Living Bacterial Hydrogels for Accelerated Infected Wound Healing

Abstract Damaged skin cannot prevent harmful bacteria from invading tissues, causing infected wounds and even serious tissue damage. Traditional treatments can not only kill pathogenic bacteria, but also suppress the growth of beneficial bacteria, thus destroying the balance of the damaged skin microbial ecosystem. Here, a living bacterial hydrogel scaffold is reported that accelerates infected wound healing through beneficial bacteria secreting antibacterial substances. Lactobacillus reuteri, a common probiotic, is encapsulated in hydrogel microspheres by emulsion polymerization and further immobilized in a hydrogel network by covalent cross‐linking of methacrylate‐modified hyaluronic acid. Owing to light‐initiated crosslinking, the hydrogel dressing can be generated in situ at the wound site. This hydrogel scaffold not only protects bacteria from immune system attack, but also prevents bacteria from escaping into the local environment, thus avoiding potential threats. Both in vitro and in vivo experiments show that it has excellent ability against harmful bacteria and anti‐inflammatory capabilities, promoting infected wound closure and new tissue regeneration. This work may open up new avenues for the application of living bacteria in the clinical management of infected wounds.


Introduction
The skin plays an extremely important role in preventing water loss and blocking the invasion of harmful substances and pathogenic microorganisms as a significant interface between the body and its surroundings, where a variety of microbial communities co-exist, including bacteria, fungi, and viruses. [1][2][3] In particular, diverse microbial communities have significant impacts on local and systemic immune responses through interactions with host epithelial and immune cells. [3][4][5][6] However, once the skin is damaged, it provides an opportunity for harmful bacteria to invade viable tissues, causing wound infections and even serious tissue damage. [7] To address the dilemma of bacterial infection, extensive efforts have been made, as reflected by the constant updates of antimicrobial agents and antimicrobial materials, including new antibiotics, antibacterial nanoparticles, cationic polymer compounds, and antimicrobial peptides. [8][9][10][11][12][13][14][15][16][17][18] Although various strategies have achieved encouraging performance in terms of inhibiting bacterial growth and accelerating skin wound healing, it is worth noting (but easily ignored) that these antimicrobial substances kill pathogenic bacteria while suppressing the growth of beneficial bacteria, thereby destroying dermal microbial ecosystem balance. [19] Therefore, it would be highly desirable and significant to develop an alternative approach for suppressing the reproduction of deleterious bacteria without inhibiting the growth of beneficial bacteria, but this has proven to be difficult.
Living bacterial therapy, which has been used to treat a variety of inflammatory and immunologic pathologic diseases by bacterial interference and immunomodulation, has received much attention over the past few years. [20,21] In nature, microorganisms compete for limited nutrition or space to survive. [22] In particular, certain beneficial bacteria can create unique local microenvironments by secreting large amounts of metabolites and antimicrobial agents, which are suitable for their own survival but which inhibit the growth of competing microorganisms. [23,24] Thus, beneficial bacteria that secrete biologically active substances with antibacterial, anticancer, or immunosuppressive capabilities have been extensively explored for diagnosis and treatment. [25][26][27] For example, Lactobacilli have been successfully utilized to treat urinary tract infections by restoring the natural vaginal flora. [28] Furthermore, Bacillus subtilis has been employed as a producer of antifungal agents for treating fungal infections. [29,30] However, the application of bacteria has been largely limited due to the intrinsic nature of bacteria, including rapid proliferation and colonization, which may result in uncontrollable invasion. [31,32] In addition, exposed bacteria often cause attacks by the immune system and cannot achieve desired therapeutic effects. [33] Therefore, it is challenging to design a scaffold that can protect bacteria from attack by the immune system, using the bioactive substances secreted by bacteria, and thereby prevent bacteria from escaping into the local environment and causing serious infections.
Hydrogels, popular biomedical polymers with 3D molecular networks, have been widely applied in the fields of drug delivery, implantation, and tissue engineering. [34] Owing to high biocompatibility and a moist healing environment, hydrogels were applied to promote tissue repair and regeneration. [35] Hydrogels not only provide physical barriers to prevent bacteria from invading damaged skin, but also absorb excessive wound exudates. Herein, we present a convenient and novel hydrogel scaffold to treat bacterial infections and accelerate wound healing, and which contains living bacteria encapsulated in hydrogel microspheres forming a hydrogel at the wound site. First, Lactobacillus reuteri, a heterofermentative lactic acid probiotic, was selected as the living bacterial strain, which has been shown to inhibit the growth of pathogenic bacteria through lower local pH and antibacterial agents produced during metabolism. [36][37][38] As is well known, hydrogels can provide good environments for nutrient transfer, cell growth, and sensing, and are widely considered to be ideal materials for encapsulating living cells. [39] To prevent the escape of encapsulated bacteria from the hydrogel, as shown in Scheme 1A, L. reuteri was wrapped into hydrogel microspheres through emulsion polymerization, which allowed the delivery of certain substances, including nutrients, biomolecules, lactic acid, and antibacterial agents. Then, a hydrogel dressing was formed at the wound site by covalent crosslinking of methacrylate-modified hyaluronic acid under light irradiation (Scheme 1B,C). It is interesting to note that in vitro experiments demonstrated that the encapsulated bacteria did not escape into the environment, and the hydrogel exhibited good resistance to pathogenic bacteria. Furthermore, lesser inflammation and faster healing were observed at the wound site in mice treated with living bacterial hydrogel, demonstrating the outstanding anti-infection and accelerated wound healing ability of such materials in vivo. Therefore, we believe that the strategy described here will open a new window for living bacterial therapy as a safer means of therapy, promoting the application of bacteria in the treatment of various diseases.

Preparation and Characterization of Hydrogel Microspheres (GHM-LR) Encapsulating Living Bacteria
Hydrogel microspheres have attracted the attention of researchers in biomedicine and tissue engineering as delivery vehicles for drugs, proteins, and cells because they provide many unique properties compared to bulk hydrogels, including injectability, modularization, and porosity. [40] Inspired by hydrogel microspheres encapsulating cells that can protect cells from en-vironmental interference through physical restriction, hydrogel microspheres can prevent the enclosed living bacteria from escaping into the local environment, which provides a safe platform for the growth of bacteria. [41] Therefore, in this study, living L. reuteri was first successfully encapsulated into hydrogel microspheres by the covalent crosslinking of methacrylated gelatin solution containing living bacteria through emulsion polymerization. Representative images of the hydrogel microparticles are shown in Figure 1A and Figure S1, Supporting Information. The prepared microspheres exhibited highly spherical and clear borders. In addition, as shown in Figure 1B, several representative images were statistically analyzed using ImageJ software to obtain a histogram of the microsphere diameters, indicating the size of most microparticles as ranging from 70 to 90 μm. The morphology of the lyophilized microspheres was obtained by scanning electron microscopy (SEM), which revealed uniform porous structures ( Figure 1C). To better visualize the growth state of bacteria in microspheres, images of L. reuteri were captured by laser scanning confocal microscopy (LSCM) after staining with a Live/Dead bacterial viability kit. As shown in Figure 1D-F and Figure S2, Supporting Information, a large number of green fluorescent bacteria were observed, demonstrating that the hydrogel microspheres were suitable for bacterial growth because of their excellent biocompatibility. In addition, it can be clearly seen that living bacteria were successfully encapsulated in the hydrogel microspheres. To further explore the proliferation of living bacteria enclosed in microspheres, flow cytometric analysis was performed. As illustrated in Figure 1G and Figure S3, Supporting Information, an obvious shift to a higher mean fluorescence intensity was observed after incubation for 12 h. To quantify the capacity of L. reuteri to grow inside the hydrogel microspheres, a viability assay was conducted by measuring luminescence levels. As shown in Figure 1H, compared to the bacteria in DeMan-Rogosa-Sharpe (MRS) medium, L. reuteri encapsulated in microspheres showed a similar growth trend. Notably, bacteria embedded in hydrogel microspheres in MRS media exhibited a lower level of luminescence than without bacterial wrapping, which may contribute to the physical limitation of microspheres on the growth of bacteria. Additionally, the hydrogel microspheres also showed good stability over 7 days in PBS, as shown in Figure 1I. Overall, the hydrogel microspheres encapsulating living bacteria were successfully prepared and proved to be suitable for bacterial growth, laying a foundation for subsequent experiments.

Preparation and Characterization of Hydrogel (MHA-LR) Containing GHM-LR
Bulk hydrogel containing living bacteria wrapped in microspheres was fabricated using a photocrosslinking reaction between methacrylated hyaluronic acid and microspheres. Furthermore, infrared spectroscopic analysis was further performed to explore the constitution of prepared hydrogels. As shown in Figure S4, Supporting Information, the appeared signals of the prepared hydrogels that attributed to the characteristic absorption of GelMA (1447 cm −1 ) and HAMA (1406 cm −1 ) precursors indicate the successful formation of the corresponding network. Rheological analysis was carried out to explore the gelation time and mechanical properties of the bulk hydrogel (MHA-LR). As shown in Figure 2A, the hydrogel was rapidly formed in approximately 10 s, and the values of G' and G'' also tended to be stable after 40 s, demonstrating the rapid gelation ability of the system and the integrity and stability of the network. [42] There was no significant difference between the hydrogel containing microspheres and hydrogels without microspheres ( Figure S5, Supporting Information). In addition, hydrogels with or without microspheres exhibited constant values of G' and G'' over the frequency range of 0.1 to 10 Hz ( Figure 2B), which could be attributed to the covalent crosslinking inherent to the network. [43] Furthermore, as shown in Figure 2C, a significant viscosity decrease was observed, displaying a non-Newtonian or pseudoplastic characteristic with a shear rate ranging from 0.01 to 100 s −1 . Phase angles of HMA and HA hydrogel were measured as a function of frequency. As shown in Figure 2D, there was no significant difference between the two samples and the phase angle was about 0.01, indicating typical elastic properties of the hydrogels.
SEM images showed that compared with HA hydrogels, porous microspheres were observed in lyophilized hydrogels containing microspheres, which was consistent with the structure of freeze-dried hydrogel microspheres, as shown in Figure 2E and Figure S6, Supporting Information. To visualize the growth of L. reuteri in MHA, images of living bacteria were obtained by LSCM after staining with the Live/Dead Kit. As observed in Figure 2F, living bacteria exhibited a good growth state, indicating that the material was nontoxic to bacteria throughout the preparation process. As mentioned above, hydrogel microspheres not only promoted living bacterial growth as a scaffold, but also prevented bacteria from escaping into the environment and causing potential infection. To verify its ability to prevent bacteria from escaping, the hydrogel (HA) directly mixed with L. reuteri and the hydrogel (MHA) containing bacterial encapsulated microspheres were placed on MRS plates and cultured for 12 h. As shown in Figure 2G, a large number of L. reuteri colonies were observed around the hydrogel, which could be explained by the proliferation of bacteria exposed to the surface of the material. However, there was no bacterial growth around the hydrogel containing microspheres encapsulating living L. reuteri, demonstrating that the microspheres prevented bacteria from escaping into the local environment. In addition, these two hydrogels mixed with living bacteria by different methods were placed in a liquid MRS medium for incubation for 12 h, and the number of bacteria in the culture medium was evaluated by reading the optical density at 450 nm after incubation with the Microbial Viability Assay Kit-WST. As seen in Figure 2H, the OD value for the HA-soaked solution was up to 0.8, suggesting the proliferation of www.advancedsciencenews.com www.advancedscience.com numerous bacteria. However, the OD value of the MHA-soaked solution was almost the same as that in PBS. In addition, we replaced L. reuteri with Escherichia coli to perform the above experiments, and the same phenomena and results were observed, as shown in Figure S7, Supporting Information. These results confirmed that hydrogels containing living bacteria encapsulated in microspheres exhibited excellent biosafety to avoid potential bacterial infection.

Antibacterial Activity and Biocompatibility Evaluation In Vitro
L. reuteri, a probiotic commonly found in humans and animals, can protect against the adverse effects of certain microorganisms and modulate immune responses. [34][35][36] According to previous reports, L. reuteri can produce lactic acid to reduce the pH of the local environment, which inhibits the growth of harmful bacteria such as Staphylococcus aureus. [34] In addition, L. reuteri can also secrete a potent antimicrobial agent called reuterin against gram-positive and gram-negative bacteria. To confirm the presence of reuterin in hydrogels containing L. reuteri, the hydrogels were soaked in an MRS medium for 12 h and the co-cultured solution was analyzed using electron ionization (EI) mass spectrometer. As shown in Figure S8, Supporting Information, the molecular ion peak of reuterin was observed at m/z 74, indicating the presence of an antimicrobial agent. Therefore, we expected hydrogels containing L. reuteri encapsulated in microspheres to exhibit good antibacterial activity against harmful bacteria, including E. coli, S. aureus, and Salmonella spp. In vitro antimicrobial tests were performed to evaluate the antibacterial activity of the materials. Compared with hydrogel fabricated without L. reuteri, obvious inhibition zones were observed on the agar plate on which the MHA-LR was placed for 12 h, as illustrated in Figure 3A. The zone diameter for the MHA-LR group was 15.1 ± 1.5, 13.2 ± 1.8, and 15.1 ± 1.6 mm against E. coli, S. aureus, and Salmonella, respectively ( Figure 3B). Moreover, the antimicrobial ability of living bacterial hydrogel was evaluated in vitro against S. aureus and E. coli. As shown in Figure S9, Supporting Information, compared with untreated S. aureus and E. coli, the vast majority of bacteria were killed after culturing with L. reuteri encapsulated hydrogels culture medium, demonstrating the excellent antibacterial ability of materials. It is consistent with the results of the zone of inhibition assay. To further explore the biofilm inhibition properties of living bacterial hydrogel, S. aureus was selected as a model bacteria to perform a LIVE/DEAD staining assay. As shown in Figure S10, Supporting Information, the S. aureus without being treated with the secretion of living L. Reuteri encapsulated in hydrogel exhibited strong green fluorescence, indicating good bacterial activity. On contrary, the treated S. aureus showed significant red fluorescence, suggesting that most of the pathogenic bacteria were killed. Given that propidium iodide penetrates only bacteria with damaged membranes, the strong red fluorescence also demonstrated that the membrane of S. aureus was damaged. These results indicated that MHA-LR exhibited excellent capability against various harmful bacteria, which could be ascribed to the lactic acid and reuterin secreted by L. reuteri within hydrogels.
To evaluate the cytocompatibility of MHA and MHA-LR, mouse fibroblast L929 cells were chosen as a model to perform live/dead assays. As shown in Figure 3C, a large number of living cells was observed in the three groups during the whole period, and there were no significant differences among them after incubation for 24, 48, and 72 h. This showed that MHA and MHA-LR gels did not affect cell growth and proliferation, demonstrating their excellent biocompatibility. Notably, living L. Reuteri wrapped in microspheres in the hydrogel had almost no negative effect on cell growth, which could be explained by the microspheres preventing the bacteria from escaping into the environment. To further assess the cytocompatibility of the materials, quantification data were obtained using the Cell Counting Kit 8 kit (CCK8). As illustrated in Figure 3D, there were no obvious differences in the OD values among the three groups, indicating no significant cytotoxic effects of MHA and MHA-LR. These results suggested that hydrogels containing living bacteria within microspheres exhibited good biocompatibility without causing side effects in terms of cell growth and proliferation, and could be expected to be used in vivo for further biological applications.

Antibacterial Activity and Biocompatibility Evaluation In Vivo
To evaluate the antibacterial ability of living bacterial hydrogel in vivo, the bacteria were separated from the infected wound after two days of treatment and was spread on agar broth plates to culture for 24 h. As shown in Figure 4A,B, a lot of bacteria were observed in the control and HA treatment groups, indicating high bacterial growth in infected tissue. However, the infected skin showed a small amount of bacterial growth after treating with living L. reuteri hydrogel, suggesting the dress's obvious antibacterial activity in vivo.
Moreover, the biocompatibility of living bacterial hydrogel was evaluated by analyzing inflammatory cytokine of mice implanted with materials containing living L. reuteri. On days 1, 3, and 5, the inflammatory cytokine including TNF-, IL-6, IL-10, and IL-1 was measured by using corresponding ELISA Kits. As shown in Figure 4C-F, there was only a slight difference in all inflammatory factors between the control group and the implanted living bacterial hydrogel group, demonstrating good biocompatibility of designed materials in vivo.

In Vivo Evaluation of Wound Healing
A full-thickness cutaneous wound model with S. aureus infection was established to evaluate the efficiency of hydrogels in promoting wound healing. Mice were divided into three groups for in vivo testing: no treatment (control group), HA hydrogel treatment (HA group), and HA hydrogels containing microsphere encapsulating living L. reuteri treatment (LRHA group). As shown in Figure 5A, the process of wound healing was monitored using images captured with a digital camera. In the control group, yellow pus was observed at the wound site on days 2 and 4, and the injury was bright red, indicating that the damaged site was undergoing inflammation. However, there was almost no yellow pus in the LRHA group during the entire treatment period compared with the HA and control groups. In addition, the wounds of mice treated with LRHA exhibited accelerated wound closure on day 2, indicating that infections caused by S. aureus were controlled ( Figure 5B). By day 4, the wound size of mice treated with LRHA dropped significantly from 38.5 ± 2.3 to 14.8 ± 2.6 mm 2 , in which the wound healing rate reached 64% ( Figure 5C,D). However, the wound sizes of control and HA group animals were still 28.3 ± 2.1 and 21.5 ± 1.8 mm 2 , respectively, with wound healing rates of 20% and 42%, respectively. After 10 days of treatment, wounds treated with LRHA were completely closed and covered with new epidermal tissue. However, 38% and 24% of the wounds remained unhealed after no treatment and treatment with HA, respectively. The wound closure time was extended to 15 and 13 days in the control and HA groups, respectively, as shown in Figure S11, Supporting Information. These results suggested that LRHA plays an important role in promoting wound healing, which is attributed to the excellent antibacterial ability of living L. reuteri secreting lactic acid and antimicrobial agents. It is worth noting that the HA-treated wounds showed a faster healing rate than the control group, which could be explained by the hydrogel having a certain promoting effect on wound healing and providing a physical barrier to prevent bacteria from invading adjacent tissues.
To further evaluate the process of wound healing, a series of pathological examinations were performed, including hematoxylin and eosin (H&E) staining, Masson's trichrome staining, and Sirius red staining. As shown in Figure 6A, tissues in the wounds of the control, HA, and LRHA groups were stained with H&E on days 7 and 10. In the control and HA groups, large numbers of inflammatory cells were observed on day 7, indicating that serious inflammation was still not controlled due to the absence of antibacterial activity of the treatments. In addition, there was almost no new epidermal tissue in either group. However, only a small number of neutrophils appeared by day 7 in the wound tissues of mice treated with LRHA, which was ascribed to the excellent capability against S. aureus of living L. reuteri wrapped in LRHA. Notably, new blood vessels and regenerated epidermis were observed, demonstrating that wound healing was accelerated by LRHA treatment. After 10 days of treatment, the defective skin was healed completely in the LRHA group, and was almost the same as normal skin ( Figure S12, Supporting Information), indicating the regeneration of dermal tissue. However, there were many inflammatory cells and an unclosed wound in the control and HA groups. Moreover, Masson's trichrome staining and Sirius red staining showed that the renewed collagen gradually increased the healing time for the three groups ( Figure 6B, Figures S13 and S14, Supporting Information). However, more regenerated collagen was deposited in the wounds treated with LRHA than in the defects treated in the other two groups, further confirming the superior wound healing effects of LRHA.
Furthermore, the renascent epithelial thickness was investigated by staining the wound tissue for cytokeratin 14 (CK14) after 7 and 10 days of treatment using different methods. As shown in Figure 7A, there was no significant difference in epithelial thickness between the control and HA groups. During the whole wound healing process, the renascent epithelial thickness became thinner and thinner for all three groups with an  increase in repair time. However, the epithelial thickness of the LRHA-treated groups was as thin as that of normal tissue (Figure S15, Supporting Information), while there was a relatively full thickness in the control and HA groups, suggesting that LRHA promoted the wound healing process. Further quantification was performed to explore the effects of LRHA on accelerating wound healing ( Figure 7B-F). Compared with the control and HA groups, the wound tissues of mice in the LRHA group exhibited shorter wound lengths, fewer inflammatory cells, a larger number of hair follicles, a higher ratio of collagen-occupied regions, and a thinner epithelial layer after 7 and 10 days of treatment. In addition, to explore the toxicity of materials in vivo, the major organs of the three groups were harvested and analyzed by H&E staining, involving the heart, liver, spleen, lungs, and kidneys. As shown in Figure S16, Supporting Information, histological analysis showed that these major organs maintained the integrity of the tissue without abnormal defects or damage, which was not significantly different from normal tissues, demonstrating the excellent biocompatibility of LRHA. These results demonstrated that LRHA has excellent antibacterial ability and outstanding performance in accelerating wound healing, which provides a new and safe strategy based on living bacteria for potential applications in wound repair in the clinic.

Conclusions
In summary, we successfully encapsulated live L. reuteri into microspheres by emulsion polymerization involving methacrylate gelatin and synthesized hydrogel dressings in situ by covalent crosslinking of methacrylate hyaluronic acid to accelerate wound healing. Encapsulated living bacteria within microspheres can grow and proliferate normally in the hydrogel, secreting lactic acid and antibacterial agents into the local environment. This hydrogel dressing not only protects bacteria from the immune system, but also prevents bacteria from escaping into the local environment, avoiding potential threats. In in vitro experiments, LRHA exhibited excellent antibacterial effects against different bacteria and with outstanding biocompatibility. Subsequently, in vivo experiments further revealed that LRHA can mitigate inflammatory cell infiltration, enhance collagen deposition, and accelerate wound healing. We anticipate that this work will open a

Experimental Section
The Preparation of HA-MA: The methacrylated hyaluronic acid was prepared according to previous literature reports. First, 2 g hyaluronic acid was dissolved in 100 mL deionized water and the pH of the solution was adjusted to 8. An excess of methacrylic anhydride was added to the solution and the mixed solution was stirred for 12 h at room temperature. Then, the product of HA-MA was obtained by ethanol precipitation and washed with ethanol three times to remove the remaining methacrylic acid and methacrylic anhydride. 1 H-NMR spectra of HA-MA were recorded on a Bruker Avance 400 MHz spectrometer ( Figure S17, Supporting Information).
The Preparation of Lactobacillus Reuteri-Loaded Gelatin Hydrogel Microparticles (LRGHM): L. reuteri was successfully wrapped into the hydrogel microspheres through the emulsion polymerization of methacrylate modified gelatin. Specifically, 60 mg methacrylated gelatin were dissolved in 1 mL PBS containing 10 8 L. reuteri and 3 mg LAP to prepare a hydrogel precursor solution. Under the irradiation of 365 nm wavelength light (10 mW cm −2 ), the mixture solution was added to 3 mL mineral oil containing 2% Span 80 in a glass flask with constant stirring. Light exposure was limited to 1 min, since hydrogel microspheres settle out of solution rapidly. L. reuteri-loaded hydrogel particles were then separated from the mixture system by centrifuging at 300 g for 2 min. Finally, gelatin microparticles containing L. reuteri were washed three times by 1× PBS and then resuspended to PBS for the follow-up study.
The Preparation of Hydrogel: HA hydrogel containing L. reuteri-loaded gelatin hydrogel particles (LRHA gel) were fabricated by the following method. In particular, 15 mg HA-MA and 3 mg LAP were added to 1 mL PBS containing LRGHM to prepare a hydrogel precursor solution. Then, the precursor was poured into Teflon disc molds and was irradiated with 365 nm wavelength LED for 5 s to obtain the LRHA gel. HA hydrogel (HA gel) without L. reuteri were prepared with the same method.
Particle Size Statistics of Gelatin Microspheres: LRGHM were diluted into the 1× PBS in a 50 mm petri dish with a suitable amount. 30 images were taken randomly by a Leica fluorescence microscopy under white light channel. The size distribution of gelatin microspheres was quantified and analyzed by ImageJ Software.
The Stability Test of Gelatin Microspheres: The obtained gelatin microspheres were immersed in PBS solution at room temperature. At a certain time, the weight of microspheres was recorded after centrifugation at 300 g for 2 min and resuspended into a PBS solution.
www.advancedsciencenews.com www.advancedscience.com Histopathological Study: Wound histology specimens were collected on days 7 and 14. All the harvested samples were fixed in 4% paraformaldehyde solution for 12 h and embedded in paraffin to prepare 5 μm thickness tissue sections. Representative specimens were stained with haematoxylin and eosin (H&E), Masson trichrome, and Sirius red to observe the histological images by microscopy. In addition, immunofluorescence staining (CK14) was also performed to evaluate the collagen deposition and angiogenesis. To evaluate the acute toxicity of the material in mice, the major organs were collected and stained with H&E for histological analysis, which included the heart, spleen, liver, kidneys, and lungs.
Statistical Analyses: All data were presented as means ± standard deviation (SD) based on experiments performed in triplicate or more. Statistical analysis was performed with Graphpad software. Unpaired two-tailed Student's t-test was performed for two-group comparisons. Statistical significance was analyzed by one-way ANOVA for more than two groups. Statistically significant differences were represented with *p < 0.05, **p < 0.01, and ***p < 0.001.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.