.99). (I) Protein expression: Significant differences were observed for all markers (df = 54). T1 treatment significantly upregulated LC3 II/I, GSDMD-N, and NLRP3, and downregulated Pro-caspase1 compared to T0 (adjusted p < 0.0001). T5 treatment significantly reversed these changes (adjusted p < 0.0001 vs. T1). Notably, for Pro-caspase1 and NLRP3/GSDMD markers, the High-dose Dual treatment (T5) restored expression levels to values statistically indistinguishable from the Control group (p > 0.95). Comparisons between doses showed that T5 was significantly more effective than T4 for upregulated markers (p < 0.0001) but equally effective for restoring Pro-caspase1 (P = 0.79). K p62 fluorescence: T1 significantly increased p62 levels (adjusted p < 0.0001). T5 reduced p62 levels (adjusted p < 0.0001) and was significantly more effective than T4 (p = 0.0017). M Apoptosis (TUNEL): The percentage of apoptotic cells was significantly increased in T1 (adjusted p < 0.0001). T5 treatment significantly reduced apoptosis (adjusted p < 0.0001 vs. T1) and showed greater efficacy than T4 (p = 0.0007). *** Indicates p < 0.001 compared to the T0 group, # indicates p < 0.05 compared to the T1 group, ##p < 0.01, ###p < 0.001. High-resolution versions of the figures are available at 10.6084/m9.figshare.30919301. NUAK1 NUAK Family Kinase 1, PDA polydopamine, PDI polydispersity index, PBS phosphate-buffered saline, PE preeclampsia. First, we evaluated the degree of oxidative stress in placentas from L-NAME-induced PE mice. ROS levels were assessed using DCFH-DA staining, while lipid peroxidation and DNA oxidative damage were examined using 4-HNE and 8-OHdG staining, respectively. The results showed that L-NAME treatment markedly increased ROS levels as well as the expression of 4-HNE and 8-OHdG in placental tissues, indicating substantial enhancement of oxidative stress. In contrast, treatment with FLN@(PDA + NUAK1) NPs significantly reduced these oxidative stress and damage markers, demonstrating the NPs’ ability to effectively mitigate placental oxidative stress in PE. Again, the high-dose group exhibited more pronounced improvements compared with the low-dose group (Fig. 8B–G). Consistent with the exacerbation of oxidative stress, activation of the NLRP3 inflammasome was observed. Western blot analysis revealed that L-NAME treatment significantly upregulated the expression of NLRP3, pro-caspase-1, GSDMD, and its N-terminal cleavage product GSDMD-NT, indicating inflammasome activation and initiation of pyroptosis. However, FLN@(PDA + NUAK1) NPs markedly reduced the expression of these inflammasome-associated proteins, suggesting effective inhibition of NLRP3 inflammasome activation, with the high-dose group showing a stronger effect than the low-dose group (Fig. 8H, I). To investigate the role of autophagic flux in the protective effects of the nanoparticles, we assessed LC3I/II expression by Western blot (Fig. 8H, I) and examined p62 expression in placental tissues via immunofluorescence. In the L-NAME-treated group, the LC3II/I ratio was markedly reduced, accompanied by a significant increase in p62 levels, indicating impaired autophagic flux. Treatment with FLN@(PDA + NUAK1) NPs restored the LC3II/I ratio and markedly decreased p62 expression, suggesting that the nanoparticles alleviate PE-related injury by promoting autophagic flux, with more pronounced effects in the high-dose group (Fig. 8H–K). Inhibition of apoptosis was further confirmed in this study. TUNEL staining revealed a significant increase in TUNEL-positive cells in the L-NAME-treated group, indicating elevated apoptosis in the placenta. Following FLN@(PDA + NUAK1) NPs treatment, the number of apoptotic cells was markedly reduced, supporting the protective effect of the nanoparticles; again, the high-dose group showed a stronger response (Fig. 8L, M). Finally, TEM was used to evaluate placental ultrastructure. In the L-NAME–treated group, placental cells exhibited pronounced mitochondrial swelling, disrupted cristae, and other typical features of oxidative stress–induced injury. In contrast, nanoparticle treatment substantially improved mitochondrial morphology, indicating that the nanoparticles protect mitochondrial integrity and effectively mitigate cellular damage associated with PE, with high-dose treatment showing superior efficacy (Fig. 8N). In conclusion, FLN@(PDA + NUAK1) NPs significantly attenuate oxidative stress, inhibit NLRP3 inflammasome activation, enhance autophagic flux, and reduce apoptosis and pyroptosis in placental tissue. These combined effects restore mitochondrial and placental structural integrity, offering robust experimental support for the therapeutic potential of FLN@(PDA + NUAK1) in the treatment of PE. Discussion PE is a multifactorial pregnancy-specific disorder typically emerging during the second or third trimester, characterized by new-onset hypertension and proteinuria57. Beyond posing an immediate risk to maternal health, PE significantly increases the long-term susceptibility to cardiovascular diseases, including heart failure and stroke, and contributes to chronic hepatic and renal dysfunction. On the fetal side, compromised placental function often results in adverse outcomes such as intrauterine growth restriction, preterm birth, and low birth weight—factors that collectively raise maternal and perinatal morbidity and mortality1,58. Despite decades of research, the pathogenesis of PE remains incompletely understood, with current evidence implicating a complex interplay among genetic predisposition, abnormal immune tolerance, endothelial dysfunction, and impaired placentation59,60. At present, clinical management largely focuses on symptomatic control and timed delivery1, which underscores the urgent need for novel therapeutic approaches that can directly target placental dysfunction. In this study, we developed and evaluated a novel ROS-responsive NP system—FLN@(PDA + NUAK1)—to investigate its therapeutic potential in PE. The findings demonstrate that these NPs exert a protective effect on placental trophoblast cells by mitigating oxidative stress, restoring mitochondrial function, and enhancing cellular viability. In comparison to traditional pharmacological interventions, FLN@(PDA + NUAK1) NPs exhibit targeted delivery, minimal systemic toxicity, and excellent biocompatibility, thereby offering a promising strategy to ameliorate placental dysfunction while minimizing maternal side effects. Importantly, our results shed light on two critical mechanistic pathways—autophagic flux and NLRP3 inflammasome activation—both of which play integral roles in the progression of PE and may serve as viable therapeutic targets. Autophagic flux is crucial in maintaining cellular homeostasis and managing oxidative stress9. Impaired autophagic flux has been implicated not only in malignancies and neurodegenerative diseases but also in obstetric complications such as PE and gestational diabetes61,62. Placental tissues, which rely on precise metabolic control, are particularly sensitive to dysregulated autophagy63,64. In this study, we have delineated the mechanisms regulating autophagic flux in the placenta during PE and explored new methods through nanotechnology to precisely modulate autophagy to enhance placental function. Our findings indicate that FLN@(PDA + NUAK1) NPs effectively promote autophagic flux and significantly reduce oxidative stress. This novel discovery, scarcely reported in prior studies, offers fresh perspectives and potential therapeutic strategies for treating PE. The NLRP3 inflammasome is a well-characterized regulator of inflammatory responses and has been extensively implicated in the pathogenesis of inflammatory disorders such as gout, type 2 diabetes, and atherosclerosis65. Here, we further investigated the role of NLRP3 in the pathological progression of PE and demonstrated its modulation through NUAK1-activated, ROS-responsive NPs. Our findings highlight a novel therapeutic avenue wherein targeted NP-mediated regulation of NLRP3 activity effectively attenuates placental inflammation and tissue damage. This approach not only underscores the central role of NLRP3 in the pathophysiology of PE but also establishes nanotechnology as a powerful platform for modulating inflammatory pathways in pregnancy-related complications. Compared with existing literature, our study provides new mechanistic insights into the role of NLRP3 in placental disease and introduces a promising therapeutic strategy for complex gestational disorders. Of note, NUAK1 is a protein kinase, and previous studies have shown that NUAK1 is closely associated with tumor initiation and progression66 and is involved in cellular oxidative stress response pathways. In the present study, we found that NUAK1 is related to the regulation of mitochondrial function, mitophagic flux, and the NLRP3 inflammasome; however, it has not yet been demonstrated whether these effects are dependent on the kinase activity of NUAK1, nor have its direct substrates been identified. Therefore, these precise molecular mechanisms require further in-depth investigation in future studies. The FLN@(PDA + NUAK1) NPs employed in this study exemplify the transformative potential of nanotechnology in modern medicine, particularly in the realms of precision medicine and targeted therapeutics. These NPs are designed to selectively accumulate in pathological tissues, thereby minimizing off-target effects on healthy cells, reducing systemic toxicity, and enhancing overall treatment efficacy. In comparison to conventional pharmacological approaches, NP-based delivery systems offer superior targeting capabilities, improved bioavailability, and the ability to modulate specific molecular pathways involved in disease progression67,68. In previous studies, the safety and efficacy of VEGF mRNA NPs have been evaluated at therapeutic doses of 0.2, 1.0, and 2.0 mg kg⁻¹. It was observed that six hours after intravenous administration, there was minimal dose-dependent inflammation, liver injury, or placental vascular permeability compared to untreated mice, indicating a favorable safety profile at these doses69. Similar NP doses of 1.0 and 2.0 mg kg⁻¹ have also been used to treat PE mouse models70. In the animal experiments conducted in this study, NPs were administered at doses of 1 or 2 mg/kg/day for multiple consecutive days. These cumulative doses fall within the range reported for similar NP-based drug delivery systems and demonstrated good safety profiles in our study. In terms of experimental methodology, this research utilized a suite of modern biotechniques, including Western Blot, ELISA, and cellular and tissue staining techniques, which ensured the accuracy of the results and provided reliability of the data. These techniques allowed us to analyze the mechanisms of action and biological effects of NPs within cells. Additionally, the use of animal models verified the biosafety and therapeutic potential of NPs, laying the groundwork for clinical applications. Despite the positive outcomes, this study has some limitations. Primarily, the experimental design relied on animal models, and due to interspecies differences, the results may not be wholly extrapolatable to humans, presenting certain limitations. Additionally, compared to mice, humans have a longer gestational period, and the current dosing and administration frequency still need to be optimized for human treatment. Due to resource constraints and ethical considerations, the number of experimental replicates was relatively limited; in future studies, we will increase the number of mice and in vitro cell experiments to further validate our findings. Moreover, the specific molecular mechanisms by which FLN@(PDA + NUAK1) NPs improve PE remain insufficiently explored in this study. Further research is needed to clarify whether the kinase activity of NUAK1 is involved, to identify its downstream substrates, and to determine which signaling pathways regulate oxidative stress, mitochondrial function and autophagic flux, and trophoblast cell migration and invasion. Moreover, although the FLN@(PDA + NUAK1) NPs demonstrated significant therapeutic potential, their long-term biocompatibility and safety still require further validation in broader preclinical studies. In-depth research into NPs’ long-term stability and potential immunogenicity is also necessary to ensure their safety and efficacy in clinical applications. The primary objective of future research is to facilitate the translation of NP-based therapies from preclinical studies to clinical application. This will require well-designed clinical trials with larger cohorts to rigorously evaluate the efficacy and safety of FLN@(PDA + NUAK1) NPs in patients with PE. Furthermore, the potential of these NPs to treat other pregnancy-related complications—such as gestational diabetes and pregnancy-induced hypertension—warrants exploration, which could significantly broaden their therapeutic scope. Functional enhancements, including surface modification, ligand conjugation, or drug loading optimization, may further improve targeting specificity and therapeutic efficacy, representing a critical direction for the next phase of development. These multidimensional advancements will support the realization of personalized medicine in obstetric care, offering safer, more precise, and more effective treatment options for patients. In summary, this study successfully developed ROS-responsive NPs, FLN@(PDA + NUAK1) NPs, and validated their potential in ameliorating pathology associated with PE through in vitro and in vivo experiments. The findings indicate that FLN@(PDA + NUAK1) NPs improve placental function by enhancing mitochondrial functionality, promoting mitochondrial autophagy, and inhibiting NLRP3 inflammasome activation, thereby reducing pyroptosis and inflammatory responses in placental trophoblast cells, and ultimately facilitating their migration and invasion (Fig. 9). Although some limitations remain, the results present new perspectives and clinically relevant pathways for intervention. Continued investigation into the clinical translation of these NPs may ultimately offer a new and effective solution for managing PE, bringing renewed hope to affected patients. Schematic illustration of the mechanism by which FLN@(PDA + NUAK1) nanoparticles improve placental function in preeclampsia by reducing oxidative stress, inflammation, and apoptosis (created in BioRender. Jiang, P. (2026) https://BioRender.com/c6vbk3u). NUAK1 NUAK Family Kinase 1, PDA polydopamine, SOD superoxide dismutase, CAT catalase, MDA malondialdehyde, IL interleukin. Methods Ethical statement All animal experiments were approved by the Animal Ethics Committee of Women’s Hospital, School of Medicine, Zhejiang University (Approval No. ZJU20240513) and were conducted in strict accordance with institutional guidelines and the ARRIVE guidelines. Chemical reagents and materials Tetraethyl orthosilicate (TEOS), ethyltrimethylammonium tosylate (CTAT), triethanolamine (TEAH3), triethanolamine (TEA), fluorescein isothiocyanate (FITC), phosphatidylcholine (PC), cholesterol (CHO), and DSPE-PEG1k were obtained from Sigma-Aldrich Co. (St. Louis, MO, USA). Ammonium nitrate (NH4NO3), anhydrous sodium sulfate, hydrogen peroxide (30% H2O2), anhydrous ethanol, dichloromethane (DCM), and petroleum ether were sourced from Beijing Chemical Reagent Co. (Beijing, China). Bis[3-(triethoxysilyl)propyl]diselenide (BTESePD) was custom-synthesized by WuXi AppTec. FA–modified DSPE-PEG (DSPE-PEG-FA) was provided by Hangzhou Allpeptide Biotechnology Co., Ltd. (Allpeptide Bio, Hangzhou, China; https://www.allpeptide.com/). Analytical-grade reagents were used in their original form, with no subsequent purification. Synthesis of diselenide-bond mesoporous silica NPs (DSeMSN) DSeMSN was synthesized via a sol-gel method. CTAT was used as the template, while the diselenide bridging group served as the silica source in a dual-silica precursor system. Under continuous stirring, 0.6 g of CTAT and 0.15 g of TEAH3 were added to 40 mL of deionized water and heated to 80 °C. After stirring for 30 min, 4.0 g of TEOS and 1.0 g of BTESePD solution were slowly added dropwise. The reaction mixture was maintained at 80 °C and stirred at 90×g for 4 h. The resulting product was collected by centrifugation at 12,900×g for 30 min and washed three times with ethanol. To remove the CTAT template, the product was refluxed in 1% NH4NO3 in ethanol for 12 h. Finally, NPs were rinsed three more times with ethanol and dried to obtain DSeMSN. Synthesis of FITC-tagged DSeMSN To facilitate subsequent visualization studies of NP uptake and distribution, DSeMSN was fluorescently labeled with fluorescein isothiocyanate (FITC). In a dark environment, 0.5 mg of FITC was dissolved in 1 mL of N,N-dimethylformamide (DMF), followed by the addition of 3-aminopropyltriethoxysilane (APTES) at a molar ratio of 12% relative to TEOS. The mixture was stirred at ambient temperature for 2 h to generate the FITC-APTES precursor solution. Separately, 0.6 g of CTAT and 0.15 g of TEAH₃ were added to 40 mL of deionized water and stirred at 80 °C for 30 min. Then, 4.0 g of TEOS, 1.0 g of BTESePD solution, and the FITC precursor were added sequentially. The reaction was maintained at 80 °C under continuous stirring at 1000 rpm for 4 h. The final product was harvested by centrifugation and rinsed three times with ethanol. To remove the CTAT template, the product was refluxed in 1% NH4NO3 in ethanol for 12 h. The FITC-labeled NPs were then washed three additional times with ethanol and dried to obtain FITC-tagged DSeMSN. Characterization of DSeMSN Fourier-transform infrared (FTIR) spectroscopy was performed using a Bruker VECTOR 22 spectrometer (Bruker, Germany) at ambient temperature. The DSeMSN specimens were blended with KBr and pressed into pellets for subsequent analysis. Spectra were recorded in the range of 4000–400 cm⁻¹ with a resolution of 2 cm⁻¹. Characteristic absorption peaks corresponding to Se–Se bonds, Si–OH groups, and C-C bonds were identified. Nitrogen adsorption-desorption isotherms were gained utilizing a Micromeritics TriStar II Plus analyzer (Micromeritics, USA). Prior to measurement, the samples were vacuum-dried at 300 °C for 12 h and cooled to ambient temperature. The specific surface area was calculated using the Brunauer–Emmett–Teller (BET) method, and pore size distribution was determined using the Barrett–Joyner–Halenda (BJH) method. X-ray photoelectron spectroscopy (XPS) was conducted using a VG ESCALAB MKII spectrometer (Thermo Fisher, USA) equipped with a Mg Kα X-ray source (1253.6 eV). Binding energies were calibrated using the C 1s peak at 284.6 eV. To assess redox responsiveness, DSeMSN samples were incubated in deionized water or 100 μM H2O2 for 24 h, and changes in the Se 3d signal were analyzed to evaluate diselenide bond cleavage. Loading of PDA and/or NUAK1 protein onto DSeMSN DSeMSN NPs, with or without FITC labeling, were suspended in polydopamine (PDA) solutions at concentrations ranging from 1 mg/mL to 5 mg/mL and stirred at ambient temperature for 2 h. The PDA-loaded NPs (DSeMSN@PDA) were then collected by centrifugation and washed three times with phosphate-buffered saline (PBS). For protein loading, DSeMSN was suspended in a NUAK1 protein solution (400 μg, 14-661-M, Sigma-Aldrich) and incubated under rotation at 4 °C for 12 h to ensure complete adsorption. Unbound protein was removed by centrifugation, yielding DSeMSN@NUAK1 NPs. To prepare NPs co-loaded with PDA and NUAK1, DSeMSN was suspended in a mixed solution containing varying concentrations of PDA and NUAK1 protein. The suspension was rotated at 4 °C for 12 h, followed by centrifugation to remove unbound components. The resulting dual-loaded NPs were designated as DSeMSN@(PDA + NUAK1). Lipid coating Lipid-coated NPs were prepared using a lipid mixture consisting of DSPE-PEG, DSPE-PEG-FA, PC, and CHO. The lipids were first dissolved in chloroform and subjected to rotary evaporation to form a thin lipid film. The dried film was rehydrated with PBS, into which DSeMSN@PDA, DSeMSN@NUAK1, or DSeMSN@(PDA + NUAK1) NPs were added. The mixture was then extruded repeatedly via a 200 nm polycarbonate membrane employing an Avanti Mini-Extruder (Avanti Polar Lipids, USA) to produce uniform lipid-coated NPs. Blank lipid NPs were referred to as FLN (FA-modified lipid NPs), while lipid-coated formulations were named FLN@PDA NPs, FLN@NUAK1 NPs, or FLN@(PDA + NUAK1) NPs, respectively. NPs prepared without FA modification (using DSPE-PEG, PC, and CHO only) were similarly processed and designated as LN@(PDA + NUAK1) NPs. Characterization of NPs Transmission electron microscopy (TEM) imaging was performed utilizing a JEM-2100F microscope (Hitachi H-7650, procured from Shanghai BAHENS Instrument Technology Co.) at an accelerating voltage of 80 kV. Image analysis software (version 1.53, National Institutes of Health, USA) was used to determine the average particle diameter. The nanoparticle hydrodynamic size and polydispersity index (PDI) were determined using a Zetasizer Nano ZS 90 (Malvern Instruments, UK). Measurements were conducted at 25 °C after dispersing the samples in deionized water at a concentration of 0.5 mg/mL. Zeta potential measurements were also conducted under the same conditions. Each measurement was repeated three times, and the average values were reported. NP stability testing Short-term stability: NPs were suspended in PBS and stored at ambient temperature. The hydrodynamic diameter and PDI were measured at time points of 0, 12, 24, 48, and 72 h using the Malvern Zetasizer Software (version 7.13, Malvern Instruments) to evaluate short-term physical stability. Long-Term Stability: To assess long-term stability, NPs were incubated in PBS at 37 °C for up to 35 days. Samples were collected at 7-day intervals, and their hydrodynamic diameter and PDI were measured via dynamic light scattering (DLS) to monitor changes in particle size distribution and aggregation over time. Drug release experiments Encapsulation Efficiency (EE) of NUAK1 Protein: The EE of NUAK1 protein in FLN@(PDA + NUAK1) NPs was measured utilizing high-performance liquid chromatography (HPLC, Agilent 1200, USA). NP samples were dissolved in PBS and filtered through a 0.22 μm membrane prior to injection into the HPLC system. Detection was performed at a wavelength of 280 nm. The EE and drug loading were calculated based on the measured protein content. ROS-Responsive Release: To evaluate the ROS-responsive release of NUAK1 in vitro, FLN@(PDA + NUAK1) NPs were suspended in PBS with or without 1 mM H2O2. The suspensions were placed into dialysis bags (HF132680, Spectrum Medical, Ruida Henghui, Beijing, China) and immersed in 30 mL of PBS at 37 °C in a water bath shaker set to 130 rpm. At predetermined time points (0, 1, 3, 6, 12, 24, and 48 h), 1 mL of release medium was harvested and replaced with an equal volume of fresh PBS. The amount of NUAK1 released was quantified by HPLC, and release profiles were plotted to assess the ROS-responsiveness of NPs. Cell culture The human early pregnancy trophoblast cell line HTR8/SVneo (CRL-3271, ATCC) and mouse macrophage RAW264.7 (TIB-71, ATCC) were propagated in RPMI-1640 (R8758, Sigma Aldrich) enriched with 10% fetal bovine serum (TMS-016), 100 U/mL penicillin (9073-60-3), and 10 μg/mL streptomycin (85886). Incubation was carried out at 37 °C in a 5% CO2 environment using a humidified chamber (Heracell™ Vios 160i CR, Thermo Scientific). When cell confluence reached 80% to 90%, passaging was performed. The cell groupings for this study were: the control group consisted of untreated normal cells (Control); the oxidative stress group included cells treated with 100 μM H2O2 (H2O2); the oxidative stress group treated with FLN@PDA NPs (H2O2 + FLN@PDA) received 100 μM H₂O₂ and 100 µg/mL FLN@PDA NPs; the oxidative stress group treated with FLN@NUAK1 NPs (H2O2 + FLN@NUAK1) also received 100 μM H₂O₂ and 100 µg/mL FLN@NUAK1 NPs; the oxidative stress group treated with FLN@(PDA + NUAK1) NPs (H2O2 + FLN@(PDA + NUAK1)) received 100 μM H₂O₂ and 100 µg/mL FLN@(PDA + NUAK1) NPs; and the oxidative stress group treated with FLN@(PDA + NUAK1) NPs along with the autophagy inhibitor 3-Methyladenine (3-MA, 5 mM, HY-19312, MCE, USA) (H2O2 + FLN@(PDA + NUAK1) + 3-MA) received 100 μM H₂O₂ and 100 µg/mL FLN@(PDA + NUAK1) NPs and 5 mM 3-MA. Cytotoxicity assay HTR8/SVneo and RAW264.7 cells were seeded in 96-well plates (5000 cells per well). Once the cells adhered and entered the logarithmic growth phase, the medium was replaced with serum-free RPMI-1640. Cells were then treated with varying concentrations of FLN@(PDA + NUAK1) NPs (ranging from 5 to 100 μg/mL) and incubated for 24 h. Cell viability was assessed using a Cell Counting Kit-8 (CCK-8, C0041, Beyotime, Shanghai, China). Following the addition of 10 μL of CCK-8 reagent to each well, the plates were incubated at 37 °C for 4 h. Absorbance was subsequently measured at 450 nm using a microplate reader to evaluate cell viability. Cellular uptake of NPs HTR8/SVneo cells were seeded in 24-well plates (50,000 cells per well) and incubated for 24 h. Subsequently, FITC-labeled NPs were added at a final concentration of 50 μg/mL, and the cells were incubated at 37 °C for 4 h. After incubation, the cells were rinsed three times with PBS to remove uninternalized NPs and then fixed with 4% paraformaldehyde (PFA, Sigma-Aldrich, USA) for 15 min. Following fixation, cells were again washed with PBS. Intracellular fluorescence distribution was visualized using a confocal laser-scanning microscope (CLSM, Leica TCS SP8, Germany). For quantitative analysis, cells were digested with trypsin and resuspended in 1 mL PBS. Investigation of endocytosis mechanisms To investigate the cellular uptake pathway of FLN@(PDA + NUAK1) NPs, HTR8/SVneo cells were pretreated with different endocytic inhibitors. Specifically, cells were incubated at 37 °C for 30 min with Dynasore (80 μM, 324410, Sigma-Aldrich), chlorpromazine (CPZ, 10 μM, BP856, Sigma-Aldrich), or EIPA. In addition, some cells were pre-incubated with excess free FA before NP administration to competitively block folate receptor-mediated uptake. FITC-labeled FLN@(PDA + NUAK1) NPs were then added at a final concentration of 50 μg/mL and incubated for a further 2 h. After treatment, the cells were washed three times with PBS, fixed with 4% paraformaldehyde for 15 min, and the intracellular fluorescence distribution of the NPs was examined using confocal laser scanning microscopy (CLSM; Leica TCS SP8, Germany). Fluorescence intensity in each group was quantified using ImageJ software to evaluate the changes in NP uptake under different inhibitory conditions. Western blot Proteins from cultured cells or placental tissue were extracted following enzymatic dissociation with trypsin (T4799-5G, Sigma-Aldrich). Lysis was performed in RIPA buffer supplemented with protease inhibitors, and protein concentration was quantified via a BCA assay (AR1189, Boster Bio, China). Equal amounts of protein were subjected to SDS–PAGE and transferred onto PVDF membranes. After blocking with 5% BSA (9048-46-8, Sigma-Aldrich) for 1 h at ambient temperature, membranes were incubated overnight at 4 °C with primary antibodies (Supplementary Table 1). After incubation, the membranes were washed three times with PBST (3 × 5 min) and then incubated with either Anti-Rat-HRP secondary antibody (Polyclonal, 31475, 1:5000; Thermo Fisher, USA) or Anti-Rabbit-HRP secondary antibody (Polyclonal, 65-6120, 1:5000; Thermo Fisher, USA) at room temperature for 1 h. The membranes were washed again with PBST three times (3 × 5 min), followed by the addition of ECL working solution (Omt-01, Beijing Omekaide Pharmaceutical Technology Co., Ltd., Beijing, China) and incubation for 1 min at room temperature. Band intensities were analyzed using ImageJ software (version 1.53). β-actin served as the internal loading control, and total protein was additionally verified by Ponceau S staining. All the uncropped western blots are available in the supplementary materials. Quantitative real-time PCR (RT-qPCR) Total RNA was extracted from cells or tissues using TRIzol reagent (10296010, Invitrogen, Thermo Fisher, USA. RNA concentration and purity were determined using a NanoDrop spectrophotometer (ND-1000, Thermo Fisher, USA). RT was performed using the PrimeScript™ RT Reagent Kit (RR086A, TaKaRa, Mountain View, CA, USA). RT-qPCR was completed utilizing SYBR® Premix Ex Taq™ (DRR820A, TaKaRa) on the LightCycler® 480 Software (version 1.5, Roche Diagnostics, Pleasanton, CA, USA). GAPDH served as the internal reference gene. Primer sequences were designed and synthesized by Shanghai General Biology Technology Co., Ltd. (listed in Supplementary table 2). Relative gene expression levels were calculated via the 2−ΔΔCt method. Detection of total and mitochondrial ROS Total ROS measurement: Intracellular ROS levels were tested using 2′,7′-dichlorodihydrofluorescein diacetate (DCFDA, HY-D0940, Abcam, USA), a cell-permeable fluorogenic probe that fluoresces upon oxidation by ROS. A 10 mM stock in methanol was diluted to 100 μM in culture medium prior to use. HTR8/SVneo trophoblasts (2 × 104) were cultured overnight on sterilized glass coverslips in six-well plates. Cells were then exposed to 100 μM H₂O₂ for 24 h to induce oxidative stress. Post-treatment, cells were washed with cold HBSS and incubated with 100 μM DCFDA at 37 °C for 30 min. After PBS rinsing, coverslips were mounted on slides, and fluorescence was detected using a multiphoton confocal microscope (Nikon A1R, ×100 objective). Fluorescence intensity was quantified using NIS-Elements software (version 5.21, Nikon). Mitochondrial ROS detection: Mitochondrial superoxide levels were assessed using MitoSOX™ Red mitochondrial superoxide indicator (M36008, Thermo Fisher, USA), and mitochondrial structure was visualized using MitoTracker™ Green (M7514, Thermo Fisher, USA). A 5 mM stock solution of MitoSOX was prepared in DMSO. Trophoblast cells (2 × 104) were seeded on coverslips, incubated overnight, and treated with 100 μM H2O2 for 24 h. Post-treatment, cells were washed with cold PBS and incubated with 2 μM MitoSOX at 37 °C for 30 min. After washing, coverslips were mounted on slides, and fluorescence was visualized with a multiphoton confocal microscope at ×100 magnification. Image analysis was conducted using NIS-Elements software. ATP content measurement Trophoblast cells (1 × 105/well) were seeded in six-well plates and subjected to the indicated treatments. ATP levels were measured using an ATP assay kit (BC0300, Solarbio, Beijing, China) according to the manufacturer’s instructions. The assay is based on the principle that hexokinase (HK) catalyzes the reaction between glucose and ATP to generate glucose-6-phosphate, which is subsequently converted to NADPH by glucose-6-phosphate dehydrogenase (G6PDH). NADPH exhibits a characteristic absorption peak at 340 nm, and its concentration is proportional to the ATP level. Absorbance at 340 nm was measured using a UV spectrophotometer (DU720, Beckman, USA) equipped with DU Series 700 Software (version 2.0), and ATP concentrations were calculated using a standard curve. Immunofluorescence staining Cells and tissue sections were rinsed with cold PBS and fixed with 4% PFA (P885233, Macklin, USA) for 15–30 min. Samples were then permeabilized with 0.1% Triton X-100 (L885651, Macklin, USA) for 15 min. After two PBS washes, samples were blocked with PBS containing 15% FBS and incubated overnight at 4 °C. Primary antibodies, including mouse anti-4-HNE (Monoclonal, MA5-27570, Thermo Fisher, USA; 1:50), rabbit anti-p62 (Polyclonal, ab109012, Abcam; 1:50), rabbit anti-Beclin 1 (Polyclonal, ab62557, Abcam; 1:50), rabbit anti-MCT1 (Polyclonal, PA5-72957, Thermo Fisher; 1:250), rabbit anti-MCT4 (Polyclonal, ab308529, Abcam; 1:50), rabbit anti-TPBPA (Polyclonal, orb186386, Biorbyt; 1:50), and anti-Laminin (Polyclonal, ab133645, Abcam; 1:50), were applied to the cells or tissues and incubated overnight at 4 °C. After three washes with TBST (TBS containing 1% Tween-20), samples were incubated with the appropriate fluorescent secondary antibodies, including goat anti-rabbit Alexa Fluor™ 555 (Polyclonal, 1:500, A-21428, Thermo Fisher, USA), goat anti-mouse Alexa Fluor™ 488 (Polyclonal, 1:500, A-11001, Thermo Fisher, USA), goat anti-rabbit Alexa Fluor™ 488 (Polyclonal, 1:500, A-11008, Thermo Fisher, USA), or goat anti-mouse Alexa Fluor™ 555 (Polyclonal, 1:500, A-21424, Thermo Fisher, USA). Nuclei were counterstained with DAPI. Fluorescence images were captured using a Zeiss Observer Z1 fluorescence microscope (Germany). Fluorescence intensity and the number of positive cells were quantified using ImageJ software. TUNEL fluorescence staining Apoptosis in cells and placental tissue was detected using a TUNEL assay kit (C1086, Beyotime, Shanghai, China). For cells, samples were fixed with immunostaining fixative (P0098, Beyotime) for 30 min. For tissues, sections were treated with 20 μg/mL proteinase K (ST532/ST533, Beyotime) at ambient temperature for 15–30 min. Endogenous peroxidase activity was quenched with 3% hydrogen peroxide. Fifty microliters of TUNEL detection solution was added to each sample and incubated in a dark, humid chamber at 37 °C for 60 min. After three PBS washes, coverslips were mounted using antifade mounting medium. TUNEL-positive cells were visualized using a Zeiss Observer Z1 fluorescence microscope under excitation wavelengths of 450–500 nm. Fluorescence intensity and the number of apoptotic cells were quantified using ImageJ. Dual-labeling mRFP-GFP-LC3 adenoviral transfection Cells were seeded in confocal dishes and cultured for 4 days prior to transfection with mRFP-GFP-LC3 adenovirus (WLA135, Shenyang WANLEI Biotechnology Co., Ltd., Shenyang, China). After treatment, cells were rinsed with PBS and fixed with 4% PFA for observation under a confocal microscope (MICA, Leica). Lysosomal activity assay Cells were incubated with the Magic Red Cathepsin B probe (ImmunoChemistry (ICT)_937, BIOPIKE, Shanghai, China) at 37 °C for 90 min in the dark, followed by PBS washing. Red fluorescence intensity was positively correlated with cathepsin B activity. TEM for cellular ultrastructure Cells were harvested and fixed with 2.5% glutaraldehyde (111-30-8, Sigma-Aldrich, USA) for 1 h in the dark, followed by post-fixation with formaldehyde (50-00-0, Sigma-Aldrich, USA). After dehydration through a graded acetone series (48358, Sigma-Aldrich, USA), cells were embedded in epoxy resin (1675-54-3, Sigma-Aldrich, USA). Polymerization was performed sequentially at 42 °C for 2 h, 52 °C overnight, and finally at 62 °C overnight. Ultrathin sections (50–70 nm) were prepared using a Leica ultramicrotome (EM UC6, Leica Biosystems, Germany). Sections were stained with 1% uranyl acetate (XHL2375, Hubei Xin Dividend Chemical Co., Ltd., Tianmen, China) for 5 min and placed in a CO₂-deficient environment for 2–4 min to induce citrate formation. Imaging was performed on a Hitachi H7650 TEM (DOLEE) at 100 kV, and digital images were captured with an AMT V700 side-mounted camera. Enzyme-linked immunosorbent assay (ELISA) Venous blood samples were allowed to clot at ambient temperature for 20–30 min and centrifuged at 2000 × g for 10 min to obtain serum. Supernatants from cultured trophoblast cells or homogenized placental tissue were also collected. All samples were stored at −80 °C prior to analysis. The levels of relevant cytokines and factors were measured using commercial ELISA kits from Abcam, following the manufacturer’s instructions: Human IL-6 ELISA Kit (ab178013) and Mouse IL-6 ELISA Kit (ab222503); Human TNF-α ELISA Kit (ab181421) and Mouse TNF-α ELISA Kit (ab208348); Human IL-18 ELISA Kit (ab215539) and Mouse IL-18 ELISA Kit (ab216165); Human IL-1β ELISA Kit (ab214025) and Mouse IL-1β ELISA Kit (ab197742); Human/Mouse Endothelin-1 ELISA Kit (ab133030); Human PLGF ELISA Kit (ab260056) and Mouse PLGF ELISA Kit (ab197748); and Human VEGFA ELISA Kit (ab222510) and Mouse VEGFA ELISA Kit (ab209882). Absorbance was measured at 450 nm using a microplate reader (Bio-Rad, USA), and the concentrations of each analyte were calculated based on standard curves. Scratch migration assay HTR8/SVneo cells were cultured in 24-well plates. A sterile 200 µL pipette tip was utilized to create a scratch in the center of each well, followed by PBS washing to remove detached cells. Cell migration into the scratch area was imaged at 0 and 24 h, and the migration area was quantified using ImageJ software. Transwell assay Transwell inserts with 8.0 µm polycarbonate membranes (CLS3422, Corning, USA) and BioCoat Matrigel-coated membranes (354480, Corning, USA) were used. HTR8/SVneo cells (2 × 104 cells/well) suspended in serum-free medium were seeded into the upper chamber, while the lower chamber contained medium supplemented with 10% FBS. Following 24 h of incubation, non-migrated cells on the upper surface were removed. Cells that had migrated through the membrane were fixed and stained with 0.1% crystal violet (C0121, Beyotime, Shanghai, China). Images of eight randomly selected fields per insert were captured using a ×10 objective lens on an inverted microscope (XDS-900, Caikon, Shanghai, China). The number of migrated cells was quantified using the “Analyze Particles” tool in ImageJ. Mitochondrial observation and detection Mitochondria were stained with 100 nM MitoTracker Green (M46750, Thermo Fisher, USA) for 20 min and then visualized using a CLSM (LSM 880 with Airyscan, Carl Zeiss, Germany) and images were processed using ZEN 2.3 software (Zeiss). Regions of interest were selected for fluorescence intensity analysis, and the number of positively stained cells was quantified using ImageJ software. JC-1 staining HTR8/SVneo cells were incubated with 10 µg/mL JC-1 dye (T3168, Thermo Fisher, USA) at 37 °C for 10 min in the dark and then imaged using a fluorescence microscope. Red-orange fluorescence indicates mitochondrial membrane polarization due to dye aggregation, while green fluorescence corresponds to the monomeric form of JC-1, indicating depolarization. Images were acquired for both fluorescence channels, and ImageJ was used to quantify fluorescence intensity in each field. PE mouse model development Wild-type BALB/c female mice (6–8 weeks old, 18–22 g) were obtained from the Experimental Animal Research Center of our institute and maintained under specific pathogen-free conditions. Mice were housed in individually ventilated cages (maximum 5 mice per cage) with environmental enrichment (e.g., nesting materials). The housing room was maintained at 25 ± 2 °C and 55 ± 5% relative humidity under a 12-h light/dark cycle with ad libitum access to standard rodent chow and water. All animal experiments were approved by the Animal Ethics Committee of Women’s Hospital, School of Medicine, Zhejiang University (Approval No. ZJU20240513) and conducted in accordance with institutional guidelines and the ARRIVE guidelines. Animal welfare and health were monitored daily by trained personnel. Monitoring criteria included body weight, physical appearance (e.g., fur condition, posture), and spontaneous behavior. Humane endpoints were predefined as severe lethargy, inability to access food/water, or >20% body weight loss; however, no animals reached these endpoints during the course of the study. For timed mating, adult BALB/c males were housed with females at a ratio of 1:2. The presence of a vaginal plug the following morning was designated as embryonic day 0.5 (E0.5). Male mice were used only for breeding and were not subjected to experimental interventions or outcome analyses. Pregnant mice were randomly assigned to experimental groups (n = 10 per group unless otherwise specified). Randomization was performed using a random number generator. Investigators were blinded to group allocation during data collection and analysis. All experimental interventions and outcome assessments were performed in pregnant female mice. Sex was not considered a biological variable because PE is a pregnancy-specific disorder, and only pregnant females were included in the analyses. To establish a PE model, pregnant mice were subcutaneously injected with L-NAME (125 mg/kg/day; NG-nitro-L-arginine methyl ester hydrochloride; 51298-62-5, MCE, USA) starting from embryonic day 10.5 (E10.5) for a duration of 8 days. FLN@PDA, FLN@NUAK1, FLN@(PDA + NUAK1) (1 mg/kg/day, 100 μL), and FLN@(PDA + NUAK1) (2 mg/kg/day, 100 μL) were administered via tail-vein injection from E10.5 to E17.5 simultaneously. Non-invasive systolic blood pressure (SBP) measurements were taken using a tail-cuff system (Softron Biotechnology, Beijing, China) on the day of mating and on days E0.5, E4.5, E8.5, E12.5, and E16.5, between 9:00 AM and 11:00 AM. Urine samples were collected before mating and on E17.5 for albumin analysis. On E17.5, for terminal tissue collection, mice were deeply anesthetized with isoflurane (induction at 4%, maintenance at 1.5–2% via inhalation) to minimize suffering. Depth of anesthesia was confirmed by the lack of pedal reflex. Euthanasia was subsequently performed via cervical dislocation by trained personnel to ensure death before tissue harvesting. Uteri and kidneys were harvested, and fetuses and placentas were weighed separately. Tissue samples were either snap-frozen and stored at −80 °C for further analysis or fixed in 4% PFA and processed for paraffin embedding using standard protocols. Fetal resorption was assessed by counting the number of viable and resorbed fetuses per uterus. The fetal resorption rate was calculated as the number of resorbed fetuses divided by the total number of fetuses (resorbed + viable). Grouping: control (T0: normal control mice); PE (T1: mice injected subcutaneously with L-NAME to induce PE, with PBS administered via tail vein as the control); PE + FLN@PDA (T2: mice injected with L-NAME and tail vein injected with FLN@PDA); PE + FLN@NUAK1 (T3: mice injected with L-NAME and tail vein injected with FLN@NUAK1); PE + FLN@(PDA + NUAK1) (T4: mice injected with L-NAME and tail vein injected with FLN@(PDA + NUAK1)). In vivo safety evaluation of NPs NP suspensions were prepared in sterile PBS at a concentration of 10 mg/mL. Mice were administered 200 μL of the suspension intravenously once daily for 10 consecutive days, while control mice received an equivalent volume of PBS. On Day 12 post-injection, blood was collected from the orbital sinus into heparinized tubes under mild anesthesia. Serum was separated by centrifugation at 1100×g for 10 min at 4 °C and stored at −80 °C for subsequent biochemical analysis. Following euthanasia on Day 12, major organs (liver, kidneys, spleen, heart, and lungs) were harvested, rinsed with PBS to remove residual blood, and fixed in 10% neutral-buffered formalin for 24 h. Fixed tissues were dehydrated in graded ethanol, cleared in xylene, and embedded in paraffin. Histological changes among treatment groups were evaluated by hematoxylin and eosin (H&E) staining. In vivo targeting observation of NPs To verify the placental targeting capability of FLN@(PDA + NUAK1) NPs, normal pregnant BALB/c mice at E12.5 (n = 10 per group) were used. Prior to NP injection, mice received daily oral L-NAME (Sigma-Aldrich, USA) until E10.5. On E12.5, mice were injected via the tail vein with Cy5-labeled FLN@(PDA + NUAK1) NPs (2 mg/kg). In vivo imaging was performed under 2.5–3.75% isoflurane (Forene®, Abbott, Copenhagen) anesthesia using the IVIS® 200 imaging system (Xenogen, Calliper Life Sciences, Hopkinton, MA) with Cy5-specific excitation (λex = 640 nm) and emission (λem = 700 nm) filters. All images were acquired using consistent settings: 1-s exposure, medium binning, f-stop of 2, and a 13 × 13 cm field of view. Imaging was conducted at 30 min and 12 h post-injection. Image analysis was performed using Living Image software (version 4.0, PerkinElmer, USA). After euthanasia, the uterus, placentas, fetuses, and major organs were collected, rinsed with PBS, and imaged on the IVIS platform for Cy5 fluorescence detection. Total fluorescence in regions of interest (ROIs) was quantified using the Living Image 4.0 software package (Caliper Life Sciences, Hopkinton, MA), and NP distribution within each organ was analyzed. Quantitative analysis of Cy5 signal in placental tissues was performed at different time points (30 min and 12 h) post-injection using the IVIS Spectrum system71. Biochemical analysis Plasma levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), creatine kinase (CK), cholesterol (CHO), triglycerides (TG), total bilirubin (TBIL), direct bilirubin (DBIL), glucose (GLU), urea (UREA), creatinine (CREA), and total protein (TP) were examined utilizing a Cobas c311 automated biochemical analyzer (Roche Diagnostics). Histopathological staining For H&E staining, tissue samples from the placenta and major organs (liver, heart, spleen, kidneys, and lungs) were fixed, paraffin-embedded, and sectioned. Deparaffinization was performed with xylene, and sections were rehydrated through graded ethanol (100%, 95%, 70%; Sigma-Aldrich, USA). Sections were stained with hematoxylin (H8070, Solarbio, Beijing, China) for 5–10 min at ambient temperature, rinsed in distilled water, then counterstained with eosin (G1100, Solarbio, Beijing, China) for 5–10 min. Slides were dehydrated, cleared, and mounted for microscopic evaluation (Olympus, Tokyo, Japan). Immunohistochemical staining for 8-hydroxy-2’-deoxyguanosine (8-OHdG) To assess DNA oxidative damage, immunohistochemical staining for 8-OHdG was conducted. Paraffin-embedded placental sections underwent antigen retrieval and were blocked with 10% goat serum (Beyotime, China) for 1 h at ambient temperature. Sections were incubated overnight in the dark with anti-8-OHdG antibody (ab48508, Abcam, UK; 1:100 dilution). The following day, goat anti-mouse IgG (ab6788, 1:1000 dilution) was applied for 30 min, followed by incubation with the streptavidin–biotin complex (SABC; Vector Laboratories, USA) at 37 °C for 30 min. Color development was performed using a DAB-staining kit (P0203, Beyotime Biotechnology, Shanghai) for 6 min, followed by hematoxylin counterstaining for 30 s. Sections were dehydrated through graded ethanol (70–100%), cleared in xylene (2 × 5 min), and mounted in neutral resin. Observations were performed under an upright microscope (BX63, Olympus, Japan), and DNA oxidative damage was quantified using ImageJ software. Statistical analysis All data were analyzed using GraphPad Prism software (version 8.0.1, GraphPad Software, San Diego, CA) and expressed as mean ± standard deviation (SD). Comparisons between two groups were conducted using unpaired t-tests, while one-way analysis of variance (ANOVA) was adopted for multiple group comparisons. Levene’s test was applied to assess the homogeneity of variance. When variances were equal, Dunnett’s t-test and LSD-t-test were used for pairwise comparisons; otherwise, Dunnett’s T3 test was applied. Pearson correlation analysis was conducted to evaluate associations among gene expression and immune cell content. A P-value < 0.05 was confirmed as statistically significant. Sample sizes were not statistically predetermined but were chosen based on similar studies in the field and our previous experience with the L-NAME-induced preeclampsia model72. No data were excluded from the analyses. Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Supplementary information Supplementary Information Peer Review File Reporting Summary Source data Source data Acknowledgements This work was supported by the National Key R&D Program of China (Grant Nos. 2022YFC2704600 and 2022YFC2704601) and the General Program of the National Natural Science Foundation of China (Grant No. 82471713). Author contributions P.J., X.Y., and Z.L. performed the experiments and data analysis. W.F. and Y.Y. were responsible for nanoparticle formulation and characterization. J.X. and H.L. contributed to cell culture and in vitro assays. L.C. assisted with animal model establishment and data collection. Q.L. conceived and supervised the study, acquired funding, and revised the manuscript. P.J. and Q.L. wrote the manuscript with input from all authors. All authors reviewed and approved the final version of the manuscript. Peer review Peer review information Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available. Data availability All data supporting the findings of this study are available within the paper and its supplementary information files. Source data are provided with this paper. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Supplementary information The online version contains supplementary material available at 10.1038/s41467-026-72775-0. References Associated Data Supplementary Materials Supplementary Information Peer Review File Reporting Summary Source data Data Availability Statement All data supporting the findings of this study are available within the paper and its supplementary information files. Source data are provided with this paper.