Self-activating anti-infection implant

Clinically, it is difficult to endow implants with excellent osteogenic ability and antibacterial activity simultaneously. Herein, the self-activating implants modified with hydroxyapatite (HA)/MoS2 coating are designed to prevent Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli) infections and accelerate bone regeneration simultaneously. The electron transfer between bacteria and HA/MoS2 is triggered when bacteria contacted with the material. RNA sequencing data reveals that the expression level of anaerobic respiration–related genes is up-regulated and the expression level of aerobic respiration–related genes is down-regulated when bacteria adhere to the implants. HA/MoS2 presents a highly effective antibacterial efficacy against both S. aureus and E. coli because of bacterial respiration–activated metabolic pathway changes. Meanwhile, this coating promotes the osteoblastic differentiation of mesenchymal stem cells by altering the potentials of cell membrane and mitochondrial membrane. The proposed strategy exhibits great potential to endow implants with self-activating anti-infection performance and osteogenic ability simultaneously.

1. Page 2, line 28. What does self-powered mean? Please clarify. 2. Page 3, line 70. Is molybdenum sulfide truly widely used in biomaterials? I would change this language as this is likely untrue. 3. Page 4, line 81. This sentence appears incomplete. 4. Page 4, line 83. This is likely not the first time a biomaterial has been proposed that is both osteoconductive and antibacterial, although it is still unclear what "self-powered" implies. 5. Page 6, line 142. "Coculturing" typically implies culture of two different species or lines; I would suggest that this is culture on a substrate. 6. The phrase "antibacterial efficiency" is used often when describing this material. The mechanism is not clear to me whether the change in bacterial colonies is due to an anti-adhesive effect versus a bacteriostatic or bacteriocidal effect. 7. In Figure 1L, it is shown that while statistically significant in effect and a two-log difference in concentration, there still remained between 7-8 log Staph aureus and E coli on the most promising material surface (HA/MoS2-Ti6). This is far from sterilizing and may clinically still lead to infection. It is misleading to express this difference as a percentage change, I recommend representing it as a log change given bacterial growth kinetics. Please comment. 8. The results on page 7 and 8 would be even more exciting if the experiment was replicated with Staph aureus mutants/knock outs that lack molecular mechanisms to adapt to anaerobic activity to corroborate the sequencing data. 9. Page 12, line 291. It is microcomputed tomography, not microtomography. 10. Page 12, line 305. Chondrocytes can be precursors to bony development (endochondral ossification) so would be careful about implying that appearance of chondrocytes is not consistent with osteogenesis. 11. Page 19, line 478. Please specify if this isolate of Staphylococcus aureus is methicillinsusceptible or -resistant. 12. Similar to my above comment on in vitro results, while there was ~two fold log reduction in presence of Staphylococcus aureus on the in vitro work, there still remains ~10^4 bacteria (see Fig 5c). Expressing this reduction as a percentage is misleading given bacterial growth kinetics. While it is reassuring that there was greater bone growth in the treatment arm, it is difficult to know if the remaining 10^4 bacteria in the long term would compromise the surgical site. In addition, proper matching controls were not performed-there should be two additional groups (Ti6 with no bacterial inoculation and HA/MoS2-Ti6 with no bacterial inoculation). It is unclear if n=6 per group was correctly powered.
Reply: Thank you so much for your suggestions. The structure of MoS2 does not change in the experiment. We have performed electron paramagnetic resonance (EPR) experiment to measure sulfur vacancies in HA/MoS2-Ti6, and there were no signal of vacancies ( Supplementary Fig. 4). In addition, we added the related content at page 6 line 4: "The electron paramagnetic resonance (EPR) experiment was further performed to characterize the vacancies in HA/MoS2-Ti6 (Supplementary Fig. 4) Reply: Thank you so much for your suggestions. We performed zeta potential to measure the specific values of cell membrane potential. And we added at page 12 line 23: "The zeta potential measurement was used to measure the bacterial membrane potential, which reflected the inherent metabolic state of the bacteria. 32 In addition, the bacterial surface potential was altered by changing the molecular nature of surface potential, which was in correlation with the metabolic process that happened in cells aerobically and anaerobically cultured. 33 As shown in Supplementary Fig. 15, when bacteria were cultured on the surface of HA/MoS2-Ti6, the zeta potential would turn more negative compared with bacteria cultured on the surface of Ti6. It was due to the metabolism transfer from aerobic respiratory to anaerobic respiratory."  24,25 Furthermore, the potential of HA/MoS2-Ti6 were lower than those of the redox c-type cytochromes on the bacterial membrane (Fig. 1k), ensuring that the electrons transferred from membrane proteins to HA/MoS2-Ti6. 26  Comment 6: "The Nrf2-mediated antioxidant defense was triggered to restore the redox disequilibrium." It is not clear how Nrf2-mediated antioxidant defense was triggered.
Reply: Thank you so much for your comment. In fact, the cellular redox potential was regulated via redox couples during cellular process, and its cellular redox potential was at -4.12 to -4.84 eV. The HA/MoS2-Ti6 had oxidizing substances, which could disturb redox equilibrium. And the content of intracellular ROS was increased. The immediate response of Nrf2-mediated antioxidant defense was triggered to redox disequilibrium.
And we added expression at page 15 line 3: "The cellular redox potential was regulated via redox couples during the cellular process, and its cellular redox potential was at -  Reply: Thank you so much for your saying that "the study is in detail and interesting".
In our manuscript, we proposed novel antibacterial mechanism based on material-bio interface interaction. The electrons of HA/MoS2 were transferred from the bacteria membrane to HA/MoS2 after contacting with bacteria due to lower redox potential.  Reply: Thank you so much for your suggestions. We had revised the statement and reference 17.
We had revised at page 4 line 1: "2D MoS2 nanostructures had a tunable electronic energy state. 14 Also, nanostructured MoS2 would benefit the separation of electron-hole pairs by decreasing the distances for electrons and holes to diffuse to the surface of the materials and also increasing the reaction sites. 12 " was not used.
Reply: Thank you so much for your suggestions. Thank you very much for the reviewer's valuable comments and saying that "The authors show in vitro evidence that supports their hypothesis in general (antimicrobial effect through interruption of metabolic pathways, genetic markers associated with osteoblastic differentiation)." The comments are very helpful for us to revise and improve our paper. We have made extensive and careful revisions accordingly, please see the following reply.
Meanwhile, we have already added the in vivo experiment of surgical defect without infection using the same biomaterials as request for more accurate experimental design.
The manuscript has significant grammatical errors.
Reply: Thank you so much for your suggestions. We had already revised our language.
Overall, this is an exciting new direction for the field of device infection. Using Reply: Thank you so much for your suggestions. We had revised the expressing reduction bacteria as absolute log, and checked in our manuscript. In addition, we redid animal experiments and n=8.
Specific comments below are as follows: Comment 1: Page 2, line 28. What does self-powered mean? Please clarify.
Reply: Thank you so much for your suggestions. We were so sorry for our unclear expression of self-powered. Self-powered systems often used to describe systems which did not need an outside energy supply to work (Nature Nanotechnology 2010,  Fig. 10-12, Fig. 2a, Fig. 1l) Fig. 5, but the ability of changing bacterial colonies of HA-Ti6 and MoS2-Ti6 was poor (Fig.1l). In additional, the SLM-Ti6 almost had not antibacterial effect, suggesting that the mechanism of the change in bacterial colonies was not due to anti-adhesive effect (Supplementary Fig. 11). In a word, the mechanism in bacterial colonies was mainly due to bactericidal effect."    Fig. 9). The c. f.  Comment 8: The results on page 7 and 8 would be even more exciting if the experiment was replicated with Staph aureus mutants/knock outs that lack molecular mechanisms to adapt to anaerobic activity to corroborate the sequencing data.
Reply: Thank you so much for your suggestions. We performed Staph aureus knock outs that lack molecular mechanisms to adapt to anaerobic activity, but we did not success to do it. In addition, pseudomonas aeruginosa was a strictly aerobic bacterium, which lacked ability to adapt to anaerobic activity. 47 We did antibacterial assay of Ti6 and HA/MoS2-Ti6 against P. aeruginosa at 6 h and 12 h, suggesting that HA/MoS2-Ti6 had no ability to clear P. aeruginosa infection (Supplementary Fig. 16). We added in our manuscript at page 13 line 23: "Furthermore, the mechanism of inducing bacteria into anaerobic respiration was evaluated. As shown in Supplementary Fig. 16, HA/MoS2-Ti6 group almost had no antibacterial ability against pseudomonas aeruginosa (P. aeruginosa) compared with Ti6 group, suggesting that the antibacterial mechanism of HA/MoS2-Ti6 group was related to the molecular mechanisms to adapt to anaerobic activity."

Supplementary Fig. 16
As     Choose #1 to perform next experiment.

Identification of gene knockout strains
We did not obtain positive clone until now. Reply: Thank you so much for your suggestions. We forgot to label the different group.
Ti6 group had more chondrocytes than HA/MoS2-Ti6 group, which was consistent with osteogenesis. And we revised in Fig. 5i-5j. We added at page 19 line 17: "As illustrated in Fig. 5i  And we also did the assay of the antibacterial assay of different samples against methicillin-resistant S. aureus (MRSA), which the HA/MoS2-Ti6 also showed great antibacterial efficiency. We did the experiment of HA/MoS2-Ti6 against MRSA in our manuscript at page 7 line 23: "The methicillin-resistant Staphylococcus aureus (MRSA) treated with HA/MoS2-Ti6 resulted in 2.36 -log reduction (Supplementary Fig. 8) Moreover, the antibacterial activity of HA/MoS2-Ti6 towards less than bacterial concentration was supplied. As shown in Supplementary Fig. 9, the c. f. u. of HA/MoS2-Ti6 group was from 7.32 log to 5.22 log compared with Ti6 group in 10 5 beginning condition, and from 8.66 log to 6.26 log compared with Ti6 group in 10 6 beginning condition.
In addition, we redid the animal experiment and added two additional groups (Ti6 with no bacterial inoculation and HA/MoS2-Ti6 with no bacterial inoculation), and n=8. We Meanwhile, H&E and Giemsa staining were used to determine the bacterial contamination of bone-tissue and bone implants after 2 weeks. The samples without bacteria of H&E and Giemsa staining were used to be control group, which was used to assess the influence of bacteria on bone tissue. A fluorescence microscope was used to analyze the histopathological microtomography." And at page 18 line 10: "The inflammatory response and remaining bacteria in the bone tissue around the implant were evaluated via hematoxylin and eosin (H&E) and Giemsa staining after 14 days of implantation ( Fig. 5a and Fig. 5b). 45 H&E staining was used to observe neutrophils, and the tissue around Ti6 with S. aureus had many neutrophils  (Fig. 5c and Fig. 5d). Compared to the Ti6 rod with S.
aureus, the bacteria in HA/MoS2-Ti6 with S. aureus group would be 2.65-log reduction.
Furthermore, the bone regeneration ability for different samples was assessed via microcomputed tomography (micro-CT) after 4 weeks of implantation. 46 Three cylindrical regions around the implant-with a diameter of 2.51 mm and a thickness of 0.40 mm-were selected, and three-and two-dimensional images were reconstructed with a special software program to reduce errors (Fig. 5e) Fig. 5f. Methylene blue-acid magenta staining and Safranin-O/Fast Green staining were used to assess the histopathological conditions around the implant after 4 weeks ( Fig. 5g-5h). As for Methylene blue-acid magenta staining, the red color represented

Reviewer #3 (Remarks to the Author):
In general, this manuscript has been strengthened by the efforts of its authors. In particular, the study design of their in vitro work has been much improved by including important control groups. By including these groups, they have generated further support to their hypothesis that their biomaterial may have both osteoinductive and antimicrobial properties. However, this new in vivo data is confusing in that it does not appear all groups were statistically compared to one another. It would be further strengthened by clarifying the statistical relationships between the 4 groups.
Ultimately, the grammatical errors throughout the text make interpreting sections challenging. For example, in the abstract alone, note grammatical errors in lines 27-28, 34, and 39. Errors are rife throughout the manuscript and interfere with understanding the authors' interpretation of their data.
While I believe I understand the authors' explanation behind their terminology of "self-powered," I still find it non-intuitive to describe their phenomenon and it appears the other reviewers agree that it is a confusing term. I would consider another phrase, such as "self-activating." Specific comments as follows: Page 3 Line 77 and Page 4 Line 88: I do not understand the sentence "And those make it applied in biomaterials." Page 4 Last Paragraph: Given the inconsistent tenses, this paragraph was difficult to read.
Page 10 Line 225: I am not familiar with the term "adhesion poverty" or several lines down "hydrophilic poverty." Page 10 Line 228: "…but the ability of changing bacterial colonies of HA-Ti6 and MoS2-Ti6 was poor (Fig.11)." What does this mean? What is the ability of changing bacterial colonies? This paragraph in general is difficult to read; it is hard to understand what conclusions the authors are trying to reach and how they have reached them given the text.
Page 18 Line 418: The fact that there are less neutrophils present on H&E does not necessarily support the claim that "the remaining bacteria in long term would not compromise the surgical site." For example, less neutrophils recruited to the site may mean that the remaining bacteria have been able to more efficiently create a biofilm matrix or enter a quiescent state such that this will become the site of a chronic infection. The histology does not necessarily support the claim of the authors. Figure 5F. Are the groups statistically significantly different from one another? My understanding is that the astrix indicate that the other three groups have significantly greater % than the infected Ti6 group. This also applies to Figures 5h and j; it is not clear which groups are being compared for statistical significance. 1

Response to Reviewer 3#
In general, this manuscript has been strengthened by the efforts of its authors. In particular, the study design of their in vitro work has been much improved by including important control groups. By including these groups, they have generated further support to their hypothesis that their biomaterial may have both osteoinductive and antimicrobial properties. However, this new in vivo data is confusing in that it does not appear all groups were statistically compared to one another. It would be further strengthened by clarifying the statistical relationships between the 4 groups.
Reply: Thank you so much for positive comment about "In general, this manuscript has been strengthened by the efforts of its authors. In particular, the study design of their in vitro work has been much improved by including important control groups. By including these groups, they have generated further support to their hypothesis that their biomaterial may have both osteoinductive and antimicrobial properties. " The reviewer also said that "this new in vivo data is confusing in that it does not appear all groups were statistically compared to one another. It would be further strengthened by clarifying the statistical relationships between the 4 groups." As request, we have clarified the statistical relationships between the 4 groups as follows: Reply: Thank you so much for your suggestions. We were sorry for our wrong description about these two terms. We have revised "adhesion poverty" into "antifouling ability" at Page 11 Line 235. Generally, the good hydrophilicity is beneficial for developing an antifouling surface. Anti-fouling ability of material means the effect of inhibition of pathogen to adhere on the surface of substrate.
And we have also revised "hydrophilic poverty" into "hydrophilicity" at Page 11 Line 236. hydrophilicity was a useful material property, and hydrophilic surface could attract water.
Page 10 Line 228: "…but the ability of changing bacterial colonies of HA-Ti6 and MoS2-Ti6 was poor (Fig.11)." What does this mean? What is the ability of changing bacterial colonies? This paragraph in general is difficult to read; it is hard to understand what conclusions the authors are trying to reach and how they have reached them given the text.
Reply: Thank you so much for your professional suggestions. In this sentence, "the ability of changing bacterial colonies of HA-Ti6 and MoS2-Ti6 was poor" means that the antibacterial ability of HA-Ti6 and MoS2-Ti6 was weak. The ability of changing bacterial colonies meant the antibacterial ability. And we also revised it in our manuscript at Page 11 Line 239: "However, the former exhibited highly effective antibacterial efficacy while the latter two had no antibacterial effect (Fig. 2l)." The final conclusion was that antibacterial activity predominantly resulted from the bactericidal ability of HA/MoS2-Ti6. And we analyzed the influence of bacteriostatic effect, bactericidal ability, antifouling effect, and the structure produced by laser cladding on antibacterial activity of HA/MoS2-Ti6, respectively.
Meanwhile, we have also revised this paragraph for more easily to read at Page 10 Line 226: "In addition, the antibacterial mechanism of HA/MoS2-Ti6 against S. aureus was further analyzed in detail (Supplementary Fig. 5, Supplementary Fig. 10-12, Fig.   3a, Fig. 2l). First, the values of MIC and MBC of HA/MoS2-Ti6 were about 5 mg mL -1 and 10 mg mL -1 , respectively. It is known that the antibacterial agents are considered bactericidal if the value of MBC is < 4 times the value of MIC. 13 The ratio of MBC to MIC was 2, indicating that HA/MoS2-Ti6 had a certain bacteriostatic effect ( Supplementary Fig. 10-11). Second, the influence of HA/MoS2-Ti6 on bacterial morphology was further assessed by SEM. The bacterial membranes were irregular or completely broken after culturing on the surface of HA/MoS2-Ti6 for 6 h, suggesting the bactericidal ability of HA/MoS2-Ti6 (Fig. 3a). Next, the influence of antifouling ability on antibacterial activity was evaluated. Generally, the good hydrophilicity is beneficial for developing an antifouling surface. In this work, as shown in Supplementary Fig. 5, the HA/MoS2-Ti6 owned a good hydrophilic surface because of many hydroxyl groups on the surface, and similarly, both HA-Ti6 and MoS2-Ti6 had good hydrophilicity. However, the former exhibited highly effective antibacterial efficacy while the latter two had no antibacterial effect (Fig. 2l). These results suggested that the antibacterial ability of HA/MoS2-Ti6 was not caused by the antifouling effect.
Additionally, the laser cladding induced structure on Ti6 (SLM-Ti6) almost had no antibacterial effect (Supplementary Fig. 12), suggesting the antibacterial ability of HA/MoS2-Ti6 was not caused by the laser induced structure. All in all, the antibacterial  Reply: Thank you so much for your comment. The results indicated that the groups statistically were significantly different from one another in Fig. 6f (originally Fig. 5f).
Based on your request, we have clarified the statistical relationships between the 4 groups in Fig. 6f, Fig. 6h and