Cu-Fe bimetallic peroxide-based nanozyme with microenvironment-triggered cascade catalysis for synergistic hydroxyl radical, nitric oxide, and oxygen generation in trimodal wound infection therapy

Abstract

Recently, chemodynamic therapy (CDT) and gas therapy (GT) have garnered significant attention for antibacterial applications. However, effectively integrating CDT and GT into a single nanocomposite for synergistic therapy remains challenging. Herein, Cu-Fe bimetallic peroxide nanoparticles (CFp)-decorated hollow polydopamine (HPDA) nanozyme (CFp/HPDA) was fabricated and used as the carrier for nitric oxide (NO) donor BNN6 loading, resulting in the NO-releasing nanozyme of CFp/HPDA@BNN6. In the acidic bacterial infection microenvironment (BIME), CFp could rapidly dissociate by releasing Cu⁺ and Fe²⁺ ions and self-supplying abundant H₂O₂. On one hand, the Fenton - like catalytic reaction enabled the generation of a high level of hydroxyl radical (•OH) for CDT. On the other hand, H₂O₂ could effectively trigger BNN6 to release NO for NO-based GT. Consequently, CFp/HPDA@BNN6 exhibited remarkable antibacterial and antibiofilm activities by exerting the bimodal GT/CDT therapy. In-depth mechanism study revealed that NO synergized the antibacterial effect of •OH by promoting bacterial ferroptosis-like death. Moreover, CFp could responsively produce a substantial amount of O₂ in the BIME, which, in combination with NO, synergistically promoted wound healing by reprogramming M1-type macrophages, downregulating hypoxia-inducible factor (HIF)-1α expression and upregulating vascular endothelial growth factor (VEGF), α-SMA and CD31 expressions. In summary, CFp/HPDA@BNN6 displayed highly efficient treatment of wound infections through the trimodal therapeutic effects of •OH, NO and O₂, offering a promising strategy for wound infection therapy.

Graphical abstract

In the present work, Cu-Fe peroxide bimetallic peroxide based nanozyme with BIME-responsive cascade catalytically releasing •OH, NO and O₂ properties was designed for synergistic trimodal therapy of wound infection. It displayed that the synergistic •OH/NO effect could effectively eliminate bacteria and the associated biofilm via the strengthened bacterial ferroptosis-like death. At the meantime, NO/O₂ synergistically promoted wound healing by repolarizing M1-type macrophages, downregulating HIF-1α expression and upregulating VEGF, α-SMA and CD31expressions.

Highlights

• •A Cu-Fe peroxide bimetallic peroxide based nanozyme was designed as effective NO-releasing carrier.
• •The nanozyme can cascade catalytically generate .•OH, NO and O₂ in response to BIME.
• •The nanozyme can effectively treat bacterial wound infections via trimodal therapeutic effects of .•OH, NO and O₂.
• •NO synergized the antibacterial effect of .•OH by promoting bacterial ferroptosis-like death.
• •NO/O₂ promoted wound healing by downregulating HIF-1α expression and upregulating VEGF, α-SMA and CD31expressions.

1. Introduction

Nowadays, human beings are facing threaten of bacterial infections, particularly in developing countries, bacterial infections display characters of higher morbidity and mortality [1]. Recent data showed that there were approximate 13.7 million infection-related deaths in 2019, particularly, the bacterial infection-induced death number accounted for 13.6 % of the global deaths, reaching a death number of 7.7 million [2]. Over the past century, antibiotics play a vital role in treating bacterial infections. However, the mistaken and abuse use of antibiotics lead to the development of drug-resistance bacteria, making it extremely difficult for effective treatment of bacterial infections. It is estimated that drug-resistance bacteria will kill 10 million people by the year of 2050 [3]. Therefore, it is of urgent demand to develop novel antibacterial alternatives with high effectiveness.

Recent work indicated that reactive oxygen species (ROS), including hydrogen peroxide (H₂O₂), superoxide anion (O₂^•-), singlet oxygen (¹O₂), hydroxyl radical (•OH), are effective in antibacterial applications [[4], [5], [6], [7]]. ROS can induce bacterial death by ways of breaking bacterial membrane, causing lipid peroxidation, disturbing bacterial metabolism, damaging bacterial DNA, etc [[8], [9], [10], [11]]. Therefore, many ROS-generating therapeutic methodologies have been developed for bacterial inactivation, such as chemodynamic therapy (CDT), photodynamic therapy (PDT), sonodynamic therapy (SDT), enzyme dynamic therapy (EDT), etc [[12], [13], [14]]. Among those ROS-based antibacterial strategies, CDT is the one that is being widely explored [12,15]. Typically, CDT employs Fe-based nanomaterials (e.g. Fe₃O₄) to generate •OH via the classical Fenton reaction between Fe²⁺ and H₂O₂ [5,16,17]. However, in the Fenton or Fenton-like reaction, the •OH generation ability of the Fenton materials is dependent on the levels of low valance state metal ion (e.g. Fe²⁺, Cu⁺) and H₂O₂. Unfortunately, it is very hard to reduce the high valance state metal ion to its low valance state form after the Fenton or Fenton-like reactions [18]. At the meantime, the H₂O₂ content in the bacterial infection microenvironment (BIME) is normally quite limited [17]. As a result, the current Fenton materials still face the insufficient •OH generation drawback, which greatly restricts the therapeutic efficiency of CDT. Therefore, developing synergistic therapeutic nanoplatform is essential to further improve the antibacterial efficiency of CDT.

Recently, nitric oxide (NO) based gas therapy (GT) has attracted enormous attention in antibacterial applications because of the multiple antibacterial mechanisms of NO and the least probability to induce bacterial drug resistance [[19], [20], [21]]. It has been demonstrated that NO can induce bacterial death by ways of damaging bacterial membrane, inhibiting bacterial respiration, cutting bacterial energy supply, breaking bacterial DNA damage, etc [22,23]. Besides, NO is considered as the excellent sensitizer in synergizing the therapeutic outcomes of many ROS-generating therapies like PDT, SDT and EDT [[24], [25], [26]]. Moreover, NO also possesses capability in promoting wound healing mainly by ways of regulating inflammatory microenvironment and inducing angiogenesis [[27], [28], [29]]. Therefore, we hypothesized that combination of NO-based GT with CDT can exerted the synergistic effect by further enhancing the antibacterial activity of CDT and providing additional wound healing effect to CDT.

For NO-based GT, NO control release remains as the big challenge. Up to know, many strategies have been developed for NO control release, generally including endogenous stimuli (e.g. acidic pH, H₂O₂, GSH)-responsive NO releasing way and exogenous stimuli (e.g. UV light, near infrared laser, X-ray)-responsive NO releasing way [19,30]. Previous work have demonstrated that applying external physical stimulation is a facile way for triggering precise NO control release. For instance, we fabricated several near infrared (NIR) laser-responsive NO-releasing carriers and demonstrated that NO releasing profiles could be precisely controlled by simply tuning the NIR laser density or irradiation way [27,[31], [32], [33], [34]]. However, the physical stimuli usually cause burst and short NO release, which is not satisfying for diseases need long term treatment. In addition, many physical stimuli like UV light and NIR laser displayed limited tissue penetration ability, which makes it hard to control NO release in deep disease tissues. In contrast, endogenous stimuli-responsive release way seems more appropriate for realizing steady and long-lasting NO release. Very recently, a novel H₂O₂-triggered NO releasing strategy was reported by us using N, N′-di-sec-butyl-N, N′-dinitroso-1, 4-phenylenediamine (BNN6) as the NO donor, where we showed that high level and long-lasting NO release could be achieved for promising antibacterial and wound healing applications [26].

Therefore, NO-based GT and CDT was integrated into a single nanoplatform for realizing high-efficient treatment on wound infections via the synergistic GT/CDT antibacterial and wound healing effect. On this purpose, Cu-Fe bimetallic peroxide nanoparticles (CFp) decorated hollow polydopamine (HPDA) nanozyme (CFp/HPDA) was fabricated and used as the carrier for BNN6 loading, obtaining NO-releasing nanozyme of CFp/HPDA@BNN6. Due to the hollow structure of CFp/HPDA, BNN6 can be effectively loaded in the nanozyme via the specific π-π stacking interaction force, showing a high BNN6 loading efficiency of 82 %. In acidic BIME, CFp of the nanozyme can rapidly dissociate by releasing Cu⁺ and Fe²⁺ ions and self-supplying substantial H₂O₂. At the one hand, high levels of •OH can be generated by the Fenton and Fenton-like catalytic reactions for CDT. Because of the particular Cu-Fe bimetallic property, CFp could always maintain Fe and Cu ions at lower oxidation states via the high-efficient catalytic loop, leading to the generation of high levels of •OH [35]. At the other hand, H₂O₂ can effectively triggered BNN6 to release NO for NO-based GT. As a result, CFp/HPDA@BNN6 displayed robust antibacterial and antibiofilm activities by exerting the bimodal GT/CDT therapy. The excellent antibacterial effects of CFp/HPDA@BNN6 was also verified on the rat model with S. aureus infection wound. In-depth mechanism study revealed that NO synergized the antibacterial effect of •OH by promoting bacterial ferroptosis-like death. Furthermore, dissociation of CFp in acidic condition can also resulted the production of O₂, which synergistically worked with the released NO to promote both in vitro and in vivo wound healing by downregulating hypoxia-inducible factor (HIF)-1α expression, upregulating vascular endothelial growth factor (VEGF), α-SMA and CD31 expressions. In all, CFp/HPDA@BNN6 displayed high treating efficiency of wound infections via the trimodal therapeutic effects of •OH, NO and O₂.

2. Experimental section

2.1. Materials

Dopamine hydrochloride and ferric chloride tetrahydrate (FeCl₂·4H₂O) were purchased from Shanghai MacLean Company. Tetraethyl orthosilicate (TEOS, 98 %) was purchased from Shanghai Macklin Reagent company. Copper (II) chloride dihydrate (CuCl₂·2H₂O), polyvinylpyrrolidone (PVP, average Mw 10000), potassium permanganate (KMnO₄, ≥99.5 %), Ammonium fluoride and Hydrofluoric acid were purchased from Tianjin Damao Chemical Reagent Co., Ltd. (Tianjing, China). Sodium hydroxide (NaOH), hydrochloric acid (HCl, 37 %), hydrogen peroxide (H₂O₂, 30 %), ammonium hydroxide (NH₃·H₂O, 28 %) and bovine serum albumin (BSA) were purchased from Aladdin (Shanghai, China). Anhydrous ethanol was purchased from Damao Chemical Reagent Factory (Tianjin, China). 5,5-Dimethyl-1-pyrroline-N-oxide (DMPO) was purchased from Japan Tongjin Chemical.

Strains of Escherichia coli (E. coli ATCC 25922) and Staphylococcus aureus (S. aureus ATCC 29213) were purchased from Guangdong Microbial Culture Collection Center (Guangzhou, China). 3,3′,5,5′-Tetramethylbenzidine detection (TMB) and Griess reagent was purchased from Shanghai Macklin Reagent company Lysogeny broth (LB) and LB agarose were purchased from Coolaber Technology Co., Ltd (Beijing, China). The terminal-deoxynucleotidyl transferase-mediated nick end labeling (TUNEL) assay kit and 2′, 7′-dichlorodihydrofluorescein diacetate (DCFH-DA) were obtained from Keygen Biotechnology Co., Ltd (Jiangsu, China). Nitric Oxide Assay Kit and Diaminofluorescein-FM diacetate (DAF-FM DA) (NO fluorescent probe) were purchased from Shanghai Biotech Co., Ltd., BBoxiProbe®O26 was purchased from Beibo Biotech Co., Ltd., and BODIPYTM581/591 C11 was purchased from Thermo Fisher Scientific (USA). The live/dead Baclight bacterial viability assay kit (L7012), fetal bovine serum (FBS), and Dulbecco's modified eagle's medium (DMEM) were bought from Thermo Fisher Scientific, Inc. (Massachusetts, USA). ThiolTrace Violet 500 was purchased from AAT Bioquest (Sunnyvale, CA, USA). Hoechst 33342, cell counting kit-8 (CCK-8) was purchased from Dojindo Laboratories (Kumamoto, Japan). Anti-Glutathione Peroxidase 4 was purchased from Abcam (Shanghai, China).

2.2. Synthesis of CFp/HPDA@BNN6

2.2.1. Preparation of hollow polydopamine nanoparticle (HPDA)

The synthesis of HPDA refereed to the previous synthetic method established by us [32]. In short, the Stöber method was firstly used to synthesize spherical silica particles (SiO₂), where 94.2 mL anhydrous ethanol, 16.2 mL pure water and 4.54 mL ammonia water were mixed and stirred for 10 min, and then 5.4 mL TEOS was added. The mixture was reacted in a 45 °C water bath for 3 h. After the reaction, the milky white precipitate was obtained by centrifugation (10000 rpm, 10 min) and washed twice with 30 mL mixed solution of water and ethanol (V_Ethanol: V_Water = 1:1), and the precipitate was collected to obtain spherical SiO₂ nanoparticles. Secondly, 350 mg dopamine hydrochloride and 120 mg/mL SiO₂ were mixed in 45 mL alkaline conditions (10 mM, pH 8.5 Tris buffer) at the mass ratio of 1:1 and allowed to react at the room temperature for 24 h to obtain polydopamine coated SiO₂ core-shell nanoparticle of SiO₂@PDA. As followed, 19 mL etching solution (1 mL hydrofluoric acid with 2 mg ammonium fluoride) was used to treat SiO₂@PDA with the etching solution/SiO₂@PDA ratio of 1 mL:10 mg for 24 h etching. Finally, HPDA was obtained by centrifugation (3500 rpm, 10 min) and lyophilization.

2.2.2. Preparation of CFp/HPDA nanozyme

The CFp/HPDA nanozyme was prepared using our previous method with some modifications [14]. Firstly, 0.3 g PVP and 0.5 g HPDA were added to a mixture containing CuCl₂·2H₂O (0.02 M) and FeCl₂·4H₂O (0.02 M) with a volume ratio of 7:3 and ultrasonically homogenized for 10 min. After stirring for 5 min, 1 mL sodium hydroxide solution (0.1 M) and 100 μL H₂O₂ (30 %) solution were added to the above mixture in sequence and stirred for 30 min. After that, the mixture was subjected to centrifugation (3500 rpm, 10 min), collected the precipitate, washed 3 times with ultrapure water and resuspended in 0.5 mL Tris buffer to obtain Cu-Fe bimetallic peroxide nanoparticles (CFp) decorated HPDA of CFp/HPDA.

2.2.3. Preparation of BNN6-loaded CFp/HPDA (CFp/HPDA@BNN6)

Briefly, 1 mL CFp/HPDA dispersion was centrifuged (3500 rpm, 10 min), the precipitate was collected and resuspended in 1 mL 400 μg/mL BNN6 solution (V_Ethanol:V_Water = 1:1), and magnetically stirred in the dark for 24 h. Then, the suspension was centrifuged (3500 rpm, 10 min), rinsed twice with deionized water and redispersed in 200 μL deionized water to obtain BNN6 loaded CFp/HPDA of CFp/HPDA@BNN6. At the meantime, control sample of HPDA@BNN6 was prepared using the similar steps as above.

2.3. Characterization

2.3.1. Transmission electron microscope (TEM) measurement

HPDA and CFp/HPDA (0.5 mg/mL) suspensions were ultrasonicated and 200 μL dispersion was dropped on a 200-mesh copper grid. After drying, the grid was placed under a transmission electron microscope (JEOL TEM-1210) to observe the material morphology at a voltage of 120 KV.

2.3.2. X-ray photoelectron spectroscopy (XPS)

The freeze-dried sample powder of CFp/HPDA was characterized by XPS apparatus (Thermo Scientific, Thermo Fisher, USA) for measuring the elemental valence state, where the test range is 0–1400 eV.

2.3.3. Dynamic laser scanning measurement

Took appropriate amount of HPDA, HPDA@BNN6, CFp/HPDA, CFp/HPDA@BNN6 and dispersed them in 1 mL ultrapure water to prepare the sample solution to be tested (three parallel measurements were set for each sample), transferred them to the sample pool in sequence, set the measurement temperature to 25 °C, and used a dynamic size analyzer (Zetasizer Nano ZS, Mallern) to measure the hydrodynamic size and Zeta potential values of the samples.

2.3.4. Elemental mapping

The elements contained in the nanozyme and their distribution were observed by TEM-Mapping technology. An appropriately diluted CFp/HPDA dispersion was dripped onto a copper mesh for naturally drying at the room temperature. After that, distribution of the contained elements on the sample was observed using a field emission transmission electron microscope with a voltage of 200 kV (JEOL-2100).

2.3.5. UV–Vis absorption spectroscopy measurements

The UV–Vis absorption spectrum of the samples was measured at room temperature. Briefly, an appropriate amount of HPDA, BNN6, CFp/HPDA, and CFp/HPDA@BNN6 solution or dispersion (0.5 mg/mL) was taken, transferred to the quartz cuvettes, placed in a UV–Vis spectrophotometer (Shimadzu) and recorded its absorption value at wavelength from 200 to 500 nm.

2.3.6. Fourier-transform infrared (FTIR) spectrum measurement

FTIR spectra of different samples were recorded on a FT-IR spectrometer (VERTEX70, Bruker, Germany) with the wavenumber range from 4000 to 400 cm⁻¹ at 4 cm⁻¹ resolution.

2.4. BNN6 loading efficiency determination

Firstly, BNN6 concentration-absorbance standard curve was established. Briefly, dissolved 5 mg BNN6 in 1 mL ethanol-water solvent (V_{water: ethanol} = 1:1) and diluted to have series concentrations of 0, 10, 15, 25, 50, 75, and 100 μg/mL. Measured the UV–Vis absorbance value of the BNN6 solution with different concentrations at 214 nm and plotted the concentration-absorbance standard curve of BNN6 (Y = 0.026 × X + 0.093). Then, collected 1 mL of the supernatant of CFp/HPDA@BNN6 by centrifugation (3500 rpm, 10 min), record its absorbance value at 240 nm, calculated the BNN6 content in 1 mL of supernatant according to the standard curve. Finally, the BNN6 loading efficiency was determined using the following formula (W is the weight):

$$\mathrm{L}\text{oading}\text{efficiency}(\%)=\frac{\mathrm{W}\left(\text{BNN}6\text{added}\right)-\mathrm{W}\left(\text{BNN}6\text{inthe}\text{supernatant}\right)}{\mathrm{W}\left(\text{BNN}6\text{added}\right)}\times 100\%$$

2.5. H₂O₂ generation detection

The H₂O₂ generation ability of CFp/HPDA@BNN6 at different pH values was verified by KMnO₄-based colorimetric assay. Briefly, 20 μL 1 M H₂O₂, 30 μL 1 M H₂O₂ and 50 μL 1 mg/mL CFp/HPDA@BNN6 (final materials concentration was 50 μg/mL) were added to 1 mL acidic KMnO₄ solution, where 1 mL acidic KMnO₄ solution was used as a blank control. After 10 min reaction, the UV–Vis absorption curve at 450–650 nm was measured by a UV–Vis spectrophotometer, and the color of the solution was photographed and recorded.

2.6. NO release detection

2.6.1. NO concentration standard curve

A NO concentration-absorbance standard curve (Y = 0.0.36 × X+0.058) was established by diluting sodium nitrite to a series of concentrations of 0 μM, 0.78 μM, 1.56 μM, 3.13 μM, 6.25 μM, 12.5 μM, 2 μM, 25 μM, 50 μM, 100 μM and then reacting with Griess reagent to measure the change in absorbance at 540 nm.

2.6.2. NO release detection

The ability of CFp/HPDA@BNN6 to release NO in pH = 5.4, 6.4 or 7.4 solutions was studied using the Griess assay. Briefly, 20 μL 1.5 mg/mL CFp/HPDA@BNN6 was added to 180 μL of pH = 5.4, 6.4, or 7.4 solution, and then Griess reagent was added. At each interval (10 min), 50 μL sample was taken out and mixed with 50 μL Griess I and 50 μL Griess II solution, and then the absorbance at 540 nm was detected using the microplate reader. Finally, the content of NO produced was calculated according to the NO standard curve.

2.6.3. H₂O₂-triggered NO release detection

Similar to the above protocols, five samples were detected by the Griess assay, which were PBS control, HPDA@BNN6, HPDA@BNN6 + 100 μM H₂O₂, CFp/HPDA and CFp/HPDA@BNN6 (150 μg/mL) in pH 6.4 condition. After that, the absorbance change at 540 nm at the corresponding time point was measured at room temperature using the UV–Vis spectrophotometer, and the generated NO concentration is calculated according to the standard curve.

2.7. •OH generation detection

2.7.1. TMB assay

The •OH production ability under different pH values was detected by the TMB assay. Briefly, mixed 100 μL CFp/HPDA@BNN6 (1.5 mg/mL) with 850 μL pH 5.4 or 6.4 or 7.4 buffer solution and 50 μL TMB (10 mM), reacted for 2 h under shaking and dark conditions. Then, used a UV–Vis spectrophotometer to measure the UV–Vis absorption curve of the above mixtures at the wavelength from 550 nm to 750 nm. The UV–Vis absorbance value at 650 nm reflected the •OH generation level of samples. To detect the time-dependent •OH generation behavior of CFp/HPDA@BNN6, UV–Vis absorbance value at 650 nm at intervals of 0 min, 10 min, 20 min, 30 min, 40 min, 50 min, 60 min were also detected.

2.7.2. Electron spin paramagnetic resonance (ESR) detection

The •OH generation was also detected using electron paramagnetic resonance spectrometer (Magnettech ESR5000). Briefly, 100 μL CFp/HPDA@BNN6 (1.5 mg/mL) and 850 μL pH 5.4 or 6.4 or 7.4 buffer solution containing 50 μL 50 μM DMPO were mixed and shaken for 10 min in the dark, and then tested using a ESR spectrometer.

2.8. O₂ generation detection

Briefly, 100 μL 100 μM H₂O₂ and 100 μL 1.5 mg/mLCFp/HPDA@BNN6 were mixed in 800 μL pH 5.4 PBS buffer, and immediately the concentration of dissolved oxygen in the solution was measured by a dissolved oxygen meter (JPB-607A) and recorded every 5 min, followed by plotting the time-O₂ concentration curve.

2.9. Antibacterial evaluation

2.9.1. Bacterial culture

First, disinfected and sterilized all items and reagents in the biosafety cabinet. Used a sterile pipette tip to draw a certain volume of LB liquid culture medium, added to the lyophilized powder of bacteria (E. coli ATCC 25922, S. aureus ATCC 29213) and incubated in a constant temperature shaker at 37 °C for 24 h. Then, diluted the bacterial suspension with sterile PBS and used a sterile inoculation loop to evenly spread the bacterial suspension on the LB solid agar medium, followed by incubating in a 37 °C biochemical incubator for 12 h. After that, used a sterile inoculation loop to pick out the single standard colony, inoculated it in LB liquid medium in a biosafety cabinet and cultured in the 37 °C constant temperature shaker for 12 h to the logarithmic phase for further use.

2.9.2. Antibacterial effect evaluation

2.9.2.1. Agar plate counting method

The antibacterial effects of CFp/HPDA@BNN6 were evaluated using the agar plate counting method. Briefly, E. coli and S. aureus in the logarithmic phase were centrifuged (6000 rpm, 5 min) and dispersed in 180 μL pH 5.4 PBS buffer solution with the final bacterial density of 1 × 10⁸ CFU/mL. Then, the bacterial suspensions were transferred to a 96-well plate, followed by adding 20 μL 1.5 mg/mL HPDA, CFp/HPDA, CFp/HPDA and CFp/HPDA@BNN6 with the final materials concentration of 150 μg/mL, where PBS buffer was used as the negative control. Three parallel experiments were repeated for each group. The plates were placed in a 37 °C constant temperature shaker for 4 h culturing. After that, the bacterial suspension was diluted to 1 × 10⁶ CFU/mL with pH 7.4 PBS buffer and spread on a sterile LB solid agar plate pretreated with UV irradiation. Again, the plates were placed in a 37 °C constant temperature incubator for 12 h culturing. Finally, photos were taken with a digital camera and the number of bacterial colonies was calculated using the ImageJ software to evaluate the antibacterial efficiency of different samples.

$$\text{Bacterial}\text{viability}(\%)=\frac{\text{Bacterial}\text{colony}\text{number}\left(\text{Sample}\right)}{\text{Bacterial}\text{colony}\text{number}\left(\text{Control}\right)}\times 100\%$$

2.9.3. SYTO-9/propidium iodide (PI) dual-staining assay

After the above treatments, bacterial suspensions were centrifuged (6000 rpm, 5 min), washed twice with PBS to remove materials and the suspended bacteria, resuspended in 200 μL PBS buffer containing 20 μL 50 μM SYTO-9 and PI fluorescent dyes and incubated at 37 °C in the dark for 60 min. After that, bacteria were washed with PBS to remove excess dyes, followed by transferring to a laser confocal dish for observation and photographing using the LSM880 confocal laser scanning microscope.

2.9.4. Antibiofilm effect evaluation

2.9.4.1. Confocal laser scanning microscope (CLSM) assay

Briefly, 100 μL E. coli and S. aureus in the logarithmic growth phase with OD₆₀₀ = 1.0 were added to the 96-well plate and placed in a constant temperature incubator at 37 °C for 48 h culturing. During the incubation process, the culture medium was replaced with fresh medium every 12 h to ensure complete bacterial biofilm formation. After that, 20 μL 1.5 mg/mL HPDA, HPDA@BNN6, CFp/HPDA and CFp/HPDA@BNN6 were added to treat the 180 μL biofilm and cultured in a biochemical incubator at 37 °C for another 12 h, where PBS buffer was used as the negative control. After that, biofilms were gently washed twice with sterile PBS to remove the materials and the suspended bacteria, followed by adding 200 μL 5 μM SYTO-9 and PI fluorescent dyes and incubated in the dark for 30 min. Finally, the bacterial biofilms were washed three times with PBS to remove the excess dyes, observed and imaged with an LSM 880 confocal laser scanning microscope.

2.9.4.2. Crystal violet staining assay

After above treatment, bacterial biofilms were cultured in a 37 °C biochemical incubator for 12 h. Then, 200 μL anhydrous methanol was added to fix the biofilm, followed by adding 200 μL 1 % crystal violet dye and incubated at 37 °C incubator in the dark for 30 min. After that, biofilms were washed several times with sterile PBS to remove the excess dye and photographed with a digital camera. Subsequently, 200 μL absolute ethanol was added to the wells and placed on a shaker for 15 min shaking until the crystal violet was completely dissolved. Finally, the optical density value at 570 nm (OD₅₇₀) was measured with a multifunctional microplate reader to evaluate the remanent biofilm biomass of each treatment.

2.10. Antibacterial mechanism evaluation

2.10.1. Bacterial morphology observation

Scanning electron microscopy (SEM) was used to observe the changes in bacterial morphology after materials treatment. Briefly, after bacteria were subjected to different treatments as above, bacterial suspensions were centrifuged (6000 rpm, 5 min) and washed several times with sterile PBS to remove the excess materials, followed by redispersing them in 2.5 % glutaraldehyde and placed in a 4 °C refrigerator for 12 h fixation. As followed, centrifuged (6000 rpm, 5 min) the above mixture to remove excess glutaraldehyde and dehydrated in sequence by a gradient 30 %, 50 %, 70 %, 80 %, 90 % and 100 % ethanol, each dehydration process lasted 10 min. Finally, bacteria were resuspended in ethanol, dripped on a single crystal silicon wafer, dried at the room temperature, subjected to gold spraying and observed with a SEM apparatus.

2.10.2. Intracellular NO release detection

NO fluorescence detection kit was used to detect the intracellular NO release level in bacteria, where DAF-FM DA was used as the fluorescent probe. Briefly, after bacteria were subjected to different treatments as above, bacteria were centrifuged (6000 rpm, 5 min) and washed twice to remove the excess materials and suspended bacteria. Then, 200 μL 10 μM DAF-FM DA fluorescent probe was added to the bacterial suspension (1 × 10⁸ CFU/mL) and incubated in a 37 °C incubator in the dark for 30 min. After centrifugation (6000 rpm, 5 min) and repeated washing with PBS, bacteria were redispersed in PBS and transferred to a laser confocal dish. The fluorescent image of the bacteria was observed using an LSM880 confocal laser scanning microscope, and the fluorescent intensity was analyzed with the ImageJ software.

2.10.3. Intracellular •OH generation detection

BBoxiProbe®O26 probe can detect intracellular •OH levels and produce green fluorescence by reacting with •OH. Firstly, 100 μL S. aureus in the logarithmic growth phase (OD₆₀₀ = 1.0) was subjected to different treatments of PBS buffer, HPDA, HPDA@BNN6, CFp/HPDA, and CFp/HPDA@BNN6 for 4 h. It is noteworthy that pH of the above suspensions was adjusted at 5.4, 100 μL 300 μg/mL materials were added. After that, samples were centrifuged (6000 rpm, 5 min) and washed twice with 200 μL PBS to remove excess materials and suspended bacteria. Then, 200 μL 10 μM BBoxiProbe®O26 probe were added to the bacterial suspensions and incubated in a 37 °C incubator in the dark for 30 min. As followed, bacterial suspensions were centrifuged (6000 rpm, 5 min), washed with PBS, dispersed bacteria in 200 μL PBS and dropped 20 μL into a laser confocal dish. The fluorescent image of the bacteria was observed using an LSM880 confocal laser scanning microscope, and the fluorescent intensity was analyzed with the ImageJ software.

2.10.4. Intracellular reactive oxygen species (ROS) detection

The ROS detection kit was used for intracellular ROS level detection, where DCFH-DA was used as the fluorescent probe. Briefly, after the above treatments, bacteria were redispersed in 180 μL PBS containing 20 μL 50 μM DCFH-DA and incubated in a 37 °C constant temperature shaker in the dark for 30 min. After washing the bacterial suspension several times with PBS, dispersed bacteria in 200 μL PBS and dropped 20 μL into a laser confocal dish, imaged and photographed by the LSM880 confocal laser scanning microscope, and the fluorescent intensity was analyzed with the ImageJ software.

2.10.5. Intracellular glutathione (GSH) detection

The GSH detection kit was used for intracellular GSH level detection, where ThiolTrace Violet 500 was used as the fluorescent probe. Briefly, after the above treatments, bacteria were redispersed in 200 μL PBS containing 10 μM ThiolTrace Violet 500 fluorescent dye, and incubated in a 37 °C constant temperature incubator in the dark for 30 min. After washing the bacterial suspension several times with PBS, dispersed bacteria in 200 μL PBS and dropped 20 μL into a laser confocal dish, imaged and photographed by the LSM880 confocal laser scanning microscope, and the fluorescent intensity was analyzed with the ImageJ software.

2.10.6. Lipid peroxidation (LPO) detection

Bacterial lipid peroxidation was evaluated using the lipid peroxidation fluorescent probe (C11-BODIPY^581/591). Briefly, after the above treatments, bacteria were redispersed in 180 μL PBS containing 20 μL 50 μM C11-BODIPY^581/591 fluorescent dye and incubated in a 37 °C constant temperature incubator in the dark for 30 min. After washing the bacterial suspension with PBS several times, dispersed bacteria in PBS and dropped 20 μL into a laser confocal dish, imaged and photographed by the LSM880 confocal laser scanning microscope.

2.10.7. Bacterial ferroptosis-like death-related proteins expression detection

Western blot experiments were used to detect expression levels of bacterial ferroptosis-like death-related proteins like Glutathione peroxidase 4 (GPX4), solute carrier family 7 member 11 (SLC7A11), long-chain acyl-CoA synthetase 4 (ACSL4) and transferrin receptor protein 1 (TFR1). Briefly, after the above treatments, centrifuged (6000 rpm, 5 min) to collect bacterial cells, followed by lysing for 50 min. After that, samples were centrifuged in a high-speed centrifuge (12000 rpm, 4 °C, 5 min), the supernatant was collected and the precipitate was discarded. Then, the protein content was determined by the BCA method and western blot experiments were performed. The samples were added to the buffer to keep the protein content of different groups consistent, boiled at 100 °C for 10 min, centrifuged at 1000 rpm for 1 min, and then added to the precast gel wells in turn. The concentrated gel was set to a constant voltage of 80 V for 30 min, and the separation gel was set to a constant voltage of 120 V for 30 min. After that, the precast gel was transferred to a methanol-activated polyvinylidene fluoride (PVDF) membrane with a constant current of 200 mA for 50 min. As followed, the PVDF membrane was taken out and blocked with bovine serum albumin (BSA) (1.25 g) dissolved in 25 mL Tris-buffer saline with Tween-20 (TBST) for 2 h, then placed on a shaker for 5 min and washed three times with TBST. Later on, anti-GPX4 (1:5000) antibody or anti-SLC7A11, anti-ACSL4, anti-TFR1 antibody was prepared and incubated with PVDF membrane on a shaker (80 rpm) overnight, and the membrane was washed with TBST 4 times for 5 min. The next day, Goat anti-rabbit antibody (1:10000) was prepared and incubated with PVDF membrane on a shaker for 2 h, and TBST was washed 3 times for 10 min. The incubated PVDF membrane was developed using a multifunctional imager.

2.10.8. Terminal-deoxynucleotidyl transferase-mediated nick end labeling (TUNEL) assay

The damage of bacterial DNA caused by the synergistic effect of NO and •OH was studied by the TUNEL staining assay. Briefly, after the above treatments, samples were washed with PBS, redispersed in 200 mL PBS containing 5 μM TUNEL dye, transferred to a laser confocal dish, imaged and photographed by the LSM880 confocal laser scanning microscope.

2.11. In vitro cell viability assay

The in vitro biocompatibility of CFp/HPDA@BNN6 on mouse embryonic fibroblasts (NIH/3T3) was evaluated by the CCK-8 assay kit. Briefly, NIH/3T3 cells were seeded in a 96-well plate at a density of 10⁴ cells/well and cultured with RPMI 1640 medium in the 37 °C for 12 h culturing. After that, the RPMI 1640 medium was removed and replaced with fresh medium containing series concentrations (0, 0.05, 0.1, 0.15 and 0.2 mg/mL) of HPDA@BNN6, CFp/HPDA, and CFp/HPDA@BNN6 for culturing another 24 h. Later on, cells were washed gently three times with PBS and added 100 μL complete medium containing 10 % CCK-8 (V/V, 1:1) for incubating in the 37 °C incubator and in the dark for 30 min. Finally, the UV–Vis absorbance value of the sample at 450 nm (A) was recorded using a multifunctional microplate reader and the cell viability was calculated based on the following formula:

$$\text{Cell}\text{viability}(\%)=\frac{\mathrm{A}\left(\text{Sample}\right)-\mathrm{A}\left(\text{blank}\right)}{\mathrm{A}\left(\text{Control}\right)-\mathrm{A}\left(\text{blank}\right)}\times 100\%$$

A (Sample) is A₄₅₀ value in the well with cells, CCK-8 reagents and sample. A (Control) is A₄₅₀ value in the well with cells and CCK-8 reagents and PBS. A (blank) is A₄₅₀ value in the well with only medium.

2.12. Cell scratch assay

Briefly, NIH/3T3 cells were plated in a 24-well plate (8 × 10⁴ cells per well) containing DMEM with 10 % FBS and incubated in a 37 °C incubator (5 % CO₂) for 12 h. Then, cells were washed 3 time with PBS and a ruler pipette tip was used to draw a straight line in the middle of the well plate to create the cell scratch. Later on, 1 mL fresh medium containing 150 μg/mL HPDA, CFp/HPDA, CFp/HPDA and CFp/HPDA@BNN6 was added and cultured in the 37 °C incubator (5 % CO₂). At intervals of 0 h, 12 h and 24 h, cells in each well were washed 3 times with PB and observed by the inverted fluorescence microscope (Nikon ECLIPSETi2-A).

2.13. Western blot experiments

HUVEC cells (2 × 10⁵ cells per well) were plated in a 6-well plate containing DMEM with 10 % FBS and incubated in a 37 °C incubator (5 % CO₂) for 12 h. After that, 1 mL fresh medium containing 150 μg/mL HPDA, CFp/HPDA, CFp/HPDA and CFp/HPDA@BNN6 was added and cultured in the 37 °C incubator (5 % CO₂) for 24 h. After that, cells were digested, lysed with radio immunoprecipitation (RIPA) protein lysis buffer and collected proteins according to the instructions of the assay kit. Then, the protein content was determined by the BCA method. The samples were added to the buffer to keep the protein content of different groups consistent, boiled at 100 °C for 10 min, centrifuged at 1000 rpm for 1 min, and then added to the precast gel wells in turn. The concentrated gel was set to a constant voltage of 80 V for 30 min, and the separation gel was set to a constant voltage of 120 V for 30 min. After that, the precast gel was transferred to a methanol-activated PVDF membrane with a constant current of 200 mA for 50 min. As followed, the PVDF membrane was taken out and blocked with BSA (1.25 g) dissolved in 25 mL tris-buffered saline with tween-20 (TBST) for 2 h, then placed on a shaker for 5 min and washed three times with TBST. Later on, vascular endothelial growth factor (VEGF) (1:5000) antibody and hypoxia-inducible factor (HIF)-1α antibody were added separately, incubated with PVDF membrane on a sh (80 rpm) overnight, and the membrane was washed with TBST 4 times for 5 min. The next day. Goat anti-rabbit antibody (1:10000) was prepared and incubated on a shaker for 2 h, and washed with TBST 3 times for 10 min. Finally, the incubated PVDF membrane was developed using a multifunctional imager.

2.14. In vivo antibacterial and wound healing effect evaluation

2.14.1. Establishment of S. aureus infected rat's wound model

Firstly, 20 female SD rats (6–8 weeks old, average weight 200 g) were selected and randomly divided into 5 different treating groups (PBS control, HPDA, CFp/HPDA, CFp/HPDA and CFp/HPDA@BNN6), with 4 rats in each group. Rats were isolated and quarantined before the experiment and raised in an SPF-level environment for 10 days. The animal experiments were conducted in accordance with the regulations of the Animal Experiment Ethics Committee of Jinan University (Approval number: 20221116-10). Then, rats were anesthetized by intraperitoneally injecting 10 % chloral hydrate (25 mg/kg), shaved the hair on rats' back, disinfected with cotton balls containing 75 % alcohol and used medical scissors to create a circular wound with a diameter of 1.5 cm on the back. As followed, 200 μL 1 × 10⁸ CFU/mL S. aureus suspension was added dropwise on rats’ wound for 2 days infection to create the S. aureus infected wound rat model.

2.14.2. Animal treatment

Every 2 days, 200 μL 150 μg/mL HPDA, HPDA@BNN6, CFp/HPDA and CFp/HPD@BNN6 was dropped on rats' wound for therapy, where 200 μL PBS was dropped in control groups. On day 0, 3, 5, 7 and 10, the exudate from the wound was collected, diluted with PBS, evenly spread on the LB agar plate, and incubated overnight at 37 °C in a biological incubator. Finally, used a digital camera to take photos of the bacterial colonies on each agar plate and used ImageJ to count the number of bacterial colonies for evaluating the in vivo antibacterial effect of each treatment. At the meantime, on day 0, 3, 5, 7 and 10, the rats’ wound in each group was measured with a precision steel ruler and photographed using a digital camera. The wound area of rats in each group was analyzed using the ImageJ software.

2.14.3. Histological analysis

After treatment, the skin tissue at the infected wound site was carefully removed with medical scissors, fixed with 4 % paraformaldehyde, and then embedded in paraffin to prepare histological sections. As followed, histological sections were stained with hematoxylin and eosin (H&E) and Masson's trichrome dyes and observed using a digital slice scanning and application system.

Myeloperoxidase (MPO) and CD68 were used to analyze the inflammation levels at the wound site, which were detected by performing the immunohistochemical staining assay. Briefly, histological sections were stained by specific MPO and CD68 antibodies. The immunohistochemical stained slices were observed and photographed using a digital slice scanning and application system.

Expression level of HIF-1α in rats’ wound tissue was detected by performing the immunohistochemical staining assay. Briefly, histological sections were stained by specific HIF-1α antibody, the stained slices were observed and photographed using a digital slice scanning and application system.

α-SMA and CD31 expressions were examined to evaluate the number of new born blood vessels on rats’ wound, which were detected by performing the immunofluorescence staining assay. Briefly, histological sections were stained by specific α-SMA and CD31 antibodies, the stained slices were observed and photographed under an inverted fluorescence microscope (Nikon ECLIPSETi2-A).

2.14.4. In vivo biocompatibility evaluation

2.14.4.1. Histological analysis

At the end of the treatment, the rats were euthanized, and the organs of heart, liver, spleen, lungs, and kidneys were collected, rinsed with PBS several times, fixed with 4 % paraformaldehyde, and then subjected to paraffin embedding and section. As followed, histological sections were stained by H&E and imaged using a digital slice scanning and application system.

2.14.4.2. Blood biochemical index analysis

At the end of the treatment, the rats’ blood were collected and placed in an anticoagulant blood collection tube for routine blood analysis.

2.15. Statistical analysis

All data in this work are presented as mean ± standard deviation, and statistical differences between different groups were analyzed using the One-way ANOVA method in GraphPad 8.3 software. The significance of differences was determined based on p values, where ∗ indicates p < 0.05, ∗∗ indicates p < 0.001, and ∗∗∗ indicates p < 0.0001.

3. Results and discussion

3.1. Synthesis and characterization of CFp/HPDA@BNN6

In the present work, a multifunctional nanozyme of CFp/HPDA@BNN6 with acid-triggered cascade catalytic releasing •OH, NO and O₂ properties was synthesized for robust antibacterial and wound healing applications via the trimodal therapeutic effects (Scheme 1). On this purpose, HPDA with a mean size of 303.5 nm (Fig. S1A and S1B) was firstly prepared using the template-etching method previously reported by us [14,32]. Then, CFp NPs were decorated on the surface of HPDA by performing the in situ synthetic method to prepare CFp/HPDA nanocomposite. In comparison with naked HPDA, CFp/HPDA displayed enhanced surface roughness and increased size at 323.5 nm (Fig. S1C and S1D), indirectly indicating the successful CFp decoration on HPDA's surface. Finally, NO donor BNN6 was loaded in CFp/HPDA through the specific π-π stacking interaction force, obtaining nanozyme of CFp/HPDA@BNN6 (Fig. 1A). The preparation process was monitored by dynamic light scattering measurements. The intensity-weighted hydrodynamic diameter of the nanozyme increased from the pristine 303.5 nm–342.8 nm after CFp decoration and BNN6 loading (Fig. 1B), while zeta potential increased from −6.13 mV to + 1.92 mV (Fig. 1C). Elemental mapping characterization was also conducted to confirm the successful preparation of CFp/HPDA@BNN6, as key elements of C, N, O, Cu, Fe could be clearly seen in the mapping images (Fig. 1D). To further confirm the BNN6 loading, UV–Vis spectroscopy measurements were conducted. As it was shown, BNN6 displayed a typical absorbance peak at 240 nm, while no typical peaks were found for HPDA and CFp/HPDA (Fig. 1E). As expected, CFp/HPDA@BNN6 displayed typical UV–Vis peak at 240 nm (Fig. 1E), demonstrating the presence of BNN6 in CFp/HPDA@BNN6. In the FTIR spectra of CFp/HPDA@BNN6, characteristic peak at 1367 cm⁻¹ assigned to N-N=O of BNN6 was observed (Fig. S2), further demonstrating the successful loading of BNN6. It was calculated that the BNN6 loading efficiency of CFp/HPDA reached at 82 %. Then, XPS was performed to study the element valence state of CFp NPs decorated on the nanozyme. The XPS survey spectra indicated the existence of C, N, O, Cu and Fe elements in the CFp/HPDA@BNN6 nanozyme (Fig. S3). In the deconvoluted O1s XPS spectra, peaks at 531.28 eV/533.48 eV assigned to metal-O and O-O groups were observed (Fig. 1F), demonstrating the presence of peroxy groups in the nanozyme. In the Cu2p XPS spectrum, two distinct peaks assigned to Cu 2p_3/2 and Cu 2p_1/2 orbitals were observed (Fig. 1G), suggesting the presence of Cu⁺ and Cu²⁺. The deconvoluted Cu 2p_3/2 and Cu 2p_1/2 peaks suggested that the amount of Cu⁺ in CFp was higher than that of Cu²⁺ (Fig. 1G). The Fe2p XPS spectrum also demonstrated the existence of Fe²⁺ and Fe³⁺, the deconvoluted Fe2p peaks also indicated the higher levels of Fe²⁺ in comparison with Fe³⁺ (Fig. 1H). The high levels of low valence state of Fe²⁺ and Cu ⁺ implying the excellent catalytic property of the nanozyme.

Scheme 1.

Schematic illustration of the synthesis, properties and wound infection treatment application of the nanozyme CFp/HPDA@BNN6. (A) Scheme shows the synthetic process of CFp/HPDA@BNN6. (B) Scheme shows the microenvironment-responsive cascade catalytically generating H₂O₂, •OH, NO and O₂ properties of CFp/HPDA@BNN6. (C) Scheme shows the antibacterial, wound healing effects and mechanisms of CFp/HPDA@BNN6.

Fig. 1.

Characterization of the nanozyme CFp/HPDA@BNN6. (A) TEM image of CFp/HPDA@BNN6. (B) Intensity-weighted hydrodynamic size distribution of HPDA, CFp/HPDA and CFp/HPDA@BNN6. (C) Zeta potential of HPDA, CFp, CFp/HPDA and CFp/HPDA@BNN6. (D) Elemental mapping image of CFp/HPDA@BNN6 showing distribution of C, N, O, Fe and Cu element. (E) UV–Vis spectra of BNN6, HPDA, CFp/HPDA and CFp/HPDA@BNN6. (F) Deconvoluted O1s XPS spectra of CFp/HPDA. (G) Deconvoluted Cu2p XPS spectra of CFp/HPDA. (H) Deconvoluted Fe2p XPS spectra of CFp/HPDA.

3.2. The acid-triggered H₂O₂ self-supply and cascade catalytic releasing •OH, NO and O₂ properties of CFp/HPDA@BNN6

In acidic environments, CFp of the nanozyme can rapidly dissociated by releasing Cu⁺ and Fe²⁺ ions and self-supplying substantial H₂O₂. Subsequently, •OH can be generated by the Fenton reactions or Fenton-like catalytic reactions between the Cu⁺/Fe²⁺ ions with H₂O₂. At the meantime, H₂O₂ can also effectively triggered BNN6 to release NO (Fig. 2A). Firstly, the H₂O₂ self-suppled property of CFp/HPDA@BNN6 was investigated using the KMnO₄-based colorimetric assay. In this assay, the pink color of the KMnO₄ can be reduced by H₂O₂ to colorless, and the characteristic UV–Vis absorbance peaks of KMnO₄ at the range of 500 nm–600 nm can be reduced as well. It showed that red color of the KMnO₄ solution did not change after CFp/HPDA@BNN6 treatment in pH 7.4 condition (Fig. S4A and S4B), neither was the UV–Vis spectra, implying that no H₂O₂ was produced in neutral pH environment. In contrast, red color of the KMnO₄ solution with CFp/HPDA@BNN6 treatment reduced in some degree in pH 6.5 condition (Fig. S4C and S4D), and obvious decrease in UV–Vis absorbance peaks ranging from 500 nm to 600 nm was observed, suggesting the generation of H₂O₂ in acidic environment. Remarkably, when pH was decreased to 5.4, CFp/HPDA@BNN6 addition turned the pink KMnO₄ solution to colorless and caused significant decrease in the characteristic UV–Vis absorbance peaks (Fig. 2B and Fig. S4E), indicating substantial H₂O₂ was generated in pH 5.4 condition. By semi-quantification, we determined that the concentration of H₂O₂ generated by 150 μg CFp/HPDA@BNN6 was about 10–20 mM (Fig. 2B), which is much higher than that in BIME [36]. Then, the •OH generation property of CFp/HPDA@BNN6 was detected by using the 3, 3′, 5, 5′-tetramethylbenzidine (TMB) assay. In this assay, the colorless TMB can be oxidized by •OH to its oxide form (ox TMB) showing blue color and displaying characteristic UV–Vis absorbance peak at 650 nm (Fig. 2C). As it was shown, TMB solution turned to blue by CFp/HPDA@BNN6 treatment in acidic pH (Fig. S5), typical UV–Vis absorbance peak at 650 nm appeared as well (Fig. 2D), suggesting •OH was generated by CFp/HPDA@BNN6 in acidic conditions. Moreover, we found a stronger UV–Vis absorbance peak at 650 nm in pH 5.4 condition as compared with that in 6.5 condition (Fig. 2D), implying CFp/HPDA@BNN6 was able to produce higher levels of •OH in pH 5.4 environment. A time-dependent •OH generation behavior was also observed for CFp/HPDA@BNN6 (Fig. 2E). The pH-dependent •OH generation property of CFp/HPDA@BNN6 was also confirmed by the electron spin resonance (ESR) characterization, where 5, 5-dimethyl-1-pyrroline-N-oxide (DMPO) was used as the trapping reagent. As it can be seen, obvious DMPO/•OH adduct signals could be visualized for CFp/HPDA@BNN6 in acidic conditions, and the sample in pH 5.4 condition displayed stronger signal peak than that in pH 6.4 condition (Fig. 2F).

Fig. 2.

Characterization of the H₂O₂, •OH, NO and O₂-generating properties of CFp/HPDA@BNN6. (A) Scheme shows the microenvironment-responsive cascade catalytically generating H₂O₂, •OH, NO and O₂ properties of CFp/HPDA@BNN6. (B) UV–Vis spectra of KMnO₄ solution with different treatments of PBS control, 10 mM H₂O₂, 20 mM H₂O₂ and CFp/HPDA@BNN6. (C) Scheme shows the mechanisms of the Griess assay and TMB assay for detecting NO and •OH, respectively. (D) UV–Vis spectra of TMB solution treated with CFp/HPDA@BNN6 in different pH conditions. (E) Increase in the UV–Vis absorbance value of TMB solution at 650 nm along with time in pH 5.4, 6.4 and 7.4, indicating the time-dependent •OH generation ability of CFp/HPDA@BNN6 in different pH conditions. (F) ESR spectra of different samples with the addition of DMPO as the trapping reagent. (G) UV–Vis spectra of different samples with addition of Griess reagents, showing the NO-releasing ability. (H) Time-dependent NO-releasing levels of CFp/HPDA@BNN6 in different pH conditions. (I) Time-dependent O₂ generation ability of CFp/HPDA@BNN6 in pH 5.4 condition.

As followed, the NO generation property of CFp/HPDA@BNN6 was evaluated using the commonly used Griess assay. In the presence of NO, the Griess reagent can be oxidized to the azo dye showing typical UV–Vis absorbance peak at 540 nm²⁷ (Fig. 2C). UV–Vis spectra showed that no typical peak was found for sample of HPDA@BNN6 (Fig. 2G), while obvious peak at 540 nm appeared with the addition of H₂O₂, indicating H₂O₂ is essential for triggering NO release. Similarly, typical peak at 540 nm could be found for CFp/HPDA@BNN6 in pH 6.4 condition, suggesting NO could be generated in acidic conditions because acid triggered CFp for H₂O₂ generation. Moreover, a time-dependent NO releasing profile was found for CFp/HPDA@BNN6 in acidic conditions (Fig. 2H). Previous work proved that Fenton reaction of CFp can also lead to the O₂ generation [35]. Therefore, the O₂-producing ability of CFp/HPDA@BNN6 was also studies by using the dissolved oxygen meter. Interestingly, a time dependent-O₂ generation profile was observed for CFp/HPDA@BNN6 in pH 5.4 conditions (Fig. 2I), suggesting the robust O₂-producing ability of CFp/HPDA@BNN6 in BIME-mimicking condition.

3.3. The synergistic NO/•OH antibacterial effect of CFp/HPDA@BNN6

Due to the simultaneous NO and •OH generation property of CFp/HPDA@BNN6, the synergistic NO/•OH antibacterial effect of the nanozyme against Gram-negative E. coli and Gram-positive S. aureus was investigated. Agar plate count assay showed that CFp/HPDA obviously reduced the colony numbers of both bacteria (Fig. 3A–C), which was assigned to the excellent antibacterial effect of the generated •OH. In contrast, HPDA@BNN6 only induced slight decrease in the colony number of S. aureus, this was because the very limited NO releasing level in H₂O₂-deficient condition (Fig. S6). Remarkably, CFp/HPDA@BNN6 treatment displayed the most significant inhibitory effect on both E. coli and S. aureus due to the synergistic NO/•OH antibacterial effect, leading to the lowest level of live bacteria after treatment (Fig. 3A–C). Then, the SYTO-9/PI dual-florescent staining assay was also performed to confirm the antibacterial effect of CFp/HPDA@BNN6. In this assay, dead bacteria were stained by the red fluorescent PI dye, while live bacteria were stained by the green fluorescent SYTO-9 dye. Consistently, CFp/HPDA@BNN6 treatment led to the highest red/green fluorescence ratio (Fig. 3D–F), confirming the robust antibacterial efficacy of the synergistic NO/•OH effect. In addition, we also observed that the antibacterial efficiency of CFp/HPDA@BNN6 was concentration-dependent (Fig. 3G–I). Particularly, the synergistic NO/•OH antibacterial effect of CFp/HPDA@BNN6 was able to completely inhibited the growth of both bacteria at the materials concentration of 200 μg/mL. Later on, the anti-biofilm activity of CFp/HPDA@BNN6 was also evaluated by performing the three-dimensional confocal laser scanning microscopy (3D CLSM) measurement and the crystal violet staining assay. 3D CLSM images showed that bacterial biofilms were largely eradicated by CFp/HPDA, while CFp/HPDA@BNN6 exerted the most severe damaging effect on both types of bacterial biofilms by inducing the strongest red fluorescent intensity (Fig. 3J and Fig. S7). Crystal violet staining assay also displayed the same trends (Fig. S8). Interestingly, biofilms of E. coli and S. aureus could be completely eradicated by the synergistic NO/•OH effect of CFp/HPDA@BNN6 at the concentration of 200 μg/mL.

Fig. 3.

Evaluation on the antibacterial and anti-biofilm effect of CFp/HPDA@BNN6. (A) Agar plate images showing the antibacterial activity of different treatments, and the corresponding histograms showing the antibacterial efficiency of different treatments against (B) E. coli and (C) S. aureus. (D) CLSM images of E. coli and S. aureus stained with SYTO-9/PI dual fluorescent dyes after different treatments (scale bar represents 10 μm), and the corresponding histogram showing the live/dead percentage of (E) E. coli and (F) S. aureus. (G) Agar plate images showing the antibacterial activity of CFp/HPDA@BNN6 with different concentration, and the corresponding histograms showing the antibacterial efficiency of CFp/HPDA@BNN6 against (H) E. coli and (I) S. aureus. (J) CLSM images of S. aureus biofilm stained with SYTO-9/PI dual fluorescent dyes after different treatments. Scale bar represents 100 μm.

3.4. The synergistic NO/•OH antibacterial mechanism of CFp/HPDA@BNN6

As followed, the synergistic NO/•OH antibacterial mechanism of CFp/HPDA@BNN6 was investigated. Firstly, the intracellular •OH and NO release inside bacteria was detected using the BBoxiProbe®O26 and DAF-FM DA probes, respectively. CLSM images indicated that high levels of intracellular •OH and NO could be generated by CFp/HPDA@BNN6 (Fig. 4A, B and 4C). Previous work indicated that •OH can effectively cause bacterial death through the ferroptosis-like pathway [37]. At the meantime, NO is considered as an ideal sensitizer to enhance the antibacterial efficiency of many therapeutic ways like photothermal therapy, photodynamic therapy, sonodynamic therapy, etc [25,31,38]. Herein, we aimed to study if NO can promote the bacterial ferroptosis-like inducing effect of •OH. DCFH-DA stained CLSM images displayed that CFp/HPDA@BNN6 treatment led to intracellular ROS boost (Fig. 4A and D), which is likely to cause LPO of bacterial membrane due to strong oxidizing nature of ROS. To confirm it, specific C11-BODIPY^581/591 probe was used for LPO detection. Strong green fluorescence was visualized in bacteria with CFp/HPDA@BNN6 treatment (Fig. 4E and Fig. S9), suggesting high levels of LPO was induced by the synergistic NO/•OH effect. TEM images showed that bacterial membrane was severely destroyed by the synergistic NO/•OH effect of CFp/HPDA@BNN6 (Fig. S10), further confirming the strong LPO-inducing effect of the nanozyme. Intracellular ROS burst generation can also cause the GSH depletion and DNA damage. ThiolTrace Violet 500 fluorescent probe stained CLSM images displayed that intracellular GSH was almost completely depleted by the CFp/HPDA@BNN6 treatment (Fig. S11). TUNEL assay indicated that bacterial DNA was severely damaged by CFp/HPDA@BNN6 (Fig. S12). GPx4 is the selenoperoxidase playing a vital role in antiperoxidant defence, GPx4 downregulation is the main character of bacterial ferroptosis-like death [39]. Hence, western blot assay was conducted to detect the GPx4 expression levels after CFp/HPDA@BNN6 treatment. Western blot results exhibited that GPx4 expression level was indeed largely reduced by •OH, and the synergistic NO/•OH effect showed the most significant downregulating effect on the GPx4 expression (Fig. 4F and G). Moreover, expression levels of ferroptosis-like related proteins of SLC7A11, ACSL4 and TFR1 were also detected by western blot analysis. Similarly, it showed that the synergistic NO/•OH effect of CFp/HPDA@BNN6 significantly downregulated SLC7A11, ACSL4 and TFR1 expressions (Fig. 4F, H, 4I and 4J). Collectively, we demonstrated that the synergistic NO/•OH effect of CFp/HPDA@BNN6 could effectively kill bacteria by inducing bacterial ferroptosis-like death, where NO synergized the bacterial ferroptosis-like induing effect of •OH.

Fig. 4.

Revealing the antibacterial mechanism of CFp/HPDA@BNN6. (A) CLSM images of bacteria stained by BBoxiProbe®O26, DAF-FM DA and DCFH-DA for detecting intracellular •OH, NO and ROS, respectively, and the corresponding histograms showing the intracellular (B) •OH, (C) NO and (D) ROS generation levels of different treatments. (E) CLSM images of bacteria stained by C11-BODIPY^581/591 probe for detecting the LPO effect of different treatments. (I) Western blot results showing the expression level of GPX4, SLC7A11, ACSL4 and TFR1 in different treating groups, and the corresponding histogram showing the relative expression levels of (G) GPX4, (H) SLC7A11, (I) ACSL4 and (J) TFR1. Labels of a-e represent (a) PBS control, (b) HPDA, (c) HPDA@BNN6, (d) CFp/HPDA and (e) CFp/HPDA@BNN6. Scale bar represents 10 μm.

3.5. The in vitro wound healing effect of CFp/HPDA@BNN6

In addition to its antibacterial activity, NO also possesses capability of promoting wound healing [26,40]. Besides, O₂ has been also demonstrated for effective wound healing by alleviating the hypoxia microenvironment of the bacterial infection site [41]. Herein, the in vitro wound healing effect of CFp/HPDA@BNN6 was investigated. Firstly, the in vitro biocompatibility of CFp/HPDA@BNN6 on NIH/3T3 cells was examined using the cell counting kit (CCK)-8 assay. It showed that no obvious toxic effect was observed for CFp/HPDA@BNN6, cell viability was higher than 75 % even at materials concentration of 200 μg/mL (Fig. 5C). Then, the in vitro wound healing activity of CFp/HPDA@BNN6 was evaluated using the commonly used cell scratch wound model. Fluorescent microscope images showed that HPDA@BNN6 could largely decrease the cell scratch gap (Fig. 5A and B), which must be due to the wound healing effect of the released NO, although limited NO was released from HPDA@BNN6 in H₂O₂-lacking condition. CFp/HPDA can also induce cell migration by reducing the cell scratch gap, which was attributed to the wound healing of the generated O₂. Remarkably, CFp/HPDA@BNN6 exhibited the most excellent wound healing effect by leading to the least cell scratch gap (Fig. 5A and B), thanks to the synergistic NO/O₂ wound healing effect. Previous work indicated that angiogenesis is essential in the wound healing process. As the key biomarkers of angiogenesis, vascular endothelial growth factor (VEGF) and hypoxia-inducible factor (HIF)-1α play important roles in regulating revascularization and cell migration [42,43]. Therefore, expression levels of VEGF and HIF-1α upon CFp/HPDA@BNN6 treatments were detected by performing western blot assay. Western blot results indicated that CFp/HPDA@BNN6 significantly upregulating the VEGF expression level and downregulating the HIF-1α expression level (Fig. 5D, E and 5F).

Fig. 5.

Revealing the in vitro wound healing effect and mechanism of CFp/HPDA@BNN6. (A) Scratch assay of NIH/3T3 cells underwent different treatments and (B) the quantitative analysis of relative scratch area at 24 h. Scale bar represents 25 μm. (C) Cell viability of NIH/3T3 cells with treatment of HPDA, CFp/HPDA and CFp/HPDA@BNN6 at different concentrations. (D) Western blot results showing the expression level of VEGF and HIF-1α in different treating groups, and the corresponding histogram showing the relative expression levels of (E) VEGF and (F) HIF-1α. Labels of a-e represent (a) PBS control, (b) HPDA, (c) HPDA@BNN6, (d) CFp/HPDA and (e) CFp/HPDA@BNN6.

3.6. The in vivo antibacterial and wound healing effect of CFp/HPDA@BNN6

Inspired by the excellent in vitro antibacterial and wound healing effect of CFp/HPDA@BNN6, the in vivo therapeutic effect of the nanozyme on rat's bacterial infected wound was investigated (Fig. 6A). During the treating period, digital images showing the wound status were recorded and bacterial number in the wound site was monitored using the agar plate count method. As it was shown (Fig. 6B and C), HPDA@BNN6 obviously promote wound healing in the initial 7 days due to the wound healing effect of the released NO. However, because of the insufficient antibacterial activity of HPDA@BNN6 (Fig. 6E and Fig. S13), large wound area remained unhealed on Day 10. In contrast, CFp/HPDA treatment further accelerated the wound healing by leaving less than 10 % unhealed wound area on Day 10, which can be attributed to the combinational therapeutic effects of •OH and O₂. Remarkably, due to the trimodal therapeutic effects of NO, •OH and O₂, CFp/HPDA@BNN6 exhibited the most excellent wound healing activity by almost completely healed the wound. Agar plates displayed that numerous bacterial colonies remained in the HPDA@BNN6 treated wounds (Fig. 6D and E), which was due to the limited generation level of NO. Bacterial colony number in the CFp/HPDA treated groups reduced obviously, demonstrating the excellent in vivo antibacterial effect of •OH. Strikingly, bacteria were completely eliminated by the synergistic NO/•OH antibacterial effect of CFp/HPDA@BNN6. Histological analysis with hematoxylin and eosin (H&E) staining and Masson staining were also performed to confirm the therapeutic effect of CFp/HPDA@BNN6. H&E-stained histological slices showed that intact epidermis was regenerated after 10 days CFp/HPDA@BNN6 treatment (Fig. 6F). Masson-stained histological slices showed that CFp/HPDA@BNN6 treatment led to the highest collagen deposition level (Fig. 6G).

Fig. 6.

Evaluation on the in vivo therapeutic effect of CFp/HPDA@BNN6. (A) Scheme shows the in vivo treating process of CFp/HPDA@BNN6. (B) Digital photos and Adobe Illustrator software-simulated diagrams show the wound healing process along with time with different treatments. (C) Histogram shows the wound healing efficiency of different treatments in different treating days. (D) Histogram shows the remanent S. aureus colony number in rats' wound after 10 days different treatments. (E) Agar plate images show the bacterial colony number in different treating groups. (F) H&E-stained histological images of wounds after 10 days different treatments. (G) Masson-stained histological images of wounds after 10 days different treatments.

Because inflammation alleviation and angiogenesis were important for promoting wound healing [[44], [45], [46]], the anti-inflammation and angiogenesis-inducing effect of CFp/HPDA@BNN6 were also investigated. Myeloperoxidase (MPO) is a key element of the innate immune system, which plays an important role in defending against bacterial invasion [47]. MPO accumulation is a sign of bacterial infection and inflammation response. Immunohistochemical analysis showed that high level of MPO was expressed in the control group (Fig. 7A and E), suggesting severe inflammation response occurred in rats' wounds after S. aureus infection. After CFp/HPDA@BNN6 treatment, expression level of MPO was significantly reduced (Fig. 7A and E), suggesting the excellent anti-inflammation effect of CFp/HPDA@BNN6. Improvement in inflammation status after CFp/HPDA@BNN6 treatment was also evidenced by the decreased expression levels of lymphocytes (LY), white blood cells (WBC) and neutrophils (NEUT) in rats’ serum (Fig. 7B, C and 7D). To elucidate the anti-inflammation of CFp/HPDA@BNN6, expression level of CD68 (the biomarker of M1-type macrophages) was also studies using the immunohistochemical analysis. It showed that CD68 expression level was significantly reduced by CFp/HPDA@BNN6 (Fig. 7A and F), suggesting that CFp/HPDA@BNN6 alleviated inflammation by reprogramming M1-type macrophages repolarization. To evaluate the angiogenesis-inducing effect of CFp/HPDA@BNN6, HIF-1α expression level in wound tissue was also detected by immunohistochemical analysis. Similar to the above in vitro results, HIF-1α expression was significantly downregulated by the CFp/HPDA@BNN6 treatment (Fig. 7A and G). In addition, α-SMA and CD31 dual-immunofluorescence staining assay were performed to evaluate the revascularization activity of CFp/HPDA@BNN6. Red (α-SMA) and green (CD31) fluorescence colocalization levels indicated the new born blood vessel numbers. Obviously, CFp/HPDA@BNN6 treatment resulted in the highest new born blood vessel numbers (Fig. 7H and I). Lastly, expression of angiogenesis-related proteins like VEGF, α-SMA, and CD31 on wound tissue were also evaluated by western blot analysis. It showed that expression levels of VEGF, α-SMA, and CD31 were all significantly upregulated by the treatment of CFp/HPDA@BNN6 (Fig. S14). Collectively, these results demonstrated the excellent angiogenesis-promoting effect of CFp/HPDA@BNN6 due to the trimodal therapeutic effects of NO, •OH and O₂.

Fig. 7.

Evaluation on the in vivo therapeutic mechanism of CFp/HPDA@BNN6. (A) Immunohistochemical images of CD68, MPO and HIF-1α in rats' wound tissues after 10 days different treatments, and the corresponding semi-quantitative expression levels of CD68 (E), MPO (F) and HIF-1α (G). Blood chemistry parameters levels of (B) lymphocyte (LY), (C) white blood cell (WBC) and (D) neutrophil counts (NEUT) in different treating groups after 10 days treatments. (H) Immunofluorescent CD31/α-SMA stained images of rat's wound tissues after 10 days different treatments and the semi-quantitative analysis on the number of blood vessels in wounds (I). Scale bar represents 100 μm.

Taken together, we revealed that CFp/HPDA@BNN6 could effectively treat S. aureus infected wound by exerting the trimodal therapeutic effects of NO, •OH and O₂, which synergistically eliminated the infected bacteria and remodeled wound inflammatory microenvironment by reprograming M1-type macrophages repolarization to the anti-inflammatory M2-type. Finally, wound healing process was promoted by downregulating HIF-1α expression and upregulating VEGF, α-SMA and CD31 expressions (Scheme 1C).

3.7. Biocompatibility evaluation of CFp/HPDA@BNN6

The in vivo biocompatibility of CFp/HPDA@BNN6 was evaluated to ensure its safe biomedical applications. H&E-stained histological slices showed that no obvious lesions could be observed on rat's main organs of heart, liver, spleen, lung and kidney after 10 days treatment (Fig. S15), indicating the excellent histocompatibility of CFp/HPDA@BNN6. Blood biochemical index testing results showed that rats' routine indexes of platelet distribution width (PDW), red blood cell (RBC), mean corpuscular volume (MCV), mean corpuscular hemoglobin concentration (MCHC), mean corpuscular hemoglobin (MCH), hemoglobin (HGB), hematocrit (HCT), monocyte percentage (Mon), mean platelet volume (MPV) all kept at the normal levels as the healthy rats (Fig. S16), suggesting the excellent blood compatibility of CFp/HPDA@BNN6. In addition, rats' weight in each group maintained at the relatively stable value of approximate 250 g during the whole treating period (Fig. S17). All these demonstrated the excellent in vivo biocompatibility of CFp/HPDA@BNN6.

4. Conclusions

In the present work, biocompatible nanozyme of CFp/HPDA@BNN6 was rationally designed and synthesized. In acidic BIME, CFp/HPDA@BNN6 could effectively decompose and simultaneously generate •OH, NO and O₂ for trimodal therapy of wound infections. At the one hand, the synergistic •OH/NO antibacterial effect of CFp/HPDA@BNN6 was able to high-efficiently inhibit bacterial activity by inducing bacterial ferroptosis-like death. Particularly, NO could synergize the bacterial ferroptosis-like induing effect of •OH. At the other hand, the synergistic NO/O₂ effect of CFp/HPDA@BNN6 could effectively promote wound healing by upregulating the VEGF expression and downregulating HIF-1α expression. As a result, CFp/HPDA@BNN6 displayed excellent power in treating in vivo wound infections via the simultaneous antibacterial, antiinflammation and wound healing activities. The findings of our work provide a potential nanozyme for high-efficient treatment of bacterial wound infections.

CRediT authorship contribution statement

Wei Liu: Validation, Investigation, Formal analysis, Data curation. Yuan Fang: Validation, Investigation, Formal analysis, Data curation. Ping Xu: Supervision, Funding acquisition, Formal analysis. Wanqin Cai: Data curation. Yunping Peng: Supervision, Funding acquisition. Wei Xue: Supervision, Resources, Funding acquisition. Siming Yu: Writing – review & editing, Writing – original draft, Supervision, Resources, Funding acquisition, Conceptualization.

Declaration of competing interest

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.

Appendix A. Supplementary data

The following is/are the supplementary data to this article:

Data availability

Data will be made available on request.