Activating MoO3 nanobelts via aqueous intercalation as a near‐infrared type I photosensitizer for photodynamic periodontitis treatment

Although molybdenum trioxide nanomaterials have been widely explored as nanoagents for biomedical applications against bacteria through photothermal therapy, chemodynamic therapy, and catalytic therapy, their utilization as photosensitizers for photodynamic therapy (PDT) have been rarely reported so far. Herein, we report the activation of MoO3 nanobelts via aqueous co‐intercalation of Na+ and H2O into their van der Waals gaps as a near‐infrared Type I photosensitizer for photodynamic periodontitis treatment. The Na+/H2O intercalation of MoO3 nanobelts can shorten its length, generate rich oxygen vacancies, and enlarge its interlayer gaps. Such structural changes thus can induce the color change from white to dark blue with a strong near‐infrared (NIR) absorption. When used as a photosensitizer, the I‐MoO3−x nanobelts exhibit much higher activities for the generation of superoxide radical (·O2−) under an 808 nm laser irradiation than that of the pristine MoO3 nanobelts. Therefore, the prepared I‐MoO3−x nanobelts show a spectral antibacterial activity against Escherichia coli and Saccharomyces aureus, thus yielding a good clinical therapeutic effect on periodontitis. Our study proves that aqueous intercalation can be a simple but powerful strategy to activate layered MoO3 nanomaterials for high‐performance PDT.

Periodontitis is an inflammatory disease caused by bacterial infection, usually causing irreversible damage to the supporting tissues of the teeth, ultimately leading to tooth loss. 1,2At present, the mechanical removal of periodontal biofilms (scaling and root planning) remains a commonly used periodontal therapy in clinical. 3However, the shape and structures of the subgingival pocket are irregular, causing the eradication of pathogens to be inexhaustible.Antibiotics have also been widely used for periodontal therapy, but the presence of drug resistance and the difficulty in maintaining effective bactericidal concentration in gingival crevicular fluid make this treatment method ineffective. 4,5Recently, antimicrobial photodynamic therapy (aPDT), which eliminates resistant microorganisms and secreted virulence factors by producing reactive oxygen species (ROS) and highly cytotoxic singlet oxygen through photosensitizers (PSs) absorbed by microorganisms, [6][7][8] has been demonstrated as an alternative to chemical antimicrobial agents for the periodontitis treatment.Although the existing porphyrin and non-porphyrin PSs were effective in aPDT, the poor solubility, slow in vivo clearance rate, and nonspecific distribution hinder their clinical applications. 9,10In addition to PSs, light sources and molecular oxygen are two other fundamental prerequisites for aPDT.In the aPDT, the Type II PSs absorb visible/nearinfrared (NIR) photons and excite to a high-energy singlet state, which react with molecular oxygen and thus produce highly toxic reactive singlet oxygen ( 1 O 2 ). 11,12However, the Type II PSs need to consume a lot of oxygen, which is not conducive to the elimination of anaerobic bacteria most associated with periodontitis.Unlike the Type II PSs, the Type I PSs in the triple excited state can generate superoxide (•O 2 − ), peroxide (O 2 2− ), or hydroxyl radical (•OH) by the interaction of the electrons or protons transfer with adjacent substrates or molecular oxygen, which is less dependent on oxygen since it can participate in subsequent disproportionation, Haber-Weiss, or Fenton reaction to produce oxygen for recycling. 13,14Therefore, the development of highly efficient Type I PSs is of great significance in the clinical treatment of periodontitis.2][33][34][35] As a typical example, Wang et al. 36 reported a sulfur atom doping strategy to prepare MoO x nanorings as a nanoagent for combined photothermal and photodynamic therapy.However, the PDT performance of MoO x nanorings is quite low and is not enough to achieve a satisfactory therapeutic effect without the combination of other treatment strategies.Promisingly, our group [37][38][39] recently reported that intercalation of layered MoO 3 nanomaterials with various guest species, including ions, dye molecules, and conductive polymers, is also a promising strategy to activate its enzyme-like catalytic activity and NIR absorption, thus achieving highly efficient catalytic activity therapy or multifunctional therapy.Therefore, intercalation is expected to be a promising strategy to boost the catalytic performance of layered MoO 3 nanomaterials for ROS generation under laser irradiation for high-performance PDT.
Herein, we report a simple liquid-phase intercalation strategy for the preparation of intercalated MoO 3−x through the co-intercalation of Na + and H 2 O into the layered MoO 3 nanobelts as an efficient near-infrared Type I photosensitizer for photodynamic periodontitis treatment.The hydrothermal-synthesized micrometerlong MoO 3 nanobelts are intercalated with Na + and H 2 O through the aqueous solution intercalation to obtain the intercalated MoO 3−x (I-MoO 3−x ) nanobelts (Figure 1).The Na + /H 2 O intercalation of MoO 3 nanobelts can induce multiple structural changes including shortening its length, partially reducing the Mo 6+ to Mo 5+ , creating rich oxygen vacancies, and enlarging its interlayer distance, yielding a color change from white to dark blue with an enhanced NIR absorption.To this end, the I-MoO 3−x nanobelts exhibit excellent Type I PDT activity for the generation of •O 2 − under the irradiation of NIR laser (Figure 1), which is superior to that of inert MoO 3 nanobelts.The I-MoO 3−x nanobelts present rapid bacteriostatic activity when used as a Type I photosensitizer for photodynamic antibacterial (both Saccharomyces aureus and E. coli) treatment.Furthermore, in vivo experiments exhibit the superior therapeutic effect of the I-MoO 3−x nanobelts on photodynamic periodontitis treatment compared to that of MoO 3 nanobelts and the traditional clinical drug.

| Synthesis and characterization of MoO 3 and I-MoO 3−x nanobelts
Single-crystalline MoO 3 nanobelts with micrometer length were successfully synthesized by a previously reported hydrothermal method. 40As shown in Figure 2A, the X-ray diffraction (XRD) pattern of MoO 3 nanobelts precisely matches the standard peaks of orthorhombic layered MoO 3 (PDF card JCPDS 05-0508), affirming its crystal structure.The scanning electron microscope (SEM) image reveals that the prepared MoO 3 simple has a uniform nanobelt with a length of ~10 μm and a width of ~100 nm (Figure 2B).This result was further verified by transmission electron microscope (TEM) image (Supporting Information: Figure S1A).The selected area electron diffraction (SAED) pattern (Supporting Information: Figure S1B) of a typical MoO 3 nanobelt (inset in Supporting Information: Figure S1B) gives a set of rectangular shape bright diffraction spots, confirming its single-crystalline nature.The single-crystalline nature of the MoO 3 nanobelt is further verified by the continuous crystal lattice fringe in the high-resolution TEM (HRTEM) image (Supporting Information: Figure S1C).After reacting the MoO 3 nanobelts with Na 2 MoO 4 •2H 2 O and Na 2 S 2 O 4 in an aqueous solution, Na + and H 2 O were co-intercalated into the van der Waals layer gaps to obtain the I-MoO 3−x nanobelts.The SEM image of I-MoO 3−x nanobelts shows that the length is 0.5-2 μm (Figure 2C), suggesting that intercalation can also shorten the length of the MoO 3 nanobelts, which is further confirmed by its TEM image (Supporting Information: Figure S2A).Similarly, the SAED pattern (Supporting Information: electron microscope (HAADF-STEM) image of the MoO 3 nanobelt reveals its single-crystalline nature and layered structure (Figure 2D).The distance between the layers is measured to be 0.73 nm (Figure 2D), which is attributed to the (020) planes of the layered α-MoO 3 .The layered structure of a typical I-MoO 3−x nanobelt is also observable in its atomic resolution HAADF-STEM image when viewed from the y-axis (Figure 2E).In contrast to MoO 3 , the I-MoO 3−x nanobelts have a wider layer spacing (~0.85 nm) for the same (020) planes with some defects, indicating that the intercalation can effectively broaden its layer spacing and generate rich defects.As shown in the electron spin resonance (ESR) spectra of the MoO 3 and I-MoO 3−x nanobelts (Figure 2F), their peaks observed at g values of 2.003 and 1.92 were consistent with the signals originating from oxygen vacancies and the Mo 5+ , respectively.The I-MoO 3−x nanobelts display a much stronger intensity for both peaks than that of the MoO 3 nanobelts, indicating the presence of much more oxygen vacancies and Mo 5+ in the former.To further elucidate the changes in structure brought about by intercalation, X-ray photoelectron spectroscopy (XPS) was utilized to investigate the MoO 3 and I-MoO 3−x nanobelts.The MoO 3 nanobelts give two peaks in their XPS Mo 3d spectrum at 235.8 and 232.7 eV (Supporting Information: Figure S4B), which were attributed to the Mo 6+ present in MoO 3 .Apart from the XPS signal of the Mo 6+ , the XPS Mo 3d spectrum of the I-MoO 3−x nanobelts gives rise to two additional shoulder peaks at 234.3 and 231.1 eV (Supporting Information: Figure S4B), indicating the existence of Mo 5+ in the I-MoO 3−x nanobelts.Correspondingly, the XPS O 1s spectrum of the I-MoO 3−x nanobelts exhibits a stronger peak at 532.3 eV than that of MoO 3 nanobelts (Figure 2G), which might be attributed to the presence of oxygen vacancies.The XPS Na 1s spectrum of MoO 3 nanobelts did not reveal any discernible peak (Supporting Information: Figure S4C), while the XPS Na 1s spectrum of I-MoO 3−x nanobelts distinctly displayed a peak at 1071.2 eV, thereby providing conclusive evidence of the existence of Na + in the I-MoO 3−x nanobelts (Supporting Information: Figure S4C).The results obtained from TEM, ESR, and XPS analyses have proved the significant structural changes in the MoO 3 nanobelts induced by Na + /H 2 O intercalation.

| Photophysical and photodynamic performance of MoO 3 and I-MoO 3−x nanobelts
The alteration in the structure of MoO 3 resulting from intercalation may give rise to significant changes in its photophysical and photodynamic properties.One obvious change that is easily observed is the color change from white to dark blue after the intercalation, as evidenced by the photos of the MoO 3 and I-MoO 3−x nanobelts (inset in Figure 2H).As shown in Figure 2H, the MoO 3 nanobelts (0.1 mg/mL) show weak absorption peaks within the spectrum of 400-1300 nm, and the intensity of their absorption gradually diminishes (Figure 2H).In contrast, the absorption of I-MoO 3−x nanobelts (0.1 mg/mL) increases gradually as the wavelength increases, leading to a strong absorption within the NIR window, which might be attributed to local surface plasmon resonance effects caused by oxygen vacancies. 42,43he I-MoO 3−x nanobelts have abundant oxygen vacancies and strong NIR absorption performance, making them potentially applicable as NIR photosensitizers for PDT.To test the ability of the I-MoO 3−x nanobelts to produce ROS under a NIR laser (808 nm), 1,3diphenylisobenzofuran (DPBF) was selected as the ROS sensing agent, which can be oxidized into cyclohexa-3, 5-diene-1,2-diylbis(phenylmethanone), leading to the disappearance of its distinctive absorption peak at 413 nm.Compared to the DPBF alone (Figure 3A), the UV absorption of DPBF treated with the MoO 3 nanobelts showed almost no significant change under an 808 nm laser irradiation for 10 min (Figure 3B), suggesting that the MoO 3 nanobelts had no photodynamic activity.After treating with the I-MoO 3−x nanobelts, a slight reduction in the UV absorption intensity of DPBF was observed (Figure 3C), while the UV absorption of DPBF significantly decreased after NIR laser (808 nm, 1 W/cm 2 ) irradiation (Figure 3D), and the intensity gradually decreased with an increase in illumination time, with a decrease of 36% after 10 min (Figure 3D).These results demonstrated that the I-MoO 3−x nanobelts can produce ROS under NIR laser irradiation, exhibiting excellent NIR photodynamic properties.For the purpose of demonstrating the type of ROS generated by the I-MoO 3−x nanobelts when subjected to NIR laser (808 nm, 1 W/cm 2 ), we utilized a singlet oxygen sensor green (SOSG) to detect the signal of 1  To enhance the understanding of the NIR-triggered ROS generation mechanism, we utilized UV-vis-NIR diffuse reflectance spectroscopy to determine the band structures of MoO 3 and I-MoO 3−x nanobelts (Supporting Information: Figure S6A,B).The bandgap of MoO 3 was measured to be 2.92 eV (Supporting Information: Figure S6C), whereas I-MoO 3−x displayed a bandgap of 1.87 eV (Supporting Information: Figure S6D), which might be due to the broadening of the van der Waals interlayer spacing of MoO 3 after Na + /H 2 O intercalation.The I-MoO 3−x nanobelts with narrower bandgap can absorb lower-energy NIR photons, making it easier for excited valence band electrons to reach the conduction band and subsequently generate electron-hole pairs.These pairs are created to form a charge-separated state, which then allows for electron transfer with the surrounding oxygen molecular to produce •O 2

−
. During the intercalation process, a large number of oxygen vacancies are generated, which can enhance the NIR optical absorption ability of I-MoO 3−x (Figure 2H), thereby enhancing its near-infrared photodynamic activity.Therefore, intercalation-induced activation of layered MoO 3 nanobelts endows it with high activity as a NIR type I photosensitizer for photodynamic periodontitis treatment.

| Photodynamic antibacterial in vitro
The effectiveness of MoO 3 and I-MoO 3−x nanobelts as antibacterial agents under NIR laser irradiation was determined by means of the colony method against both S. aureus and E. coli.Under dark conditions, even high concentrations (1 mg/mL) of MoO 3 and I-MoO 3−x nanobelts do not have a significant impact on bacterial (S. aureus and E. coli) growth (Supporting Information: Figure S7), demonstrating their low toxicity in the absence of light.Promisingly, the bacteria in the I-MoO 3−x + NIR group were effectively suppressed, with a killing rate of 99.9999% for both S. aureus and E. coli., compared to the MoO 3 + NIR group (Supporting Information: Figure S7), indicating that the I-MoO 3−x nanobelts possessed the excellent NIR photodynamic antibacterial activity.
To further investigate the photodynamic antibacterial effect of I-MoO 3−x nanobelts, its antibacterial properties with different concentrations (10, 20, 40, 50, 60, 80, and 100 μg/mL) under different irradiation times (3, 5, and 10 min) were also verified using the plate counting method.As for S. aureus (Figure 4A,B), the I-MoO 3−x nanobelts with concentrations of 10 and 20 μg/mL exhibit less significant antibacterial effects, and even after prolonged exposure to NIR laser (808 nm, 1 W/cm 2 , 10 min), there are still 89.01%and 86.79% remaining for bacterial survival, respectively.When the concentration of the I-MoO 3−x nanobelts is increased to 40 μg/mL, 73.84% of the bacteria are killed after irradiation for 5 min, and the mortality rate will reach 92.92% if the light exposure time is extended to 10 min.Importantly, 50 μg/mL of the I-MoO 3−x nanobelts exhibit almost 100% antibacterial I-MoO 3−x is the key to efficiently inhibiting bacteria.The photodynamic antibacterial efficiency of I-MoO 3−x nanobelts was further verified by co-staining with SYTO9 and propidium iodide (PI).The phosphate buffer solution (PBS), PBS + NIR, and I-MoO 3−x groups exhibited robust green fluorescence and an absence of red fluorescence (Figure 4C), suggesting that the bacteria maintained their physiological integrity.In comparison, it was observed that the I-MoO 3−x + NIR group led to a significantly higher proportion of deceased bacteria, which matches well with the results in the colony method.
The SEM images were used to examine the changes in the morphology of S. aureus after the various treatments.The bacteria exhibit sustained morphology in the PBS group, both before and after the NIR laser treatment (Figure 4D).In addition to intact bacteria, nanobelt types of I-MoO 3−x have also been clearly observed (Figure 4D), proving that I-MoO 3−x has no significant effect on bacterial growth in the absence of NIR.However, after NIR irradiation for 10 min, S. aureus exhibits a significant reduction in growth, further demonstrating the excellent antibacterial performance of the I-MoO 3−x nanobelts.

| Photodynamic periodontitis treatment in vivo
To evaluate the antibacterial activity of I-MoO 3−x in vivo under infectious conditions in clinical treatments, periodontitis as a representative chronic infectious disease triggered by microorganisms in the dental plaque was established in rats by the continuous "∞-ligature method" and sugar-rich diet feeding (Supporting Information: Figure S11). 44The process of model preparation and drug administration is shown in Figure 5A.The effect of alleviating periodontal inflammation response in vivo was evaluated by the gingival bleeding index (GBI) score, threedimensional microcomputed tomography (CT) scanning of alveolar bone, and staining of periodontal tissue pathological slices.Bacteria aggregation around the tooth will cause inflammation in gingiva tissue, which leads to bleeding. 45ompared with the other treatment groups, including control, I-MoO 3−x , clinical drug chlorhexidine (CHX), and healthy, the GBI score of the I-MoO 3−x + NIR group was significantly decreased (Figure 5B,C).Although there were individual bleeding spots in the gingiva of the I-MoO 3−x + NIR group, there was no significant difference in GBI between the laser group and the healthy group.The GBI of each group was consistent with the bacterial content in the oral cavity tested by standard plate counting above.7][48][49] Different solutions were locally delivered to dental plaque twice a day with a brief 5-minute exposure time, and NIR laser exposure was coupled with the 5-minute time in half of the I-MoO 3−x groups.After 2 weeks of topical treatment, the area of dental plaque on the surface of teeth stained with basic fuchsin in the group of I-MoO 3−x + NIR was close to the healthy group, which was significantly decreased compared to the other treatment groups (Figure 5D,E).Correspondingly, the standard plate counting method was conducted to evaluate the bacterial activity in the remaining dental plaque.After 24 h of incubation under aerobic (Figure 5F,G) and anaerobic (Supporting Information: Figure S12) conditions, the I-MoO 3−x + NIR group exhibited no obvious colonies compared to the other groups including control, I-MoO 3−x , and CHX groups, proving that the I-MoO 3−x nanobelts could also significantly inhibit both aerobic and anaerobic bacteria under a NIR laser irradiation.These results suggested that microorganisms in the dental plaque were killed by the I-MoO 3−x + NIR, and even the dental plaque itself was dispersed.
3D micro-CT reconstructed images showed the increased vertical distance between the cementoenamel junction of teeth and the crest of alveolar bone in the control (periodontitis) group confirming the loss of alveolar bone (Figure 6A).Quantification analyses of micro-CT through bone volume/tissue volume (BV/TV) and trabecular number (Tb.N) confirmed that the therapeutic effect of I-MoO 3−x under NIR irradiation was higher than that of the control group, but there was no statistical difference with that of the healthy group.(Figure 6B,C).Compared with the periodontitis group, alveolar bone loss (ABL) was significantly reduced in the I-MoO 3−x + NIR treatment group, almost the same as in the healthy group.The corresponding quantitative results of ABL showed the same trend (Figure 6D).
Hematoxylin-eosin (H&E) staining and Massonstained periodontium presented a normal periodontal structure in the healthy group (Figure 6E).The control (periodontitis) group showed a loss of alveolar bone in the region between the upper (the black arrow in Figure 6E H&E staining) molars and bifurcation of the root of the molar (the blue arrow in Figure 6E H&E staining).Meanwhile, the breakdown of the epithelial barrier and clear destruction of collagen fibrous (the red arrow in Figure 6E Masson staining) tissues were observed in the control (periodontitis) group.1][52][53] Previous studies have confirmed that bacteria compromise the physical barrier of the oral mucosa by infecting the oral mucosal tissue. 50,54In our study, the same intact keratinized squamous epithelial was observed in the I-MoO 3−x + NIR group as in the healthy group.However, there was a minor loss of collagen and rete-ridge definition, and more gingival microvasculature was detected in the I-MoO 3−x + NIR group (the black arrow in Figure 6E Masson staining), which may be related to the stimulation of the ligature. 55TRAP staining was performed to detect the number of osteoclasts.TRAPpositive cells are characterized by multinucleated giant cells and claret-red in the cytoplasm.Compared with the control (periodontitis) group, few TRAP-positive cells were observed in the I-MoO 3−x + NIR group (Figure 6E).The decreased number of osteoclasts demonstrates evident demotion of bone destruction. 5,56This could explain the results of the alveolar bone loss in micro-CT.These results confirmed that topical I-MoO 3−x + NIR treatment is quite effective in oral antibacterial and alleviating inflammation of periodontitis.Importantly, compared with the healthy control group, no visible organ damage or inflammatory lesions were observed in the main organ H&E slices of rats in the I-MoO 3−x + NIR group (Supporting Information: Figure S13), suggesting that the I-MoO 3−x nanobelts have no significant side effects or toxicity.

| CONCLUSION
In summary, we have developed a simple aqueous solution intercalation strategy for the preparation of I-MoO 3−x nanobelts by co-intercalating Na + and H 2 O into their van der Waals gaps, which can be used as a NIR type I photosensitizer for photodynamic periodontitis treatment.The co-intercalation of Na + /H 2 O can reduce the length of MoO 3 nanobelts, generate rich oxygen vacancies, and enlarge its interlayer gap, thus boosting its optical absorption in the NIR region.Comparing the MoO 3 nanobelts, the I-MoO 3−x nanobelts have a narrower bandgap and abundance of defects, and can absorb lower-energy NIR photons, thus allowing the excited valence band electrons to reach the conduction band more easily and produce electron-hole pairs to facilitate electron transfer with oxygen molecular to generate •O 2 − .Therefore, the prepared I-MoO 3−x nanobelts possess a spectral antibacterial activity against E. coli and S. aureus, thus yielding a positive therapeutic effect for periodontitis.Our study has demonstrated that liquid-phase intercalation is a simple but effective way to activate layered MoO 3 nanomaterials for high-performance NIR photodynamic periodontitis treatment, which may be applicable to cancer therapy.
Figure S2B) of a typical I-MoO 3−x nanobelt (Supporting Information: Figure S2B, inset) displays a similar set of rectangular-shaped bright diffraction spots, indicating the unchanged single-crystalline nature after intercalation (Supporting Information: Figure S2C, inset).The XRD pattern of the I-MoO 3−x nanobelts is well in agreement with that of the hydrated molybdenum bronze [Na(H 2 O) 2 ] 0.25 MoO 3 (Figure 2A), 41 and the formula of I-MoO 3−x is calculated to [Na(H 2 O) 2 ] 0.27 MoO 3 by using an inductively coupled plasma optical emission spectrometer (ICP-OES), which can be considered as the Na + /H 2 O co-intercalated MoO 3 compound.The EDS element mapping of I-MoO 3−x nanobelts further suggests that sodium ions are uniformly distributed (Supporting Information: Figure S3).The MoO 3 and I-MoO 3−x nanobelts were further characterized by various techniques to gain a better understanding of the intercalation-induced structural changes.From the y-axis perspective, the atomic resolution high-angle annular dark-field scanning transmission F I G U R E 1 Schematic illustration of the preparation of I-MoO 3−x nanobelts and their application as a near-infrared photosensitizer for photodynamic periodontitis treatment.
ray diffraction (XRD) patterns of MoO 3 and I-MoO 3−x nanobelts with insert references.Scanning electron microscope images of (B) MoO 3 and (C) I-MoO 3−x nanobelts.Atomic-resolution high-angle annular dark-field scanning transmission electron microscope images of a typical (D) MoO 3 and (E) I-MoO 3−x nanobelt.(F) Electron spin resonance spectra, (G) high-resolution XPS O 1s, and (H) UV-vis-NIR spectra of MoO 3 and I-MoO 3−x nanobelts.Inset in (H) are the photos of the aqueous dispersions of (left) MoO 3 and (right) I-MoO 3−x nanobelts.NIR, near-infrared.
O 2 .As shown in Supporting Information: Figure S5A,B, almost no fluorescence signal of SOSG can be detected with the addition of the I-MoO 3−x nanobelts, even after irradiation by an 808 nm laser for 10 min, indicating that the I-MoO 3−x nanobelts cannot produce 1 O 2 under NIR laser irradiation.Moreover, we measured the ESR spectra of MoO 3 and I-MoO 3−x nanobelts in water or dimethyl sulfoxide (DMSO) solution using 5,5dimethyl-1-pyrroline-N-oxide (DMPO) as a scavenger for •OH or •O 2 − under dark or irradiation of an 808 nm NIR laser for 10 min.As shown in Supporting Information: Figure S5C, there is no effective signal in the ESR spectra of MoO 3 and I-MoO 3−x nanobelts in water even after illumination, which proves that •OH cannot be generated in this process.Upon substitution of water with DMSO, a distinctive signal indicative of •O 2 − emerged in the ESR spectrum of the I-MoO 3−x nanobelts after treatment by an 808 nm laser for 10 min (Figure 3F), suggesting that the I-MoO 3−x nanobelts could act as the NIR Type I photosensitizer to generate •O 2 − under a NIR laser irradiation.

F
I G U R E 3 UV/vis absorption spectra of DPBF treated with (A) NIR, (B) MoO 3 + NIR (808 nm, 1 W/cm 2 ), (C) I-MoO 3−x , and (D) I-MoO 3−x + NIR (808 nm, 1 W/cm 2 ) for different times.(E) The relative absorption intensity at 410 nm of DPBF with different treatments.(F) Electron spin resonance spectra of •O 2 − trapped by DMPO in the presence of MoO 3 or I-MoO 3−x under dark or irradiation of NIR laser (808 nm, 1 W/cm 2 ).DMPO, 5,5-dimethyl-1-pyrroline-N-oxide; DPBF, 1,3-diphenylisobenzofuran. NIR, near-infrared.activity after irradiation for 10 min, which will be further shortened to 5 and 3 min when its concentration increases to 60-80 and 100 μg/mL, respectively.As shown in Supporting Information: Figure S8, the bacteriostatic concentration and time required for the I-MoO 3−x nanobelts to inhibit the growth of E. coli are similar to those for S. aureus.These results indicate that the antibacterial effect of the I-MoO 3−x nanobelts becomes more obvious with an increase in the concentration and irradiation time of I-MoO 3−x nanobelts, which can achieve the rapid (10 min) antibacterial activity at lower concentration (50 μg/mL).In addition, Porphyromonas gingivalis (P.gingivalis) is selected as an anaerobic bacteria to test antibacterial activity of I-MoO 3−x nanobelts.As shown in Supporting Information: Figure S9, the I-MoO 3−x nanobelts (50 μg/mL) also exhibit excellent photodynamic antibacterial effects against anaerobic bacteria under the irradiation by a NIR laser (808 nm, 1 W/cm 2 ) for 10 min.To eliminate the impact of local heating on bacterial growth, we have measured the photothermal heating curves of MoO 3 and I-MoO 3−x with different concentrations under the same NIR laser radiation (Supporting Information: Figure S10).Compared to MoO 3 , I-MoO 3−x can slowly increase in temperature under laser irradiation, and the temperature can rise to around 47 °C at a concentration of 100 μg/mL for 10 min.After combining I-MoO 3−x (100 μg/mL) with bacteria (both S. aureus and E. coli), we put the sample in an incubator and increased the temperature to 50 °C for 10 min.As shown in Supporting Information: Figure S10C, the growth of both S. aureus and E. coli was almost unaffected, while the bacteria in the I-MoO 3−x + NIR group were almost completely killed (Figure 4A and Supporting Information: Figure S7), proving that the photodynamic activity of F I G U R E 4 (A) Photographs of Saccharomyces aureus on agar plates treated with different concentrations of I-MoO 3−x (10, 20, 40, 50, 60, 80, and 100 μg/mL) with or without the irradiation of NIR laser (808 nm, 1 W/cm 2 ) for different times.(B) Survival rate of S. aureus in (A) was calculated with the plate count method.(C) Confocal images of S. aureus treated with PBS and I-MoO 3−x (50 μg/mL) with or without the irradiation of NIR laser (808 nm, 1 W/cm 2 ) for 10 min.(D) The SEM images of S. aureus treated with PBS, PBS + NIR, I-MoO 3−x , and I-MoO 3−x + NIR.NIR, near-infrared.

F
I G U R E 5 (A) Timeline of the establishment of the periodontitis model and drug administration.(B) Photographs of the gingival bleeding of the oral cavity after treatment under microscopy.(C) Quantitative results of the gingival bleeding index (GBI, n = 5).(D) Photographs of the basic fuchsin-stained plaques under microscopy.(E) Quantitative results of the basic fuchsin-stained plaque biofilm area (n = 3).(F) Digital photos of oral cavity bacteria re-culture after treatment under aerobic culture conditions.(G) Quantitative results of the standard plate counting assay from (F) (n = 3).Significant differences: *p < 0.05, **p < 0.01, ***p < 0.001.F I G U R E 6 (A) Representative micro-CT images of maxillary alveolar bone surrounding the maxillary second molars (M2) after the treatment: the above image in each group displays the three-dimensional reconstruction image and the image below in each group represents the bucco-palatal sagittal slice of micro-CT.The quantitative analysis of bone-related parameters: (B) BV/TV and (C) Tb.N. (D) Quantitative result of the distance between CEJ and the crest of alveolar bone along three buccal roots of the first molar M1 and two of the second molar M2.Significant differences: *p < 0.05, **p < 0.01, ***p < 0.001 (n = 3).(E) H&E-stained, Masson-stained, and TRAP-stained tissue sections of alveolar bones between molars.