Supramolecular Polymerization Powered by Escherichia coli : Fabricating a Near-Infrared Photothermal Antibacterial Agent in Situ

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE3 Oct 2022Supramolecular Polymerization Powered by Escherichia coli: Fabricating a Near-Infrared Photothermal Antibacterial Agent in Situ Zihe Yin, Yuchong Yang, Jinpeng Yang, Guobin Song, Hao Hu, Peng Zheng and Jiang-Fei Xu Zihe Yin Key Lab of Organic Optoelectronics and Molecular Engineering, Department of Chemistry, Tsinghua University, Beijing 100084 Google Scholar More articles by this author , Yuchong Yang Key Lab of Organic Optoelectronics and Molecular Engineering, Department of Chemistry, Tsinghua University, Beijing 100084 Google Scholar More articles by this author , Jinpeng Yang Key Lab of Organic Optoelectronics and Molecular Engineering, Department of Chemistry, Tsinghua University, Beijing 100084 Google Scholar More articles by this author , Guobin Song State Key Laboratory of Coordination Chemistry, Chemistry and Biomedicine Innovation Center (ChemBIC), School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023 Google Scholar More articles by this author , Hao Hu Key Lab of Organic Optoelectronics and Molecular Engineering, Department of Chemistry, Tsinghua University, Beijing 100084 Google Scholar More articles by this author , Peng Zheng State Key Laboratory of Coordination Chemistry, Chemistry and Biomedicine Innovation Center (ChemBIC), School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023 Google Scholar More articles by this author and Jiang-Fei Xu *Corresponding author: E-mail Address: [email protected] Key Lab of Organic Optoelectronics and Molecular Engineering, Department of Chemistry, Tsinghua University, Beijing 100084 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.021.202101490 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail An Escherichia coli reduction-powered supramolecular polymerization is reported, leading to the fabrication of a near-infrared (NIR) photothermal antibacterial agent in situ. To this end, a bifunctional monomer containing two viologen moieties was designed. When incubating E. coli with the bifunctional monomer and cucurbit[8]uril, viologen moieties were reduced to viologen cation radicals by E. coli, and a supramolecular polymer with supramolecular dimers of viologen cation radicals integrated into the main chain was fabricated on the surface of E. coli. The NIR photothermal conversion property of the supramolecular dimer of viologencation radicals endowed the supramolecular polymer with photothermal antibacterial ability, and this performance was further improved by the local enrichment effect of supramolecular polymers and their enhanced adsorption onto the bacteria surface. Moreover, only certain bacteria, such as E. coli, possess the reducing ability to power supramolecular polymerization, whereas many other bacteria, such as Bacillus subtilis, Pseudomonas aeruginosa, and Staphylococcus aureus, do not possess this ability. Therefore, the supramolecular polymer exhibits outstanding bacterial inhibition efficiency (>99.9%) with high specificity toward E. coli under 1064 nm NIR irradiation. It is anticipated that this biologically powered in situ supramolecular polymerization strategy presents great potential in fabricating smart biomedical supramolecular materials with adaptivity and programmability. Download figure Download PowerPoint Introduction In an out-of-equilibrium biological system, many metabolic activities and physiological functions are energy-consuming processes. For bacteria, a transmembrane redox potential generated by procaryote respiration not only supports biological functions,1,2 but also can be employed as an energy source for various artificial applications, such as bioelectrosynthesis, initiation of polymerization, and microbial fuel cells.3–10 Notably, the reducing ability of bacteria originating from this redox potential has been applied to the in situ fabrication of biomedical materials at the places where they function.7,8,11 Compared with conventional biomedical materials, which are generally readily constructed and then transported to the working area, in an in situ fabrication strategy, the targeting ability could be realized, and then the transporting process could be omitted, thus improving the specificity and adaptivity of biomedical materials.12–17 As an integration of polymer science and supramolecular chemistry, supramolecular polymers exhibit superiority in various biomedical applications18–25 such as drug delivery,26–30 bioimaging and diagnosis,31,32 phototherapy,33–35 and tissue engineering.36–38 If supramolecular polymerization could be implemented in biological environments, biodegradability and stimuli-responsiveness originating from the dynamic and reversible noncovalent interactions may be imparted to the supramolecular biomedical materials that are fabricated in situ. Therefore, we wondered whether the reducing ability of bacteria could be utilized to power a supramolecular polymerization process. Our design idea arises from the fact that some electron-deficient dyes with large π-conjugated structures can be reduced to organic free radicals in a local reducing environment produced by some facultative anaerobic bacteria such as Escherichia coli.11,39,40 On the one hand, the self-assembly of these organic free radicals into supramolecular free radicals may serve as the driving force for supramolecular polymerization.41–50 On the other hand, supramolecular free radicals are expected to possess remarkable near-infrared (NIR) absorption,51–54 which may endow the supramolecular polymers with an outstanding photothermal conversion property and photothermal antibacterial performance. Therefore, by integrating the formation of NIR photothermal conversion motifs into a bacteria-powered supramolecular polymerization process, NIR photothermal antibacterial supramolecular polymers are hopeful to be constructed. Such an antibacterial agent would be selectively activated by the reduction with E. coli, whereas it could exactly kill thebacteria that induced the activation before. Such a “suicide” process of the bacteria may impart high antibacterial specificity to the agent. To this end, we designed and synthesized a bifunctional monomer named isviologen-diazabicyclo[2.2.2]octane-viologen (VDV), which contained two viologen moieties as end groups linked by a rigid and positively charged 1,4-diazabicyclo[2.2.2]octane unit. It was discovered that E. coli possessed the ability for reducing viologen into viologen cation radical under anaerobic conditions. As shown in Schemes 1a and 1b, when incubating E. coli with equimolar amounts of VDV and cucurbit[8]uril (CB[8]), the viologen moieties could be reduced to viologen cation radicals, and an in situ supramolecular polymerization could occur driven by the 2:1 host–guest complexation between viologen cation radicals and CB[8], generating a supramolecular polymer (VR-SP) with supramolecular dimers of viologen cation radicals integrated into the main chain. Moreover, the supramolecular dimer of viologen cation radicals exhibited absorption at the NIR biowindow, which might endow the VR-SP with photothermal antibacterial activity. By forming such an NIR photothermal antibacterial agent, E. coli could be killed by the local increase in temperature under NIR irradiation. This photothermal therapy is a powerful antimicrobial method with high penetration depth, few side effects, and low risk in generating drug resistance.55–71 Furthermore, the photothermal antibacterial efficiency could be enhanced significantly by the local enrichment effect of supramolecular polymers and their enhanced adsorption onto the surface of bacteria, since the positively charged VR-SP would exhibit strong interactions with the negatively charged surface of the bacteria. In addition, after completely inhibiting E. coli, VR-SP would be spontaneously degraded, as the viologen cation radicals are easily oxidized (commonly by oxygen) without the continuous reduction powered by living bacteria, thus turning off the antibacterial activity automatically. We therefore envisioned that, based on the supramolecular polymerization powered by E. coli, an NIR photothermal antibacterial agent with high antibacterial efficiency, specificity, and degradability could be fabricated in situ. Scheme 1 | (a) Chemical structures of the bifunctional monomer VDV, CB[8], and the supramolecular dimer of viologen cation radicals, and a graphic representation of the photothermal antibacterial supramolecular polymer VR-SP. (b) A schematic representation for the photothermal therapy using an NIR photothermal antibacterial agent fabricated through in situ supramolecular polymerization powered by E. coli. Download figure Download PowerPoint Experimental Method Unless otherwise noted, materials were obtained from commercial suppliers and were used without further purification. All types of bacteria were grown in Luria–Bertani (LB) culture medium at 37 °C for 8 h before further application. The detailed synthetic routes of VDV and 1-methyl-1′-(3-(trimethylammonio)propyl)-viologen tribromide (VMA), as well as detailed experimental methods are shown in Supporting Information. UV–vis spectra were measured using a HITACHI UH4150 spectrophotometer (Hitachi, Ltd., Tokyo, Japan) or a GE Ultrospec9000 spectrophotometer (General Electric Company, Boston, Massachusetts, United States). VDV (0.20 mM), or VDV (0.20 mM) and CB[8] (0.20 mM) were dissolved in LB medium. Then, 3.5 mL LB medium with dissolved agent was transferred into a quartz cuvette (1.00 cm). Then 100 μL of the suspension of bacteria (E. coli, Bacillus subtilis, Enterococcus faecalis, Staphylococcus aureus, or Pseudomonas aeruginosa) was added. The samples were sealed and incubated at 37 °C for 20 h, and then the UV–vis spectra were recorded. Electron paramagnetic resonance (EPR) spectra were obtained using a JEOL JES-FA200 spectrometer. VDV (0.20 mM), or VDV (0.20 mM) and CB[8] (0.20 mM) were dissolved in LB medium. Then 100 μL of the suspension of bacteria (E. coli or B. subtilis) was added into 3.5 mL LB medium with dissolved agent. The medium was blown by a pipette and transferred into a capillary tube (Φ = 0.5 mm). After being filled with the medium, the capillary tube was sealed by melted paraffin. The samples were incubated at 37 °C for 40 h and then measured with the EPR spectrometer. The redox potential of VDV and VDV-CB[8] was measured by differential pulse voltammetry. The differential pulse voltammetry measurements were carried out with the three-electrode system using a CHI660E electrochemical workstation (CH Instruments, Inc., Austin, Texas, United States). A glass carbon disk (0.07 cm2) was applied as the working electrode with a Pt counter electrode and a saturated Ag/AgCl electrode (KCl saturated) reference electrode. The working electrode was polished with 0.3 and 0.05 μm alumina (Al2O3) successively and washed with deionized water before use. NaCl solution (50 mM) was used as supporting electrolyte. After degassing by passing N2 through the solution for at least 30 min, differential pulse voltammetry measurements were conducted on VDV (0.20 mM), or VDV (0.20 mM) and CB[8] (0.20 mM) solution. The redox potential of the bacterial cultural medium was measured using a hand-held ORP probe (Shanghai Sanxin SX712). The E. coli suspension (15 μL) was added into a cuvette containing LB medium (2 mL). The composite electrode of the ORP probe (using saturated Ag/AgCl electrode as reference electrode) was also immersed in the medium to monitor redox potential changes. The cuvette was sealed to prevent interference from air, and the bacteria were then incubated at 37 °C for 12 h. After that, the redox potential of the medium of E. coli was detected. For E. faecalis, S. aureus, B. subtilis, and P. aeruginosa, the redox potential was measured with the same method. A Nanowizard 4 atomic force microscopy (AFM; JPK, Berlin, Germany) coupled to an inverted microscope (Olympus IX73, Olympus Corporation, Tokyo, Japan) was used to perform the AFM imaging on live E. coli. The AFM was equipped with a cell-culture chamber to keep the temperature appropriate for E. coli (37 ± 1 °C). And a poly-d-lysine-coated mica disc was used as the substrate for E. coli immobilization and subsequent imaging. A PFQMN-Lc-A-CAL AFM cantilever (Bruker Corporation, Billerica, Massachusetts, United States) with a nominal spring constant of 0.1 N/m was used. For detailed scanning parameters and sample preparation procedures, please refer to Supporting Information. An FC-1064-3000-MM Laser Light Source (Wavicle Laser) was applied to produce a 1064 nm NIR laser in the photothermal therapy under NIR irradiation, and the NIR laser-induced heat experimental phenomenon was recorded by a Fluke Ti450 Infrared Camera. VDV (0.20 mM), or VDV (0.20 mM) and CB[8] (0.20 mM), or VMA (0.40 mM) and CB[8] (0.20 mM) were dissolved in LB medium, and 3.5 mL LB medium with dissolved agents was transferred into a quartz cuvette (for control group, LB medium without dissolved agent was applied). Then 30 μL of the suspension of bacteria (E. coli or B. subtilis) was added. The samples were sealed and incubated at 37 °C for 20 h. After incubation, the samples were irradiated by a 1064 nm NIR laser (2.0 W/cm2) for 12.5 min at 37 °C, and the temperature of the cuvette was recorded by Fluke Ti450 Infrared Camera every 2.5 min. The inhibition ratio was determined by calculating the number of colony-forming units (CFUs). The cell viability of both BEAS-2B and NCM460 cells was measured in vitro by the Cell Counting Kit-8 (CCK-8) assay. The cells were planted in 96-well plates with a density of about 5 × 103–8 × 103 cells per well for 24 h. Then the cells were treated with the medium containing equimolar amounts of VDV and CB[8] or VDV alone ranging from 0.10 to 0.40 mM for 24 h. After that, the cells were treated with fresh medium containing CCK-8 for 1 h to replace the previous medium. A microplate reader (EnVision, PerkinElmer, Inc., Waltham, Massachusetts, United States) was used for the measurement of absorbance at 450 nm, which was applied to calculate the cell viability. For detailed cell culture conditions, please refer to Supporting Information. Results and Discussion The formation of viologen cation radicals from VDV by E. coli reduction was characterized by UV–vis spectroscopy. As shown in Figure 1a, when incubating E. coli with VDV or equimolar amounts of VDV and CB[8] for 12–20 h, the color of the solution changed from light yellow to blue or dark violet, which are the characteristic colors of viologen cation radical monomer or supramolecular dimer of viologen cation radicals, respectively. With the existence of CB[8], absorption bands ranging from the UV to the NIR regionpeaking at 366, 542, and 872 nm were observed, which were attributed to the characteristic absorption of the supramolecular dimer of viologen cation radicals. In contrast, the characteristic absorption peaks of viologen cation radical monomer at 396 and 601 nm were observed in the absence of CB[8]. These results suggest the formation of viologen cation radicals. Figure 1 | (a) UV–vis spectra and images of the solution of VDV with or without CB[8] after incubating with E. coli at 37 °C for 20 h. (b) EPR spectra of VDV with or without CB[8] after incubating with E. coli at 37 °C for 40 h. (c and d) UV–vis spectra and images of the solution of VDV and CB[8] after incubating with E. coli, E. faecalis, B. subtilis, P. aeruginosa, or S. aureus at 37 °C for 20 h (VDV: 0.20 mM, CB[8]: 0.20 mM, medium: LB medium). Download figure Download PowerPoint The generation of viologen cation radicals by E. coli reduction was further confirmed by EPR. As shown in Figure 1b, an EPR signal with g factor of 2.0034 was observed after incubating E. coli with VDV. When incubating E. coli with equimolar amounts of VDV and CB[8], an EPR signal with the same g factor but significantly lower intensity was observed. Given that no EPR signal was observed when incubating E. coli without any additives, the observed signals were ascribed to viologen cation radical monomer; the lower signal intensity observed in the presence of CB[8] indicated that most of the generated viologen cation radicals self-assembled with CB[8] to form the supramolecular dimer of viologen cation radicals.48,72 In contrast, no EPR signal was observed when incubating VDV or VDV-CB[8] with aerobic bacteria B. subtilis, indicating the negligible reducing ability of B. subtilis to viologen ( Supporting Information Figure S10). To understand whether E. coli was the only bacteria possessing the reducing ability, five types of bacteria including E. coli, E. faecalis, B. subtilis, S. aureus, and P. aeruginosa were incubated with VDV-CB[8], and UV–vis spectra were recorded after 20 h incubation. As shown in Figure 1c, only E. coli presented remarkable reducing ability to generate viologen cation radicals. E. faecalis showed a much weaker reducing capacity than E. coli. For the other three types of bacteria, no absorption of viologen cation radicals was observed (Figure 1d), indicating their low reducing ability. To study the reducing ability of different types of bacteria quantitatively, the redox potential of the medium of different types of bacteria after incubation was measured. As shown in Table 1, the redox potential of E. coli culture medium reached as low as −551 mV. As for E. faecalis, the redox potential of the medium was −429 mV. The other bacteria media exhibited higher redox potential (Table 1). Compared with the first one-electron reduction potential of VDV (−558 mV) and VDV-CB[8] (−481 mV) measured by differential pulse voltammetry ( Supporting Information Figure S12), we judged that E. coli could reduce VDV to generate viologen cation radicals with or without CB[8]. E. faecalis could only reduce a small proportion of VDV with the existence of CB[8], whereas the other three types of bacteria could not reduce VDV. All the above results reveal that only E. coli possesses the remarkable reducing ability to generate viologen cation radicals, and the viologen cation radicals self-assemble into supramolecular dimer of viologen cation radicals in the presence of CB[8]. Table 1 | The Redox Potential of the Medium of Different Types of Bacteria after Incubationa Bacteria Redox Potential (mV) E. coli −551 E. faecalis −429 B. subtilis −171 P. aeruginosa −289 S. aureus −300 aAll redox potential values were measured related to saturated Ag/AgCl electrode. The formation of supramolecular polymers in solution driven by the host–guest complexation between viologen cation radicals and CB[8] was investigated by diffusion-ordered NMR spectroscopy (DOSY). Equimolar amounts of VDV and CB[8] were dissolved in D2O and then VDV was reduced via a photoinduced electron transfer process under 254 nm UV irradiation. After the complete reduction of viologen moieties, the diffusion coefficient of the species in solution measured by DOSY decreased significantly from 1.80 × 10−10 to 6.35 × 10−11 m2·s−1, indicating the formation of supramolecular polymers. According to the Stokes–Einstein equation, the degree of polymerization (DP) was approximately 23 (see Supporting Information). This result indicates that VR-SP is formed in aqueous solution successfully. To investigate whether VR-SP could be fabricated and adsorbed on the surface of E. coli, AFM imaging was employed. E. coli was incubated with equimolar amounts of VDV and CB[8] for 20 h to prepare VR-SP on the surface. After the color of the medium changed from light yellow to dark violet, E. coli was transferred into the petri dish, and fixed onto a poly-d-lysine coated mica disc. After that, E. coli was imaged directly by AFM. Compared with the untreated E. coli shown in Figure 2a, the edge of the E. coli incubated with VDV-CB[8] was uneven, as shown in Figures 2b, 2c, and 2e, suggesting the formation and adsorption of supramolecular polymers on the surface of E. coli. To further support that the adsorbed supramolecular polymers were responsible for the observed uneven edge of E. coli, the buffer in culture dish with the imaged bacteria in Figure 2c was carefully replaced by phosphate-buffered saline (PBS) without disturbing the bacteria. As shown in Figures 2d and 2f, the E. coli again exhibited the smooth edge, which could be explained as the oxidation of VR-SP by air; therefore, the adsorbed supramolecular polymers were degraded. Considering that VDV alone could not lead to the uneven edges of bacteria ( Supporting Information Figure S17), AFM observations suggest that VR-SP is formed and adsorbed on the surface of bacteria. Figure 2 | AFM images of (a) untreated E. coli and (b and c) E. coli incubated with 0.20 mM VDV and 0.20 mM CB[8]. (d) AFM image of the E. coli in image (c) after replacing the solution in the petri dish with PBS buffer. (e) Magnified image of the labeled area in (c). (f) Magnified image of the labeled area in (d). Download figure Download PowerPoint The DP of VR-SP generated on the surface of E. coli was further estimated by end-group analysis measured by EPR. Viologen cation radical monomer, which is the end group of VR-SP, can be characterized by EPR, whereas the supramolecular dimer of viologen cation radicals exhibits no EPR signal. Combining the EPR measurements of E. coli with VR-SP adsorption and viologen cation radical monomer, the DP of VR-SP was estimated to be approximately 17 (see Supporting Information). In comparison, the DP of the supramolecular polymer formed in aqueous solution at the same concentration was also measured by the end-group analysis using EPR, which was estimated to be approximately 21. This result indicated that VR-SP could be formed on the surface of E. coli though the bacteria surface might disturb the host–guest complexation between viologen cation radical and CB[8] a little. Therefore, by combining the AFM imaging and end-group analysis, we conclude that supramolecular polymers can be fabricated on the surface of bacteria through in situ supramolecular polymerization powered by E. coli. Based on the E. coli-powered supramolecular polymerization in situ, and the NIR absorption of the supramolecular dimer of viologen cation radicals, we further explored the bacterial inhibition activity of VR-SP by photothermal therapy. The photothermal conversion property of VR-SP prepared by chemical reduction was first studied. As shown in Supporting Information Figure S19, a significant temperature increase of VR-SP solutionwas observed under 1064 nm irradiation at the second NIR (NIR-II) window, indicating the effective photothermal conversion of VR-SP. The photothermal conversion efficiency was 22.0% (see Supporting Information). Next, the photothermal antibacterial experiments were conducted. As shown in Figures 3a and 3b, under NIR irradiation at 1064 nm, the temperature of the medium with VR-SP adsorbed E. coli increased significantly. It rose from 35.7 to 57.6 °C after 12.5 min irradiation, indicating that VR-SP successfully converted optical energy into heat. When E. coli was incubated with VDV alone, the increase of temperature (to 51.5 °C) was lower than the VR-SP group after irradiation for the same time. When E. coli was treated with monofunctional viologen model compound VMA (Figure 3c) and CB[8], which could be reduced to generate supramolecular dimer of viologen cation radicals but not supramolecular polymers, the temperature increase was limited (to 49.4 °C), similar to that of the control group without any agent (to 46.4 °C) under the same irradiation condition. The significantly higher temperature increase of the VR-SP group could be ascribed to the local enrichment and enhanced adsorption of VR-SP onto the surface of E. coli. As shown in Supporting Information Table S1, the zeta potential of E. coli significantly increased from −57.7 mV of untreated E. coli to −50.4 mV when incubated with equimolar amounts of VDV and CB[8]. As a comparison, when incubated with only VDV, the zeta potential increased slightly to −55.6 mV. These results indicate that the adsorption amount of VR-SP onto the surface of E. coli is higher than that of VDV because of the structure of supramolecularly polymerized cations of supramolecular polymers. As for bacteria with low reducing ability, B. subtilis for example, since no viologen cation radicals were generated and no supramolecular polymers could be formed, the increase of temperature was similar after irradiation in the presence or absence of VDV-CB[8] (43.7 °C or 46.8 °C with or without VDV-CB[8], respectively). The results of these groups indicate that the polymeric structure of VR-SP is essential to reach a higher temperature in photothermal therapy. Figure 3 | (a) Temperature changes of the medium with E. coli or B. subtilis, or with E. coli or B. subtilis treated with different agents under NIR-II (1064 nm) irradiation. (b) Thermal images of the medium with E. coli or B. subtilis, or with E. coli or B. subtilis treated with different agents after 12.5 min NIR-II (1064 nm) irradiation. (c) The chemical structure of VMA (VDV: 0.20 mM, CB[8]: 0.20 mM, VMA: 0.40 mM, medium: LB medium). Download figure Download PowerPoint The quantitative antibacterial activity was demonstrated by means of plate coating. As shown in Figure 4, the inhibition efficiency of VR-SP to E. coli was 99.9% after 12.5 min of 1064 nm irradiation, which was significantly higher than that of VDV (44.7%), and VMA-CB[8] (54.0%). In contrast, equimolar amounts of VDV and CB[8] exhibited no inhibition of B. subtilis after irradiation under the same condition, since no supramolecular polymers were formed. It is notable that after 12.5 min irradiation under air, the absorption peaks of supramolecular dimer of viologen cation radicals in the UV–vis spectrum disappeared within several minutes ( Supporting Information Figure S26), and the EPR signal of viologen cation radicals almost disappeared ( Supporting Information Figure S27), both of which indicate the degradation of VR-SP after finishing the inhibition of E. coli. This is reasonable since dead E. coli cannot reduce viologen continuously to prevent the air oxidation-induced depolymerization. Therefore, these results validate that VR-SP is an NIR photothermal antibacterial agent with outstanding bacterial inhibition efficiency, high specificity towards E. coli, and good degradability. Figure 4 | CFU ratio of E. coli or B. subtilis incubated with or without different agents after 12.5 min NIR-II irradiation (1064 nm). Inset: partial magnified figure of the CFU ratio of VR-SP adsorbed E. coli group (red dash line labeled). Data shown are mean ± SD from n = 3 (SD = standard deviation, ns, p > 0.05; **, p < 0.01, tested by employing the pair-sample t test) compared with the E. coli or B. subtilis group without any treatment (agent or irradiation). Download figure Download PowerPoint To explore the potentials of VR-SP as biological materials, the cytotoxicity of equimolar amounts of VDV and CB[8] on human cells was investigated. To this end, human lung normal cell line BEAS-2B and colorectal normal cell line NCM460 were selected as representative cells. As shown in Figures 5a and 5b, when treated with equimolar amounts of VDV and CB[8] ranging from 0.10 to 0.40 mM, the cell viability of both BEAS-2B cell and NCM460 cell measured by CCK-8 assay was still around 100%. A similar tendency was also observed on the cells treated with VDV alone ( Supporting Information Figure S28). Therefore, neither VDV-CB[8] nor VDV alone exhibited clear cytotoxicity even at a relatively high concentration (0.40 mM), which was twice the concentration used in antibacterial experiments. Therefore, VR-SP may be a potential safe agent for photothermal therapy. Figure 5 | Cytotoxicity of equimolar amounts of VDV and CB[8] to (a) BEAS-2B cell and (b) NCM460 cell. Download figure Download PowerPoint Conclusion We have successfully fabricated an NIR photothermal antibacterial agent through E. coli reduction-