Mesenchymal stem cell membrane-coated kaempferol biomimetic nanoformulation for the treatment of atherosclerosis
Qianting Zhang, Xiangxiu Wang, Hongping Zhang, Keqiao He, Daojun Pu, Hong Chen, Anna Malashicheva, Wenbo Han, Chuanrong Zhao, Guixue Wang
TL;DR
Researchers developed a new nano-drug delivery system using mesenchymal stem cell membranes to target atherosclerotic sites and reduce plaque buildup.
Contribution
This is the first report of using mesenchymal stem cell membranes to encapsulate kaempferol for treating atherosclerosis.
Findings
Modified MSCM-encapsulated kaempferol nanoparticles specifically target endothelial cells in inflammatory environments.
Intravenous injection of the nanoparticles reduced lipid plaque load and improved plaque structure in a mouse model.
The nanoparticles showed biosafety with no significant effects on blood count, lipid balance, or organ function.
Abstract
Atherosclerosis and its complications are highly prevalent worldwide, and managing oxidative stress in endothelial cells to alleviate abnormal inflammatory damage is a critical therapeutic approach. Nanomedicine delivery systems offer promising solutions by overcoming the limitations of surgical interventions and the off-target effects of oral drugs. In this study, we developed a modified mesenchymal stem cell membrane (MSCM)-encapsulated nanoparticle drug delivery system that effectively delivers kaempferol to atherosclerotic sites. These biomimetic nanoparticles were able to specifically target endothelial cells in an inflammatory environment while evading macrophage-mediated endocytosis. Moreover, the modified MSCM-encapsulated kaempferol nanoparticles (KPM) had a protective effect on oxidatively damaged endothelial cells. In vivo, the modified nanoparticles successfully migrated…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
FIG. 1
FIG. 2
FIG. 3
FIG. 4
FIG. 5
FIG. 6
FIG. 7- —National Natural Science Foundation of China 10.13039/501100001809
- —National Natural Science Foundation of China 10.13039/501100001809
- —National Natural Science Foundation of China 10.13039/501100001809
- —Science and Technology Innovation Project of Jinfeng Laboratory
- —Fundamental Research Funds for the Central Universities 10.13039/501100012226
- —Chongqing Traditional Chinese Medicine Innovation Team Construction Project
- —Chongqing Science and Technology Joint Medical Research Project
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsNanoparticle-Based Drug Delivery · Atherosclerosis and Cardiovascular Diseases · Curcumin's Biomedical Applications
INTRODUCTION
I.
Atherosclerosis (AS) is a chronic inflammatory disease characterized by lipid deposition in the vessel wall of large- and medium-sized arteries, leading to the formation of a plaque microenvironment with elevated concentrations of reactive oxygen species (ROS) accompanied by the downregulation of various lipid peroxidation markers, such as glutathione peroxidase 4 (GPX4) and nuclear factor erythroid 2-related factor 2 (Nrf2).1–4 Increasing evidence suggests that inhibition of oxidative stress in endothelial cells has a positive effect on the treatment of AS. Oxidative stress activates inflammatory cells, triggering an inflammatory response that contributes to the formation and stabilization of AS plaques while also causing endothelial dysfunction, which is often considered the initiating stage in the development of AS. The natural compounds found in traditional Chinese medicine—including alkaloids, flavonoids, glycosides, and polysaccharides—exhibit potential anti-inflammatory and antioxidant properties, suggesting their therapeutic applicability in the treatment of atherosclerosis.5,6 Kaempferol (Kae), a natural flavonoid compound found in the leaves, stems, roots, and flowers of Zingiber officinale, acts as an antioxidant. It scavenges ROS, inhibits lipid peroxidation, and reduces ROS levels, thereby alleviating oxidative stress and mitigating AS. Kae is known for its broad therapeutic effects, including neuroprotective, antioxidant, and anticancer effects, and has been studied in various lipid oxidation-related disorders, including AS, stroke, Alzheimer's disease, and cancer.7,8 Furthermore, research has shown that Kae promotes autophagy by inhibiting the PI3K/Akt/mTOR pathway in human endothelial cells, which helps to alleviate the apoptosis induced by oxidized low-density lipoprotein (ox-LDL).9 Despite its promising therapeutic potential, the clinical application of Kae is limited by several factors, including poor solubility, low stability, suboptimal bioavailability, and inadequate concentrations in serum and target tissues.10–13 Therefore, strategies to extend its half-life, enhance drug utilization, and ensure targeted delivery to lesion sites are critical for maximizing the anti-atherosclerotic effects of Kae.
Nanodelivery systems offer significant advantages, particularly in overcoming challenges related to drug solubility and formulation properties.14–17 The widespread use of biomimetic cell membranes provides a promising solution for in vivo escape and targeted delivery to specific sites. Various cell membranes have emerged as excellent natural biomaterials for these purposes.18–20 Erythrocyte membrane-encapsulated biomimetic nanoparticles, for example, have been shown to evade macrophage-mediated phagocytosis in the bloodstream and accumulate specifically in atherosclerotic plaques.21 Similarly, membranes derived from macrophages,22 leukocytes,23 monocytes,24 and platelets have demonstrated additional therapeutic functions, including antimicrobial properties.25,26 Mesenchymal stem cells (MSCs) have also gained attention because of their ability to target inflammatory or malignant tissues in experimental animal models.27 Direct in vivo import of stem cells represents an innovative approach to disease treatment, leveraging their low immunogenicity and excellent tumor-homing properties for tissue repair and therapy.28,29 However, MSCs play an intricate role in disease progression and metastasis.30 The therapeutic application of MSCs is often limited to localized effects and may even pose risks, such as the induction of metastasis.31 In this context, MSC membranes (MSCMs) offer a safer alternative as drug delivery vehicles. MSCMs have been explored for use in disease treatment32 and can be further functionalized for targeted tumor therapy, as demonstrated by Yang et al.27 MSCs are derived from various sources, with umbilical cord stem cell-derived MSCs (UCMSCMs) offering clear advantages. These include ease of extraction, noninvasive collection, rapid expansion, and limited differentiation potential.33–35 While the therapeutic use of MSCs often relies on their secretory effects,36,37 these effects are primarily localized and can pose risks, such as metastasis induction.35,38,39 By modifying the surface of nanoparticles with MSC membranes, the particles can more closely mimic physiological states, avoiding immune clearance while facilitating interaction with target cells. This biofilm modification is expected to enhance the multifunctionality and adaptability of cell membrane-encapsulated nanoparticles, enabling better performance in complex biological environments. The expression of CD47 on UCMSCMs is particularly notable, as it enables the membrane to send a “do not eat me” signal to the immune system, similar to other biofilms. Additionally, MSCM retains the homing properties of MSCs, with integrins (β and α) and CD44 playing critical roles in cell adhesion, migration, and homing.40 Furthermore, the surface of the MSCM contains members of the CXCR and CCR families, which are coupled with G-protein receptors. These receptors enable MSCM to interact with effector ligands, contributing to their therapeutic potential.41
In this study, a combination of coprecipitation and emulsification methods was employed to construct antioxidant-loaded KPs, addressing the limitations of low oral bioavailability and poor water solubility. Mesenchymal stem cell (MSC) membranes were subsequently encapsulated to enhance the anti-inflammatory targeting ability, prolonged circulation, bioavailability, and biosafety of the nanoparticles. The successful preparation of nanomedicines with favorable physical and biochemical properties was confirmed through dynamic light scattering (DLS), Western blot (WB), and transmission electron microscopy (TEM) characterization. In vitro cell phagocytosis assays and small animal imaging experiments demonstrated that KPM effectively evades immune detection and targets inflammatory endothelial cells. The antioxidant and anti-atherosclerotic effects of KPM were validated by measuring in vitro ROS levels, ex vivo and in vivo malondialdehyde (MDA) concentrations, NRF2/GPX4 expression, and staining of aortic arch tissue sections. Furthermore, ex vivo and in vivo safety experiments confirmed the safety and reliability of KPM. In conclusion, KPM has significant advantages, including immune escape, targeted delivery to inflammatory endothelial cells, antioxidative stress activity, and anti-atherosclerotic effects, making it a promising candidate for clinical applications in the treatment of atherosclerosis.
RESULTS AND DISCUSSION
II.
Preparation and characterization of KP/KPM
A.
Antioxidant-loaded nanoparticles were prepared by combining coprecipitation and emulsification methods. First, Kae-loaded PLGA nanoparticles (KPs) were synthesized, followed by the extraction of MSCM. The MSCM was then encapsulated around the KP via a combination of ice bath sonication and coextrusion to form the KPM [Fig. 1(a)]. Dynamic light scattering (DLS) analysis revealed that after coating with MSCM, the particle diameter increased from approximately 148.9 to 167.2 nm, and the zeta potential decreased from approximately −1.00 to −4.89 mV, corresponding to the thickness of the macrophage bilayer membrane [Fig. 1(b)]. The DLS curves were smooth and flat, with no stray peaks, indicating a homogeneous particle size, stable potential, and good dispersion. Nanoparticles with a size of less than 200 nm and a negative potential are generally considered more favorable for cellular uptake. Detailed data on the synthesized nanoparticles are provided in supplementary material Table S1.
Preparation and characterization of biomimetic nanoparticles. (a) Schematic representation of the preparation of KP/KPM. KP was synthesized via coprecipitation and emulsification, and MSCM was extracted via hypotonic cracking. MSCM and KP were mixed and extruded to form KPM. (b) Representative plots of the particle size and potential for KP/KPM. (c) Electron micrograph of KP/KPM, scale bar = 200 nm. (d) Particle size of KP/KPM at different times in aqueous media. (e) Release curves of KPs/KPMs in PBS at different times. The data are expressed as the means ± standard deviations (n = 3). Statistical significance was calculated by one-way ANOVA and LSD or Dunnett T3 post hoc tests.
TEM analysis revealed that the nanoparticles exhibited a roughly spherical morphology with a mean size of 200 nm [Fig. 1(c)]. A hydrated membrane was clearly visible around the KPM, confirming that the MSC membrane fully encapsulates the KP (Fig. S1). To assess the stability of the nanoparticles, their particle sizes in PBS medium were monitored for up to 72 h, which revealed that both KP and KPM exhibited good stability [Fig. 1(d)]. The release rate of the nanoparticles was measured in a PBS environment at 37 °C, and after 72 h, the Kae release rates were 43.83% ± 1.53% for KP and 33.49% ± 1.25% for KPM. These findings suggest that the biofilm coating slows the release of the drug [Fig. 1(e)]. Additionally, KP and KPM demonstrated high drug entrapment efficiency, with an encapsulation efficiency of 79.97% ± 0.19% and a drug loading rate of 7.27% ± 0.02%. These results collectively confirmed the successful encapsulation of the PLGA cores with the MSCM.
Targeted delivery and safety assessment
B.
To further evaluate the quality of the MSC membranes, we characterized the proteins present in the MSCs, MSC membranes, MSC proteins, and KPMs. As shown in Fig. 2(a), the major protein content in all four samples exhibited minimal changes, indicating that the proteins in the MSC membranes were largely preserved. To assess the retention of key membrane proteins, we characterized the expression of CD47 and β1 integrins. The results revealed that both CD47 and β1 integrins were predominantly retained in the MSC membranes [Fig. 2(b)]. Previous studies have demonstrated that CD47 plays a crucial role in helping cells evade immune clearance in vivo,42 whereas β1 integrins are involved in targeting inflammatory endothelial cells. Additionally, CXCR4, which is overexpressed on the inflammatory endothelium, can bind to CXCL12.43
*Targeted delivery and safety assessment of biomimetic nanoparticles. (a) Characterization of proteins in MSCs, MSC membranes and MSC proteins and KPM by polyacrylamide gel electrophoresis. (b) Protein blotting analysis of CD47 and β1 in MSCs, MSC membranes and MSC proteins, and proteins in KPM. (c) DP and DPM were injected through the tail vein of the mice, and the circulation time was observed. (d) 1 h after DP and DPM were injected through the tail vein of the mice, blood vessels were dissected and removed for small animal imaging. (e) Fluorescence statistics of vascular small animal images. (f) After tail vein injection of DP or DPM, major organs were dissected and removed for small animal fluorescence imaging. The data are expressed as the means ± standard deviations (n = 3). Statistical significance was calculated by one-way ANOVA and LSD or Dunnett T3 post hoc tests, *P < 0.05; *P < 0.01.
Fluorescence imaging in mice [Figs. 2(c) and 2(f)] further revealed that the fluorescence intensity of the MSCM-loaded nanoparticles (KPM) decreased more slowly than that of the control nanoparticles (DP), indicating a prolonged circulation time. In the heart, liver, spleen, lungs, and kidneys of the imaged mice, DPs accumulated primarily in the liver and kidneys, suggesting rapid clearance and a shorter circulation time. In contrast, KPM showed prolonged circulation in the bloodstream, likely due to the presence of CD47 on the surface, which helps the virus evade rapid clearance by the liver and kidneys.
To assess targeting, we used ApoE^−/−^ mice, in which lipid deposition induces endothelial inflammation—a well-recognized feature of AS.44,45 The aortic arch, which is often a prominent site of AS lesions,46 was specifically targeted. Following tail vein injection and imaging after a period of circulation [Figs. 2(d) and 2(e)], we found that KPM predominantly accumulated in the aortic arch, which is consistent with the previous studies on the targeted delivery of nanoparticles to inflammatory endothelial regions.
Phagocytosis of nanoparticles by cells
C.
On the basis of previous studies that identified the vascular targeting effect of β1 integrin in AS mice, this study aimed to validate its role in targeting inflammatory endothelial cells. We treated both normal and inflammatory endothelial cells with DP or DPM, followed by a 2-h phagocytosis assay. Confocal microscopy was then used to observe the results. The data showed that, in inflammatory endothelial cells, the phagocytosis of nanoparticles was significantly greater after encapsulation with MSCM than in the normal endothelium. In normal endothelial cells, there was no significant difference in the amount of DP or DPM phagocytosed. However, after activation, both DP and DPM enhanced phagocytosis, with DPM demonstrating greater uptake than DP, confirming the targeting effect of the MSC membrane [Figs. 3(a) and 3(b)].
*Phagocytic efficiency of biomimetic nanoparticles. (a) Laser confocal plots of DP and DPM uptake by normal and inflamed endothelium. (b) Flow fluorescence images of DP and DPM uptake by normal and inflamed endothelium. (d) Flow fluorescence maps of the uptake of DP and DPM by the normal endothelium and inflamed endothelium after the addition of the β1 antibody. (f) Flow fluorescence plots of DPM uptake by inflamed endothelium after treatment with different concentrations of antibody. Figures (c), (e), and (g) correspond to the statistical plots of figures (b), (d), and (f), respectively. The letter A in front of each fluorescent nanoparticle represents the activation of the treated endothelium, i.e., the inflammatory endothelium. (h) Laser confocal plots of DP and DPM uptake by normal and inflamed macrophages. (i) Flow fluorescence plots of the uptake of DP and DPM by macrophages and deactivated macrophages. Figure (j) shows the statistical graph in figure (h). ADP refers to DP phagocytosis by activated macrophages, and ADPM refers to DPM phagocytosis by activated macrophages. The data are expressed as the means ± standard deviations (n = 3). Statistical significance was calculated by one-way ANOVA and LSD or Dunnett T3 post hoc tests, *P < 0.05; *P < 0.01.
To further confirm that the targeting effect was mediated by β1 integrin, we preincubated cells with an appropriate concentration of β1 antibody prior to phagocytosis. The results revealed a significant decrease in the phagocytosis efficiency of DPM [Fig. 3(c)], and the inhibition was dose dependent, with higher antibody concentrations leading to more pronounced inhibition. These findings indicate that the targeting of inflammatory endothelial cells by the MSC membrane is, indeed, mediated by β1 integrin. Furthermore, when we used a β1 antibody to inhibit the phagocytosis of DPs, we observed a similar decrease in uptake by inflammatory endothelial cells; however, the inhibition was less pronounced than that with DPM (Fig. S2). This finding suggests that β1 may influence additional pathways, which, in turn, affect nanoparticle phagocytosis.
To investigate the escape function of CD47, we used macrophages for immune simulation. Confocal microscopy and flow cytometry analysis revealed that DPs were captured by macrophages as foreign bodies, whereas DPM showed significantly lower uptake. When macrophages are treated with LPS to induce polarization, phagocytosis is greatly enhanced. Our results revealed that in M2-type macrophages, DPM, although it captured more M2-type macrophages than did normal macrophages, was still significantly less abundant than DP. These findings suggest that, in both normal and LPS-activated macrophages, the MSC membrane provides a protective “escape” function for nanoparticles, reducing their recognition and uptake by macrophages [Figs. 3(h) and 3(i)].
Antioxidant properties in vivo
D.
To explore the antioxidant activity of the drugs, ApoE^−/−^ mice were treated via tail vein injection. In untreated ApoE^−/−^ mice, the levels of Nrf2 and GPX4 were significantly decreased. Compared with both KP and Kae, KP and KPM resulted in greater upregulation of Nrf2/GPX4 [Figs. 4(a) and 4(b)], suggesting that KPM was more effective at reversing oxidative stress in vascular tissues. Nrf2/GPX4 is a key regulatory pathway in oxidative stress, and Kae can alleviate oxidative damage to endothelial cells by activating the Nrf2/GPX4 signaling pathway. Increased activation of this pathway helps to counteract oxidative stress, which is particularly pronounced in atherosclerotic (AS) mice because of the unique structure of the lesion site.47,48
*Antioxidant properties of biomimetic nanoparticles in vivo. (a) Fluorescence triple-labeling colocalization of GPX4, CD31, and DAPI in ApoE−/− mouse blood vessels. (b) Fluorescence triple-labeling colocalization of Nrf2, CD31, and DAPI in ApoE−/− mouse blood vessels. (c) ApoE−/− MDA content in vascular aortic arch tissue after different drug treatments. (d) ApoE−/− MDA content in the blood of mice after different treatments. (e) MDA content of endothelial cells after different drug treatments. (f) Fluorescence Image of the ROS activity of endothelial cells after different treatments. (g) Quantitative statistics of ROS in endothelial cells after different treatments. The data are expressed as the means ± standard deviations (n = 3). Statistical significance was calculated by one-way ANOVA and LSD or Dunnett T3 post hoc tests, *P < 0.05; *P < 0.01.
Kae, a natural antioxidant and anti-inflammatory agent, is suitable for the treatment of AS. However, as a free drug, Kae is rapidly cleared from the body upon injection. In contrast, KPs, as nanoparticles, benefit from the enhanced permeability and retention (EPR) effect, although they lack immune escape and targeting capabilities. In comparison, KPM, encapsulated in MSC membranes, combines the EPR effect of nanoparticles with the ability of CD47 to promote immune evasion and the ability of β1 to target inflammatory endothelial cells, creating more favorable conditions for antioxidant action.
Oxidative stress in endothelial cells plays a pivotal role in the progression of AS.49,50 To model oxidative stress in endothelial cells, we pretreated the cells with H_2_O_2_.51–53 As shown in Fig. 4(c), endothelial cell viability was significantly impacted at a H_2_O_2_ concentration of 0.4 mM, which was then optimized to 0.2 mM for subsequent experiments to maximize oxidative stress without affecting cell viability. Oxidative stress leads to excessive ROS production, which directly damages endothelial cells and impairs their vasodilatory function. Thus, reducing ROS levels is critical for repairing endothelial damage and alleviating AS progression. Our results revealed that KPM treatment significantly reduced ROS levels more effectively than KP treatment did [Figs. 4(d) and 4(e)].
In addition, the levels of three types of MDA, a marker of oxidative stress, were assessed: blood [Fig. 4(f)], tissue [Fig. 4(g)], and cellular [Fig. 4(h)]. KPM was found to significantly reduce MDA levels across all three measurements. Therefore, KPM primarily inhibits oxidative stress by increasing the expression of GPX4 and Nrf2 in tissues and reducing MDA and ROS activity, providing a protective effect against endothelial damage in AS.
Therapeutic effects in atherosclerotic mice
E.
The therapeutic effect was primarily assessed on the basis of fat deposition at the root of the vascular arch. ApoE^−/−^ mice were used to model abnormal lipid accumulation in the vasculature.54 After treatment, the Oil Red O (ORO) staining area was significantly reduced in all treatment groups, with KPM showing a markedly greater therapeutic effect than KP [Figs. 5(a) and 5(b)]. The presence of obvious atherosclerotic lesions in the blood vessels confirmed the successful establishment of the model. KPM treatment led to a substantial reduction in the plaque area—approximately 40%—demonstrating its significant therapeutic effect [Figs. 5(c) and 5(d)].
*KPM reduces lipid deposition in the aortic arch of ApoE−/− mice. (a) Oil red O staining of the aortic arch of ApoE−/− mice in different treatment groups. (b) Statistical graph of the data in figure (a). (c) Oil red O staining of aortic arch root sections from ApoE−/− mice in different treatment groups. (d) Statistical graph of the data in figure (c). The data are expressed as the means ± standard deviations (n = 3). Statistical significance was calculated by one-way ANOVA and LSD or Dunnett T3 post hoc tests, *P < 0.05; *P < 0.01.
According to the previous studies, atherosclerosis is associated with increased collagen deposition in plaques, leading to thickening of the plaque structure, further expansion of the plaque area, and disruption of vascular blood flow.55 Masson's trichrome staining was used to observe the collagen content in the plaques of the mice in the different treatment groups. Compared with the control treatment, treatment with Kae, KP, or KPM resulted in a reduced collagen content in the plaques [Figs. 6(a) and 6(b)]. Additionally, necrotic nuclei in the aorta were detected via toluidine blue staining. In the AS model, large areas of nuclear necrosis with cavitation were observed at the root of the hypertrophic aortic arch, indicating advanced lesions. However, after treatment, the necrotic areas were significantly reduced, and the degree of cavitation decreased, suggesting that one month of treatment effectively alleviated nuclear necrosis [Figs. 6(c) and 6(d)].
*Therapeutic effects of KPM in atherosclerotic mice. (a) Masson staining of aortic arch root sections from ApoE−/− mice in different treatment groups. (b) Statistical graph of the data in figure (a). (c) TB staining of aortic arch root sections from ApoE−/− mice in different treatment groups. (d) Statistical graph of the data in figure (c). (e) Immunohistochemical effect of CD68 on aortic arch root sections from ApoE−/− mice in different treatment groups. (f) Statistical graph of the data in figure (e). (g) Immunohistochemical effect of MMP9 on aortic arch root sections from ApoE−/− mice in different treatment groups. (h) Statistical graph of the data in figure (g). The data are expressed as the means ± standard deviations (n = 3). Statistical significance was calculated by one-way ANOVA and LSD or Dunnett T3 post hoc tests, *P < 0.05; *P < 0.01.
Previous studies have demonstrated that matrix metalloproteinase 9 (MMP9), which is secreted by macrophages and foam cells, plays a critical role in vascular matrix remodeling during atherosclerosis and contributes to the progression of this disease.56,57 Mice treated with the different formulations presented a decrease in relative macrophage numbers [Figs. 6(e) and 6(f)] and downregulation of MMP9 expression [Figs. 6(g) and 6(h)], further confirming that KPM effectively mitigates the progression of atherosclerosis.
Biosafety assessment
F.
Biocompatibility is important for the survival of an individual. We first tested the toxicity of the three types of cells that make up the blood vessels and found that the nanomedicine had no significant nanotoxicity on any of the three types of cells when a concentration of 10 μg was used; therefore, a concentration of 10 μg was used for treatment and exploration in subsequent experiments [Figs. 7(a)–7(c)]. Moreover, after one month of treatment, HE staining of major organs, such as the heart, liver, spleen, lungs, and kidneys, was carried out, and the five different treatments had no side effects on the function of major organs, indicating that the nanomedicine we developed had no effect on the safety of the organisms at the level of the individual organisms [Fig. 7(d)]. In addition, the hemolysis test, which explores the degree of compatibility of different nanomaterials with blood, revealed that none of them produced hemolysis compared with the positive control, and the rate of hemolysis was in accordance with the national standard, which was less than 5% [Fig. 7(e) and Fig. S3]. Body weight monitoring of different groups of mice during the administration period revealed that the administration did not cause feeding disorders in the mice. The significant weight loss that occurred with the first injection could be due to stress, but this loss was not due to toxicity, as all four groups of mice, including the PBS control group, lost weight. In contrast, after the injection, the body weights of the mice all increased [Fig. 7(f)], which was the result of constant high-fat chow feeding during the treatment period, indicating that the mice were able to continue to thrive after the administration of the drug without causing biotoxicity that would lead to weight loss. In addition, the four lipid profiles did not significantly differ, indicating that the nanomedicine did not cause substantial changes in blood lipids [Figs. 7(g)–7(j)]. Routine blood parameters and liver and kidney functions were sufficient to identify potentially harmful factors, and KPM treatment did not cause abnormalities in routine blood parameters or liver and kidney functions, indicating that there are no potentially harmful factors associated with KPM (supplementary material Tables S2 and S3).
*Biosafety assessment of biomimetic nanoparticles in vivo and ex vivo. (a)–(c) Effects of different drugs at different concentrations on smooth muscle cell, endothelial cell, and macrophage activities. (d) HE staining graphs of the starring organs of ApoE−/− mice in different drug treatment groups. (e) Hemolysis experiments in different treatment groups. (f) Changes in the body weights of ApoE−/− mice in different drug administration groups during the treatment period. (g)–(j) Lipid tetrads in ApoE−/− mice in different treatment groups. The data are expressed as the means ± standard deviations (n = 3). Statistical significance was calculated by one-way ANOVA and LSD or Dunnett T3 post hoc tests, *P < 0.05; *P < 0.01.
DISCUSSION AND CONCLUSIONS
III.
Our study introduces a novel MSCM-mimetic biopharmaceutical drug delivery system for AS treatment. Unlike many existing biofilm-targeted drug delivery strategies, this approach offers broad-spectrum functionalization driven by cellular function, with dual inflammatory tropism.58,59 This enables targeted drug delivery without the need for specific target molecules or complex bioconjugation processes. The system leverages cell function-driven immune escape, inflammation targeting, and damage homing to achieve efficient delivery to affected tissues.60,61 Once the drug-loaded nanoparticles accumulate at the inflammatory site, the drug is released slowly, providing effective therapeutic action.
The presence of membrane antigens (e.g., CXCR4 and β1) on the MSCM further enhances the ability of the nanoparticles to bind and sequester a variety of proinflammatory cytokines and chemokines, which are crucial in the atherosclerotic process. This sequestration helps neutralize inflammation and contributes to the therapeutic effect. While previous studies have focused on cell membrane-based immune escape, as well as the overexpression of integrin α and β families at AS lesion sites,62 our study highlights the additional benefit of macrophage membranes for inflammatory targeting in conjunction with immune escape.63,64
The KPMs developed in this study express CXCR4, which, in addition to immune escape and inflammation targeting, increases the likelihood of targeting AS lesions.16,65 Under inflammatory conditions, the chemokine CXCL12 is upregulated to attract immune cells (e.g., lymphocytes and monocytes) to the site of inflammation.66 CXCR4 on MSCM binds to CXCL12,67 facilitating the migration and infiltration of immune cells to the lesion site. Therefore, under inflammatory conditions, CXCL12 expression on endothelial cells may be upregulated to promote immune cell recruitment, enhancing the therapeutic potential of KPM.
The co-expression of integrin β family proteins and CXCR4 may provide superior targeting effects compared with those of other cell membrane types. However, a direct comparison with other cell membranes was not included in this study, but such an experiment could be valuable for future research. Additionally, in the AS environment, blood flow dynamics are altered, especially in the low/oscillatory shear stress microenvironment at plaque sites.49,68,69 These changes may influence CXCL12 expression and function, which can significantly affect immune regulation, cell behavior, and even tumor progression.70 Future studies should explore how mechanical forces alter CXCL12 expression and how these changes impact cell and tissue function. These findings have important implications for optimizing nanomedicine therapeutic strategies and drug delivery systems.
In conclusion, this study developed biomimetic nanoparticles synthesized with PLGA and Kae as the core and MSCM as the shell. The material characterization results indicated that the KPM nanoparticles exhibited favorable biological properties, including an appropriate size for the EPR effect, controlled slow-release characteristics, stability for formulation preservation, and good biocompatibility. At the cellular level, the experimental results demonstrated that the MSCM plays a crucial role in evading macrophage-mediated capture while also serving as a primary binding site for targeting inflammatory endothelial cells. At the animal level, KPM was shown to be effective in treating AS by specifically targeting the aorta, prolonging the circulation time, and reducing oxidative stress. This was achieved by upregulating the expression of antioxidant factors such as Nrf2 and GPX4 while simultaneously lowering ROS and MDA levels. Previous studies have confirmed the effectiveness of Kae in protecting the vascular endothelium from oxidative stress and inflammation-induced damage, both in vitro and in vivo, and have been shown to ameliorate the expression of proinflammatory cytokines and the upregulation of related inflammatory signaling pathways induced by endothelial dysfunction,71 for example, by upregulating the Nrf-2/HO-1 signaling pathway or downregulating the NF-κB and MAPK signaling pathways to reduce the levels of ROS and inflammatory factors, thereby protecting vascular endothelial cells from oxidative stress.72,73 Furthermore, Kae exhibited no additional side effects, and its therapeutic efficacy was significant during treatment. These findings are consistent with the effects of colchicine used by Gao et al. and rapamycin used by Wang et al. in the treatment of AS.21,74 However, this study also has several limitations, including a small sample size, which may affect the generalizability of the results. Larger-scale studies are needed to confirm whether KPM can be used for the prevention and treatment of human AS. Additionally, comparisons with other therapeutic agents should be made to identify the most suitable drug and establish standardized criteria for evaluation.
METHODS
IV.
Materials
A.
Human mesenchymal stem cells (MSCs) were kindly provided by Chongqing Honghui Umbilical Cord Stem Cell Center (Chongqing, China) and were isolated from the human umbilical cord. Rat smooth muscle cells (A_7_R_5_), mouse monocyte macrophage leukocytes (RAW 264.7), and a human umbilical vein endothelial cell line (HUVEC) were purchased from Pricella (Wuhan, China). RPMI 1640 medium and DMEM were purchased from Gibco, phosphate-buffered saline (PBS) and αMEM were purchased from Servicebio (Wuhan, China), and bovine serum from BI (Israel) was used. Serum was purchased from BI (Israel). Kaempferol (Kae), polyvinyl alcohol (PVA), and poly(lactic acid)-hydroxyacetic acid copolymer (50:50) (PLGA) were purchased from Meilunbio (Guangzhou, China). Dialysis bags (3500 kd) were purchased from Biosharp (Guangzhou, China), and a CCK8 kit, a cell membrane extraction kit, a ROS kit, and DIO were purchased from Beyotime (Jiangsu, China).
Cell culture
B.
The cells and HUVECs were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum; A_7_R_5_ and RAW 264.7 cells were cultured in DMEM supplemented with 10% fetal bovine serum; MSCs were maintained in αMEM supplemented with 10% fetal bovine serum; and the cells were passaged 3–7 times. All the cells were maintained at 37 °C in a humidified incubator containing 5% CO_2_.
Cell membrane extraction
C.
When the MSCs reached 90% confluence, the cells were washed twice with ice-cold PBS buffer, trypsin-digested, and resuspended in PBS. The cells from five dishes of T75 cell dishes were pooled, and after centrifugation of the cell suspension at 1000 × g for 5 min, the cell precipitate was collected, and the homogenate was resuspended in 1 ml of buffer (pH 7.4) containing Cell Membrane Extraction Reagent A, a mixture of phosphatases, and protease inhibitors. The mixture was incubated for 10 min. The mixture was placed in liquid nitrogen in a 37 °C water bath for 5 cycles of freeze-thawing. The supernatant was further centrifuged at 700 × g for 10 min at 4 °C, and the cell membrane precipitate was subsequently resuspended in PBS (pH 7.4) and stored at −80 °C for further experiments.
Synthesis of KP and KPM
D.
First, 10 mg of PLGA and 1 mg Kae was dissolved in 1 ml of DMSO, slowly added to a 1% PVA aqueous solution via the coprecipitation method, dialyzed in pure water via a dialysis bag, and kept away from light overnight. The resulting nanoparticles were subsequently obtained as KPs. The above-mentioned extracted cell membrane was mixed with 8 ml of KP according to the amount obtained from five dishes, sonicated in an ice bath for 3 min, and then squeezed back and forth 15 times by using a membrane extruder with 200 and 400 μm membranes to obtain KPM. Drug loading and concentrations of Kae in KP and KPM particles were measured by UV-Vis spectrophotometry. All particles were diluted by PBS to a final Kae concentration of 100 μg/ml for all in vitro and in vivo studies. A total of 2.5 μl of DIO was mixed with an equal volume of staining enhancement solution, 995 μl of DMSO containing 10 mg/ml was added, the mixture was ultrasonicated for 3 min, and DP and DPM were obtained via coprecipitation and membrane squeezing methods as described above.
Characterization of KP/KPM
E.
The amount of encapsulated Kae was quantified via UV-vis spectrophotometry. A Kae stock solution (1 mg/ml) was prepared in dimethyl sulfoxide (DMSO) and serially diluted to a concentration gradient of 1, 2.5, 5, 10, 20, and 25 μg/ml. The absorbance of these standard solutions was measured at 365 nm using a UV-vis spectrophotometer to establish a linear standard curve. Subsequently, lyophilized particle powder was accurately weighed and then completely dissolved in a known volume of DMSO to release the encapsulated drug. The absorbance of the resulting solution was measured at 365 nm. The concentration of Kae was calculated by interpolating the absorbance value into the established standard curve.
To examine the KP and KPM morphology, 10 μl of diluted sample solution was dropped onto a carbon-coated copper grid and deposited until the surface was dry. The samples on the grid were stained with 2% (v/v) phosphotungstic acid for 2 min, and the excess stain was removed by blotting with filter paper for analysis via transmission electron microscopy (TEM, HT7700, Hitachi, Japan). The hydrodynamic size distributions, zeta potentials, and polydispersity indices (PDIs) of KP and KPM were assessed via a Zetasizer Nano ZS dynamic light scattering (DLS) instrument (Malvern, UK).
Isolation of membrane proteins and Western blotting
F.
The protein concentrations of the MSC lysates and the MSCM, KP, and KPM fractions were quantified via the BCA assay. Aliquots of these total proteins were mixed with upwelling buffer and separated on 10% SDS–PAGE gels via a Bio-Rad electrophoresis system (USA). After electrophoresis, one piece of the gel was taken and stained with Caumas Brilliant Blue staining solution, washed, imaged, and observed for protein retention to explore the feasibility of the membrane lifting method; the proteins from the other gel were transferred to a PVDF membrane. The PVDF membrane was then blocked with 5% skim milk for 1 h and incubated with primary antibody at 4 °C overnight. After three washes with TBST, the membrane was incubated with secondary antibody for 2 h, followed by three washes with TBST. The PVDF membranes were incubated with electrochemiluminescent (ECL) substrate and then exposed under a Bio-Rad ChemiDox touch imaging system to identify CD47, β1 and CXCR4 expression via protein blotting.
Encapsulation packaging and in vitro drug release
G.
The encapsulated package loading was determined via the method of Wang et al.21 with reference to the general formula for the material release rate. To measure the drug release profile, 1 ml of KP or KPM solution was dialyzed in 10 ml of PBS at pH 7.4 via a dialysis bag.15,21,75 The amount of Kae was checked by placing it into 10 ml of medium, and 2 ml of dialysis buffer was taken at each time point at different time intervals in a shaker at 37 °C and replenished with 2 ml of fresh buffer. For the above-mentioned experiments, the amount of Kae was determined by measuring the absorbance value at a wavelength of 365 nm via a UV–Vis spectrophotometer (AOE Instruments, Shanghai, China).
Toxicity
H.
EC, A_7_R_5_, and RAW 264.7 cells were seeded at 10 000 cells/ml in 96-well plates, and after 24 h of stable growth, the cells were treated with different concentrations of Kae or equivalents of KP and KPM for 24 h. Then, serum-free medium containing 10 μl of CCK8 solution was added in the form of fluid exchange, the cells were incubated at 37 °C for 1–2 h, and their absorbance value was determined at 450 nm using a microplate reader. The toxicity of the drug was determined by measuring the absorbance value at 450 nm via the survival formula to calculate the survival rate of the cells.
In vivo and ex vivo safety tests
I.
All animal experiments were conducted in accordance with relevant guidelines and regulations and were approved by the Animal Welfare and Ethics Committee of Chongqing University (Approval No. CQULA-2022JC-12-180). For the hemolysis experiment, after the mice were anesthetized, blood was collected from the eye socket and anticoagulated 1:9 with EDTA-2K (10×); the anticoagulated blood was diluted 4:5 with 0.9% NaCl solution; 2 ml clean centrifuge tubes were used; 2 ml Kae, MSC membranes, KP, KPM, PBS, and distilled water were added, with three parallels in each group; 40 μl diluted blood was added to each tube; the samples were incubated at 37 °C for 1 h; at the end of the warming bath, the samples were centrifuged at 2000 × g for 5 min; and the absorbance value was measured at 545 nm. Each centrifuge tube was filled with 40 μl of diluted blood and incubated at 37 °C for 1 h. At the end of the warm bath, the centrifuge was placed at 2000 × g and centrifuged for 5 min. The centrifuged supernatant was collected, and the absorbance value was measured at 545 nm to calculate the hemolysis rate.
HE staining
J.
Frozen sections of the remaining heart and other major organs of the mice were prepared. The sections were stained with hematoxylin staining solution for nucleus staining, 1% hydrochloric acid for alcohol differentiation, drops of eosin solution for cytoplasmic staining, gradient alcohol dehydration, xylene solution for transparency, neutral gum for sealing, and imaging under a microscope.
Cellular uptake
K.
ECs were cultured and stabilized for 12 h at a density of 0.5 × 10^5^ cells/well in 24-well plates with round cell crawlers and then treated with 4 μg/ml for 24 h to form inflammatory cells, and the controls were untreated blank control cells. DP and DPM were diluted to a concentration of 100 μg/ml in serum-free 1640 medium, phagocytosed for 2 h, washed with PBS, fixed with 4% paraformaldehyde, stained with DAPI, sealed, and photographed via confocal microscopy for observation. Another six-well plate was used, and each well was seeded into a well plate at 1 × 10^5^ cells, cultured and stabilized for 12 h, treated with 4 μg/ml for 24 h to form inflammatory cells, and the control group consisted of untreated blank cells. DP and DPM were diluted to a concentration of 100 μg/ml with serum-free 1640 medium and phagocytosed for 2 h. Then, the cells were washed with PBS and digested with trypsin, and the cells were collected for flow cytometry analysis.
Inhibition of uptake
L.
RAW 264.7 cells were cultured at a density of 0.5 × 10^5^ cells/well in 24-well plates with round cells crawling for 12 h for stabilization and treated with 2 μg/ml for 24 h to form inflammatory cells, and untreated blank control cells were used as controls. DP and DPM were diluted to a concentration of 100 μg/ml in serum-free DMEM and phagocytosed for 2 h. The mixture was then washed with PBS, fixed with 4% paraformaldehyde, stained with DAPI, sealed, and photographed via confocal microscopy for observation. Another six-well plate was used, and each well was seeded into a well plate at a density of 1 × 10^5^ cells, cultured, stabilized for 12 h, and then treated with 2 μg/ml for 24 h to generate inflammatory cells. The control consisted of untreated blank cells. DP and DPM were diluted to a concentration of 100 μg/ml with serum-free DMEM and phagocytosed for 2 h. Then, the cells were washed with PBS, blown with PBS, and collected for flow analysis.
Another six-well plate was used, and each well was seeded into a well plate at 1 × 10^5^ endothelial cells, cultured and stabilized for 12 h, and then treated with 4 μg/ml for 24 h to generate inflammatory cells. The control consisted of untreated blank cells. The β1 antibody was diluted to different concentrations in serum-free 1640 medium and added to the plates for binding for 4 h. Then, DP and DPM were diluted to a concentration of 100 μg/ml in serum-free 1640 medium and phagocytosed for 2 h. Afterward, the cells were washed with PBS and trypsin digested, and the cells were collected and analyzed via flow cytometry.
Antioxidant efficacy
M.
The endothelial cells were pretreated with KP or KPM for 12 h, treated with 0.2 μM hydrogen peroxide for 24 h, treated with ROSup for 30 min as a positive control, incubated with DCFH-DA for 30 min, washed with PBS, and then placed under a fluorescence microscope to observe the level of ROS. The oxidation model was established according to the above steps, and the MDA content was determined via the addition of different drugs via a kit.
Co-immunofluorescence staining for DAPI, GPX4, and CD31 was performed on aortic arch sections from KP- and KPM-treated ApoE^−/−^ mice, while immunofluorescence staining for DAPI, Nrf2, and CD31 was performed, and the results were compared with those from the WT and PBS control groups to observe the levels of antioxidants in KPs and KPMs. MDA levels in the blood and aortic arch of ApoE^−/−^ mice from different drug treatment groups were measured via a kit.
Mouse model and anti-AS efficacy
N.
Male ApoE^−/−^ mice were fed high-fat chow for 2 months for modeling, divided into different treatment groups, and given tail vein injections every 3 days at the end of the second month, with a volume of 200 μl for each injection. The control group was injected with an equal volume of PBS, and the body weights were recorded during the period of treatment without the cessation of high-fat feeding. After one month of treatment, anesthesia was applied, and blood collection was performed for routine blood and lipid profile, while the five viscera and aortic arch samples were collected and fixed with 4% paraformaldehyde. Whole blood vessels were subjected to gross oil red or frozen sections for oil red O staining, Masson staining, TB staining, immunohistochemical determination of CD68, αSMA, and MMP9 expression, and statistical analysis of the treatment effect via ImageJ software.
Animal-targeting imaging and long-lasting circulation
O.
The successfully generated mice were injected with 200 μl of DP or DPM via the tail vein to observe the retention time and targeting function of the nanoparticles in vivo instead of with KP or KPM. The retention time was passed at certain time intervals, blood was collected from the tail vein, and the fluorescence intensity was measured under a fluorescence enzyme marker. The degree of fluorescence enrichment in the mice was observed via an in vivo imaging system. After the targeting experiments were cycled through 12 h and anesthetized and treated, the heart, liver, spleen, lungs, kidneys, and blood vessels were sampled and imaged with a small animal imaging apparatus.
Statistics and analyses
P.
The data are expressed as the means ± standard deviations (n = 3). Statistical significance was calculated by one-way ANOVA and LSD or Dunnett T3 post hoc tests, ^*^P < 0.05; ^**^P < 0.01; and ns, no significance.
SUPPLEMENTARY MATERIAL
See the supplementary material for three figures (supplementary Figs. 1–3) and three tables (supplementary Tables 1–3), interleaved in the sequence of their reference within the manuscript.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1U. Forstermann, N. Xia, and H. Li, “Roles of vascular oxidative stress and nitric oxide in the pathogenesis of atherosclerosis,” Circ. Res. 120(4), 713–735 (2017).10.1161/CIRCRESAHA.116.30932628209797 · doi ↗ · pubmed ↗
- 2Z. Wu, M. Zhou, X. Tang, J. Zeng, Y. Li, Y. Sun, J. Huang, L. Chen, M. Wan, and C. Mao, “Carrier-free trehalose-based nanomotors targeting macrophages in inflammatory plaque for treatment of atherosclerosis,” ACS Nano 16(3), 3808–3820 (2022).10.1021/acsnano.1c 0839135199998 · doi ↗ · pubmed ↗
- 3Y. Xu, Q. Zeng, B. Sun, S. Wei, Q. Wang, and A. Zhang, “Assessing the role of Nrf 2/GPX 4-mediated oxidative stress in arsenic-induced liver damage and the potential application value of Rosa roxburghii Tratt [Rosaceae],” Oxid. Med. Cell. Longevity 2022, 9865606.10.1155/2022/9865606 PMC 907355035528517 · doi ↗ · pubmed ↗
- 4S. Li, Z. Xu, Y. Wang, L. Chen, X. Wang, Y. Zhou, D. Lei, G. Zang, and G. Wang, “Recent advances of mechanosensitive genes in vascular endothelial cells for the formation and treatment of atherosclerosis,” Genes Dis. 11(3), 101046 (2024).10.1016/j.gendis.2023.06.01638292174 PMC 10825297 · doi ↗ · pubmed ↗
- 5Y. Guo, M. Jia, Q. He, J. Zeng, Y. Zhang, Y. Gao, Z. Zhao, A. Malashicheva, W. Miao, J. Zhao, G. Wang, and Y. Wang, “A carrier-free triple-drug co-assembled nanoformulation for synergistic treatment of cerebral ischemia-reperfusion injury,” ACS Appl. Mater. Interfaces 17(23), 33618–33632 (2025).10.1021/acsami.5c 0588240432247 · doi ↗ · pubmed ↗
- 6Y. Zhao, G. Zang, T. Yin, X. Ma, L. Zhou, L. Wu, R. Daniel, Y. Wang, J. Qiu, and G. Wang, “A novel mechanism of inhibiting in-stent restenosis with arsenic trioxide drug-eluting stent: Enhancing contractile phenotype of vascular smooth muscle cells via YAP pathway,” Bioact. Mater. 6(2), 375–385 (2021).10.1016/j.bioactmat.2020.08.01832954055 PMC 7484501 · doi ↗ · pubmed ↗
- 7Y. Yuan, Y. Zhai, J. Chen, X. Xu, and H. Wang, “Kaempferol ameliorates oxygen-glucose deprivation/reoxygenation-induced neuronal ferroptosis by activating Nrf 2/SLC 7A 11/GPX 4 axis,” Biomolecules 11(7), 923 (2021).10.3390/biom 1107092334206421 PMC 8301948 · doi ↗ · pubmed ↗
- 8P. Rajendran, T. Rengarajan, N. Nandakumar, R. Palaniswami, Y. Nishigaki, and I. Nishigaki, “Kaempferol, a potential cytostatic and cure for inflammatory disorders,” Eur. J. Med. Chem. 86, 103–112 (2014).10.1016/j.ejmech.2014.08.01125147152 · doi ↗ · pubmed ↗