A novel fertility-sparing nanocomposite for endometriosis treatment via oxidative stress and inflammation alleviation

Wenwen Liu: Writing – review & editing, Writing – original draft, Validation, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Jie Shen: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation. Jiaqi You: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation. Tao Wang: Software, Methodology, Investigation, Formal analysis. Pusheng Yang: Software, Methodology, Investigation, Formal analysis. Yaxin Miao: Software, Methodology, Investigation, Data curation. Xiaotong Peng: Software, Methodology, Investigation, Data curation. Yimin Yu: Methodology, Investigation. Chengyu Liu: Methodology, Investigation. Ang Li: Writing – review & editing, Supervision, Project administration, Funding acquisition, Formal analysis, Conceptualization. Jing Sun: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition, Conceptualization.

We prepared polydopamine nanocomposites (PDA NCs) and loaded them with dydrogesterone, following the methods reported in the literature [ 33 ]. By measuring the absorbance changes of the nanocomposites and supernatant, we found that, on average, 1 mg NCs could load approximately 10.26 μg dydrogesterone. To enhance their biocompatibility and targeting to endometriotic lesions, we functionalized the nanocomposites with methoxy polyethylene glycol amine (mPEG-NH2) and laminin (LN), synthesizing PDA-PEG-LN and DYD@PDA-PEG-LN nanocomposites, as shown in Fig. 2 A [ 34 , 35 ]. Fig. 2 Synthesis and characterization of nanocomposites (A) Schematic illustration showing the synthesis process of PDA-PEG-LN and DYD@PDA-PEG-LN. (B) SEM and TEM images depicting PDA, PDA-PEG, PDA-PEG-LN and DYD@PDA-PEG-LN. (C) Size distribution and (D) zeta potential measurements for PDA, PDA-PEG, PDA-PEG-LN, and DYD@PDA-PEG-LN. (E) FTIR spectra of DA and PDA. (F) FTIR spectra of dydrogesterone, PDA, PDA-PEG, PDA-PEG-LN, and DYD@PDA-PEG-LN. (G) UV–vis absorption spectra of PDA, PDA-PEG, PDA-PEG-LN, and DYD@PDA-PEG-LN Fig. 2 Synthesis and characterization of nanocomposites (A) Schematic illustration showing the synthesis process of PDA-PEG-LN and DYD@PDA-PEG-LN. (B) SEM and TEM images depicting PDA, PDA-PEG, PDA-PEG-LN and DYD@PDA-PEG-LN. (C) Size distribution and (D) zeta potential measurements for PDA, PDA-PEG, PDA-PEG-LN, and DYD@PDA-PEG-LN. (E) FTIR spectra of DA and PDA. (F) FTIR spectra of dydrogesterone, PDA, PDA-PEG, PDA-PEG-LN, and DYD@PDA-PEG-LN. (G) UV–vis absorption spectra of PDA, PDA-PEG, PDA-PEG-LN, and DYD@PDA-PEG-LN In Fig. 2 B, the results from scanning electron microscopy (SEM) and transmission electron microscopy (TEM) revealed that all the nanocomposites had a uniform spherical morphology. As depicted in Fig. 2 C, the dynamic light scattering (DLS) measurement results indicated that the PDA, PDA-PEG, PDA-PEG-LN, and DYD@PDA-PEG-LN nanocomposites were stable in water, with particle sizes of 125.3 nm, 134.6 nm, 177.7 nm and 185.5 nm, respectively. In Fig. 2 D, the zeta potential values of them were nearly −26.8 mV, −7.4 mV, −10.7 mV, and −8.7 mV, respectively. The Fourier transform infrared (FTIR) spectra provided additional evidence for the formation of PDA NCs ( Fig. 2 E). Absorption band between 3500 and 3200 cm −1 could be detected in both dopamine (DA) and PDA NCs due to the stretching vibrations of N-H, O-H and NH 2 . In the PDA NCs, the peaks at 1608 cm −1 and 1514 cm −1 were due to the stretching vibrations of C=C in the benzene ring, and the peak at 1514 cm −1 also corresponded to the bending vibrations of N-H. The main chemical bonds in dopamine were the benzene ring, phenolic hydroxyl groups, and amino groups. The benzene ring exhibited distinct sharp peaks at 1617 cm −1 and 1501 cm −1 . After polymerization to form polydopamine, the structure of the benzene ring changed, and new C=C double bonds formed. These C=C peaks were located in the wavenumber range of 1650 to 1500 cm −1 . Since the absorption peaks from different sources of C=C overlapped with each other, the absorption peak in this range broadened, and that was used to determine the successful synthesis of polydopamine. As shown in Fig. 2 F, PDA-PEG-LN nanocomposites retained characteristic absorption peaks of PDA at 3392 cm −1 , 1612 cm −1 , and 1505 cm −1 . Additionally, the C-H stretching vibration of PEG methylene resulted in an absorption band in the range of 3000-2800 cm −1 , and the C-O stretching vibration peak of polyethylene glycol was observed at 1094 cm −1 . The appearance of the amide C-N stretching vibration peak at 1402 cm −1 , the Ar-O stretching vibration peak at 1202 cm −1 , and the amine C-N stretching vibration peak at 1132 cm −1 confirmed the successful loading of LN. The characteristic absorption peaks of PDA-PEG-LN were still present in DYD@PDA-PEG-LN, and the characteristic absorption peak of dydrogesterone was observed around 1660 cm −1 , attributed to the stretching vibrations of C=O and C=C in dydrogesterone. Based on these observations, we can confirm the successful preparation of DYD@PDA-PEG-LN nanocomposites. As illustrated in Fig. 2 G, the UV–vis spectra of particles at different preparation stages were documented. DYD@PDA-PEG-LN NCs exhibited enhanced absorbance in the 260–300 nm range compared to PDA-PEG-LN NCs, which corroborated the successful encapsulation of dydrogesterone. Before verifying the effects of DYD@PDA-PEG-LN nanocomposites on endometrial stromal cells and endometriosis lesions, we explored the biocompatibility and blood compatibility of non-drug-loaded nanocomposites in vitro. Initially, we used the standard methyl thiazolyl tetrazolium (MTT) assay to assess the relative viability of human endometrial stromal cells (HESCs) under various concentrations of PDA-PEG-LN NCs. As shown in Fig. 3 A, no significant cytotoxicity was observed and the cell viabilities in all groups were above 90 %. Subsequently, cells at different incubation concentrations were stained using Calcein AM and propidium iodide (PI). Under fluorescent excitation, dead cells stained by PI will emit red light, while live cells stained by Calcein AM emitting green light. As shown in Fig. 3 B, almost all cells in these groups appeared green, which proved that PDA-PEG-LN NCs did not alert the morphology and viability of HESCs. By measuring the lactate dehydrogenase (LDH) concentration in the supernatant after co-incubation with PDA-PEG-LN NCs, the release of cellular LDH caused by PDA-PEG-LN NCs was assessed. The results ( Fig. 3 C) show that there was no significant difference in LDH release levels between the experimental groups and the control group, proving that the PDA-PEG-LN NCs did not induce potential cell damage. We also measured the ATP levels and found ( Fig. 3 D) that there was no significant difference in ATP levels among the groups, indicating that the PDA-PEG-LN NCs did not affect the mitochondrial function of the cells. The hemolysis assay was used to verify the interaction between PDA-PEG-LN NCs and blood components. After co-incubating the NCs with red blood cells for 3 h, no significant hemolysis was observed at any concentration ( Fig. 3 E). The above experiments demonstrated that PDA-PEG-LN NCs have good biocompatibility in vitro, which was consistent with previous research findings [ 27 ]. Fig. 3 Biosecurity and blood compatibility of nanocomposites (A) Relative viability of HESC cells and (B) fluorescence images of cells stained with Calcein AM and PI after incubation with PDA-PEG-LN. Scale bar: 500 μm. (C) LDH release levels and (D) ATP levels of HESCs with the addition of PDA-PEG-LN. (E) Hemolysis assessment after PDA-PEG-LN treated experiments. (F) Changes in mouse body weights, (G) serum levels of ALT, AST, and ALP, and (H) H&E staining images of the heart, liver, spleen, lung, and kidney after a 21-day treatment. Scale bar: 100 μm. Not significant (ns): p > 0.05. Inset of panel E: photographic image of hemolysis. Fig. 3 Biosecurity and blood compatibility of nanocomposites (A) Relative viability of HESC cells and (B) fluorescence images of cells stained with Calcein AM and PI after incubation with PDA-PEG-LN. Scale bar: 500 μm. (C) LDH release levels and (D) ATP levels of HESCs with the addition of PDA-PEG-LN. (E) Hemolysis assessment after PDA-PEG-LN treated experiments. (F) Changes in mouse body weights, (G) serum levels of ALT, AST, and ALP, and (H) H&E staining images of the heart, liver, spleen, lung, and kidney after a 21-day treatment. Scale bar: 100 μm. Not significant (ns): p > 0.05. Inset of panel E: photographic image of hemolysis. We administered DYD@PDA-PEG-LN NCs into mice via subcutaneous injection and tail vein injection at a dose of 10 mg/kg body weight, once every two days. After 21 days, we compared the subcutaneous injection group, tail vein injection group, and the control group to demonstrate their biocompatibility in vivo. As depicted in Fig. 3 F, there were no significant differences in body weight changes among these groups during injection process, and the levels of serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), and alkaline phosphatase (ALP) of all groups were not significantly different after 21 days of injection ( Fig. 3 G). We extracted the main organs each group and performed hematoxylin and eosin (H&E) staining. There were no significant morphological changes or signs of inflammation found in the hearts, livers, spleens, lungs, and kidneys of the mice, regardless of injecting subcutaneously or intravenously ( Fig. 3 H). The aforementioned experiments confirmed that DYD@PDA-PEG-LN NCs also had good biosecurity in vivo. Reactive oxygen species (ROS) are related to a variety of signaling pathways and cellular processes like proliferation, differentiation, and apoptosis [ 36 ]. In endometriosis, high levels of oxidative stress can not only promote chronic inflammation and the progression of the disease but may also damage ovarian function. The PDA NCs, acted as a potent antioxidant, capable of scavenging ROS [ 37 , 38 ], and also possessed the ability to alleviate inflammation and inflammation-induced damage [ 33 , 39 ]. Therefore, after demonstrating the good biocompatibility of PDA-PEG-LN NCs, we verified their ability to scavenge superoxide anions (O2· − ) and hydroxyl radicals (·OH), reduce intracellular ROS and the expression of inflammatory cytokines after stimulation. O2· − scavenging ability of PDA-PEG-LN NCs was reflected by measuring the inhibition ratio of photoreduction of NBT. The solution containing riboflavin, methionine, nitro blue tetrazolium (NBT) was irradiated under ultraviolet (UV). The reduction of NBT resulted in a stronger absorbance signal detectable around 600 nm, which represented the generation of a high concentration of O2· − [ 40 ]. After adding PDA-PEG-LN NCs to the above solution and UV radiation, the signal around 600 nm in the absorption spectra decreased. As depicted in Fig. 4 A, the absorbance signal decreased with the increase in PDA-PEG-LN NCs concentration. When the concentration was 0.05 mg/ml, almost 60 % O2· − could be removed. The absorbance did not continue to decrease at concentrations of PDA-PEG-LN NCs greater than 0.5 mg/ml, which may be due to the completely removal of O2· − in the solution. Terephthalic acid (TA) generated fluorescent 2-hydroxyterephthalic acid in the presence of hydroxyl radicals (·OH), which could be detected fluorescence signals around 425 nm [ 41 ]. As shown in Fig. 4 B, added PDA-PEG-LN NCs at varied concentrations in the solution significantly reduced the fluorescence values. Similar to the O2· − scavenging experiment, the fluorescence intensity also decreased with the increase in the concentration of PDA-PEG-LN NCs, indicating that these nanocomposites also had good clearance effects for ·OH. Fig. 4 Antioxidant and anti-inflammatory capabilities of PDA-PEG-LN NCs Scavenging efficiencies of PDA-PEG-LN NCs for (A) O2· − and (B) ·OH. (C) Uptake kinetics of ICG-labeled PDA-PEG-LN NCs by HESCs. ROS levels in HESCs treated with (D) Rosup or (E) IL-1β, in the absence or presence of PDA-PEG-LN NCs. Expression of inflammatory mediators (F) IL-6 and (G) TNF-α in IL-1β treated HESCs with the presence of PDA-PEG-LN NCs. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. Fig. 4 Antioxidant and anti-inflammatory capabilities of PDA-PEG-LN NCs Scavenging efficiencies of PDA-PEG-LN NCs for (A) O2· − and (B) ·OH. (C) Uptake kinetics of ICG-labeled PDA-PEG-LN NCs by HESCs. ROS levels in HESCs treated with (D) Rosup or (E) IL-1β, in the absence or presence of PDA-PEG-LN NCs. Expression of inflammatory mediators (F) IL-6 and (G) TNF-α in IL-1β treated HESCs with the presence of PDA-PEG-LN NCs. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. We further explored whether PDA-PEG-LN NCs have effects on intracellular ROS elevated by external stimuli. Prior to this, we verified the uptake of PDA-PEG-LN NCs by HESCs. We added ICG-labeled PDA-PEG-LN NCs to the culture medium of HESCs and detected the fluorescence signals within the cells at 1st, 2nd, 4th, 6th, 24th, and 48th hours using flow cytometry. As shown in Fig. 4 C, the fluorescence signals, indicating nanocomposite being uptake, were detectable after 1 h of incubation and intensified with prolonged incubation. This suggested that extended incubation durations led to enhanced internalization of nanocomposites. Next, we employed flow cytometry and dichlorofluorescein diacetate (DCFH-DA) probes to monitor changes of intracellular ROS concentrations. We allowed the HESC cells to fully take up PDA-PEG-LN NCs and added exogenous Rosup reagent to the culture medium to stimulate the generation of intracellular ROS. As shown in Fig. 4 D, the addition of PDA-PEG-LN NCs did not alter the ROS levels in normal cells, and the Rosup reagent significantly increased the ROS content in HESCs. Importantly, cells that had ingested PDA-PEG-LN NCs displayed a lower ROS elevation upon stimulation compared to the control cells.In endometriosis patients, especially those with decreased ovarian reserve function [ 25 ], the expression of inflammatory factors such as TNF-α, IL-6, and IL-1β was increased [ [42] , [43] , [44] ]. Inflammatory cytokines may further increase the ROS levels in turn, thereby leading to a vicious cycle of oxidative stress and chronic inflammation. We established an endometriosis inflammation model by incubating HESCs with IL-1β for 24 h [ 45 ]. We found that IL-1β also can increase intracellular ROS, and PDA-PEG-LN NCs were able to mitigate this ROS increase ( Fig. 4 E). IL-1β enhanced the secretion of inflammatory factors by HESCs. Concurrently, PDA-PEG-LN NCs effectively reduced the expression of IL-6 ( Fig. 4 F) and TNF-α ( Fig. 4 G). The above results indicated that PDA-PEG-LN NCs possessed effective ROS scavenging capabilities in both in vitro and intracellular environments. In addition, they can inhibit the activation and secretion of inflammatory mediators. This may be attributed to the abundant reductive functional groups within the PDA-PEG-LN NCs [ 34 , 46 , 47 ]. In endometriotic lesions, macrophages may be a key factor in maintaining the chronic inflammatory microenvironment. Macrophages are distinguished by their dual classic phenotypes: the pro-inflammatory M1 macrophages and the anti-inflammatory M2 macrophages [ 48 ]. Consequently, we investigated the potential impact of PDA-PEG-LN NCs on macrophage phenotype and the expression of related genes. We stimulated THP-1 cells with PMA to induce their transformation into adherent M0 macrophages. Subsequently, the M0 macrophages were induced towards the M1 phenotype using PMA, LPS and IFN-γ, and towards the M2 phenotype using PMA, IL-4, and IL-13. During the polarization of M1 and M2, PDA-PEG-LN NCs were added to the cultures to investigate their effects on macrophage polarization and the expression of associated inflammatory genes. First, we used flow cytometry and quantitative real-time polymerase chain reaction (qRT-PCR) to measure the expression levels of CD68 and CD14 in each group. In Fig. 5 A, the expressions of CD68 and CD14 in the induced cells were higher than in THP-1, confirming that these induced cells were macrophages. In Fig. 5 B, the positive rate and the fluorescence intensity of the surface marker CD86 on M1 macrophages, was significantly higher than the M0 and M2 groups. And the addition of PDA-PEG-LN NCs reduced the positive rate and fluorescence intensity. We further assessed the gene expression through qRT-PCR. There was a marked upregulation of key inflammatory genes in M1 macrophages, including TNF-α, IL-6, CXCL10, CD80, and CD86, when compared to their counterparts in M0 and M2. The addition of PDA-PEG-LN NCs also reduced the expression of these genes in M1 macrophages. In the course of M2 macrophage induction ( Fig. 5 C), the CD206 positive rate and fluorescence intensity were significantly elevated in M2 macrophages, which markedly surpassed M0 and M1 macrophages. As demonstrated by qRT-PCR, the expression of genes CCL22, CD206, and TGF-β in M2 macrophages was higher compared to M0 and M1 macrophages. M2 macrophages treated with PDA-PEG-LN NCs exhibited a notable upregulation of CD206 expression, as determined by both flow cytometry and qRT-PCR. The qRT-PCR results also indicated that PDA-PEG-LN NCs could upregulate the expression of the CCL22 gene in M2 macrophages, while the expression of TGF-β remained comparable to normally induced M2 macrophages. Therefore, we believe that these nanocomposites can effectively downregulate the expression of inflammatory genes in M1 macrophages, inhibiting the polarization of M0 to M1, and upregulate the expression of genes related to M2 macrophages, promoting the polarization of M0 to M2 macrophages. These results have validated our hypothesis that PDA-PEG-LN NCs may influence the polarization of macrophages. Fig. 5 The impact of nanocomposites on macrophage polarization and the expression of related genes (A) M0 marker expression in THP-1 cells was assessed following induction by PMA, PMA + IFN-γ+ LPS, PMA + IFN-γ+ LPS + NCs, PMA + IL-4+ IL-13, and PMA + IL-4+ IL-13+NCs. CD68 expression was detected using flow cytometry, while gene levels of CD68 and CD14 were assessed by qRT-PCR. (B) M1 marker expression in THP-1 cells was assessed following induction by PMA, PMA + IL-4+ IL-13, PMA + IFN-γ+ LPS, and PMA + IFN-γ+ LPS + NCs. CD86 expression was detected using flow cytometry, and gene levels of TNF-α, IL-6, CXCL10, CD80, and CD86 were assessed by qRT-PCR. (C) M2 marker expression in THP-1 cells was assessed following induction by PMA, PMA + IFN-γ+ LPS, PMA + IL-4+ IL-13, and PMA + IL-4+ IL-13+NCs. CD206 expression was detected using flow cytometry, and gene levels of CCL22, CD206, and TGF-β were assessed by qRT-PCR. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. Fig. 5 The impact of nanocomposites on macrophage polarization and the expression of related genes (A) M0 marker expression in THP-1 cells was assessed following induction by PMA, PMA + IFN-γ+ LPS, PMA + IFN-γ+ LPS + NCs, PMA + IL-4+ IL-13, and PMA + IL-4+ IL-13+NCs. CD68 expression was detected using flow cytometry, while gene levels of CD68 and CD14 were assessed by qRT-PCR. (B) M1 marker expression in THP-1 cells was assessed following induction by PMA, PMA + IL-4+ IL-13, PMA + IFN-γ+ LPS, and PMA + IFN-γ+ LPS + NCs. CD86 expression was detected using flow cytometry, and gene levels of TNF-α, IL-6, CXCL10, CD80, and CD86 were assessed by qRT-PCR. (C) M2 marker expression in THP-1 cells was assessed following induction by PMA, PMA + IFN-γ+ LPS, PMA + IL-4+ IL-13, and PMA + IL-4+ IL-13+NCs. CD206 expression was detected using flow cytometry, and gene levels of CCL22, CD206, and TGF-β were assessed by qRT-PCR. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. There was an elevated presence of M1 macrophages, and both M1 and M2 macrophages displayed an enhanced proinflammatory phenotype, in the endometrium of individuals with endometriosis [ 49 , 50 ]. In endometriotic lesions, the increase in inflammatory cytokines, such as TNF-α, IL-6, and IL-1β [ [51] , [52] , [53] ], which promoted endometriosis, also recruited macrophages. These macrophages released various inflammatory substances attracting more macrophages, and contributed to a self-sustaining cycle [ 54 ]. Therefore, we proposed a hypothesis that treatment of endometriosis with these nanocomposites may break the vicious cycle of chronic inflammation, thereby preventing the initiation and progression of the disease. Endometriosis is an estrogen-dependent inflammatory disease. Dydrogesterone, acting as a progesterone, can exert its effect at the lesion site, reducing the hormone-dependent proliferation and inducing the atrophy of ectopic endometrium through promoting apoptotic pathways [ 55 ]. The reduction in ROS and inflammation, through a synergistic effect, may enhance the efficacy of dydrogesterone, potentially allowing for a lower dosage and mitigating its side effects [ [56] , [57] , [58] , [59] ]. First, we co-cultured HESCs with dydrogesterone or DYD@PDA-PEG-LN NCs to assess the relative viability of HESCs under different concentrations of the drug using MTT assay. In Fig. 6 A, when the concentrations of dydrogesterone were 0.1 μg/ml and 1 μg/ml, the relative viability of the cells significantly decreased, and the effect of DYD@PDA-PEG-LN NCs on HESCs was more pronounced than that of an equivalent dose of dydrogesterone alone. Subsequently, we used the EdU Cell Proliferation Kit with Alexa Fluor 555 and flow cytometry, to assess the proportion of proliferating cells in each group. In Fig. 6 B, it could be observed that the use of PDA-PEG-LN NCs did not affect cell proliferation, while dydrogesterone and DYD@PDA-PEG-LN NCs reduced the proportion of proliferating cells. However, the effect of DYD@PDA-PEG-LN NCs was significantly better than dydrogesterone. Next, we detected the proportion of apoptotic cells using the FITC-Annexin V/PI Apoptosis Detection Kit and flow cytometry. During the early stage of apoptosis, the cells were marked by FITC-Annexin V, thereby emitting green fluorescence. The nuclei of necrotic cells or cells in the late stage of apoptosis were bound by Propidium Iodide (PI), exhibiting red fluorescence. As shown in Fig. 6 C, free dydrogesterone nanocomposites did not induce an increase in cell apoptosis. Dydrogesterone as well as DYD@PDA-PEG-LN NCs, increased the proportion of apoptosis cells and necrotic cells. Notably, DYD@PDA-PEG-LN NCs led to a higher number of apoptosis cells in both early and late stages. We further used the Mitochondrial Membrane Potential Assay Kit with JC-1 and the fluorescence microscope to detect the reduction in the mitochondrial membrane potential, a hallmark of the early stages of apoptosis. When the mitochondrial membrane potential was low, JC-1 existed as JC-1 monomers, producing green fluorescence; when it was high, JC-1 formed JC-1 aggregates that emitted red fluorescence. In Fig. 6 D, both dydrogesterone and DYD@PDA-PEG-LN NCs diminished the mitochondrial membrane potential. DYD@PDA-PEG-LN NCs were particularly effective in promoting early-stage apoptosis, corresponding to the results in Fig. 6 C. We also assessed the impact on cell migration using a scratch assay ( Fig. 6 E). The scratch width changes at 24 and 48 h, revealed that both dydrogesterone and DYD@PDA-PEG-LN NCs could inhibit the migration, with the nanocomposites showing greater inhibitory effects. In summary, DYD@PDA-PEG-LN NCs demonstrated superior efficacy in inducing apoptosis as well as suppressing both cellular proliferation and migration, outperforming dydrogesterone. Fig. 6 The effects of DYD@PDA-PEG-LN NCs on HESC cells (A) Relative viability of HESC cells following incubation with DYD@PDA-PEG-LN NCs or dydrogesterone. (B) Proportion of proliferating HESCs measured following the addition of PDA-PEG-LN NCs, dydrogesterone, or DYD@PDA-PEG-LN NCs. (C) Apoptosis in HESCs after incubation with PDA-PEG-LN NCs, dydrogesterone, or DYD@PDA-PEG-LN NCs, as determined by Annexin Ⅴ and PI staining. (D) Fluorescence images of JC-1 indicating HESCs' mitochondrial membrane potential following treatment with PDA-PEG-LN NCs, dydrogesterone, or DYD@PDA-PEG-LN NCs. Scale bar: 100 μm. (E) Scratch width measurements of HESCs following treatment with PDA-PEG-LN NCs, dydrogesterone, and DYD@PDA-PEG-LN NCs. Scale bar: 200 μm. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. Fig. 6 The effects of DYD@PDA-PEG-LN NCs on HESC cells (A) Relative viability of HESC cells following incubation with DYD@PDA-PEG-LN NCs or dydrogesterone. (B) Proportion of proliferating HESCs measured following the addition of PDA-PEG-LN NCs, dydrogesterone, or DYD@PDA-PEG-LN NCs. (C) Apoptosis in HESCs after incubation with PDA-PEG-LN NCs, dydrogesterone, or DYD@PDA-PEG-LN NCs, as determined by Annexin Ⅴ and PI staining. (D) Fluorescence images of JC-1 indicating HESCs' mitochondrial membrane potential following treatment with PDA-PEG-LN NCs, dydrogesterone, or DYD@PDA-PEG-LN NCs. Scale bar: 100 μm. (E) Scratch width measurements of HESCs following treatment with PDA-PEG-LN NCs, dydrogesterone, and DYD@PDA-PEG-LN NCs. Scale bar: 200 μm. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. When applied to endometriosis models, antioxidants may not only exhibit antioxidant and anti-inflammatory properties but also suppress the proliferative activity of ectopic endometrial cells [ 60 , 61 ], and simultaneously they were conducive to promoting cell apoptosis [ 62 , 63 ]. In typical endometrial cycles, the concentration of apoptotic proteins peaked in the late luteal phase. But in individuals with endometriosis, the expected increase was notably absent. This may be due to the activation of cellular protective enzymes under oxidative stress, which in turn impeded the expression of proteins crucial for apoptosis [ 64 , 65 ]. In our research, we observed that while PDA-PEG-LN NCs, on their own, did not exert a direct cytotoxic effect on HESCs, DYD@PDA-PEG-LN NCs significantly augmented the dydrogesterone's capacity to suppress HESC proliferation and migration, as well as to promote apoptosis. We hypothesized that this synergistic enhancement stemmed from the inherent antioxidant and anti-inflammatory properties of the PDA-PEG-LN NCs, which may amplify the therapeutic efficacy of DYD on HESCs by modulating its mechanism of action. Overall, we proposed a potential strategy for improving the treatment of endometriosis, by leveraging the complementary actions of nanocarriers and therapeutic agents. Through the aforementioned in vitro experiments, we demonstrated the effects of DYD@PDA-PEG-LN NCs on HESC cells. Then we further validated their therapeutic efficacy in vivo. As shown in Fig. 7 A, to establish a mouse model of endometriosis, we transplanted human ectopic endometrial tissue into immunodeficient mice, with one graft per mouse, as previously described [ 20 , [66] , [67] , [68] , [69] ]. On the 7th and 28th days post-surgery, the grafts were removed and subjected to immunohistochemical (IHC) staining. In Fig. 7 B and C, both estrogen receptor 1 (ESR1) and progesterone receptor (PGR) stained positively, demonstrating the grafts' reactivity to estrogen and progesterone. Positive KI-67 staining proved the grafts’ mitotic activity. High VEGF levels indicated active angiogenesis and increased capillary bed permeability, potentially enhancing macrophage recruitment and nanocomposite penetration [ 66 ]. Then, we used this model to evaluate the distribution of DYD@PDA-PEG-LN NCs in vivo, thereby verifying their therapeutic effects on endometriotic lesions. Fig. 7 The successful preparation of the endometriosis mice and the distribution of nanocomposites in mice (A) Schematic illustration of endometriosis model. IHC staining of ESR1, PGR, KI-67, and VEGF in the grafts on (B) the 7th and (C) the 28th days post-surgery. In vivo fluorescence imaging of endometriosis mice after (D, E) subcutaneous or (F, G) intravenous injection with ICG-labeled DYD@PDA-PEG-LN or DYD@PDA nanocomposites. (H, I) In vivo fluorescence imaging of sham-operated mice after subcutaneous or intravenous injection with ICG-labeled DYD@PDA-PEG-LN NCs. Fig. 7 The successful preparation of the endometriosis mice and the distribution of nanocomposites in mice (A) Schematic illustration of endometriosis model. IHC staining of ESR1, PGR, KI-67, and VEGF in the grafts on (B) the 7th and (C) the 28th days post-surgery. In vivo fluorescence imaging of endometriosis mice after (D, E) subcutaneous or (F, G) intravenous injection with ICG-labeled DYD@PDA-PEG-LN or DYD@PDA nanocomposites. (H, I) In vivo fluorescence imaging of sham-operated mice after subcutaneous or intravenous injection with ICG-labeled DYD@PDA-PEG-LN NCs. As shown in the schematic illustration of Fig. 7 D, on the 7th day after transplantation, we administered DYD@PDA-PEG-LN NCs with ICG fluorescence locally beneath the skin of the endometriosis mice's grafts, at a dose of 10 mg/kg body weight. To verify whether the PEG and LN modifications would affect the nanocomposites' distribution, we also injected ICG-labeled DYD@PDA NCs into endometriosis mice at the equivalent dose. We used an in vivo imaging system to record the fluorescence signals within the mice at 1st, 2nd, 4th, 6th, 24th, and 48th hours post-injection. As shown in Fig. 7 E, at the 1st, 2nd, and 4th hours, strong fluorescence signals were observable on the graft sites. At the 6th hour, the fluorescence signals of the DYD@PDA-ICG NCs group were slightly lower than that of the DYD@PDA-PEG-LN-ICG NCs group. Furthermore, at the 24th and 48th hours, the fluorescence signals of the DYD@PDA-ICG NCs group were significantly lower. This phenomenon could be attributed to the PEG and LN modifications, which allowed the nanocomposites to better evade detection by the immune system, consequently resulting in a prolonged half-life within the biological environment. Next, we administered the two nanocomposites into endometriosis mice, via tail vein injection at the same dosage, as shown in Fig. 7 F. In Fig. 7 G, both DYD@PDA-PEG-LN-ICG NCs and DYD@PDA-ICG NCs were able to accumulate at the lesion sites. Notably, at the 2nd and 4th hours, fluorescence signals could be observed at the endometriotic lesions in both groups, albeit with lower signals in the DYD@PDA-ICG NCs group. Over time, both groups exhibited a decrease in fluorescence signals at the lesion sites, yet the signals of DYD@PDA-PEG-LN-ICG NCs remained stronger at 6th, 24th, and 48th hours. The fluorescence signals in DYD@PDA-ICG NCs group were obviously diminished, with the signals being almost undetectable at the 48-h mark. The lesion sites featured an abundance of highly permeable neovasculature and a notable accumulation of macrophages [ 66 , 70 ]. Additionally, the PDA nanocomposites possessed certain adhesive properties [ 46 ]. These factors likely contributed to the nanocomposites' retention in the endometriosis tissues. The superior targeting capability of DYD@PDA-PEG-LN NCs may be related to the LN modification, which enhanced the nanocomposites' binding affinity to the upregulated ANTXR2 receptor on ectopic endometrial cells [ 19 , 29 ]. As shown in Fig. 7 H, to exclude the impact of surgical procedures, a sham-operated group was established by performing incisions and sutures in the same area, without the implantation of ectopic endometrium. Similarly, on the 7th day post-surgery, subcutaneous injections to the incision site or tail vein injections of DYD@PDA-PEG-LN-ICG NCs were administered. At the 1st and 2 nd h after subcutaneous injections, the distribution of DYD@PDA-PEG-LN-ICG NCs in the sham group was broader ( Fig. 7 I), while in the endometriosis mice, the nanocomposites were more concentrated around the endometrial implants ( Fig. 7 E). Starting from the 4th hour, the fluorescence signals at the incision sites in the sham group ( Fig. 7 I), were consistently lower than that in the endometriosis group ( Fig. 7 E). At the 24th and 48th hours, the fluorescence signals in the sham group were almost undetectable. However, in the sham group following intravenous administration of the nanocomposites, no nanocomposites' aggregation at the incision sites was observed from 1 to 48 h post-injection ( Fig. 7 I). This phenomenon suggested that the primary cause of nanocomposites’ retention following local injection was the presence of transplanted endometrial tissues. The sham group mice also demonstrated retention of nanocomposites at the wound site, likely due to internalization by immune cells recruited to the wound, increased permeability of the neovasculature during the healing process [ 71 , 72 ], and the adhesive properties of the nanocomposites [ 46 ]. In other words, while the surgical procedures and the adhesive properties of the nanocomposites might influence retention at the wound site following subcutaneous injection, the presence of implanted endometrial tissues was the primary factor. Furthermore, the surgical procedures alone, in the absence of grafts, did not result in the targeted aggregation of DYD@PDA-PEG-LN-ICG NCs at the wound site following intravenous injection. This observation confirmed the excellent targeting ability of DYD@PDA-PEG-LN NCs for ectopic endometrium in the endometriosis model mice. The aforementioned experiments have demonstrated that DYD@PDA-PEG-LN NCs not only have good therapeutic effects in vitro but also exhibit good targeting effects in vivo. Next, we evaluated the therapeutic effects of DYD@PDA-PEG-LN NCs on ectopic endometrial lesions in endometriosis model mice after 21 days of treatment. We administrated PBS, PDA-PEG-LN NCs, dydrogesterone, and DYD@PDA-PEG-LN NCs into endometriosis mice via subcutaneous or intravenous injections. The treatment regimen began on the seventh day following the transplantation and continued for a 21-day period, ending on the 28th day post-surgery. We documented the variations in the area of subcutaneous grafts among mice in all groups. Upon the treatment's conclusion, the grafts were excised and weighed. In Fig. 8 A–E, we presented the grafts before and after treatment, along with a comparison of the area and weight of the grafts among the four groups. Notably, treatment with dydrogesterone or DYD@PDA-PEG-LN NCs, resulted in significant decreases in both graft area and weigh, compared to the control group. Despite equal amounts of dydrogesterone in both groups, the nanocomposite group exhibited a substantially greater reduction, with no recurrence in the 8-week study ( Fig. 8 F). The similar outcomes following treatment via tail vein injection were illustrated in Fig. 8 G–L. This implied that DYD@PDA-PEG-LN NCs might amplify the therapeutic efficacy of dydrogesterone or induce graft atrophy through distinct mechanisms, thereby contributing to a more pronounced reduction in graft size and weight. Additionally, the standard dosage of dydrogesterone for treating endometriosis was around 0.5 mg/kg/day, whereas the dydrogesterone in the DYD@PDA-PEG-LN NCs group was approximately 0.05 mg/kg/day. Treating endometriosis with DYD@PDA-PEG-LN NCs might allow for a reduced dydrogesterone dosage while still achieving improved therapeutic effects. Furthermore, this could potentially lead to a reduced incidence of side effects. Fig. 8 The therapeutic effects of DYD@PDA-PEG-LN NCs on endometriosis lesions (A–E) Comparisons of graft area and weight following subcutaneous injection of PBS, PDA-PEG-LN NCs, dydrogesterone, and DYD@PDA-PEG-LN NCs. (A) Treatment plan for subcutaneous injection. (B) Grafts before implantation into the mice. (C) Grafts excised from mice after a 21-day subcutaneous injection. (D) Area of subcutaneous grafts in each group assessed every 4 days. (E) Weight of the excised grafts following the 21-day injection in each group. (F) Area of subcutaneous grafts in each group after subcutaneous injection discontinuation. (G–K) Comparisons of graft area and weight following intravenous injection of PBS, PDA-PEG-LN NCs, dydrogesterone, and DYD@PDA-PEG-LN NCs. (G) Treatment plan for intravenous injection. (H) Grafts before implantation into mice. (I) Grafts excised from mice after a 21-day intravenous injection. (J) Area of subcutaneous grafts in each group assessed every 4 days. (K) Weight of the excised grafts following the 21-day injection in each group. (L) Area of subcutaneous grafts in each group after intravenous injection discontinuation. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. Fig. 8 The therapeutic effects of DYD@PDA-PEG-LN NCs on endometriosis lesions (A–E) Comparisons of graft area and weight following subcutaneous injection of PBS, PDA-PEG-LN NCs, dydrogesterone, and DYD@PDA-PEG-LN NCs. (A) Treatment plan for subcutaneous injection. (B) Grafts before implantation into the mice. (C) Grafts excised from mice after a 21-day subcutaneous injection. (D) Area of subcutaneous grafts in each group assessed every 4 days. (E) Weight of the excised grafts following the 21-day injection in each group. (F) Area of subcutaneous grafts in each group after subcutaneous injection discontinuation. (G–K) Comparisons of graft area and weight following intravenous injection of PBS, PDA-PEG-LN NCs, dydrogesterone, and DYD@PDA-PEG-LN NCs. (G) Treatment plan for intravenous injection. (H) Grafts before implantation into mice. (I) Grafts excised from mice after a 21-day intravenous injection. (J) Area of subcutaneous grafts in each group assessed every 4 days. (K) Weight of the excised grafts following the 21-day injection in each group. (L) Area of subcutaneous grafts in each group after intravenous injection discontinuation. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. IHC staining for ESR1, PGR, KI-67, and VEGF was conducted on the grafts from the aforementioned group. We observed that the ectopic endometrial grafts treated with DYD@PDA-PEG-LN NCs showed a more significant decrease in ESR1 expression compared to other groups, regardless of whether the treatment was administrated via subcutaneous ( Fig. 9 A) or intravenous ( Fig. 9 B) injection. The reduction in steroid hormone receptors implied a diminished hormonal responsiveness and an impaired ability to grow in response to estrogen. Additionally, the decrease expression of KI-67 and VEGF also indicated a reduction in the proliferative capacity and angiogenic activity of the lesion tissues post-treatment. The reduction in the size of endometriosis grafts, as depicted in Fig. 8 , further confirmed these findings. Moreover, both subcutaneous and intravenous injection of DYD@PDA-PEG-LN NCs demonstrated therapeutic efficacy for the grafts, further proving their ability to effectively aggregate and remain within the lesion tissues in mice. Fig. 9 The immunohistochemical staining and their statistical analysis results of each group IHC staining of ESR1, PGR, KI-67, and VEGF in the grafts after (A) subcutaneous or (B) intravenous injection of PBS, PDA-PEG-LN NCs, dydrogesterone, as well as DYD@PDA-PEG-LN NCs. (C–F) The relative positive rate of (C) ESR1, (D) PGR, (E) KI-67, and (F) VEGF in grafts following subcutaneous injection. (G–J) The relative positive rate of (G) ESR1, (H) PGR, (I) KI-67, and (J) VEGF in grafts following intravenous injection. Scale bar: 100 μm. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. Fig. 9 The immunohistochemical staining and their statistical analysis results of each group IHC staining of ESR1, PGR, KI-67, and VEGF in the grafts after (A) subcutaneous or (B) intravenous injection of PBS, PDA-PEG-LN NCs, dydrogesterone, as well as DYD@PDA-PEG-LN NCs. (C–F) The relative positive rate of (C) ESR1, (D) PGR, (E) KI-67, and (F) VEGF in grafts following subcutaneous injection. (G–J) The relative positive rate of (G) ESR1, (H) PGR, (I) KI-67, and (J) VEGF in grafts following intravenous injection. Scale bar: 100 μm. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. Through in vitro experiments, we have found that PDA-PEG-LN NCs can effectively scavenge ROS, reduce the expression of inflammatory factors, and inhibit the polarization of M1 macrophages while promoting M2 macrophages. Therefore, we speculated that DYD@PDA-PEG-LN NCs may have a similar mechanism when treating endometriosis in vivo. To evaluate oxidative stress in endometriotic lesions, we employed the ferric-orange xylenol (FOX) system to measure hydroperoxide levels in the grafts, which indirectly reflected the ROS changes at the lesions. As shown in Fig. 10 A, both subcutaneous and intravenous injections of PDA-PEG-LN NCs or DYD@PDA-PEG-LN NCs significantly reduced hydroperoxide content compared to the PBS group. However, no significant reduction was observed in the dydrogesterone group. These results suggested that both PDA-PEG-LN NCs and DYD@PDA-PEG-LN NCs could ameliorate oxidative stress in endometriotic lesions. Subsequently, we collected serum from normal mice (non-operated) and model mice post-treatment. Then we measured serum levels of inflammatory cytokines TNF-α, IL-1β, and IL-6, as depicted in Fig. 10 B, C, and 10D, respectively. Treatment with PDA-PEG-LN NCs, dydrogesterone, and DYD@PDA-PEG-LN NCs all led to a reduction in inflammatory cytokine levels compared to the PBS group, with DYD@PDA-PEG-LN NCs showing the greatest efficacy. To further explore the treatment mechanism, we conducted immunofluorescence (IF) staining on the grafts from each group. The results for the ectopic endometrium following 21 days of subcutaneous injections were displayed in Fig. 10 E, while those following intravenous injections were shown in Fig. 10 F. We observed that the ectopic endometrium treated with PDA-PEG-LN NCs or DYD@PDA-PEG-LN NCs, regardless of the injection method, exhibited decreased expression of iNOS and increased expression of CD206. This suggested a shift from pro-inflammatory M1 macrophages to anti-inflammatory M2 macrophages in the treated tissues. In contrast, the ectopic endometrium treated with only PBS or dydrogesterone showed no significant changes in M1 and M2 macrophage markers. These findings were consistent with our in vitro experiments, and suggested that PDA-PEG-LN NCs may modulate macrophage phenotypes within the grafts. The observed changes in macrophage phenotype could also contribute to the observed decrease in inflammatory factor levels in mice. Fig. 10 Changes in the levels of hydroperoxide, TNF-α, IL-1β, IL-6 and the proportion of M1 and M2 macrophages phenotypes in grafts of each group (A) Hydroperoxide content in the grafts after subcutaneous or intravenous injection of PBS, PDA-PEG-LN NCs, dydrogesterone, or DYD@PDA-PEG-LN NCs. Serum levels of inflammatory cytokines, including (B) TNF-α, (C) IL-1β, and (D) IL-6, after treatments. IF staining for F4/80, iNOS, and CD206 in the grafts after (E) subcutaneous or (F) intravenous injection treatments. Scale bar: 100 μm. (G–I) The relative integrated fluorescence signal of (G) F4/80, (H) iNOS, and (I) CD206 in grafts following subcutaneous injection. (J–L) The relative integrated fluorescence signal of (J) F4/80, (K) iNOS, and (L) CD206 in grafts following intravenous injection. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. Fig. 10 Changes in the levels of hydroperoxide, TNF-α, IL-1β, IL-6 and the proportion of M1 and M2 macrophages phenotypes in grafts of each group (A) Hydroperoxide content in the grafts after subcutaneous or intravenous injection of PBS, PDA-PEG-LN NCs, dydrogesterone, or DYD@PDA-PEG-LN NCs. Serum levels of inflammatory cytokines, including (B) TNF-α, (C) IL-1β, and (D) IL-6, after treatments. IF staining for F4/80, iNOS, and CD206 in the grafts after (E) subcutaneous or (F) intravenous injection treatments. Scale bar: 100 μm. (G–I) The relative integrated fluorescence signal of (G) F4/80, (H) iNOS, and (I) CD206 in grafts following subcutaneous injection. (J–L) The relative integrated fluorescence signal of (J) F4/80, (K) iNOS, and (L) CD206 in grafts following intravenous injection. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. Progesterone can treat endometriosis by downregulating estrogen receptor expression, and dydrogesterone had a similar mechanism as shown in the IHC results above. Progesterone also possesses anti-inflammatory effects in the normal uterus, and the decline in its levels during non-pregnant states can lead to an increase in local pro-inflammatory cytokines and chemokines. Additionally, elevated pro-inflammatory factors within endometriotic lesions can reduce PGR expression and weaken the progesterone's effect on these lesions. Persistent chronic inflammation may also lead to permanent progesterone resistance, which has been found in both the ectopic and normal endometrium of endometriosis patients [ 73 ]. Approximately one-third symptomatic endometriosis patients failed to respond to progesterone therapy, and inflammation and oxidative stress might play a key role in the progesterone resistance of ectopic endometrium [ 74 , 75 ]. We found that dydrogesterone could downregulate ESR1 and inflammation levels, but DYD@PDA-PEG-LN NCs demonstrated superior therapeutic effects. This may be due to the nanocomposites' ability to alleviate oxidative stress and break the vicious cycle of local inflammation, thereby reducing progesterone resistance and enhancing the therapeutic efficacy of dydrogesterone. In the normal endometrium, macrophages were predominantly of the M2 type [ 76 ]. In the endometrium of women with endometriosis, the proportion of M2 macrophages was reduced, and there was a shift towards the M1 phenotype [ 49 ]. Furthermore, M1 macrophages in the endometrium of individuals with endometriosis displayed an enhanced pro-inflammatory phenotype. The environment in the lesions may promote the polarization of M2 macrophages towards the M1 type [ 50 ], contributing to the chronic inflammation associated with the disease. Although endometriosis shared several characteristics with malignant tumors, and the presence of M2 macrophages in the lesions was considered to facilitate the occurrence and progression of endometriosis [ [77] , [78] , [79] ]. In our study, DYD@PDA-PEG-LN NCs shifted the macrophage phenotype in the ectopic endometrium closer to that of normal women's endometrium, demonstrating excellent therapeutic efficacy on endometriosis. Both in vitro and in vivo experimental results have shown that DYD@PDA-PEG-LN NCs can achieve better therapeutic effects compared to the use of dydrogesterone alone. These superior therapeutic effects also implied the potential for lower treatment dosages and reduced incidences of adverse reactions in clinical therapy. Furthermore, DYD@PDA-PEG-LN NCs may also improve patient compliance during clinical treatment. Existing treatment methods may affect the natural conception process or damage normal ovarian tissue, which may lead to a delayed age of fertility and contribute to a further decline in ovarian function. Dydrogesterone can be used by women who are pregnant or trying to conceive, and the standard therapeutic dose does not suppress ovulation. Therefore, we further assessed whether DYD@PDA-PEG-LN NCs would impair the ovarian function and fertility of mice, and explored their potential teratogenic effects on mouse embryos, as shown in the schematic illustration of Fig. 11 A. Fig. 11 The impact of DYD@PDA-PEG-LN NCs on ovarian function and fertility (A) Schematic illustration of experimental plan. (B) H&E staining images of ovaries from mice following subcutaneous or intravenous injection of DYD@PDA-PEG-LN NCs. Scale bar: 200 μm. (C) The number of follicles and (D) serum AMH levels were assessed in these mice. Reproductive outcomes were assessed, including (E) the cohabitation time required for conception and (F, G) the number of embryos after a 21-day treatment. (H) H&E staining images of embryos. Scale bar: 2 mm. (I) H&E staining images of placentas. Scale bar: 100 μm. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. Fig. 11 The impact of DYD@PDA-PEG-LN NCs on ovarian function and fertility (A) Schematic illustration of experimental plan. (B) H&E staining images of ovaries from mice following subcutaneous or intravenous injection of DYD@PDA-PEG-LN NCs. Scale bar: 200 μm. (C) The number of follicles and (D) serum AMH levels were assessed in these mice. Reproductive outcomes were assessed, including (E) the cohabitation time required for conception and (F, G) the number of embryos after a 21-day treatment. (H) H&E staining images of embryos. Scale bar: 2 mm. (I) H&E staining images of placentas. Scale bar: 100 μm. Not significant (ns): p > 0.05, ∗: p < 0.05, ∗∗: p < 0.01, ∗∗∗: p < 0.001. Serum AMH levels are commonly used to reflect ovarian reserve function. We collected ovaries and serum from normal mice and those treated with DYD@PDA-PEG-LN NCs via subcutaneous and tail vein injections at a dosage of 10 mg/kg body weight. Over a period of 21 days, with injections administered every two days, we observed changes in the ovarian structure of the mice using H&E staining and measured their AMH levels. As shown in Fig. 11 B, no thickening of the ovarian cortex or structural disarray was observed in the mice across all groups. The number of follicles was not significantly different from that of the ovaries in the control group mice ( Fig. 11 C), and no apparent signs of ovarian atrophy were found. At the same time, there were no significant differences in AMH levels among the mice in all groups ( Fig. 11 D), which meant that the 21-day treatment with DYD@PDA-PEG-LN NCs did not adversely affect ovarian reserve function. Consequently, we further verified whether the use of DYD@PDA-PEG-LN NCs would affect the fertility of the mice. We co-housed female mice that had received DYD@PDA-PEG-LN NC injections and control female mice with male mice until pregnancy was confirmed in the females. On the 14.5th day of gestation, the pregnant mice were euthanized to accurately count the number of embryos in each group. The embryos and placentas from the mice in each group were collected and subjected to H&E staining. As shown in Fig. 11 E, F and 11G, mice in all groups successfully conceived without any miscarriages, and there were no significant differences in the cohabitation time or the number of embryos, compared to the control group. Therefore, we believe that although PDA-PEG-LN NCs were employed to load with dydrogesterone, DYD@PDA-PEG-LN NCs did not affect the natural conception process or the number of embryos in mice. Considering that exposure to exogenous nanocomposites may be associated with embryonic malformations or damage to the placental barrier in pregnant mice, we also conducted pathological histological examinations on the embryos and placentas from mice in all groups. In Fig. 11 H, pregnant mice that had received DYD@PDA-PEG-LN NCs injections did not exhibit significant morphological abnormalities in their embryos compared to the control group. No abnormal manifestations were observed in the placental decidua, trophoblast, or spiral arteries ( Fig. 11 I). This demonstrated that DYD@PDA-PEG-LN NCs not only had good biocompatibility within mice but also did not damage ovarian function or fertility, showing no significant reproductive toxicity. We believe that this approach will offer a new perspective for young women with endometriosis, balancing the therapeutic effects of endometriosis with fertility needs.

We dissolved 0.2g of dopamine hydrochloride (Sigma-Aldrich) in 200 ml of ultrapure water and adjusted the solution pH to 9 with 210 μl of 5M NaOH solution. Under conditions of a 60 °C heating in oil bath, we used a magnetic stirrer to stir the mixture at 350 rpm to facilitate oxidative self-polymerization. After stirring for 6 h, Polydopamine nanocomposites (PDA NCs) were obtained and washed three times with deionized water to remove other ions and unpolymerized dopamine monomers. The PDA NCs were collected by centrifugation at a speed of 18500 rpm (Thermo Fisher Scientific) and then dried using a freeze dryer. The PDA NCs were mixed with mPEG–NH 2 at a ratio of 1:2 and stirred overnight. The mixture was then washed with ultrapure water, and PDA-PEG NCs were collected by centrifugation. Subsequently, the laminin (LN) solution (50 μg/ml) was mixed with an equal volume of PDA-PEG NCs (0.5 mg/ml) and stirred at 37 °C for 3 h. The PDA-PEG-LN NCs were collected after washing with ultrapure water and centrifugation. At room temperature, 10 mg of PDA NCs were mixed and incubated with 0.25 mg of dydrogesterone (Aladdin) in 2 ml of methanol using magnetic stirring for 24 h. To assess the amount of loaded drug, the absorbance of the obtained particles and supernatant after centrifugation at 295 nm was measured using a microplate reader. The amount of DYD loaded onto the PDA NCs and the concentration of remaining DYD in the supernatant were measured through a standard curve. This standard curve was obtained by diluting a DYD solution in methanol (0.25 mg/ml) to various concentrations ranging from 250 μg/ml to 0.9766 μg/ml, with the R 2 correlation coefficient of 0.99989. The resulting DYD@PDA NCs were washed three times with ultrapure water and then sequentially loaded with mPEG–NH 2 and laminin using the aforementioned methods. The DYD@PDA-PEG-LN NCs were washed three times with ultrapure water, freeze-dried, and then stored at −80 °C. To synthesis ICG-labeled DYD@PDA-PEG-LN NCs and DYD@PDA NCs, ICG (1 mg/ml) was mixed with DYD@PDA-PEG-LN NCs or DYD@PDA NCs (1 mg/ml) at a volume ratio of 1:5, stirred, and incubated overnight in the dark. After being washed three times with ultrapure water and centrifuged to collect, they were freeze-dried and stored at −80 °C. Scanning electron microscopy (SEM, ZEISS Sigma 300) and transmission electron microscopy (TEM, JEOL JEM-F200) were utilized to characterize the morphology of NCs at various stages. A Malvern Zetasizer instrument was employed to measure the hydrodynamic size distribution and zeta potentials of NCs at different stages. Human endometrial stromal cells (HESCs) were obtained from Procell and cultured in DMEM medium (Servicebio) containing 10 % FBS and 1 % penicillin-streptomycin, under a constant temperature and humidity environment (37 °C, 5 % CO 2 ). Cytotoxicity Studies : HESCs were cultured in a 96-well plate at a density of 10^4 cells per well and incubated for 6 h. PDA-PEG-LN NCs at varying concentrations (0, 0.01, 0.025, 0.05, 0.075, 0.1, 0.25, 0.5, 0.75, 1 mg/ml) were added, and after incubation for 24 or 48 h, the cells were treated with an MTT assay kit (Beyotime Biotechnology). The absorbance at 570 nm were measured using a microplate reader to reflect the cytotoxicity of PDA-PEG-LN NCs at different concentrations (n = 3). Cellular Viability Observation : After culturing HESCs in a 6-well plate for 12 h, different concentrations of NCs were added for co-incubation for 24 h. Following incubation, the cells were washed with PBS solution and then stained with Calcein AM and PI (Beyotime Biotechnology) to distinguish live and dead cells, respectively. Fluorescence images were collected using a fluorescence microscope (Leica). Release of Lactate Dehydrogenase (LDH) : To measure the level of LDH released by cells co-incubated with NCs, cells not treated with NCs were lysed, centrifuged to remove the pellet, and the supernatant was collected. Supernatants from cells after 24 or 48 h of incubation with NCs were also collected, centrifuged to remove the pellet, and the total LDH levels inside the cells as well as the released LDH concentration in the supernatant after co-culture were measured using an LDH detection assay kit (n = 3). ATP Level Determination (Beyotime Biotechnology) : HESCs were co-incubated with NCs at various concentrations for 24 or 48 h. After washing the cells, they were lysed and ATP detection working solution was added. The luminescence values of each group were measured using a multifunctional microplate reader, reflecting the ATP levels of each group (n = 3). Hemolysis Assay : Erythrocytes (RBCs) were separated from serum and diluted with saline to four times their original volume. A mixture of 0.2 ml of the diluted RBCs and 0.8 ml of 0.9 % NaCl solution served as the negative control group, while a mixture of 0.2 ml of the diluted RBCs and 0.8 ml of water served as the positive control group. Subsequently, the RBCs were co-incubated with 0.9 % NaCl solution containing different concentrations of NCs, and the absorbance at 541 nm of each group was measured using a multifunctional microplate reader after 3 h (n = 4). Mice were administered DYD@PDA-PEG-LN NCs via subcutaneous or intravenous injections at a dosage of 10 mg/kg body weight every two days, and their body weight was recorded (n = 6). After 21 days, comparisons were made between the subcutaneously injected group, the tail vein injected group, and the control group to demonstrate the biocompatibility of DYD@PDA-PEG-LN NCs in vivo. Simultaneously, we collected serum from each group and determined the levels of ALT, AST, and ALP using a fully automatic biochemical analyzer (n = 6). We then euthanized the mice, removed their main organs, namely the heart, liver, spleen, lungs, and kidneys, fixed them in 4 % paraformaldehyde, embedded them in paraffin, sliced them, and performed H&E staining (n = 6). The scavenging efficiency for O2·− was reflected by the photoreduction inhibition degree of NBT by NCs. In a 25 mM PBS solution (pH 7.4), riboflavin (20 μM), methionine (12.5 mM), and NBT (75 μM) were added, followed by the addition of different concentrations of NCs (0, 0.01, 0.025, 0.05, 0.075, 0.1, 0.25, 0.5, 0.75, 1 mg/ml). After exposure to ultraviolet light for 15 min, the absorbance changes at 600 nm for all groups’ supernatant, including a blank control without NCs and UV irradiation, were measured to reflect the scavenging efficiency of the NCs for O2·− (n = 4). The scavenging efficiency of NCs for ·OH was indicated by the reduction in 2-hydroxyterephthalic acid concentration. In a 25 mM phosphate-buffered saline (PBS) solution (pH 7.4), terephthalic acid (0.5 mM, dissolved in 2 mM NaOH), H2O2 (10 mM), and NCs at various concentrations were added. After incubating in the dark for 18 h, the fluorescence signal of 2-hydroxyterephthalic acid (Ex: 320 nm, Em: 425 nm) was measured to reflect the concentration changes of ·OH, and thus determining the scavenging efficiency of the NCs for ·OH (n = 4). After culturing HESCs in a 6-well plate for 12 h, 0.1 mg/ml of PDA-PEG-LN-ICG NCs were added for co-incubation. At 1st, 2nd, 4th, 6th, 24th, and 48th hours post-incubation, the cells were washed three times, digested with trypsin to create a cell suspension, and then the fluorescence intensity of the cells was measured using a flow cytometer, reflecting the uptake of PDA-PEG-LN-ICG NCs by HESCs (n = 3). Firstly, HESCs were co-cultured with PDA-PEG-LN NCs at a concentration of 0.1 mg/ml for 48 h to allow for adequate cellular uptake. Afterward, the dichlorofluorescein diacetate (DCFH-DA) probe was added, and the potential impact of PDA-PEG-LN NCs on the levels of reactive oxygen species (ROS) within normal cells was determined using flow cytometry. To detect the intracellular ROS levels after external stimulation, the DCFH-DA probe was added to both the control group cells and the cells that had taken up PDA-PEG-LN NCs. Following this, exogenous Rosup reagent was added to the cell culture medium of the two groups, and after 20 min, the changes in ROS levels were measured using flow cytometry (n = 3). Similarly, after adding 5 ng/ml of IL-1β to the culture medium of both groups for 24 h to stimulate the cells, the changes in intracellular ROS were determined using the DCFH-DA probe and flow cytometry (n = 3). Finally, the supernatants from each group of cells were collected, and the levels of IL-6 and TNF-α were measured by enzyme-linked immunosorbent assay (n = 4). THP-1 cells were obtained from Procell and cultured in RPMI-1640 medium (Servicebio) containing 10 % FBS, 1 % penicillin-streptomycin, and 0.05 mM β-Mercaptoethanol (Solarbio) under a constant temperature and humidity environment (37 °C, 5 % CO 2 ). The THP-1 cells were stimulated with 100 ng/ml PMA (MCE), and after 48 h, they were induced to become M0 macrophages. The culture medium was then replaced with fresh medium containing 100 ng/ml PMA, 100 ng/ml LPS (Sigma), and 20 ng/ml IFN-γ (MCE), and after 72 h, M1 macrophages were obtained. For the induction of M2 macrophage, the culture medium was refreshed with fresh medium containing 100 ng/ml PMA, 20 ng/ml IL-4 (MCE), and 20 ng/ml IL-13 (MCE), and after 72 h, M2 macrophages were obtained. During the induction of M1 and M2 macrophages, 0.1 mg/ml of PDA-PEG-LN NCs were added to the culture medium respectively. After 72 h, the impact of PDA-PEG-LN NCs on the polarization of M1 or M2 macrophages was assessed. After fixing and permeabilizing each group of cells (Elabscience), staining was performed using FITC anti-human CD68 (Biolegend). The signal intensity of FITC fluorescence in each group of cells, as detected by flow cytometry, reflected the content of the M0 marker CD68 (n = 3). Following the successful induction of M0 macrophages, APC anti-human CD86 from Biolegend was added to each group of cells for surface staining (n = 3), while PE anti-human CD206 from BD Pharmingen was used for staining fixed and permeabilized cells (n = 3). The fluorescence values of APC or PE in the flow cytometry results indicated the content of the M1 marker CD86 or the M2 marker CD206, respectively, in each group of cells. Subsequently, RNA was extracted from each group of cells using a column-based RNA extraction method. After genomic DNA removed, reverse transcription was carried out using PrimeScript RT reagent Kit with gDNA Eraser (Takara). The resulting cDNA solution added PCR primers corresponding to the target genes, was then subjected to quantitative real-time polymerase chain reaction (qRT-PCR). The qRT-PCR was conducted using the TB Green Premix Ex Taq II (Takara) on the Applied Biosystems 7500 system to detect changes in macrophage-related genes across each group (n = 3). MTT Assay : After co-culturing HESCs with dydrogesterone or DYD@PDA-PEG-LN NCs carrying an equivalent dose of dydrogesterone for 48 h, the MTT method was used to assess the relative viability of HESCs under different concentrations of dydrogesterone (n = 6). EdU Proliferation Assay : The EdU Cell Proliferation Kit with Alexa Fluor 555 was used, incorporating the thymidine analogue EdU (5-ethynyl-2′-deoxyuridine) during DNA synthesis. Subsequently, a Click reaction was used to label EdU with the fluorescent dye Alexa Fluor 555. After digesting and resuspending the cells, the proportion of EdU-555 was detected by flow cytometry, thereby measuring the proportion of cell proliferation (n = 3). Apoptosis Detection : The proportion of apoptotic cells after co-incubation was detected using the FITC-Annexin V/PI Apoptosis Detection Kit and flow cytometry. During the early stages of apoptosis, phosphatidylserine (PS) was translocated to the cell surface and was labeled by FITC-Annexin V, thereby emitting green fluorescence. In contrast, the nuclei of necrotic cells or cells in the late stages of apoptosis that had lost membrane integrity were bound by Propidium Iodide (PI), thus exhibiting red fluorescence (n = 4). Mitochondrial Membrane Potential Detection : Changes in the mitochondrial membrane potential of cells in each group were detected using the Mitochondrial Membrane Potential Assay Kit with JC-1 and a fluorescence microscope. A decrease in mitochondrial membrane potential was also one of the early hallmarks of apoptosis. When the mitochondrial membrane potential was low, JC-1 could not aggregate in the mitochondrial matrix and existed in the form of JC-1 monomers, producing green fluorescence. Conversely, when the mitochondrial membrane potential was high, JC-1 aggregated in the mitochondrial matrix to form polymers known as J-aggregates, which could produce red fluorescence. Cell Migration Assay : HESCs cells that had been co-incubated with PDA-PEG-LN NCs, dydrogesterone, or DYD@PDA-PEG-LN NCs for 48 h were replated, and then a scratch wound was made using a 200-μL pipette tip. The changes in the width of the scratch were observed after 24 h and 48 h to reflect the cells' migratory ability. An ethical review committee of Tongji Hospital affiliated with Tongji University reviewed and approved the details of animal experiments (ID:2024‐DW‐SB-276). The typical ectopic endometrial tissues were collected at Shanghai First Maternity and Infant Hospital. The procedures acquired approval from the Medical Ethics Committee of Shanghai First Maternity and Infant Hospital (KS22356), and all participants had signed the written informed consent. The female BALB/c nude mice aged 6 weeks and weighing approximately 14g, were raised in a Specific Pathogen Free (SPF) barrier environment with constant temperature and humidity, a 12-h light/dark cycle. Their feed and water had been sterilized by high-pressure. We selected typical ectopic endometrial tissues from patients with ovarian endometriosis who opted for surgical treatment and transplanted these subcutaneously into the interscapular region of female BALB/c nude mice (one implant per mouse). Patient Inclusion Criteria: (1) Women of reproductive age between 20 and 45 years old. (2) Regular menstrual cycles. (3) Underwent laparoscopic or open surgery due to dysmenorrhea or infertility, with a postoperative pathological diagnosis of endometriosis. (4) Patients diagnosed with stage III-IV endometriosis. The excised tissues were transferred to a laminar flow hood, rinsed with PBS three times, and cut into tissue blocks approximately 6 mm by 6 mm in size. The nude mice were anesthetized with 2 % isoflurane inhalation, and the dorsal skin was disinfected with iodine tincture. The sterile ophthalmic scissor was used to make an approximately 0.5 cm incision in the interscapular area. A careful blunt dissection was carried out within a 1 cm radius of the subcutaneous space to embed the endometrial tissue as deeply as possible, and then the incision was sutured with 4-0 silk suture. After disinfecting the incision, the mouse was returned to the cage for continued breeding. Starting on the day of surgery and every 48 h thereafter, estradiol benzoate was injected into the leg muscles of the nude mice to maintain the growth of the lesions. Sham-operated mice underwent incision and suture at the same site without the implantation of endometrial tissue. At the 7th days and 28th days post-surgery, the grafts were excised, fixed with 4 % paraformaldehyde and then embedded in paraffin for sectioning. Subsequently, all tissue sections underwent immunohistochemical (IHC) staining for ESR1, PGR, KI-67, and VEGF. After deparaffinization and antigen retrieval, blocking, incubation with primary and secondary antibodies, section washing, and DAB color development, the IHC staining results for ESR1, PGR, KI-67, and VEGF were observed under a microscope. On the 7th day after surgical transplantation, DYD@PDA-PEG-LN-ICG NCs or DYD@PDA-ICG NCs were injected into the mice at a dose of 10 mg/kg body weight, administered either subcutaneously or intravenously, to achieve local or systemic delivery of the nanocomposites. Fluorescence images of the mice were recorded using an in vivo imaging system at 1st, 2nd, 4th, 6th, 24th, and 48th hours post-injection (n = 3). To exclude the impact of surgical factors on the targeting ability of DYD@PDA-PEG-LN-ICG NCs, sham-operated mice that underwent incision and suture without implantation of ectopic endometrial tissue were used. On the 7th day after surgery, the mice were injected with DYD@PDA-PEG-LN-ICG NCs locally around the surgical incision and via tail vein injection. Similarly, the fluorescence imaging results of the sham-operated mice were recorded using an in vivo imaging system at 1st, 2nd, 4th, 6th, 24th, and 48th hours post-injection (n = 3). We randomly divided the endometriosis model mice into two groups: the subcutaneous injection group and the intravenous injection group. Each group was further divided into four subgroups, namely the PBS group (injected with PBS), the PDA-PEG-LN group (10 mg/kg body weight), the dydrogesterone group (102.6 μg/kg body weight), and the DYD@PDA-PEG-LN group (10 mg/kg body weight). Injection began on the 7th day after transplantation, were administered every two days, and ended after 21 days (on the 28th day after transplantation). The area of the subcutaneous grafts in each group of mice was recorded every 4 days after the endometrial transplantation (n = 6). After the treatment, the grafts in each group were excised and weighed (n = 6). The excised grafts were then fixed with 4 % paraformaldehyde, embedded in paraffin, sectioned, and subjected to immunohistochemical (IHC) staining for ESR1, PGR, KI-67, and VEGF. After deparaffinization and antigen retrieval, blocking, incubation with primary and secondary antibodies, section washing, and DAB color development, the IHC staining results for ESR1, PGR, Ki-67, and VEGF were observed under a microscope. To document the recurrence of lesions in each group following the 21-day treatment period, the subcutaneous area of the grafts was recorded every four days. We collected the grafts from the endometriosis model mice following PBS, PDA-PEG-LN NCs, dydrogesterone, or DYD@PDA-PEG-LN NCs over a period of 21 days. We prepared tissue homogenates of these grafts and lysed some of them with RIPA buffer. The protein concentration in the grafts was determined using the Detergent Compatible Bradford Protein Assay Kit. Then, the ferric-orange xylenol (FOX) system was used to assess the levels of hydroperoxides in each group [ 31 ], which reflected the changes in ROS within the lesion tissues. FOX working solution: Working solution 1 was prepared by adding ammonium ferrous sulfate to a 250 mM sulfuric acid solution to achieve the concentration of 25 mM ammonium ferrous sulfate. Working solution 2 was composed of 62.5 μM xylenol orange and 150 mM sorbitol. The ratio of working solution 1 to 2 was 1:100, and this working solution should be prepared fresh for immediate use. We mixed the test solution diluted appropriately with the working solution in a 1:2 ratio, then incubated them in the dark for 30 min. After centrifugation at 3500 rpm for 10 min, the absorbance of the supernatant at 580 nm were measured using a microplate reader. A standard curve was used to calculate the amount of hydroperoxides in the test solution, and it was obtained by diluting the 30 % hydrogen peroxide solution to various concentrations ranging from 0.0136 μg/ml to 0.17 μg/ml. The standard curve had an R2 correlation coefficient of 0.99996. Finally, the hydrogen peroxide content of the grafts was expressed in μg/mg of protein (n = 6). The serum was collected from normal mice (non-operated), endometriosis model mice injected with PBS, PDA-PEG-LN NCs, dydrogesterone, or DYD@PDA-PEG-LN NCs for 21 days. The levels of TNF-α, IL-1β, and IL-6 in the mouse serum were measured using the enzyme-linked immunosorbent assay (ELISA) method (n = 6). The excised ectopic endometrial grafts were fixed with 4 % paraformaldehyde, embedded in paraffin, sectioned, and then immunofluorescence staining was performed. All macrophages in the grafts were labeled with the F4/80 antibody. Additionally, M1 macrophages and M2 macrophages were specifically marked with iNOS or CD206 antibodies, respectively. All the cells were stained with DAPI for nuclear visualization. The staining results were observed using a fluorescence microscope. The ovaries and serum were collected from mice that had not received any injections, as well as from mice that had been treated with DYD@PDA-PEG-LN NCs via subcutaneous or intravenous injections at a dose of 10 mg/kg body weight over a period of 21 days. The structural changes in the mouse ovaries were observed through H&E staining, and the serum AMH levels were determined using the enzyme-linked immunosorbent assay (ELISA) to reflect the ovarian reserve function of mice in each group (n = 6). As previously described [ 24 , 32 ], we assessed the fertility of normal ICR mice and the ICR mice that received DYD@PDA-PEG-LN NCs via subcutaneous or intravenous injections for 21 days. One or two female mice were placed with a male mouse of the same age in the same cage, and to ensure successful conception, the male mice were exchanged among the cages every four days. The morning when a vaginal plug was observed in the female mice was considered to be day 0.5 of pregnancy. We recorded the time from the start of cohabitation to successful conception for mice in each group, and performed euthanasia on the pregnant mice on day 14.5 of gestation to accurately count the number of embryos in each group. Subsequently, the embryos and placentas from mice in each group were fixed with 4 % paraformaldehyde, embedded in paraffin, sectioned, and subjected to H&E staining (n = 3). Data were expressed as mean ± standard error of mean (SEM). The statistical analyses between two groups were performed by using student's t -test. One-way analysis of variance (ANOVA) was used for comparisons among three or more groups. P < 0.05 was considered significantly different. The symbol “∗” denoted p < 0.05, “∗∗” denoted p < 0.01, and “∗∗∗” denoted p  0.05. Statistical analysis was conducted using GraphPad Prism 8 Software.

In summary, we have developed a novel treatment method for endometriosis by leveraging the targeting and therapeutic effects of DYD@PDA-PEG-LN NCs on endometriotic lesion tissues. This approach not only ameliorates chronic inflammation and inhibits estrogen-dependent growth but also does not affect ovarian reserve function and fertility. In this study, through a series of in vitro and in vivo experiments, we have demonstrated that DYD@PDA-PEG-LN NCs possess good antioxidant and anti-inflammatory capabilities, and inhibit the polarization of pro-inflammatory M1 macrophages while promoting the expression of anti-inflammatory M2 macrophages. In the endometriosis model mice, DYD@PDA-PEG-LN NCs can also target and promote the atrophy of the lesions while reducing the dosage of dydrogesterone. Furthermore, DYD@PDA-PEG-LN NCs exhibit high biocompatibility and low systemic toxicity, and treating endometriosis with them does not impair ovarian reserve function and fertility.

Endometriosis (EM) refers to the occurrence, growth, infiltration and repeated bleeding of endometrial tissue glands and stroma beyond the covered endometrium of the uterine cavity or outside the uterine, subsequently causing pain, infertility, nodules or masses. It is an estrogen-dependent inflammatory disease affecting 5–10 % of women at reproductive age [ 1 , 2 ]. Up to 50 % of infertile women and 50–80 % of women with pelvic pain suffer from endometriosis. Clinically, there is a widely accepted association between endometriosis and infertility [ 3 ]. There are three phenotypes of endometriosis, namely ovarian endometriosis (the most common), superficial peritoneal endometriosis, and deep infiltrating endometriosis [ 4 ]. The treatment goal of endometriosis is to eliminate lesions, alleviate pain, improve fertility, and prevent recurrence. Clinically, the main treatments for endometriosis are medication and surgery. Medical treatment cannot cure endometriosis fundamentally and it primarily works by creating a low-estrogen environment (such as using gonadotropin-releasing hormone agonists, GnRH-a), a high-androgen environment, or a high-progestin environment. By achieving a stable steroid hormone environment, it inhibits the proliferation of ectopic endometrial cells [ 5 ]. The treatment with GnRH-a is very effective, but it can cause frequent and almost intolerable side effects due to the hypoestrogenic state [ 6 ]. Androgen-like treatment drugs may cause androgen-related side effects such as adverse effects on serum cholesterol levels [ 7 ]. Combined oral contraceptives or progestin therapy, while well-tolerated and safe, may lead to irregular bleeding or an increase in the size of uterine fibroids [ 5 ]. Contraceptive effect of these medicine will affect the natural fertility process, which is contradictory to the patient's fertility needs. Dydrogesterone (DYD) is a derivative of endogenous progesterone, exhibiting higher activity and selectivity for progesterone receptors. It can be used by women who are pregnant or trying to conceive, and the standard therapeutic dose does not suppress ovulation [ 8 ]. Nevertheless, the efficacy of dydrogesterone in treating endometriosis is inferior to other hormonal drugs. Its daily dosage of 30 mg may be associated with adverse reactions, which leads to dydrogesterone not being the preferred treatment option for endometriosis [ 9 ]. Surgery is generally considered the best treatment for ovarian and deep infiltrating endometriosis. For superficial peritoneal endometriosis, there is still no consensus [ 10 ]. Surgical treatment is a double-edged sword when removing ovarian endometriosis lesion, because it may damage normal ovarian tissue and reduce ovarian reserve function [ 11 ], affecting the patient's fertility. Whether medical or surgical treatment will result in the increase of the patient's reproductive age, which can further decrease fertility in turn [ 12 ]. The increase in age has a negative impact on the quantity and quality of oocytes [ 13 ], which leads to higher rates of infertility and lower pregnancy rates [ 14 ]. Advanced maternal age is also one of the risk factors for recurrent implantation failure [ 15 ]. As the average ages of first marriage and childbearing continue to rise [ 16 ], doctors should take the patient's age into account, and strengthen the protection of the patient's ovarian function and fertility when treating endometriosis. Consequently, there is an urgent need to develop a new treatment strategy that balances the therapeutic effects of endometriosis with the protection of fertility. Nanocomposites can protect the cargo they carry and precisely target to enable a reduced dosage of medication [ 17 ]. Currently, various tissue-targeting nano-therapeutic delivery systems have been successfully applied in treatment [ 18 ]. Endometriosis, like cancer, is an angiogenesis-dependent disease, although it is a benign condition. Recent years, used as therapeutic agents or delivery carriers, a variety of composite nanomaterials for the treatment of endometriosis have been developed [ 19 , 20 ]. Most of these studies focus on the clearance of lesions, but neglect that endometriosis patients are mostly women of reproductive age, so these developed nanomaterials may have certain impacts on ovarian reserve function or fertility. In this study, we prepared a nano-composite material that does not affect ovarian reserve function or fertility while effectively treating endometriosis. Endometriotic lesions are characterized by elevated oxidative stress [ 21 ], with ectopic endometrial cells usually containing large amounts of reactive oxygen species (ROS) [ 22 ]. Increased oxidative stress promotes the inflammation and progression of endometriosis. Moreover, it may be associated with infertility [ 23 ], and the reduction of embryo numbers [ 24 ]. Chronic inflammation may be a potential pathological mechanism driving endometriosis [ 3 ] and involved in the impairment of ovarian reserve function [ 25 ]. Alleviating oxidative stress and the inflammatory environment may cure endometriosis lesions and associated infertility [ 26 ]. In addition, complex dysregulated hormonal signaling is implicated in the initiation and progression of endometriosis [ 3 ]. Dydrogesterone can induce atrophy of ectopic endometrium without affecting orthotopic endometrial tissue or suppressing ovulation [ 8 ]. Given the antioxidant and anti-inflammatory capabilities of polydopamine (PDA) [ 27 ], we formulate nanocomposites composed of dydrogesterone and polydopamine (DYD@PDA). Furthermore, we PEGylate the nanocomposites to increase their passive targeting to the lesion tissues while decreasing the incidence of adverse reactions during treatment [ 28 ]. Compared to normal endometrium, anthrax toxin receptor 2 (ANTXR2) is upregulated in endometriotic cells [ 29 ]. Laminin (LN) has been confirmed as an endogenous ligand for ANTXR2 and may also be related to cell adhesion [ 30 ]. Therefore, we further modified the surface of the nanocomposites with laminin to obtain DYD@PDA-PEG-LN nanocomposites, improving the targeting capability to endometriotic lesions and enabling a reduced dosage of dydrogesterone. As shown in Fig. 1 , DYD@PDA-PEG-LN nanocomposites can clear reactive oxygen species, reduce inflammation, modulate macrophage phenotypes, and induce apoptosis of endometriotic lesions. Using an endometriosis mouse model, we have demonstrated that these nanocomposites can be retained and effectively target ectopic endometrial grafts, whether administered locally or systemically. Biosecurity assessments indicate high biocompatibility and biodegradability, and DYD@PDA-PEG-LN nanocomposites will not damage ovarian reserve function or fertility. We anticipate that this study can offer a new perspective for developing safe and effective endometriosis treatments that do not affect ovarian reserve function and fertility. Furthermore, we hope that future therapeutic research will not only address the elimination of endometriotic lesions but also consider the impact of aging and include the potential for treatment-induced damage to ovarian function and fertility into the evaluation criteria. Fig. 1 Schematic representation of the application of DYD@PDA-PEG-LN nanocomposites in treating endometriosis lesions The diagram illustrates the targeting and accumulation of DYD@PDA-PEG-LN nanocomposites in ectopic endometrial tissues, and shows their effects on inducing lesion atrophy by improving oxidative stress, reshaping the phenotype of macrophages, ameliorating chronic inflammation, and inhibiting estrogen-dependent proliferation of endometriotic lesions. Fig. 1 Schematic representation of the application of DYD@PDA-PEG-LN nanocomposites in treating endometriosis lesions The diagram illustrates the targeting and accumulation of DYD@PDA-PEG-LN nanocomposites in ectopic endometrial tissues, and shows their effects on inducing lesion atrophy by improving oxidative stress, reshaping the phenotype of macrophages, ameliorating chronic inflammation, and inhibiting estrogen-dependent proliferation of endometriotic lesions.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.