Dynamic Nanoscavengers With Boosted Antioxidative Properties for Hepatic Ischemia‐Reperfusion Injury Therapy

ABSTRACT

Inspired by nature, eumelanin‐like polymeric antioxidants have been widely used for the treatment of various oxidative stress‐related diseases in recent years. In response to functional requirements, extensive studies have been conducted to improve the antioxidant properties of eumelanin‐like materials, but the dilemma brought by the oxidative polymerization strategy has never been broken. Herein, the natural eumelanin precursor, levodopa, was integrated with formylphenylboronic acid through dynamic covalent chemistry to give antioxidant nanoparticles. Benefiting from this synthetic scheme, the resulting dynamic nanoscavengers avoided the loss of antioxidant moieties during the fabrication process and could achieve interesting ROS‐responsive degradation behaviors. These nanoscavengers exhibited over ten times the antioxidant capacity of conventional eumelanin‐like materials and demonstrated improved efficacy in hepatic ischemia‐reperfusion injury therapy, with treatment groups showing more optimal levels of oxidative stress markers and inflammation‐related factors. This study provides new insights into the design and development of a new generation of melanin‐inspired antioxidant materials.

Dynamic nanoscavengers based on eumelanin monomer can responsively degrade in acidic or ROS environments, thereby exhibiting enhanced antioxidant capacity, and have been used in the treatment of hepatic ischemia‐reperfusion injury therapy.

1. Introduction

Eumelanin‐like materials have been one of the promising biomimetic materials in the past decade [1, 2, 3, 4]. They have many properties similar to natural eumelanin, such as light protection, antioxidation, and metal chelation [2]. In addition to the fields of energy and environment, eumelanin‐like or eumelanin‐inspired materials have also been shining in biomedical applications due to their excellent biocompatibility and antioxidative properties [5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26].

During the recent years, eumelanin‐like materials, such as polydopamine (PDA) and polylevodopa (P(L‐DOPA)), have been employed for the treatment of some oxidative stress‐related diseases, such as wound damage repair and inflammation regulation, relying on their excellent free radical scavenging properties [6, 17, 27, 28, 29]. In order to meet the ever‐changing application requirements, several studies have been carried out to further improve the antioxidant performance of eumelanin‐like materials. The core ideas of previous research mainly include three points: First, increasing the proportion of phenolic hydroxyl groups in the product through reduction treatment [28]; second, adjusting the stacking structure of polymer chains to increase the accessibility of phenolic hydroxyl groups [29]; and third, introducing other antioxidant groups [17].

Unfortunately, none of these works has gotten rid of the oxidative polymerization method for eumelanin‐like materials preparation, which is not friendly to the antioxidant properties of the product. On one hand, some functional groups will be oxidized and lost prematurely during the preparation process [1]. On the other hand, the polymer networks within the resulting eumelanin‐like materials are usually compact and dense, and a large number of functional groups are hidden inside the nanomaterial. Therefore, those previous studies could only mitigate the adverse effects of oxidative polymerization on eumelanin‐like materials, but did not really solve the problem.

The antioxidative function of eumelanin‐like materials is inherited from natural melanin precursors, levodopa (L‐DOPA). Developing suitable new polymerization strategies for those natural precursors may pave the way for solving the above problems. L‐DOPA has an amino group and a catechol structure, which can undergo dehydration condensation with an aldehyde group and a boric acid group, respectively [30, 31]. The two reactions involved belong to common dynamic covalent chemistry, and the imine bond and boronate–catechol linkages formed can undergo reverse hydrolysis under certain conditions [32, 33, 34]. In addition, the boronate–catechol linkages can also be broken in response to reactive oxygen species (ROS) [35]. Taking these into consideration, formylphenylboronic acid (FPBA) was selected as the monomer and polycondensed with L‐DOPA to obtain the dynamic covalent nanoscavengers, P(DOPA‐FPBA). Compared with many previously reported eumelanin‐like materials, this class of nanoscavengers did not suffer from oxidative loss of phenolic hydroxyl groups and could be degraded in acidic or ROS environments, exhibiting outstanding antioxidant properties, with a free radical scavenging capacity that can even be ten times higher.

Hepatic ischemia‐reperfusion injury (HIRI) is a serious issue in patients experiencing hepatic trauma, liver transplantation, hepatic resection, or infectious shock, potentially resulting in hepatocellular dysfunction, rejection, and liver failure [36]. While the mechanisms behind HIRI development have not been fully elucidated, extensive evidence indicates that it is linked to excessive production of ROS and subsequent reoxidation during perfusion recovery [37]. Unfortunately, research on the use of antioxidants to treat HIRI is still limited, and most work has not fully considered the impact of antioxidant properties on therapeutic efficacy. Herein, the as‐prepared dynamic nanoscavengers beyond eumelanin‐like materials have been employed for HIRI therapy and envisioned to exhibit better treatment effects as compared to conventional eumelanin‐like NPs. It is believed that nanoscavengers based on dynamic covalent chemistry can play an important role in the treatment of other types of oxidative stress‐related diseases and bring new inspiration to the iteration of eumelanin‐like materials.

2. Results and Discussion

2.1. Synthesis and Characterization of P(DOPA‐FPBA) NPs

Initially, 4‐FPBA was chosen as the comonomer with L‐DOPA because it would obviously result in less steric hindrance during polycondensation compared with 3‐FPBA and 2‐FPBA (Figure 1A). Since the reaction was a dehydration process and the monomers were polar, DMF was selected as the reaction solvent. Both the Schiff base reaction and the formation of boronate–catechol linkages are base‐catalyzed processes, but it was believed that these reactions between L‐DOPA and 4‐FPBA were easy to occur in DMF, so only heating condition was adapted to accelerate the kinetic process (Figure 1A). In order to prevent L‐DOPA from being oxidized by air in a heated state, the reaction was carried out under a nitrogen atmosphere.

FIGURE 1.

(A) Polycondensation of L‐DOPA and 4‐FPBA; (B) Summary of the formulation and physical parameters of the formed P(DOPA‐FPBA) NPs; (C) Representative SEM images (the four on the left) of P(L‐DOPA) NPs, scale bar = 250 nm, and TEM and EDS mapping images (the rightmost) of P(DOPA‐FPBA)‐2; (D) FTIR spectrum of P(DOPA‐FPBA)‐2; (E) XPS spectrum of P(DOPA‐FPBA)‐2; (F) The atomic ratio of P(DOPA‐FPBA)‐2 obtained by XPS and the theoretical atomic ratio of monomers; (G) High‐resolution XPS spectra of B 1s regions for P(DOPA‐FPBA)‐2 and 4‐FPBA; (H) High‐resolution XPS spectra of C 1s regions for P(DOPA‐FPBA)‐2 and 4‐FPBA; (I) High‐resolution XPS spectra of N 1s regions for P(DOPA‐FPBA)‐2 and L‐DOPA.

DMF acted as a good solvent for 4‐FPBA but a poor solvent for L‐DOPA, resulting in a white and turbid mixture during the early stages. However, as the polymerization reaction proceeded, the reaction solution gradually turned into yellow turbidity within 20 min (Figure S1). It is worth noting that when 3‐FPBA or 4‐FPBA was condensed with L‐DOPA, the reaction liquid first changed from white opaque to yellow transparent, and then became yellow turbid after several hours (Figure S1). This was a typical precipitation polymerization process, L‐DOPA was first converted into oligomers and fully dissolved in DMF, and then the product precipitated due to the increase in polymerization degree [38].

Due to steric hindrance in the reaction, the product precipitation time varied when 3‐FPBA and 4‐FPBA participated in the polymerization. As for the scenario of using 2‐FPBA copolymerization, the reaction liquid still had no product precipitation after 12 h (Figure S1), which may be due to the great steric resistance of the growth of the polymerization chain. The above phenomenon proved that it was reasonable to choose 4‐FPBA as a comonomer for L‐DOPA.

In order to determine the chemical reactions involved, nuclear magnetic resonance (NMR) analysis was performed on the copolymerization solution of L‐DOPA and 4‐FPBA. In the ¹H NMR spectrum of the crude reaction solution (Figure S2), it can be found that signals of protons in 4‐FPBA appeared at the high field (δ = 9.87, 7.80–7.84, 7.01–7.05 ppm). Meanwhile, similar peaks in the boron NMR spectrum that correspond to the imine and boron ester bond structures were observed. There was also a new peak at high fields in the ¹¹B NMR spectrum (δ = 14.69 ppm). The new peak above corresponded to the imine and boronate–catechol linkages, indicating that the copolymerization reaction proceeded as designed. In addition, electrospray ionization mass spectrometry (ESI‐MS) analysis of the polycondensation reaction solution also confirmed the presence of L‐DOPA and 4‐FPBA oligomers formed via dehydration (Figure S3).

The copolymerization products of L‐DOPA and 4‐FPBA were separated by centrifugation and recorded as P(DOPA‐FPBA)‐i after washing (i = 1‐4, Figure 1B). In addition, note that benzaldehyde and PBA could not react with L‐DOPA to obtain centrifugation products, which proved the importance of the synergistic effect between the Schiff base reaction and the phenylboronic acid‐polyphenol reaction (Table S1). Scanning electron microscopy (SEM) and dynamic light scattering (DLS) results showed that the P(DOPA‐FPBA) formed nanoparticles with size adjustability, which was positively correlated with the monomer concentration (Figures 1B,C and S4A). The particles exhibited good dispersion stability in water, as indicated by zeta potentials below −20 mV (Figures 1B and S4B), likely due to the ionization of carboxyl and other groups.

The chemical composition and morphology of P(DOPA‐FPBA) were further studied. Transmission electron microscopy (TEM) and energy dispersive spectrometer (EDS) mapping confirmed the spherical shape of P(DOPA‐FPBA) and revealed the presence of B, C, N, and O (Figure 1C). Besides, as shown in the Fourier‐transform infrared spectroscopy (FTIR) spectrum (Figure 1D), the peaks at 2845 cm⁻¹, 2751 cm⁻¹, and 1360 cm⁻¹ of the P(DOPA‐FPBA) were greatly reduced compared with the monomer. The above peaks corresponded to the stretching vibrations of C = O and B‐OH, and their reduction meant a decrease in aldehyde and boronic acid groups. In order to accurately characterize the structure of P(DOPA‐FPBA), x‐ray photoelectron spectroscopy (XPS) analysis was performed. The results confirmed that the polymer contained the same elements as the monomers and allowed determination of the atomic ratios of B, N, and O (Figure 1E,F). The N/B ratio was calculated as 0.89, indicating that L‐DOPA and 4‐FPBA were polymerized in almost equal amounts, while the O/B ratio was 4.27, indicating that the polymerization reaction did involve dehydration. Note that the amount of dehydration was close to three equivalents compared to the monomers, which was very consistent with the Schiff base and phenylboronic acid‐polyphenol reaction. In the high‐resolution XPS spectra of B 1s regions for P(DOPA‐FPBA), it could be found that the signal peak had a lower binding energy than that of 4‐FPBA (Figure 1G), indicating increased electron cloud density on boron due to boronate–catechol linkage formulation. Besides, P(DOPA‐FPBA) lacked a distinct peak at 287.6 eV corresponding to the formyl groups in C 1s regions (Figure 1H). The 399.4 eV peak in N 1s regions suggested the presence of imine structures in P(DOPA‐FPBA) (Figure 1I). It was speculated that the peak at 401.7 eV of P(DOPA‐FPBA) was caused by the boron‐nitrogen interaction, which might be one of the driving forces for the precipitation of the polymer chain from DMF. In fact, the P(DOPA‐FPBA) NPs could indeed be induced to disintegrate by imidazole, which was considered to be a competitor for the boron‐nitrogen interaction (Figure S5) [39]. The P(DOPA‐FPBA) NPs could be partially dissolved in a mixed solvent of DMSO and water, and imine structures and boronate–catechol linkages were indeed found through NMR and ESI‐MS analysis (Figures S6 and S7).

2.2. Responsive Behavior and Antioxidant Properties of P(DOPA‐FPBA) NPs In Vitro

The aforementioned results indicated that dehydration polycondensation reactions did occur between L‐DOPA and 4‐FPBA, and the obtained P(DOPA‐FPBA) NPs contained multiple imine and boronate–catechol linkages. Due to the existence of these structures, P(DOPA‐FPBA) NPs were expected to degrade under acidic or ROS environments, thereby continuously exposing the internal phenolic hydroxyl groups for antioxidant activity (Figure 2A).

FIGURE 2.

(A) Schematic diagram of the difference in antioxidant behavior between the conventional eumelanin‐like material and dynamic P(DOPA‐FPBA) NPs; Degradation behavior in response to (B) pH and (C) H₂O₂; (D) ABTS·⁺ scavenging activities of PDA, P(L‐DOPA), P(DOPA‐FPBA)‐2 and Vc over time, antioxidant dosage was 100 µg; (E) ABTS·⁺ scavenging activities of PDA, P(L‐DOPA), P(DOPA‐FPBA)‐2 and Vc across varied dosages, data were collected over 7 h; (F) ABTS·⁺ scavenging capacities of PDA, P(L‐DOPA), P(DOPA‐FPBA)‐2 and Vc; (G) DPPH scavenging activities of PDA, P(L‐DOPA), P(DOPA‐FPBA)‐2 and Vc over time, antioxidant dosage was 50 µg; (H) DPPH scavenging activities of PDA, P(L‐DOPA), P(DOPA‐FPBA)‐2 and Vc across varied dosages, data were collected at the 7 h; (I) DPPH scavenging capacities of PDA, P(L‐DOPA), P(DOPA‐FPBA)‐2 and Vc.

Under neutral conditions, P(DOPA‐FPBA) NPs would be slightly degraded due to the hydrolysis of the imine and boronate–catechol linkages (Figure 2B). When the pH of the system decreased, the dispersed solution would become more transparent (Figure S8A,B), the residual rate would also decrease, and it approached the degradation equilibrium in 2 h. SEM images showed that the NPs were completely destroyed under acidic conditions (Figure 2B). However, it is worth noting that the degradation experiment was carried out in a closed system. If the local concentration of degradation products decreases in the organism due to molecular metabolism, antioxidant consumption, etc., the remaining P(DOPA‐FPBA) NPs will theoretically be driven to continue hydrolyzing. Moreover, it is well known that boronate–catechol linkages can collapse in a ROS environment, so H₂O₂ could also accelerate the degradation of P(DOPA‐FPBA) NPs (Figure 2C) [35]. After H₂O₂ stimulated the degradation of P(DOPA‐FPBA) NPs, the solution not only became more transparent, but also turned brown because the degradation products were further oxidized (Figure S8A,C). Based on in vitro degradation experiments, it can be speculated that under in vivo conditions, the P(DOPA‐FPBA) NPs would eventually transform into boric acid and phenolic derivatives under the combined effects of hydrolysis and ROS [40]. Boric acid is not enzymatically metabolized in animals and mainly exists in its original form, with its excretion primarily occurring via the kidneys into urine [41]. Regarding the phenolic derivatives, based on previous studies, it is speculated that they may also be excreted in urine via glomerular filtration and renal tubular secretion [42].

The responsive degradation of P(DOPA‐FPBA) NPs might be beneficial to their antioxidant performance due to internal groups that could be continuously exposed (Figure 2A). In the typical ABTS·⁺ scavenging experiment, P(DOPA‐FPBA) NPs cleared ABTS·⁺ faster and more effectively than PDA and P(L‐DOPA) with similar particle sizes, outperforming even the classical small molecule antioxidant Vc (Figure 2D, Figure S9). Sample usage affected the ABTS·⁺ scavenging capacity, whereas P(DOPA‐FPBA) NPs consistently showed higher scavenging efficiencies (Figure 2E). Their antioxidant properties, equivalent antioxidant capacity was calculated, which was defined as the ratio of the amount of ABTS·⁺ or DPPH scavenged (mol) to the amount of the repeating units of antioxidants (mol). As shown in Figure 2F, P(DOPA‐FPBA) NPs could scavenge ABTS·⁺ 5.9 times as much as the repeating units, which was 175% of P(L‐DOPA), 275% of PDA, and 383% of Vc. In the DPPH scavenging assay, P(DOPA‐FPBA) NPs also demonstrated a remarkable performance. At the same dosage, P(DOPA‐FPBA) NPs could scavenge more DPPH than PDA and P(L‐DOPA) (Figure 2G,H). After normalization to repeat units, the DPPH scavenging capacity of P(DOPA‐FPBA) NPs was 5.0 mol/mol, which was 246% of P(L‐DOPA), 1322% of PDA, and 260% of Vc (Figure 2I). This substantial improvement suggests that the dynamic covalent polycondensation strategy, which avoids oxidation and enables degradation of P(DOPA‐FPBA) NPs, resulted in a greater involvement of phenolic hydroxyl groups in the antioxidant process.

Motivated by the scavenging results, we subsequently evaluated the antioxidant capacities of P(DOPA‐FPBA), P(L‐DOPA) and PDA NPs at the cellular level. First of all, the cytotoxicities of the above three NPs were assessed by the alamar blue assay utilizing a mouse hepatocyte AML‐12 cell line. Cell viabilities remained high in all nanomaterial treatment groups at 24 and 48 h across concentrations of 5–100 µg/mL, indicating the excellent cytocompatibility of P(DOPA‐FPBA), P(L‐DOPA), and PDA NPs (Figures 3A and S10).

FIGURE 3.

(A) Viabilities of AML‐12 cells after treatment with P(DOPA‐FPBA), P(L‐DOPA) and PDA NPs for 48 h; (B) Protective effects of P(DOPA‐FPBA), P(L‐DOPA) and PDA NPs against H₂O₂‐mediated oxidative stress in AML‐12 cells; (C) Quantitative detection of intracellular ROS levels; (D) Fluorescence image of live/dead cell staining and intracellular ROS stained by DCFH‐DA, scale bar, 100 µm; (E) Protective effects of P(DOPA‐FPBA), P(L‐DOPA) and PDA NPs against H₂O₂‐mediated cell apoptosis. *p < 0.05, **p < 0.01, ***p < 0.001.

Subsequently, the protective effects of P(DOPA‐FPBA), P(L‐DOPA), and PDA NPs against H₂O₂‐induced oxidative stress in AML‐12 cells were further evaluated. The cells treated with 300 µM H₂O₂ or not were presented as positive control (PC) group and negative control (NC) group, respectively. Cell viability in the PC group was measured to be ∼ 50%. By contrast, all NPs pretreatment groups exhibited the rescued activities on AML‐12 cells, as evidenced by the consistent increase in cell viabilities along increasing material concentrations. For instance, at 20 and 50 µg/mL material concentrations, cell viabilities were measured to be approximately 52% and 64% upon treatment with PDA NPs, 65% and 77% upon treatment with P(L‐DOPA) NPs, and 76% and 83% upon treatment with P(DOPA‐FPBA) NPs (Figure 3B). Evidently, the P(DOPA‐FPBA) NPs pretreatment better mitigated H₂O₂‐induced cytotoxicity compared to PDA and P(L‐DOPA) NPs pretreatments. The live/dead staining results (Figure 3D) also corroborated the cell viability trends.

To further evaluate the relevant protection mechanisms, we performed the fluorescent dye (DCFH‐DA) staining assay to assess ROS levels in each group (Figure 3C,D). The cells treated solely with H₂O₂ exhibited notably high ROS levels, as indicated by the green fluorescence emitted from the DCFH‐DA fluorescent dye under oxidative conditions. In contrast, treatment of AML‐12 cells with P(L‐DOPA) or P(DOPA‐FPBA) NPs in the presence of H₂O₂ significantly suppressed DCFH‐DA fluorescence, especially in the cells treated with P(DOPA‐FPBA) NPs. Further quantitative analysis demonstrated the fluorescence intensities of DCFH‐DA in AML‐12 cells exposed to PDA, P(L‐DOPA), and P(DOPA‐FPBA) NPs were determined to be 89.7%, 64.1%, and 33.9% relative to that of the cells treated with H₂O₂ only, respectively. Overall, the P(DOPA‐FPBA) NPs exhibited higher cell protection efficacies against oxidative stress compared with PDA and P(L‐DOPA) NPs. Consistent with our speculation, further quantitative flow cytometric results demonstrated that the apoptosis/necrosis rate of cells treated with H₂O₂ only was markedly higher (∼1.4‐fold, 2.5‐fold and 4.6‐fold) than that of PDA, P(L‐DOPA), and P(DOPA‐FPBA) NPs treatment groups, respectively (Figure 3E). Evidently, H₂O₂ induces peroxidation of DNA, protein, and lipids, triggering apoptotic cell death. In contrast, treatments with PDA, P(L‐DOPA), and P(DOPA‐FPBA) NPs alleviate oxidative stress by scavenging ROS, thus lowering the apoptosis/necrosis ratio. Furthermore, among these nanomaterials, P(DOPA‐FPBA) NPs demonstrated the greatest effectiveness in rescuing cells from oxidative stress‐induced apoptosis.

2.3. Treatment of HIRI

Giving the promising antioxidant effect of P(DOPA‐FPBA) NPs, we further investigated their potential protective role against HIRI in a murine model. The HIRI model was created by occluding 70% of the liver blood flow with an atraumatic clip for 1 h, followed by a 6 h restoration of normal blood flow. Mice were intravenously injected with P(DOPA‐FPBA), P(L‐DOPA), or PDA NPs (0.25 mg/kg) at 1 h before operation. Mice in the sham group were pre‐injected with PBS and surgery by similar surgical procedure but without occlusion of the liver vessels. Blood and liver samples were collected, and corresponding liver functions at 6 h after surgery were analyzed. The serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were determined to assess liver damage severity. Compared with the control and sham groups, the ALT and AST levels were significantly elevated in the HIRI group, indicating severe liver damage via occlusion of the liver vessels and restoration of normal blood flow (Figure 4A,B). However, treatment with P(DOPA‐FPBA), P(L‐DOPA) or PDA NPs significantly suppressed the levels of ALT and AST, demonstrating the effective protection of the tailored NPs against HIRI. Moreover, the mice treated with P(DOPA‐FPBA) NPs exhibited stronger suppression on the levels of ALT and AST in contrast to those in the mice treated with P(L‐DOPA) or PDA NPs, highlighting the excellent liver‐protecting function of P(DOPA‐FPBA) NPs. The H&E staining results of liver tissue were utilized to evaluate the therapeutic effectiveness of P(DOPA‐FPBA) NPs (Figure 4C). Severe liver tissue damages in HIRI mice treated with PBS was verified by the characterization including hepatocyte necrosis, cytolysis, and vacuolar degeneration. PDA treatment group slightly attenuated the liver damage of HIRI mice, while treatment of mice with P(L‐DOPA) or P(DOPA‐FPBA) NPs, significantly exhibited decreased cell necrosis and vacuolar degeneration in liver tissues, especially in the mice treated with P(DOPA‐FPBA) NPs. The histopathological features observed in H&E‐stained liver tissue sections were consistent with the changes in serum ALT and AST levels. To assess biosafety, P(DOPA‐FPBA) NPs were administered intravenously to healthy mice at the equivalent dosage as that employed in the in vivo HIRI experiment. Serum ALT, AST, and BUN levels in mice remained within normal ranges at both 1 day and 7 days post‐injection (Figure S11). H&E staining of major organs revealed no apparent pathological changes compared to control group, confirming the excellent in vivo biocompatibility of P(DOPA‐FPBA) NPs (Figure S12).

FIGURE 4.

(A) ALT (B) AST levels in the serum of mice in different experimental groups; (C) H&E staining images of liver tissues in different experimental groups; (D) Fluorescence images of liver tissues in different experimental groups after Hoechst 33342 and DHE staining; (E) SOD and (F) MDA levels in liver tissues of mice in different experimental groups. *p < 0.05, **p < 0.01, ***p < 0.001, compared with the control group; #p < 0.05, ##p < 0.01, ###p < 0.001, compared with the HIRI group.

HIRI is a pathophysiological event primarily driven by oxidative stress. Hence, the dihydroethidium (DHE) staining assay was conducted on mice in each treatment group to assess ROS levels (Figure 4D). The control and sham groups exhibited minimal DHE fluorescence due to the absence of ischemia‐reperfusion injury. In contrast, the HIRI group exhibited intense red fluorescence in damaged liver tissues, indicating the production of massive ROS including superoxide anions. The suppressed red fluorescence signals in damaged liver tissues were observed in mice treated with P(DOPA‐FPBA), P(L‐DOPA) or PDA NPs due to the ROS‐scavenging capabilities of these nanomaterials. Notably, the mice treated with P(DOPA‐FPBA) NPs exhibited lower red fluorescence signals in damaged liver tissues in contrast to those treated with P(L‐DOPA) or PDA NPs. This result confirmed the superior antioxidative properties of P(DOPA‐FPBA) NPs over P(L‐DOPA) and PDA NPs. Consistently, superoxide dismutase (SOD) activities in liver tissues were significantly suppressed and malondialdehyde (MDA) in liver tissues was elevated in the HIRI group compared to the control and sham groups (Figure 4E,F). Remarkably, P(DOPA‐FPBA) NPs treatment significantly restored the SOD activities and decreased MDA levels in HIRI‐affected liver tissues. Besides, Bio‐TEM imaging of liver sections confirmed the endocytic uptake of P(DOPA‐FPBA) NPs into pinocytotic vesicles within hepatic macrophages (Figure S13). As these macrophages are major producers of pro‐inflammatory cytokines and ROS during HIRI [43, 44], the efficient ROS‐scavenging capacity of the NPs enables their macrophage uptake to confer antioxidant effects and mitigate HIRI.

TdT‐mediated dUTP nick‐end labeling (TUNEL) staining further revealed that P(DOPA‐FPBA) NPs and P(L‐DOPA) NPs treatments both markedly reduced hepatocyte apoptosis as compared to the HIRI group (Figure 5A). However, PDA NPs showed negligible alleviation on HIRI‐induced hepatocyte apoptosis at the same administered dose. TEM imaging displayed severely disrupted inner mitochondrial membranes with broken and reduced cristae in the HIRI group, whereas the P(DOPA‑FPBA) NPs‑treated group maintained well‐organized cristae and clear inner membrane invaginations, suggesting effective protection against ROS‑mediated structural damage during ischemia‑reperfusion (Figure 5B). Consistently, changes in mitochondrial apoptosis‐related protein expression were also observed: BCL‐2 was down‐regulated (to 34% of the control level), while BAX and cleaved caspase‐3 (CC‐3) were up‐regulated (to 2.16‑fold and 2.21‑fold, respectively) in the HIRI group. However, treatment with P(L‐DOPA) and P(DOPA‐FPBA) NPs, particularly P(DOPA‐FPBA) NPs, reversed these trends (Figure 5C,D,E). In addition, we preliminarily investigated whether ROS scavenging affects other cell death pathways. As shown in Figure S14, HIRI led to a marked reduction in GPX4 expression (indicative of ferroptosis) and an increase in LC3 puncta (indicative of autophagy activation) [45, 46, 47, 48]. Notably, treatment with P(DOPA‐FPBA) NPs significantly restored GPX4 levels and attenuated LC3 aggregation. These results suggest that the ROS‐scavenging capacity of P(DOPA‐FPBA) NPs may contribute to the inhibition of both ferroptosis and autophagy in vivo [49, 50]. Collectively, these findings suggest that P(DOPA‐FPBA) NPs effectively alleviate the injury of HIRI via scavenging ROS levels and modulating antiapoptotic pathways in damaged liver tissue. The dynamics and ROS‐responsive cleavage of boronate–catechol linkages in P(DOPA‐FPBA) NPs realize the full utilization of the antioxidant activity of phenolic hydroxyl groups, providing significant therapeutic benefits for HIRI treatment.

FIGURE 5.

(A) TUNEL staining results of liver tissues of mice in different experimental groups, scale bar, 50 µm; (B) Representative TEM images showing the ultrastructure of mouse liver mitochondria from each group. Scale bars: 500 nm; Expression of mitochondrial apoptosis‐related proteins: (C) BCL‐2, (D) BAX and (E) Cleaved caspase‐3 in mouse liver tissues across experimental groups (n = 3) *p < 0.05, **p < 0.01, ***p < 0.001, compared with the control group; #p < 0.05, ##p < 0.01, ###p < 0.001, compared with the HIRI group.

Encouraged by the aforementioned results, we further conducted an immunofluorescence imaging investigation using macrophage‐specific marker F4/80 to examine the impact of P(DOPA‐FPBA) NPs on macrophage activation given that macrophages play a crucial role in initiating inflammation and inducing HIRI progression. As shown in Figure 6A, HIRI led to significant monocyte/macrophage staining compared to the control and sham groups. Similarly, the PDA treatment group exhibited negligible inhibition on the extent of monocyte/macrophage staining. On contrast, liver tissue samples from mice treated with P(L‐DOPA) and P(DOPA‐FPBA) NPs demonstrated a marked reduction in monocyte/macrophage staining, indicating that these NPs could inhibit monocyte/macrophage activation caused by HIRI. Besides, minimal activation of monocytes/macrophages was observed in liver tissue slices from P(DOPA‐FPBA) NPs‐treated group due to the superior ROS scavenging activitiy of P(DOPA‐FPBA) NPs compared to PDA and P(L‐DOPA) NPs.

FIGURE 6.

(A) F4/80 and Ly6G staining results of liver tissues of mice in different experimental groups, scale bar, 100 µm; (B) TNF‐α, (C) IL‐1β and (D) IL‐6 levels in liver tissues of mice in different experimental groups. *p < 0.05, **p < 0.01, ***p < 0.001, compared with the control group; #p < 0.05, ##p < 0.01, ###p < 0.001, compared with the HIRI group.

Furthermore, neutrophil infiltration was also immunostained utilizing the surface antigen Ly6G as the neutrophil marker. Consistently, P(DOPA‐FPBA) NPs treatment group also greatly suppressed the neutrophil recruitment, thereby preventing hepatocyte apoptosis compared to the HIRI and other treatment groups (PDA and P(L‐DOPA) NPs). These findings suggested that P(DOPA‐FPBA) NPs efficiently inhibited HIRI‐induced macrophage activation and neutrophil infiltration, resulting in the alleviative hepatocyte apoptotic process. To further investigate the anti‐inflammatory effects of these NPs, we measured inflammatory markers using ELISA. As shown in Figure 6B–D, liver tissues from the HIRI group exhibited markedly higher levels of tumor necrosis factor‐α (TNF‐α), interleukin‐1 beta (IL‐1β), and interleukin‐6 (IL‐6) compared to the control and sham groups (p < 0.001). However, treatment with P(DOPA‐FPBA) NPs significantly reversed these changes and exhibited greater reductions in TNF‐α, IL‐1β, and IL‐6 levels compared with the PDA and P(L‐DOPA) NPs treatment groups. It is well known that HIRI‐induced inflammation involves the activation and infiltration of pro‐inflammatory cells and oxidative stress‐related hepatocyte damage. Our results demonstrated P(DOPA‐FPBA) NPs with outstanding antioxidant abilities address HIRI by initially scavenging elevated ROS levels in ischemic liver tissues and attenuating oxidative stress, subsequently suppressing monocyte/macrophage activation, neutrophil infiltration, and pro‐inflammatory cytokine expression. Ultimately, the tailored P(DOPA‐FPBA) NPs mitigate the inflammatory response in the damaged liver, alleviate HIRI, and safeguard liver function.

3. Conclusion

L‐DOPA and 4‐FPBA could be polycondensed to obtain antioxidative nanoparticles through dynamic covalent chemistry. Since P(DOPA‐FPBA) NPs did not undergo the oxidation of phenolic groups like conventional eumelanin‐like materials during the preparation process, they exhibited better antioxidant capacity than PDA and P(L‐DOPA) NPs, as well as excellent performance in the treatment of HIRI. P(DOPA‐FPBA) NPs were expected to be a promising alternative to eumelanin‐like antioxidants as dynamic nanoscavenger, which could be further applied to a wide range of treatment scenarios. In addition, the presence of multiple carboxyl groups in L‐DOPA enables its transformation into diverse material forms (e.g., gels, elastomers) through Schiff base reactions and phenylboronic acid‐polyphenol interactions. It is anticipated that this work would provide new opportunities towards the design and preparation of melanin‐inspired functional materials.

4. Experimental Section

4.1. Chemicals and Materials

L‐DOPA (98%, Energy Chemical), 2‐formylphenylboronic acid (2‐FPBA, 98.0%, Energy Chemical), 3‐formylphenylboronic acid (3‐FPBA, 98.0%, Energy Chemical), 4‐formylphenylboronic acid (4‐FPBA, 98.0%, Energy Chemical), N,N‐Dimethylformamide (DMF, Chron Chemicals), phenylboronic acid (PBA, 99.0%, J&K Chemicals), benzaldehyde (99.0%, Energy Chemical), nickel chloride hexahydrate (99.0%, Energy Chemical), dopamine hydrochloride (DA·HCl, 98.0%, Energy Chemical), ammonia aqueous solution (28.0∼28.0%, Chron Chemicals), 30% hydrogen peroxide (H₂O₂, Aladdin), imidazole (99.0%, Adamas‐beta), 2,2‐diphenyl‐1‐picrylhydrazyl (DPPH, 97.0%, Tokyo Chemical Industry), Ammonium 2,2'‐(hydrazine‐1,2‐diylidene)bis(3‐ethyl‐2,3‐dihydrobenzo[d]thiazole‐6‐sulfonate) (ABTS, 98.0%, Tokyo Chemical Industry), potassium persulfate (98.0%, Chron Chemicals), ascorbic acid (Vc, 99.0%, J&K Chemicals). Bicinchoninic acid (BCA) Protein Assay Kit was purchased from Solarbio Science and Technology Co., Ltd. (Beijing, China). Resazurin was purchased from Aladdin Reagent Company (Shanghai, China). 2',7'‐dichlorodihydrofluorescein diacetate (DCFH‐DA) and Dihydroethidium (DHE) were obtained from Sigma‐Aldrich (St. Louis, MO). Hoechst 33342 was obtained from Beyotime Biotechnology Co. Ltd. (Shanghai, China). Calcein‐AM/PI Double Stain Kit and Annexin V‐FITC Apoptosis Detection Kit were purchased from Yeasen Biotechnology Co., Ltd. (Shanghai, China). Mouse IL‐6 ELISA Kit, Mouse IL‐1β ELISA Kit, and Mouse TNF‐α ELISA Kit were purchased from Jonlnbio Industrial Co., Ltd. (Shanghai, China). Antibodies specific for BCL‐2, BAX, and C‐caspase 3 were obtained from Proteintech Group (Wuhan, China). GPX4, LC3B, F4/80, and Ly6G antibodies, along with FITC‐ and Cy3‐conjugated secondary antibodies, were obtained from Servicebio Biotechnology Co. Ltd. (Wuhan, China). All other chemicals were of the highest reagent grade available.

4.2. Synthesis of P(DOPA‐FPBA) NPs

Equimolar amounts of L‐DOPA and formylphenylboronic acid were dispersed in DMF preheated to 85°C and stirred for 12 h under nitrogen atmosphere. Next, the reaction solution was cooled to room temperature with continuous stirring, and the reaction product was separated by centrifugation and washed three times with DMF and water. See Figure 1B for reagent dosage.

4.3. Synthesis of P(L‐DOPA) NPs

180 mg of L‐DOPA was dissolved in 85°C water, 4.39 mg of nickel chloride hexahydrate was added, and the mixture was heated and stirred for 20 h. After the reaction, the product was separated by centrifugation and washed three times with water.

4.4. Synthesis of PDA NPs

275 mg of DA·HCl was dissolved in a mixture of 100 mL of water and 40 mL of ethanol, and then 0.5 mL of ammonia was added, and the mixture was stirred at room temperature for 12 h. The reaction product was separated by centrifugation and washed three times with water.

4.5. H₂O₂ and pH Responsive Degradation Experiments

The nanoparticles were dispersed in H₂O₂ solution or PBS buffer (10 mM), and the amount of centrifugal product was measured regularly. The centrifugation condition was 15,000 rpm for 10 min, and extreme care was required during washing to avoid visible product loss. The residue ratio was defined as the mass ratio of the centrifugal product to the feed. The results were obtained from three parallel experiments. Owing to the complex nature of polyphenol oxidation products, conventional quantitative methods including HPLC were not adopted herein.

4.6. DPPH Scavenging Experiment

In DPPH scavenging testing, aliquots (3 mL) of the fresh DPPH ethanol solution (1 mM) were individually mixed with varying masses of nanoparticles, and then the absorbance at 517 nm of the ten‐fold diluted mixture was measured periodically to calculate the DPPH radical scavenging capacity of the nanoparticles. The results were given by three parallel experiments.

4.7. ABTS·⁺ Scavenging Experiment

ABTS·⁺ working solution was prepared by dissolving 19.2 mg ABTS and 4 mg potassium persulfate in 5 mL water, stirring in the dark (16 h, 25°C), followed by ten‐fold dilution. In ABTS·⁺ scavenging assays, 3 mL of the working solution was mixed with different amounts of nanoparticles, and then the absorbance at 734 nm of the ten‐fold diluted mixture was measured periodically to calculate the ABTS·⁺ scavenging capacity of the nanoparticles. The molar absorptivity coefficient of ABTS·⁺ used was adopted from a past study [51]. The results were obtained from three parallel experiments.

4.8. Cell Culture and Animal Studies

The mouse hepatocyte cell line AML‐12 cells (RRID:CVCL_0140) were purchased from the American Type Culture Collection (ATCC) and cultured in DMEM/F‐12 medium supplemented with 10% FBS, 1% ITS premix, 40 ng/mL dexamethasone. Cells were maintained at 37°C in a humidified atmosphere with 5% CO₂. Male C57BL/6J mice (18–20 g) were purchased from Liaoning Changsheng Biotechnology Co., Ltd. (Benxi, China) [SCXK (Liao) 2020‐0001]. Mice were housed at 25°C and 55% of humidity under natural light/dark conditions and were allowed free access to food and water. All animal procedures were performed according to the guidelines of Dalian Medical University, under an SPF‐level animal facility license [SYXK (Liao) 2018‐0007], and approved by the Institutional Animal Care and Use Committee (AEE20032).

4.9. Cell Viability

AML‐12 cells were seeded in 96‐well plates (8000 cells/well) overnight, followed by exposure to PDA, PDOPA, and PDOPAB NPs (5, 10, 20, 50, and 100 µg/mL) for 24 and 48 h. Subsequently, the cytotoxicity activities were assessed using resazurin assay by monitoring changes in fluorescence intensity (excitation/emission: 540/590 nm) of the culture medium over 2 h. In order to detect the antioxidant capabilities of tailored nanomaterials, AML‐12 cells were pretreated with PDA, P(L‐DOPA), and P(DOPA‐FPBA) NPs at different concentrations (5, 10, 20, 50, and 100 µg/mL) for 24 h, then rinsed with PBS, and cultured with H₂O₂ (300 µM) for 24 h. The control group was treated with H₂O₂ only. The viability of cells was assessed using resazurin assay as described above.

4.10. Live/Dead Cell Staining

AML‐12 cells were treated with PDA, P(L‐DOPA), and P(DOPA‐FPBA) NPs (20 µg/mL) for 2 h, followed by rinsing with PBS and incubation with H₂O₂ (300 µM) for another 6 h. At last, Calcein‐AM and PI were utilized to stain the cells for visualization of the live and dead/late apoptotic cells following the manufacturer's suggested protocol.

4.11. Intracellular ROS Measurements

The level of intracellular ROS was determined utilizing a fluorescent probe, DCFH‐DA. AML‐12 cells were seeded in 24‐well plates at a density of 3 × 10⁴/well for 24 h incubation. The cells were treated with PDA, P(L‐DOPA), and P(DOPA‐FPBA) NPs for 2 h prior to H₂O₂‐stimulation (300 µM). After another 6 h incubation, the cells were stained with DCFH‐DA (20 µM) for 30 min in the dark. The cells were rinsed with cold PBS to remove the excessive fluorescent probe. The fluorescence images were captured by an inverted fluorescence microscopy (LSM900, Zeiss). To quantitatively evaluate the level of intracellular ROS, The cells were pre‐incubated with tailored NPs (20 µg/mL) for 2 h. Then, the cells were treated with H₂O₂ (300 µM) and incubated for another 24 h. The cells were stained with DCFH‐DA as described above. The fluorescence intensity was determined at 490/525 nm (excitation/emission) utilizing a multifunctional microplate reader (Tecan, Switzerland). The values were normalized based on the protein content in each sample.

4.12. Cell Apoptosis Assay

AML‐12 cells were seeded into 6‐well plates (3 × 10⁵ cells/well) and incubated overnight. Subsequently, the cells were pretreated with diverse NPs (20 µg/mL) for 2 h, then rinsed with PBS and incubated with H₂O₂ (300 µM) for 24 h. Finally, the treated cells were harvested, rinsed twice with cold PBS, and stained with Annexin V‐FITC and PI, followed by being analyzed using a flow cytometer (Agilent NovoCyte Advanteon).

4.13. Preparation of HIRI Model in Mice

HIRI model was induced as follows: the male C57BL/6J mice were fasted for 12 h before the surgical operation. Subsequently, the mice were anesthetized, followed by being placed on a heated surgical pad. A midline laparotomy was performed, clamping the portal triad to the left and median liver lobes for 1 h, followed by reperfusion for 6 h by removing the clamp. In the sham group, the livers were exposed to the same surgical procedure except for vascular occlusion.

4.14. Therapeutic Efficacy in a Hepatic IRI Model

The mice were randomly divided into six groups (n = 5): (1) control group, (2) Sham group, (3) HIRI group, (4) PDA NPs + HIRI group, (5) PDOPA NPs + HIRI group, and (6) PDOPAB NPs + HIRI group. PDA, P(L‐DOPA), and P(DOPA‐FPBA) NPs (0.25 mg/kg) were intravenously injected 1 h before surgery. After 6 h reperfusion, the blood samples were collected to measure the serum levels of alanine transaminase (ALT) and aspartate transaminase (AST) according to the manufacturer's protocol. Finally, the mice were sacrificed, and the main organ tissues were collected for further analysis.

4.15. H&E Staining and TUNEL Staining

Main organs were fixed in 4% paraformaldehyde, paraffin‐embedded, and sectioned into 5 µm slices for H&E staining. Pathological slides were then imaged and observed with an optical microscope. Apoptosis in liver sections from each group was further assessed by TUNEL assay, following the manufacturer's instructions.

4.16. DHE Staining

Left liver lobes were cryopreserved in O.C.T., cryosectioned at −20°C, washed with TBST, stained with 20 µM DHE for 1 h, and then with Hoechst 33342 (10 µg/mL) for 10 min before fluorescence microscopy.

4.17. Antioxidant Indicators and Cytokines Measurements

Livers from each group were frozen at −80°C until assayed. Liver homogenates were then prepared and centrifuged at 4°C to obtain supernatants. SOD and MDA levels were evaluated according to the manufacturer's instructions. Cytokine levels (TNF‐α, IL‐1β, and IL‐6) in the supernatants were also measured by ELISA according to the manufacturer's protocol.

4.18. Immunofluorescence Staining

Paraffin‐embedded liver sections were deparaffinized with xylene and hydrated in graded ethanol series. The slides were treated with 3% H₂O₂ for 25 min, followed by blocking with 3% BSA for 30 min. Afterwards, the liver sections were immunostained using primary antibodies (F4/80 and Ly6G), followed by incubation with FITC and Cy3‐conjugated secondary antibodies. Nuclei were counterstained with DAPI, and sections were visualized using a fluorescence microscopy (BX53, Olympus).

4.19. Immunohistochemical Staining

After routine processing, liver sections were incubated with primary antibody (LC3B or GPX4) at 4°C overnight, then with HRP‐conjugated secondary antibody at RT for 50 min, followed by DAB, hematoxylin counterstaining, dehydration, mounting, and light microscopy.

4.20. Bio‐TEM Imaging of Liver Sections

Liver tissues were fixed in glutaraldehyde and osmium tetroxide, then dehydrated, infiltrated, and embedded in epoxy resin. Ultrathin sections were subsequently prepared and double‐stained with uranyl acetate and lead citrate for TEM observation.

4.21. Western Blotting Assay

Liver tissue proteins extracted using RIPA buffer were quantified by BCA assay and separated by 10% SDS‐PAGE. Proteins were then transferred to PVDF membranes, blocked, and incubated overnight at 4°C with primary antibodies. Membranes were subsequently incubated with Poly‐HRP‐labeled secondary antibody for 2 h at 37°C. Following TBST washes, protein bands were visualized using chemiluminescence detection on an Amersham Imager 600 system, with β‐actin as a loading control for band intensity normalization. Finally, western blot images were processed and analyzed by densitometry using the open‐source software Fiji (version 2.3.0) [52].

4.22. Statistical Analysis

Data were presented as mean ± standard deviation (SD). Statistical comparisons were determined by the analysis of variance (ANOVA) among ≥ 3 groups or Student's t‐test between 2 groups. P values less than 0.05 were considered to be statistically significant.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: smll74752‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.