Multifunctional nano-delivery system based on DNase I and photodynamic therapy for combatting enterococcus faecalis biofilm infections

Abstract

Persistent or refractory apical periodontitis is primarily caused by microbial retention, as conventional root canal treatment often fails to eliminate infections completely, and systemic antibiotic therapy is insufficient to achieve effective concentrations for eradicating bacterial biofilms within root canals. This highlights the urgent need for novel therapeutics offering safe and effective antimicrobial strategies. Antimicrobial photodynamic therapy (aPDT) is a promising approach for root canal disinfection. However, commonly used photosensitizers such as Ce6 suffer from poor water solubility and strong aggregation tendencies, resulting in limited penetration into infected sites. In this study, we developed a DNase I-Lip@Ce6 nanodelivery system by combining deoxyribonuclease I (DNase I) with liposome-encapsulated Ce6. The liposomal carrier facilitated efficient delivery of Ce6 into target bacterial cells, while DNase I degraded extracellular DNA in the biofilm matrix, weakening its protective barrier. This synergistically enhanced Ce6 penetration and therapeutic efficacy, leading to the successful eradication of planktonic Enterococcus faecalis and in vitro biofilms. This strategy offers a novel approach for the precision treatment of persistent oral infections and holds strong potential for clinical translation.

Introduction

Periapical infections are common dental diseases, typically caused by bacterial infections within the root canal¹. Among the microbial pathogens involved, Enterococcus faecalis has been identified as one of the most prevalent species in persistent or recurrent apical periodontitis, with detection rates reaching up to 70% in previously treated cases². Within the root canal system, E. faecalis has the ability to form resilient biofilms and exhibits intrinsic resistance to harsh environmental conditions and antimicrobial agents, making it a major challenge in endodontic treatment^3,4,5. Once biofilm formation occurs, the embedded bacteria can become thousands of times more resistant to antibiotics than planktonic bacteria^6,7. This increased resistance is primarily due to the biofilm matrix, which impedes the penetration of antimicrobial drugs. Moreover, even if antimicrobial agents manage to kill some bacteria, the remaining dead bacteria provide attachment sites for new bacteria, facilitating the regeneration of the biofilm^8,9. As a result, the degradation of the biofilm matrix has emerged as a promising strategy to enhance the effectiveness of treatment.

Currently, common root canal treatments include mechanical preparation and chemical irrigation, but their effectiveness is significantly limited by the complexity of the root canal system¹⁰ and the persistence of biofilm. Mechanical instruments are unable to completely eliminate intracanal infections¹¹, while sodium hypochlorite (NaOCl), the preferred root canal irrigant, exhibits potent histolytic and antimicrobial properties that can disrupt biofilms¹². However, NaOCl negatively affects dentin hardness and elasticity^13,14, is cytotoxic to vital tissues¹⁵, and can cause severe injury to skin or mucosa in the event of accidental spillage. While antibiotics are used to treat oral infections, achieving effective concentrations in vivo to eradicate bacterial biofilms in root canals remains challenging. The growing problem of bacterial resistance and concerns regarding antibiotic misuse further complicate the issue. Therefore, there is an urgent need to identify more effective treatments to overcome these limitations.

Antimicrobial photodynamic therapy (aPDT) has gained widespread attention as an emerging therapeutic approach. Photodynamic therapy (PDT), which has been gradually developed since the early 20th century, gained significant recognition due to its successful application in antitumor therapy. In recent years, notable progress has been made in aPDT research for antimicrobial therapy¹⁶. aPDT sterilizes pathogens by combining an appropriate light source with a photosensitizer to generate reactive oxygen species (ROS), particularly singlet oxygen ($\mathrm {^1O_2}$), which oxidatively damages bacterial cells^17,18. Chlorin e6 (Ce6), a commonly used photosensitizer, exhibits a high ROS quantum yield and low dark toxicity¹⁹, making it advantageous for root canal disinfection, especially against drug-resistant pathogens such as Enterococcus faecalis. However, Ce6 still faces significant limitations in clinical applications, primarily due to its poor water solubility and strong tendency to aggregate²⁰. These drawbacks compromise its uniform distribution within the narrow and complex anatomy of the root canal and hinder its penetration into deep canal regions and biofilm interiors. These factors collectively diminish its therapeutic efficacy. To address these challenges, nanodelivery systems are urgently needed to improve the solubility, stability, and biofilm penetration of Ce6, thereby enhancing its practical performance in aPDT-based root canal disinfection²¹.

Among biofilm components, extracellular DNA (eDNA) is considered a critical factor in biofilm structural stability, bacterial protection, and pathogenicity²². Numerous studies have demonstrated that exogenously applied DNase I can specifically degrade extracellular DNA (eDNA), thereby weakening the protective function of the biofilm matrix. When combined with antimicrobial agents, DNase I further enhances the agents’ ability to kill microbial cells^23,24. This not only disrupts the barrier surrounding embedded bacteria but also facilitates deeper penetration of antimicrobials^25,26.

This study aims to develop a DNase I-functionalized liposomal nanoplatform (DNase I-Lip@Ce6, hereafter referred to as DLC) that combines enzymatic biofilm disruption with enhanced photodynamic therapy. In this system, DNase I is covalently conjugated to the outer surface of the liposome, enabling it to degrade the biofilm matrix upon contact, while Ce6 is encapsulated within the lipid bilayer to ensure stable and efficient delivery to bacterial cells. The platform demonstrated effective in vitro inhibition of intracanal bacteria and biofilms, offering a promising therapeutic strategy for clinical root canal disinfection.

Materials and methods

Primary reagents and instrumentation

Ce6 was purchased from Adamas (Shanghai, China). 1,2-Distearoyl-sn-glycero-3-phosphocholine (DSPC), 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-Polyethylene Glycol-N-Hydroxysuccinimide (DSPE-PEG-NHS), and cholesterol were purchased from Macklin (Shanghai, China). Live/Dead Bacterial Viability Kit (with SYTO 9 and propidium iodide dye) was purchased from Sigma-Aldrich (USA). The live/dead BacLight viability kit was obtained from Invitrogen (USA), and 9,10-anthracenediyl-bis(methylene)dicarboxylic acid (ABDA) was sourced from Rhawn (Shanghai, China). 2’,7’-dichlorofluorescein diacetate (DCFH-DA) was purchased from Aladdin (Shanghai, China), and Singlet Oxygen Sensor Green (SOSG) was obtained from Meilunbio (Dalian, China). The BCA protein concentration assay kit was purchased from Beyotime (Shanghai, China), and Brain Heart Infusion (BHI) broth was sourced from Hopebio (Qingdao, China). DMEM medium and fetal bovine serum were purchased from Gibco (USA). Phosphate buffer solution (PBS), Cell Counting Kit-8, DNase I, DiBAC4(3) membrane potential fluorescent probe, crystal violet, FITC-Phalloidin, and 4’,6-diamidino-2-phenylindole dihydrochloride (DAPI) were all purchased from Solarbio (Beijing, China). A 660 nm LED light source (40 mW/$\mathrm cm^2$) was used in the experiments.

Cell lines

Enterococcus faecalis (ATCC 29212) strains were purchased from the Zhejiang Provincial Microbial Strain Preservation Center and cultured under anaerobic conditions (80% $\mathrm N_2$, 10% $\mathrm H_2$, and 10% $\mathrm CO_2$) at 37$^\circ$C for amplification using Brain Heart Infusion (BHI) broth or BHI agar plates. After 24 hours of incubation in BHI liquid medium, the bacterial suspensions were measured for optical density at 600 nm ($\mathrm OD_{600}$) and adjusted to a final concentration of $1 \times 10^8$ CFU/mL with phosphate-buffered saline (PBS) for subsequent experiments. The L929 murine fibroblast cell line and human gingival fibroblasts (HGF) were purchased from Cyagen Biosciences (Guangzhou). Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% (v/v) fetal bovine serum (FBS) and 1% (v/v) penicillin-streptomycin, and were maintained at 37$^\circ$C with 5% $\mathrm CO_2$.

Experimental methods

Synthesis of DNase I-Lip@Ce6 nanoparticles

Liposomal nanoparticles (Lip@Ce6) were prepared as follows: DSPC, cholesterol (Chol), and DSPC-PEG-NHS (mass ratio 5:3:2, total mass 10 mg) were accurately weighed and dissolved in anhydrous ethanol. Meanwhile, 1 mg of chlorine e6 (Ce6) was dissolved in tetrahydrofuran (THF). The two solutions were then mixed and transferred to a vessel containing 10 mL of ultrapure water. The mixture was subjected to ultrasonic treatment at 30% power for 5 minutes (VCX 750 Ultrasonic Cell Disruptor, Sonic, USA) to obtain a homogeneous Lip@Ce6 dispersion under light-protected conditions in an ice water bath.

Liposomes were subsequently functionalized via EDC/NHS-mediated covalent conjugation²⁷. Specifically, 10 mL of Lip@Ce6 suspension was mixed with 1.25 mg of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC, 6.5 $\mu$M) and 0.75 mg of N-hydroxysuccinimide (NHS, 6.5 $\mu$M), and the mixture was incubated in a water bath at a constant temperature for 30 minutes to activate the carboxyl groups on the liposome surface. Then, 10 mg of bovine pancreatic deoxyribonuclease I (DNase I, 0.325 $\mu$M) was added, and the solution was stirred for an additional 30 minutes to promote the covalent binding of DNase I to the NHS-activated liposome surface. Finally, the reaction mixture was dialyzed using a dialysis membrane (8000–14000 Da) to remove unreacted reagents and byproducts, yielding the DNase I-modified liposomal complex (DNase I-Lip@Ce6).

Physicochemical properties of DNase I-Lip@Ce6 nanoparticles

The hydrodynamic particle size distribution and surface zeta potential of free Ce6 and DNase I-Lip@Ce6 complexes were systematically characterized using a Malvern particle size analyzer (NS300, Malvern, UK). The morphology of DNase I-Lip@Ce6 was observed via transmission electron microscopy (TEM) (HT7700, Hitachi, Japan), while the characteristic absorption spectra of free Ce6 and liposome-encapsulated Ce6 (Lip@Ce6) were measured using a UV-visible spectrophotometer(P4, Mapada, Shanghai, China). To quantify the encapsulation efficiency and drug loading capacity of Ce6 in the liposomes, methanol cleavage of the lipid bilayer was performed to release the encapsulated drug, followed by the determination of Ce6 concentration in the solution using UV-Vis spectroscopy ($\lambda$ = 400 nm). The encapsulation efficiency (EE%) and drug loading capacity (DL%) were calculated according to Eqs. (1) and (2), respectively. The chemical functional groups and molecular interactions of DNase I-Lip@Ce6 were further investigated using Fourier transform infrared spectroscopy (FT-IR) (Vetex 70v, Bruker, Germany) in the wave number range of 4000–500 $\mathrm cm^{-1}$.

$$\begin{aligned} Encapesulation\ Efficiency(EE\%)=\frac{Amout\ of\ encapsulated\ drug}{Initial\ amount\ of\ drug} \times 100\% \end{aligned}$$

(1)

$$\begin{aligned} Drug\ Loading(DL\%) = \frac{Amount\ of\ encapsulated\ drug}{Amount\ of\ encapsulated\ drug + Amount\ of\ liposomes} \times 100\% \end{aligned}$$

(2)

The conjugation efficiency of DNase I was evaluated by separating the bound enzyme from the free form using ultracentrifugation. The liposomes were centrifuged at 40,000 rpm and 4$^\circ$C for 40 minutes. The supernatant (free DNase I) and pellet (liposome-bound DNase I) were collected separately. The concentration of free protein in the supernatant was quantified using a BCA protein assay kit. The conjugation efficiency (%) was calculated as Eq. (3):

$$\begin{aligned} Conjugation\ efficiency(\%) = \frac{Initial\ amount\ of\ DNase\ I - Amount\ of\ free\ DNase\ I}{Initial\ amount\ of\ DNase\ I} \times 100\% \end{aligned}$$

(3)

ROS generation

To evaluate the reactive oxygen species (ROS) generation capacity of the nanoparticles, ABDA and a singlet oxygen fluorescence probe (SOSG) were used for quantitative analysis. The procedure was as follows: 1 $\mu$L of ABDA dimethyl sulfoxide (DMSO) solution (stock concentration of 10 mM) was added to 1 mL of DLC (5 $\mu$g/mL) suspension and mixed thoroughly. The UV-visible absorption spectra in the range of 350–450 nm were recorded under 660 nm laser irradiation (with a power density of 40 mW/$\mathrm cm^2$) at 0, 1, 2, 3, 4, and 5 minutes. Phosphate-buffered saline (PBS) was used as a blank control.

To specifically detect the singlet oxygen ($^{1}{\textrm{O}}_{2}$) yield at different concentrations, the SOSG probe (stock concentration 10 mM, 1.0 $\mu$L) was used to replace ABDA, creating the same reaction system. Fluorescence intensity changes were measured at an excitation wavelength of 488 nm and emission wavelength of 530 nm after irradiation with the 660 nm laser (40 mW/$\mathrm cm^2$) for 5 minutes. All experiments were repeated independently three times.

Bacterial binding and intra-biofilm ROS generation

To quantitatively assess the enhancement of bacterial binding by liposome encapsulation, flow cytometry was used to analyze the interaction between DLC and Enterococcus faecalis. The bacterial suspension of Enterococcus faecalis ($1 \times 10^8$ CFU/mL) was prepared, centrifuged (5000 rpm, 5 min), and resuspended in PBS. An equal volume of DLC solution (5 $\mu$g/mL) was added to 500 $\mu$L of the bacterial suspension, followed by a 5-minute incubation in the dark. Unbound nanoparticles were then removed by centrifugation, the mixture was washed with PBS, and resuspended. The fluorescence intensity was measured using a flow cytometer (LongCyte2, Challenbio, Beijing, China), and the histogram distribution was analyzed using FlowJo software. Control groups included PBS and free Ce6 (5 $\mu$g/mL).

To further assess the generation of reactive oxygen species (ROS) within the biofilm, the DCFH-DA fluorescent probe was employed. Enterococcus faecalis biofilms were incubated with 1 mL of DCFH-DA solution ( 10 $\mu$M , 1 $\mu$L) for 30 minutes in the dark, followed by three washes with PBS to remove excess probe. The fluorescence intensity of DCF was measured at an excitation wavelength of 488 nm and emission wavelength of 530 nm using flow cytometry to quantify intracellular ROS levels. The experimental groups included: PBS, PBS + L, Ce6, Ce6 + L, DLC, DLC + L, DNase I, and DNase I + L (L: light irradiation at 660 nm).

Antimicrobial activity

An appropriate experimental concentration was first selected. Equal volumes (500 $\mu$L) of Enterococcus faecalis suspension (1 $\times$ 108 CFU/mL) were mixed with DLC or Ce6 at varying concentrations (1.25, 2.5, 5, and 10 $\mu$g/mL) in 24-well plates and incubated at 37$^\circ$C for 5 minutes. The mixtures were then irradiated with a 660 nm laser at 40 mW/$\mathrm cm^2$ for 5 minutes. Following treatment, the bacterial suspensions were serially diluted tenfold, and 10 $\mu$L of each dilution was spotted onto BHI agar plates. The plates were incubated at 37$^\circ$C for 24 hours. Untreated bacterial suspensions served as the control. The experiment was repeated three times, with each sample tested in triplicate.

The in vitro antimicrobial activity of DLC was quantified using the standard colony counting method. Briefly, a suspension of Enterococcus faecalis (500 $\mu$L, $1 \times 10^8$ CFU/mL) was mixed with an equal volume of DLC solution (5 $\mu$g/mL Ce6) in a 24-well plate and incubated for 5 minutes at 37$^\circ$C, protected from light. The mixture was then irradiated with a 660 nm laser (40 mW/$\mathrm cm^2$) for 5 minutes. Following irradiation, 100 $\mu$L of the mixture was taken for gradient dilution, and 10 $\mu$L of the dilution was plated on brain heart infusion (BHI) agar plates. The plates were incubated at 37$^\circ$C for 24 hours, and the colony-forming units (CFU) were counted. Control groups included PBS, PBS+L, DLC, Ce6, and Ce6+L (5 $\mu$g/mL), with each group tested in triplicate.

Bacterial membrane integrity was assessed using SYTO 9/PI staining. The suspension was centrifuged at 8000 rpm for 5 minutes, washed with PBS, and resuspended in a staining solution containing SYTO 9 (1.67 $\mu$M) and PI (20 $\mu$M), followed by incubation for 30 minutes in the dark. The suspension was then centrifuged again to remove any free dye. The ratio of live to dead cells was observed by fluorescence microscopy. The experimental conditions were consistent with those of the colony counting method and were repeated three times.

Antimicrobial mechanism

To investigate the effect of DLC-mediated photodynamic action on the ultrastructure of bacteria, the bacterial suspension was prepared as previously described. The bacteria were then collected by centrifugation at 5000 rpm for 5 minutes, fixed in 2.5% glutaraldehyde at 4$^\circ$C for 12 hours, and washed three times with PBS. The samples underwent ethanol gradient dehydration (30%, 50%, 70%, 90%, and 100%, with 10 minutes at each concentration), followed by critical point drying and gold coating. The surface morphology of the bacteria was observed using scanning electron microscopy (Gemini 300, Zeiss, Germany). The control groups included PBS, PBS+L, DLC, Ce6, and Ce6+L.

Intracellular protein leakage following photodynamic treatment was measured using the BCA method. The supernatant from aPDT-treated bacteria was collected, and the absorbance at 560 nm was determined using the BCA protein quantification kit. The amount of intracellular protein leakage was calculated based on the standard curve. The control groups were PBS, PBS+L, DLC, Ce6, and Ce6+L.

Membrane potential changes were assessed using the DiBAC4(3) probe. The treated bacterial suspension (500 $\mu$L) was incubated with DiBAC4(3) at a final concentration of 5 $\mu$M for 30 minutes at 37$^\circ$C, away from light. Fluorescence intensity was measured using flow cytometry with 488 nm excitation and 530 nm emission, and the fluorescence signal shift was analyzed using FlowJo. The experimental conditions were consistent with those of the protein leakage assay.

Biofilm eradication effect

To evaluate the biofilm eradication effect, live/dead cell viability assays were conducted. Enterococcus faecalis suspension (50 $\mu$L, $1 \times 10^8$ CFU/mL) was added to a 24-well plate containing round coverslips and 450 $\mu$L of Brain Heart Infusion (BHI) broth. The plate was incubated at 37$^\circ$C for 7 days to allow biofilm formation, with medium replacement every 48 hours. After biofilm formation, the supernatant was removed, and 500 $\mu$L of DLC was added and incubated in the dark for 5 minutes. The plate was then irradiated with a 660 nm laser (40 mW/$\mathrm cm^2$) for 5 minutes. The plates were washed twice with PBS, incubated with SYTO9 (1.67 $\mu$M) and propidium iodide (PI, 20 $\mu$M) staining solution for 30 minutes in the dark, washed again to remove unbound dye, and the fluorescence images were acquired using a confocal laser scanning microscope (CLSM) (LSM 900, Zeiss, Germany). The control group consisted of PBS. Additional control groups included PBS, PBS + L, DLC, Ce6, Ce6 + L, DNase I, and DNase I + L.

Furthermore, antimicrobial activity was quantified by colony-forming unit (CFU) analysis. After treatment of the mature biofilms as described, 1 mL of sterile PBS was added and sonicated (40 kHz, 10 min) to disperse the biofilm. The homogenized bacterial suspension was then vortexed for 1 minute. After gradient dilution, 10 $\mu$L of the dilution was plated on BHI agar plates.

Finally, biofilm removal efficiency was assessed by crystal violet staining. The treated biofilm was washed twice with PBS, fixed with 4% glutaraldehyde for 20 minutes, air-dried, and stained with 0.1% (w/v) crystal violet for 20 minutes. After washing with PBS to remove unbound dye, the stained biofilm was solubilized in 95% ethanol, transferred to a 96-well plate, and absorbance at 595 nm ($\mathrm OD_595$) was measured using a microplate reader (Infinite 200 Pro, Tecan, Switzerland) The biofilm disruption rate was calculated using the following Eq. (4) :

$$\begin{aligned} Biofilm\ Disruption\ Rate(\%) = \frac{OD_{control} - OD_{sample}}{OD_{control}} \times 100\% \end{aligned}$$

(4)

Biological safety assessment

Fresh blood was collected from rats, and erythrocytes were separated by centrifugation at 1500 rpm for 15 minutes. The supernatant was discarded, and the erythrocyte pellet was washed three times with sterile saline until the supernatant became clear. The erythrocytes were then resuspended in saline to prepare a 2% (v/v) erythrocyte suspension. An equal volume (500 $\mu$L) of erythrocyte suspension was mixed with an equal volume of DLC solution (1.25, 2.5, 5, and 10 $\mu$g/mL), and incubated at 37$^\circ$C for 3 hours. Saline was used as the negative control, and 1% Triton X-100 as the positive control. After incubation, the mixture was centrifuged at 1500 rpm for 10 minutes, and the supernatant was transferred to a 96-well plate. Absorbance at 540 nm ($OD_{540}$) was measured using a microplate reader. The hemolysis rate was calculated using the following Eq. (5):

$$\begin{aligned} Hemolysis(\%)=\frac{OD_{sample}-OD_{negative\ control}}{OD_{positive\ control}-OD_{negative\ control}} \times 100\% \end{aligned}$$

(5)

To further evaluate the biosafety of DLC, cytotoxicity experiments were conducted. Mouse L929 fibroblasts were seeded in 96-well plates at a density of $5 \times 10^3$ cells/well. After 24 hours, the medium was replaced with DMEM containing different concentrations of DLC (1.25, 2.5, 5, and 10 $\mu$g/mL), and incubated for 24, 48, and 72 hours, respectively. After removing the supernatant, 100 $\mu$L of DMEM containing 10% CCK-8 solution was added to each well and incubated for an additional 2 hours. A cell-free DMEM + CCK-8 group was included as a blank control to subtract background absorbance. Relative cell viability was calculated by measuring absorbance at 450 nm using a microplate reader.

Cytotoxicity was further assessed using the LIVE/DEAD cell viability assay. L929 cells were seeded in 48-well plates ($1 \times 10^4$cells/well) and incubated for 24 hours. The medium was replaced with fresh medium containing different concentrations of DLC (1.25, 2.5, 5, and 10 $\mu$g/mL), and incubated for another 24 hours. The supernatant was replaced with 150 $\mu$L of AM/EthD-1 solution for 30 minutes in the dark. After PBS washing, live cells (green fluorescence, Calcein-AM) and dead cells (red fluorescence, EthD-1) were observed using an inverted fluorescence microscope.

Human gingival fibroblasts (HGF) were co-cultured with different concentrations of DLC to assess their effects on HGF cell adhesion. HGF cells were seeded in 48-well plates at a density of $4 \times 10^3$ cells per well, and after 24 hours, DMEM containing different concentrations of DLC (1.25, 2.5, 5, and 10 $\mu$g/mL) was added. The cells were cultured for another 24 hours. After removing the culture medium, the cells were washed three times with PBS, fixed with 4% paraformaldehyde for 20 minutes, and permeabilized with 0.5% Triton X-100 for 10 minutes. The cytoskeleton was stained with FITC-Phalloidin (100 nM), and the nuclei were stained with DAPI (1 $\mu$g/mL). The cells were then washed with PBS after incubation in the dark. Cell morphology and adhesion were observed using inverted fluorescence microscopy.

Statistical analysis

Unless otherwise specified, all references to the concentration of DLC in the data refer to the concentration of Ce6 encapsulated within the nanomaterial. All experiments were performed in at least triplicate, and data are presented as mean ± standard deviation (SD). Statistical analyses were conducted using GraphPad Prism 9.5.1. One-way or two-way analysis of variance (ANOVA) was applied to assess differences between groups, followed by Tukey’s post hoc test for multiple comparisons. ($\alpha$ = 0.05, *p < 0.05, **p < 0.01, ***p < 0.001).

Results and discussion

Synthesis and characterization of DNase I-Lip@Ce6

Encapsulation efficiency measurements revealed that the liposomes had an encapsulation efficiency of 83.64% and a drug loading capacity of 8.5% (w/w) for the photosensitizer chlorine e6 (Ce6), indicating efficient drug loading. Dynamic light scattering (DLS) analysis showed that the hydrated particle size of the DLC was 132.2 ± 4.3 nm with a polydispersity index (PDI) of 0.203 ± 0.026 (Fig. 1A). Transmission electron microscopy (TEM) images further confirmed the monodispersed spherical structure of the liposomes (Fig. 1B), suggesting successful encapsulation and uniform dispersion of the hydrophobic Ce6. Additionally, the zeta potential indicated high and stable water solubility (Fig. 1C). The particle size and dispersion of DLC remained stable after 14 days of storage at room temperature (25$^\circ$C) (Fig. 1D), demonstrating excellent colloidal stability.

Ultraviolet-visible (UV-Vis) absorption spectra demonstrated that DLC exhibited characteristic absorption peaks at 405 nm, 504 nm, and 660 nm, consistent with free Ce6 (Fig. 1E). This indicates that the encapsulation process did not alter the conjugated structure or the photophysical properties of Ce6. Given that the penetration depth of 660 nm light in biological tissues is significantly greater than that of 405 nm, 660 nm was selected as the excitation wavelength for photodynamic therapy (PDT) in subsequent experiments. In the infrared spectrum of DLC, absorption peaks at 1645.2$\mathrm cm^{-1}$ and 1234.2$\mathrm cm^{-1}$ were attributed to the carbonyl stretching vibration of the amide bond—formed by the condensation of the active ester group on the liposome surface with the amino group of DNase I—and to theC-N stretching vibration, respectively. These findings confirm that DNase I was successfully conjugated onto Lip@Ce6 (Fig. 1F). By determining the protein concentration, we further evaluated the conjugation efficiency of DNase I on the liposome surface. The calculated conjugation efficiency of DNase I to the liposomes was 49.4%.

Fig. 1

Synthesis and Characterization of DLC Nanoparticles. (A) Particle size distribution of DLC; (B) Transmission electron microscopy (TEM) images; (C) Zeta potential of Ce6 and DLC; (D) Stability of DLC over 14 days; (E) UV-Vis spectra of Ce6 and DLC; (F) FTIR spectra of various materials. Data are presented as the mean ± standard deviation, n = 3.

aPDT-driven ROS generation

Fig. 2

ROS Generation and Binding Ability of DLC to Bacteria. The generation of ROS and the binding ability of DLC to bacteria were assessed using 9,10-Anthracenediyl-bis(methylene)dicarboxylic acid (ABDA). Time-dependent absorbance spectra of the reaction were recorded for: (A) PBS, (B) Ce6 and (C) DLC under 660 nm light irradiation (at a Ce6 concentration of 5 $\mu$g/mL, 40 mW/$\mathrm cm^2$). (D) $^{1}{\textrm{O}}_{2}$ generation curves for each group. (E) Relative fluorescence intensity of SOSG reactions with different concentrations of Ce6 and DLC under 660 nm light irradiation (40 mW/$\mathrm cm^2$, 5 min). (F) Internalization and tight binding of free Ce6 and DLC by Enterococcus faecalis; (G) Quantitative fluorescence intensity of Ce6 in Enterococcus faecalis. (H) Fluorescence signals of Enterococcus faecalis biofilms in different treatment groups as measured by the DCFH-DA probe. (I) Quantitative fluorescence intensity of intracellular ROS in different groups. Quantitative values were obtained from flow cytometry data of three independent experiments (data are expressed as mean ± standard deviation, ns: not significant, $***\ p$ < 0.001).

Reactive oxygen species (ROS) play a crucial role in antimicrobial therapy. In antimicrobial photodynamic therapy (aPDT), photosensitizers activated under specific light conditions generate substantial amounts of ROS, which can damage bacterial cell membranes, proteins, and DNA, ultimately leading to bacterial death. Therefore, the amount of ROS generated serves as a key indicator for evaluating the effectiveness of aPDT. In this experiment, UV-Vis spectral analysis (Fig. 2B, C) revealed that the characteristic absorption peak intensity of free Ce6 and DLC at 400 nm decreased progressively with increasing irradiation time of the 660 nm laser (40 mW/$\mathrm cm^2$). The absorption intensity of ABDA (10 $\mu$M) in the DLC group decreased more rapidly and to a greater extent than in the free Ce6 group (Fig. 2D), suggesting that photosensitizer encapsulation significantly enhanced the efficiency of ROS-mediated oxidative damage. No significant absorbance changes were observed in the control group (PBS) under the same conditions (Fig. 2A), thereby ruling out non-specific interference.

To further quantify the efficiency of singlet oxygen ($^{1}{\textrm{O}}_{2}$) generation, a high-sensitivity fluorescent probe, SOSG (10 $\mu$M), was used. Irradiation with a 660 nm laser (40 mW/$\mathrm cm^2$) for 5 minutes resulted in significantly higher fluorescence intensities at an excitation wavelength of 500 nm and an emission wavelength of 530 nm in the DLC group compared to the free Ce6 group, despite having the same Ce6 concentration. This indicated that liposomal encapsulation effectively enhances the $^{1}{\textrm{O}}_{2}$ production rate (Fig. 2E). Possible mechanisms suggest that free Ce6, due to its hydrophobicity, is prone to molecular aggregation, which leads to the dissipation of excited-state energy and a reduced $^{1}{\textrm{O}}_{2}$ quantum yield^28,29. In contrast, liposomal carriers inhibit Ce6 aggregation through a spatial site-barrier effect³⁰, maintaining its monodispersed state. Additionally, the bilayer structure of the liposomes can reduce the collision of $^{1}{\textrm{O}}_{2}$ with the solvent³¹. As a result, the $^{1}{\textrm{O}}_{2}$ quantum yield of DLC was significantly higher than that of free Ce6, demonstrating the optimization of photosensitizer performance through nanocarrier encapsulation.

Bacterial binding and ROS generation in biofilms

Antimicrobial agents must penetrate deep into biofilms and be effectively internalized by planktonic bacteria to exert antimicrobial effects. However, the poor hydrophobicity and limited dispersion of Ce6 can lead to fluorescence quenching and inadequate retention in both biofilms and bacteria. To assess whether DLC could enhance the uptake of Ce6 by planktonic bacteria and generate more reactive oxygen species (ROS) in biofilms, the uptake of Ce6 by Enterococcus faecalis was quantified using flow cytometry. The results showed that the fluorescence intensity of bacteria in the DLC-treated group was significantly higher ( 10-fold, p < 0.001) than in the free Ce6 group (Fig. 2F, G), indicating that liposomal encapsulation significantly improved the bacterial internalization of Ce6. Previous studies have suggested that the amphiphilic structure of liposomes may facilitate their interaction with the bacterial cytosol via the membrane fusion pathway, overcoming the aggregation and fluorescence quenching of free Ce6 due to its hydrophobicity³².

In addition, we elucidated the potential antimicrobial mechanism by evaluating intracellular ROS levels in the biofilm. Endogenous ROS levels were detected using the 2,7-dichlorodihydrofluorescein diacetate (DCFH-DA) probe. As shown in Fig. 2H, no significant ROS production was detected in Ce6, DLC, or DNase I-treated E. faecalis biofilms in the absence of light irradiation, confirming that photoactivation is required to trigger ROS production. Upon 660 nm laser (40 mW/$\mathrm cm^2$) irradiation for 5 minutes, the fluorescence intensity of the DLC+L group was significantly higher than that of the Ce6+L group (Fig. 2I), indicating that the DLC+L group was effectively activated by disrupting the biofilm matrix. DLC enhanced Ce6 permeability and ROS diffusion efficiency by reducing biofilm matrix density.

In summary, the synergistic effect of liposomal encapsulation and enzymatic degradation strategies enabled DLC to: (1) enhance photosensitizer targeting through increased bacterial uptake, (2) optimize ROS distribution in the deeper layers of the biofilm through matrix depolymerization, and (3) maintain a high singlet oxygen yield by reducing Ce6 aggregation.

In vitro antibacterial effect

Based on its excellent ROS generation efficiency and significantly enhanced bacterial internalization, the bactericidal activity of DLC-mediated aPDT against Enterococcus faecalis was further investigated. As shown in Fig. 3A, DLC-mediated aPDT effectively killed E. faecalis in a dose-dependent manner. Even at a low concentration of 1.25 $\mu$g/mL, the bacterial count was reduced from 8.43 ± 0.05 log to 4.15 ± 0.11 log (p < 0.001), corresponding to a log reduction of 4.28 ± 0.15 (>99%). The antimicrobial effect was clearly evident on agar plates, where no visible bacterial colonies were observed at a concentration of 5 $\mu$g/mL, indicating a complete bacterial eradication with a log reduction of 8.43 ± 0.05 (>99.99%). This concentration was selected for subsequent in vitro experiments. Moreover, at the same concentration, the DLC + L group exhibited stronger bacterial growth inhibition compared to the Ce6 + L group.

The plate coating method was employed to further synthesize and evaluate the antibacterial activity of different material groups. As shown in Fig. 3B, the PBS-treated group exhibited similar colony counts regardless of light exposure, indicating that 660 nm light irradiation alone had no antimicrobial effect. Similarly, the group treated with Ce6 or DLC displayed colony counts greater than 8 log units in the absence of light, which was comparable to the control group (p> 0.05), suggesting that the nanomedicine’s antimicrobial activity was not activated under these conditions. In contrast, the presence of light irradiation triggered significant antimicrobial activity in the DLC group. As illustrated in Fig. 3C, the colony count in the DLC+L group was significantly reduced after 5 minutes of irradiation (reduction > 99.99%, p < 0.001). Liposomal encapsulation of Ce6 facilitated its efficient uptake by bacteria and enhanced the production of cytotoxic $^{1}{\textrm{O}}_{2}$, thereby significantly boosting its photoantimicrobial activity. These results confirm the potent E. faecalis killing efficacy of DLC in vitro.

Bacterial viability was further assessed through live/dead staining experiments. SYTO 9 dye stained all bacteria, showing green fluorescence regardless of cell membrane integrity, while propidium iodide (PI) only penetrated bacteria with compromised membranes, producing red fluorescence. After incubation with DLC and irradiation with 660 nm light, the live/dead experiments revealed significant red fluorescence, indicating that the photodynamic effect caused substantial damage to the bacterial cell membranes, leading to extensive bacterial death. In contrast, the Ce6 +L group exhibited both green and red fluorescence, suggesting limited antibacterial activity against E. faecalis. Moreover, in the control group, green fluorescence predominated in both the L and Ce6 groups, with nearly no red fluorescence observed, indicating that these materials had no damaging effect on bacterial cell membranes in the absence of light (Fig. 3D).

To investigate the bactericidal mechanism of DLC-mediated aPDT, scanning electron microscopy (SEM) was employed to observe the morphological changes in Enterococcus faecalis. As shown in Fig. 3E, the PBS control group exhibited typical pairs or short-chain arrangements of bacteria, with intact cell walls and smooth surfaces. In the L, Ce6, and DLC groups, the bacterial morphology remained normal, with no obvious signs of damage. However, after irradiation with 660 nm laser (40 mW/$\mathrm cm^2$) for 5 minutes, the surface of some bacteria in the free Ce6 group showed irregular folds and localized depressions. In contrast, the DLC+L treatment group exhibited more pronounced bacterial damage, including membrane rupture, leakage of cytoplasmic contents, and cell collapse. The degree of membrane damage was positively correlated with the extent of protein leakage observed (Fig. 3F). By eliminating the potential influence of DNase I, which was bound to the DLC, this trend clearly indicated the strong membrane-disrupting ability of the DLC+L treatment.

To further quantify the impact of photodynamic effects on bacterial membrane integrity, we used the DiBAC4(3) probe to detect changes in membrane potential. This probe is a lipophilic anionic fluorescent dye that exhibits increased fluorescence during membrane depolarization ($\lambda$ex/$\lambda$em = 488/530 nm), whereas hyperpolarization leads to decreased fluorescence due to ion efflux³³. Flow cytometry analysis revealed a significant increase in fluorescence intensity in the DLC+L treatment group compared to the control group (Fig. 3G, H), indicating membrane depolarization. This phenomenon was likely due to the disruption of the lipid bilayer structure of the cell membrane by ROS, which led to the dysfunction of ion channels or pumps. In addition, ROS compromised membrane integrity and increased permeability, allowing positively charged ions to move freely across the membrane, which further exacerbated the depolarization of the Enterococcus faecalis cell membrane.

This finding aligned with the widely accepted reactive oxygen species (ROS)-based antimicrobial mechanism of photodynamic therapy (PDT), indicating that the modified photosensitizer-induced generation of $\mathrm {^1O_2}$ can effectively trigger oxidative stress in bacteria, leading to membrane disruption and metabolic dysfunction. Taken together, these results suggested that DLC enhanced the photoreactivity of Ce6 and significantly promoted $\mathrm {^1O_2}$ release under light irradiation, thereby exhibiting potent in vitro antibacterial activity.

Fig. 3

In vitro Antibacterial Activity of DLC. (A) Representative images of Enterococcus faecalis colonies formed on Brain Heart Infusion (BHI) agar plates after treatment with different concentrations of Ce6 and DLC, followed by 660 nm light irradiation (40 mW/$\mathrm cm^2$, 5 minutes). (B) Representative images of E. faecalis colonies on agar plates after treatment with various groups. (C) Bacterial viability. (D) Live/dead fluorescence images. (E) Scanning electron microscope (SEM) images. Red arrows indicate damaged bacterial cell membranes. (F) Total protein released by E. faecalis after different treatments. (G) Intracellular fluorescence signals of E. faecalis treated with different groups, as measured by the DIBAC4(3) probe. (H) Quantitative fluorescence intensity. Data are expressed as the mean ± standard deviation, n = 3. ns: not significant, $***\ p$ < 0.001.

In vitro biofilm removal effect

After confirming the outstanding antimicrobial activity of the DLC nanoplatform against planktonic Enterococcus faecalis, this study further explored its efficacy in eradicating bacterial biofilms. Our preliminary investigation involved assessing the biofilm eradication performance of DLC against Enterococcus faecalis biofilms using live/dead bacterial staining. As shown in Fig. 4A, the biofilms of Enterococcus faecalis in the unilluminated and L groups displayed strong green fluorescence from SYTO 9, while red fluorescence from propidium iodide (PI) was scarcely observed. In Ce6 + L-treated biofilms, both limited green and red fluorescence signals were visible, indicating that only a portion of the biofilm was disrupted. In contrast, the DLC + L-treated biofilms exhibited intense PI fluorescence, indicating near-complete eradication of the biofilm.

Additionally, the biofilms after treatment were collected for colony-forming unit (CFU) analysis. As shown in Fig. 4B, in the PBS, Ce6, DLC, and DNase I-treated groups, a large number of colonies were widely dispersed on the agar plates, indicating that the nanocomposite and its individual components were ineffective in independently eradicating Enterococcus faecalis biofilms in the absence of light. The photodynamic activity of Ce6 began to take effect upon exposure to light, as evidenced by a significant reduction in the number of remaining colonies in the Ce6 + L group, with a logarithmic reduction of 2.59 ± 0.10 (from 8.43 ± 0.02 to 5.84 ± 0.09, with a>90% killing rate, p < 0.001). Furthermore, no surviving colonies were observed in the DLC + L group, indicating a dramatic decrease in bacterial viability, with a logarithmic reduction of 2.53 ± 0.48 (from 8.43 ± 0.02 to 4.24 ± 0.09, with a >99.99% killing rate, p < 0.001) (Fig. 4C).

To validate the above results, crystal violet staining was used to further quantify the biofilm biomass removal efficiency of DLC (Fig. 4D). The results showed that after Ce6 + L treatment, the biofilm structure remained largely intact, whereas in the DLC + L treatment group, the biofilm biomass was significantly reduced by 66% (p < 0.001) (Fig. 4E), indicating that the synergistic effect of liposome encapsulation and enzymatic degradation could effectively disrupt the three-dimensional structure of the biofilm. Moreover, although DNase I was ineffective in eliminating biofilm bacteria, the combination of DNase I and DLC treatment effectively reduced the biofilm mass, with DLC + L exhibiting strong anti-biofilm activity, suggesting the synergistic effect of DNase I and Lip@Ce6. The remarkable biofilm removal effect could be attributed to the degradation of bacterial extracellular DNA (eDNA) by DNase I, which weakens the bacterial extracellular matrix (ECM), making the biofilm structure more permeable and thus significantly enhancing the bactericidal efficacy of Ce6³⁴. The crystal violet assay further supported this explanation, clearly showing that DNase I-coated materials not only more effectively eliminated bacterial extracellular components but also continuously reduced biofilm biomass, indicating that DLC could degrade the extracellular matrix and enhance the bactericidal activity of Ce6 against bacterial cells.

Fig. 4

In vitro Biofilm Eradication Effect of DLC. (A) Live/dead staining of Enterococcus faecalis biofilms after various treatments, (B) Representative images of the sample plates, and (C) Bacterial viability. (D) Representative images of crystal violet-stained Enterococcus faecalis biofilms after different treatments, and (E) Biofilm biomass. Data are presented as mean ± standard deviation, n = 3, ns: no significance, $***\ p$ < 0.001.

Biocompatibility and hemocompatibility

The biocompatibility of antimicrobial photosensitizers is a crucial evaluation criterion for their clinical translation. An ideal antimicrobial photosensitizer should exhibit minimal cytotoxicity to mammalian cells within its effective concentration range. In this study, the biocompatibility of DLC was evaluated using the following experimental systems: CCK-8 assay to assess the impact of DLC on the proliferation of mouse fibroblasts. When treated with DLC at concentrations of 1.25, 2.5, 5, and 10 $\mu$g/mL, L929 cells showed good biocompatibility (Fig. 5A). After 72 hours of incubation in the dark, the average cell viability remained close to 100%, even at a concentration of 10 $\mu$g/mL, where the cell viability still exceeded 75%. Live/dead cell staining (Calcein-AM/EthD-1) showed that the cell density and morphology (spindle-shaped and well-spread) of L929 cells treated with different concentrations of DLC were not significantly different from those of the control group (Fig. 5B). EthD-1 staining revealed no significant red fluorescence (with dead cells comprising <5%), confirming that the cell membrane integrity was not compromised. Human gingival fibroblasts (HGF) further confirmed these results, as the cell cytoskeleton (FITC-phalloidin) and nuclear morphology (DAPI staining) remained normal (Fig. 5C). Additionally, the hemocompatibility of DLC was evaluated using a rat erythrocyte hemolysis assay. After 3 hours of treatment with various concentrations of DLC, the hemolysis rate was <3% (Fig. 5D, E), significantly lower than the positive control group (1% TritonX-100, hemolysis rate 100%), meeting the safety requirements for clinical applications.

Based on the above comprehensive findings, the DNase I-Lip@Ce6 (DLC) nanoplatform demonstrated strong antibacterial and biofilm eradication efficacy, and favorable biocompatibility, making it a promising and safe adjunctive photosensitizer for clinical endodontic disinfection.

Fig. 5

Biocompatibility and Hemocompatibility of DLC. (A) CCK-8 assay of L929 cell viability after treatment with various concentrations of DLC for 24, 48, and 72 hours. (B) Live/dead cell staining of L929 cells co-cultured with different concentrations of DLC for 24 hours. (C) Fluorescence images of HGF cells co-cultured with the corresponding concentrations of DLC for 24 hours. Green represents actin filaments stained with FITC-phalloidin, and blue represents nuclei stained with DAPI. Data are expressed as mean ± SEM. For (A) and (D), n = 5, and for (B) and (C), n = 3. ns: no significance, $*\ p$ < 0.05, $**\ p$ < 0.01, $***\ p$ < 0.001.

Conclusion

In this study, we developed a multifunctional nano-delivery system, DNase I-Lip@Ce6, which integrates enzymatic degradation and photodynamic therapy (PDT) to enhance biofilm eradication in Enterococcus faecalis-induced apical periodontitis. This platform caused significant membrane damage in E. faecalis, markedly improved antibacterial efficacy, and effectively eliminated mature biofilms, resulting in a 66% reduction in biofilm biomass. Compared with existing therapies, our system simultaneously disrupts the protective biofilm matrix and delivers reactive oxygen species (ROS) in a controlled and localized manner, thereby overcoming limitations such as poor tissue penetration and bacterial resistance. These advantages highlight its potential as an effective and safe antimicrobial strategy for root canal disinfection. In future work, we will evaluate the in vivo therapeutic efficacy and biosafety of this platform and explore its application in other biofilm-associated infections.