Response to mechanical loading in rat Achilles tendon healing is influenced by the microbiome

Response to mechanical loading in rat Achilles tendon healing is influenced by the microbiome a1111111111 a1111111111 We have previously shown that changes in the microbiome influence how the healing ten- don responds to different treatments. The aim of this study was to investigate if changes in the microbiome influence the response to mechanical loading during tendon healing. 90 Sprague-Dawley rats were used. Specific Opportunist and Pathogen Free (SOPF) rats OPENACCESS were co-housed with Specific Pathogen Free (SPF) rats, carrying Staphylococcus aureus Citation: Dietrich-Zagonel F, Hammerman M, and other opportunistic microbes. After 6 weeks of co-housing, the SOPF rats were contam- Eliasson P, Aspenberg P (2020) Response to inated which was confirmed by Staphylococcus aureus growth. Clean SOPF rats were used mechanical loading in rat Achilles tendon healing is as controls. The rats were randomized to full loading or partial unloading by Botox injections influenced by the microbiome. PLoS ONE 15(3): e0229908. https://doi.org/10.1371/journal. in their calf muscles followed by complete Achilles tendon transection. Eight days later, the pone.0229908 healing tendons were tested mechanically. The results were analysed by a 2-way ANOVA Editor: Hazel R. C. Screen, Queen Mary University with interaction between loading and contamination on peak force as the primary outcome of London, UNITED KINGDOM and there was an interaction for both peak force (p = 0.049) and stiffness (p = 0.033). Fur- Received: October 18, 2019 thermore, partial unloading had a profound effect on most outcome variables. In conclusion, the response to mechanical loading during tendon healing is influenced by changes in the Accepted: February 16, 2020 microbiome. Studies aiming for clinical relevance should therefore consider the microbiome Published: March 10, 2020 of laboratory animals. Copyright:© 2020 Dietrich-Zagonel et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and Introduction reproduction in any medium, provided the original author and source are credited. Tendon healing is dependent on the immune system [1] which has a profound effect on the Data Availability Statement: All relevant data are strength of the tendon callus [2, 3]. The immune system, and particularly T cells, changes in within the paper and its Supporting Information response to alterations in the microbial environment in the gut and skin [4, 5]. To the extent files. that T cells play a role in tendon healing, microbial changes in the gut can indirectly influence Funding: This study was funded by the Swedish this healing. Recent data show that changes in the gut microbiota can alter the levels of CD4 Research Council (VR 02031-47- 5; PA). The + and CD3+ T cells within the tendon callus [1]. Furthermore, changes in the microbiome can Swedish National Centre for Research in Sports influence the response to different immunomodulatory treatments, such as local PRP injec- (P218-0140; PA), Linko ¨ping University and tions or systemic corticosteroid treatment, during tendon healing [1, 3]. To the best of our Ostergo ¨tland Country Council (LIO-698351; PA) knowledge, these two studies are the first to confirm a connection between the microbiome and Lions Research Foundation (no specific grant number; PE). The funders had no role in study and tendon healing outcomes. PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 1 / 9 PLOS ONE Microbiome, loading and tendon healing. design, data collection and analysis, decision to Mechanical loading also profoundly influences tendon healing. In a rat Achilles tendon publish, or preparation of the manuscript. healing model, complete load protection can reduce the strength of the healing tendon by a factor of 5 [6]. The magnitude of loading appears to activate different mechanisms which will Competing interests: The authors have declared that no competing interests exist. ultimately lead to different effects on the mechanical properties [7]. The Achilles tendon heals through callus formation and the initial tissue contains weak matrix with primarily leukocytes infiltrating this matrix [8]. Full loading, when applied to this matrix, has been shown to induce microdamage and alter the immune cell composition and ultimately promote a pro-inflamma- tory response [7, 9]. Hence, in part, the response to mechanical loading seems to be closely connected to immune cell reactions and thereby possibly to changes in the microbiome [8– 10]. We therefore hypothesized that changes in the microbiome, would also influence the response to mechanical loading. Results The SPF rats were only used for co-housing as they came from a different breeder facility. Therefore, only SOPF animals (clean and contaminated) were used for the actual experiment and data collection. Co-housing model and contamination procedure 22 out of the 30 co-housed SOPF rats showed Staphyloccocus aureus, growth grade 2 or more. The remaining 8 rats with little or no growth were excluded from the analysis (Fig 1). Microbial community status 6 weeks of co-housing lead to higher levels of Staphylococcus aureus and Escherichia coli in the gut microbiota of the contaminated SOPF rats compared to the clean SOPF rats, where the lev- els of these bacteria were below the detectable limits. However, the total amount of bacteria was higher in the clean rats in comparison to the contaminated rats, including higher levels of gram-positive bacteria (Staphylococcus spp and Enterococcus spp) as previously described [1] (Table 1). Mechanical testing Partial unloading by Botox had profound effects on most outcome variables (Table 2) and lead to a reduced material and structure properties (Table 3). This was seen in both clean and con- taminated rats. The interaction between loading and contamination (by a 2-way ANOVA) was statistically significant for the predetermined primary variable peak force and also for stiffness (p = 0.049 and 0.033 respectively, Table 3). Albeit, looking at the two factors (loading and Fig 1. Bacteria growing on CHROMagar Staph Chrom plates, from oral swabs of contaminate rats. The pink colour means growth of S. aureus. The green colour means growth of other bacteria. Numbers mean: 3 dominant presence of S. aureus; 2 more than half; 1 means minimal; and 0 means no detected Staphyloccocus aureus. Only rats with growth grade 2 and 3 were used for data collection. https://doi.org/10.1371/journal.pone.0229908.g001 PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 2 / 9 PLOS ONE Microbiome, loading and tendon healing. Table 1. The gut bacterial flora of SOPF rats. Enterococcus spp /g Escherichia coli /g Staphylococcus aureus/g Staphylococcus spp/g Total amount of bacteria/g SOPF 3100000 < 100 < 100 2800 100000000 clean 1500000 < 100 < 100 380 410000000 920000 < 100 < 100 400 140000000 5700000 < 100 < 100 < 100 160000000 110000 < 100 < 100 6200 470000000 Mean 2266000 Non detectable Non detectable 2445 256000000 SOPF 140000 36000 280000 < 100 40000000 contaminated 160000 1600000 1600000 < 100 42000000 2000000 44000 320000 < 100 40000000 240000 460000 140000 < 100 90000000 3600000 28000 300000 < 100 70000000 Mean 1228000 433600 528000 Non detectable 56400000 Values are from SOPF clean (n = 5) and contaminated (n = 5) rats. Mean values are presented for each bacterial strain and the total amount of bacteria from faecal samples. https://doi.org/10.1371/journal.pone.0229908.t001 contamination) individually, it showed that loading had a pronounced response on all vari- ables while contamination had no effect in the 2-way ANOVA. A post hoc analysis showed a small reduction in stiffness (p = 0.042) after contamination in partially unloaded rats, while contamination in fully loaded rats tended to increase the peak force compared to the clean rats (p = 0.056, Table 3 and Fig 2). Discussion Our predetermined primary variable was the interaction between loading status and micro- biome on tendon callus strength and we found such an effect. There was also an interaction for stiffness. However, the two individual factors showed only a pronounced effect of loading and not by contamination. When analysing separately clean and contaminated animals, load- ing status alone (full loading vs partial unloading) had a clear effect on all material and struc- tural properties. When microbiome effects were analysed separately, contamination leads to more diverse results depending on the loading status. In a previous study using the same co-housing design as in this study, changes in the micro- biome lead to an altered response to dexamethasone treatment during tendon healing. Table 2. Mechanical results from clean and contaminated rats. Full loading X Botox p-value Clean Contaminated Structural properties Transverse area (mm ) <0.001 <0.001 Gap distance (mm) <0.001 <0.001 Peak force (N) <0.001 <0.001 Stiffness (N/mm) <0.001 <0.001 Energy uptake (N/mm) <0.001 <0.001 Material properties Peak stress (Mpa) <0.001 <0.001 Estimate of E-modulus (Mpa) <0.001 <0.001 https://doi.org/10.1371/journal.pone.0229908.t002 PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 3 / 9 PLOS ONE Microbiome, loading and tendon healing. Table 3. Full loading vs Botox in SOPF animals. Clean Contaminated Full loading Botox Full Botox Full Botox p-value 2-way ANOVA Ratio: Cont/ p-value Ratio: Cont/ p-value loading loading Interaction clean t-test clean t-test Mean - Mean - Mean - Mean - (Cont-Loading) SD SD SD SD Structural Transverse area 16.80 10.68 17.73 10.83 0.668 (0.563–0.001) 1.06 0.51 1.01 0.88 properties (mm ) (4.0) (2.4) (3.6) (2.0) Gap distance (mm) 9.90 (0.8) 3.52 (0.5) 10.28 3.96 (0.6) 0.918 (0.097–0.001) 1.04 0.32 1.13 0.06 (1.2) Peak force (N) 26.73 9.68 (2.9) 30.7 (6.7) 8.51 (2.9) 0.049 (0.278–0.001) 1.15 0.06 0.87 0.35 (3.7) Stiffness (N/mm) 3.35 (0.9) 2.18 (0.5) 3.77 (0.7) 1.71 (0.4) 0.033 (0.908–0.001) 1.12 0.18 0.78 0.04 Energy uptake 70.26 13.04 72.45 12.62 0.789 (0.856–0.001) 1.03 0.79 0.97 0.85 (N/mm) (15.2) (4.4) (27.5) (5.7) Material Peak stress (Mpa) 1.65 (0.3) 0.95 (0.4) 1.78 (0.4) 0.78 (0.2) 0.149 (0.855–0.001) 1.08 0.37 0.81 0.23 properties Estimate of 2.00 (0.5) 0.76 (0.3) 2.29 (0.8) 0.64 (0.2) 0.151 (0.571–0.001) 1.15 0.22 0.84 0.27 E-modulus (Mpa) Clean animals: Full loading (n = 15), Botox (n = 15); Contaminated: Full loading (n = 14), Botox (n = 8). Values are mean and standard deviation (SD). https://doi.org/10.1371/journal.pone.0229908.t003 Furthermore, the controls showed a different response between the clean and contaminated animals where contaminated rats had considerably decreased peak stress and estimate of elas- tic modulus [1]. One difference in the set-up between our new study and this previous one is the duration of tendon healing. The previous study was evaluated 12 days after tendon transec- tion instead of 8 days after. This could be an explanation to the less pronounced effect of con- tamination alone in this study. We have previously used 8 days of healing for several studies on unloaded and loaded heal- ing tendons [6, 7]. However, 8 days is only a few days after the inflammatory phase of healing, which last approximately 3–5 days in rats [11]. To observe a more pronounced response of the microbiota in the mechanical testing we probably need to wait further into the proliferatory and remodelling phase. We have previously seen that there is no mechanical effect 3 days after Fig 2. Mechanical data for clean and contaminated rats. Peak force and stiffness at 8 days after tendon transection with or without loading (Botox). Partial unloading by Botox reduced tendon force and stiffness in the clean and contaminated animals. Contamination decreased the stiffness on the partially unloaded animals. Boxes include median, interquartile range and total range (whiskers). ( ) Means statistically significant difference (p<0.05). https://doi.org/10.1371/journal.pone.0229908.g002 PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 4 / 9 PLOS ONE Microbiome, loading and tendon healing. one single loading episode, however there is a measurable effect 7 days after [10]. Furthermore, new unpublished data shows that loading has the most pronounced effects on tendon healing during 1–2 weeks of healing, but not at 4 weeks (unpublished data). It seems possible that a 12 days healing time in this study could have yielded clearer answers to the secondary outcome variables too. The strong effect of Botox demonstrates the paramount importance of mechanical loading for tendon healing. This has been shown in numerous previous studies using Botox and other unloading devices in this rat transection model, and does not need further discussion here [6, 11–14]. However, it is worth noticing that the response to contamination tended to diverge between Botox treated and fully loaded rats. The ratio for peak force, between clean and con- taminated rats, was 0.87 in the Botox treated animals compared to 1.15 in the fully loaded ani- mals. Previous studies have shown that full loading vs partial unloading activates different mechanisms which ultimately will lead to different mechanical response, more specifically, full loading seems to prolong the proinflammatory response [7]. Our results imply that a change in the rat microbiome using our co-housing method modi- fies the rat microbiota, as confirmed by both oral swabs tests and faecal analyses. Unpublished data, after only 3 weeks of co-housing, showed that albeit the bacteria had been transferred to the new animals, a longer time was needed to develop a change in the bacterial flora of the ani- mals. This was confirmed by a lack of change in T-cell populations between the clean and con- taminated rats in contrast to what we saw after 6 weeks of co-housing [1]. There was also a lack of effect on the peak force. It seems therefore that a longer period of co-housing is needed for the immune system to “mature” upon contact with new bacteria. Co-housing, for 6 weeks as in this study, will ultimately lead to higher levels of CD3+ and CD4+ T cells in the healing tendon callus [1]. In bone healing, CD4+ T cells are suggested to play a beneficial role in tissue regeneration [15]. This might explain the tendency to stronger tendons in fully loaded contaminated rats compared to the clean (p = 0.056). The early callus is dominated by different leukocytes, and the composition of these cells undergoes dramatic change during the early healing phases in our model [8–10]. Full loading, in contrast to partial unloading, can result in tissue microdamage followed by a pro-inflammatory response [7]. As full loading, but not partial unloading, creates microdamage, it could explain the divergent response to contamination in with different degrees of loading. Immunomodulatory changes by the microbiome have probably more profound effects on the healing of fully loaded tendons. The Botox model for partial unloading is however probably more similar to the clinical situation after an Achilles tendon rupture where the patients have the lower leg immobilized in an orthosis or cast. The response to small loads in Botox treated animals probably acts primarily through mechanotransduction while the response in fully loaded animals acts probably through a combination of a mechanotransduction and micro- damage. Partial unloading by Botox can also change the composition of T cells in the tendon callus [9] but when we compare partial unloaded clean and contaminated rats, contamination lead to a significant decrease in the stiffness of the healing tendon. However, a limitation with this study is that the fully loaded animals did not get a sham injection in the calf muscle, but we believe that it is unlikely that this minor muscle needling itself would influence the tendon healing. Another limitation of the study is that the mechanisms of how the gut microbiota influence tendon healing was not elucidate. We only used a standard commercial test for faecal analysis in this study and no extensive bacteriological analysis was performed. This specific test was chosen to confirm a change in the microbiota, and not to verify which strain of bacteria that are important for tendon healing. The objectives with this study was to understand if the response to different loading conditions were affected by the microbiome and albeit the effects PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 5 / 9 PLOS ONE Microbiome, loading and tendon healing. are minor, we did confirm a link between microbiome changes and the loading status during tendon healing. Taken together, the microbiome influences tendon healing and the response to mechanical loading. The use of different levels of clean animals could possibly lead to difficulties in repro- ducing results. Future experiments aiming for clinical relevance should probably take the microbiome into account when studying the response to loading in animal models. Conclusion The gut microflora has an effect on mechanical loading for tendon healing stimulation. Materials and methods Experimental design In total 90 rats were used for this study: 60 specific opportunist and pathogen free (SOPF) rats and 30 specific pathogen free (SPF). The SPF rats were bred under less clean conditions and were known to carry Staphylococcus aureus. 30 of the SOPF rats were co-housed one by one with an SPF rat for 6 weeks for contamination by microorganisms carried by the SPF rats. The remaining SOPF rats were kept in a separate part of the department to ensure that they were remaining clean but were also housed in pairs. Only SOPF rats (clean and contaminated) were used for data collection. These animals were randomized to either full loading or partial unloading (mild loading) before the contamination process was performed (Fig 3). Partial unloading was achieved by Botox injections in the calf muscles. Thereafter, all rats underwent Achilles tendon transection and eight days later, tendons were harvested and mechanically Fig 3. Experimental set-up. https://doi.org/10.1371/journal.pone.0229908.g003 PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 6 / 9 PLOS ONE Microbiome, loading and tendon healing. tested. Our pre-determined primary outcome variable was an interaction between loading and contamination for peak force, measured by a 2-way ANOVA. Animals and housing. Female SOPF and SPF Sprague-Dawley rats, weighing on average of 275 grams (SD 23), 11 to 12 weeks old, were purchased from Janvier, Le Genest-Saint-Isle, France, from two different breeding facilities. All experiments were approved by the local ethi- cal committee (Regional Ethics Committee for Animal Experiments in Linko ¨ping, ref. 15–15 and 592). The experiments were adhered to institutional guidelines for care and treatment of laboratory animals, and all efforts were made to minimize suffering. All animals were housed in acrylic cages, two by two, and placed on ventilated racks with humidity (55%), temperature (22˚C) and light-dark cycle (12h each). Standard food pellets and water was given ad libitum. Co-housing model and contamination procedure. The SOPF rats were, according to the breeder, free of Bordetella bronchiseptica, CAR bacillus, Clostridium piliforme (tyzzer), Coryne- bacterium kutscheri, Dermatophytes, Encephalitozoon cuniculi, Helicobacter spp, Klebsiella oxy- toca/pneumoniae, Mycoplasma pulmonis, Pasteurellaceae sp., Pneumocystis spp, Proteus spp (mirabilis, vulgaris), Pseudomonas aeruginosa, Salmonella spp., Staphylococcus aureus, Strepto- bacillus moniliformis, Streptococci ß-hemolytic, and Streptococcus pneumoniae. While all SPF rats carried Staphylococcus aureus and Pseudomonas aeruginosa at arrival. The Staphylococcus aureus was confirmed by oral swab test and bacterial culture as this was the intended bacterial strain that we wanted to contaminate the SOPF rats with. Contamination was performed by co-housing one SOPF rat with one SPF rat, as described previously [1, 16]. After 6 weeks of co- housing, contamination was confirmed in the SOPF rats by testing for Staphylococcus aureus by an oral swab test and bacterial culture. Bacteria were grown on CHROMagar Staph Chrom plates for 24 hours and the result was graded as 0, 1, 2 or 3, where 0 means no detected Staphy- loccocus aureus, 1 means minimal, 2 more than half, and 3 dominant presence. Only rats with growth grade 2 and 3 were used for analysis. Botox injections for partial unloading. 30 rats (15 clean and 15 contaminated) were anesthetized with isoflurane gas followed by botulinum toxin (Botox, Allergan, Irvine, CA) in the right hind leg. The injections were performed into the gastrocnemius lateralis, medialis and soleus muscles (dose of 1 U/muscle, in total 3 U/animal and a volume of 0.06 mL). The animals were thereafter allowed free cage activity. Botox effectiveness was confirmed prior to the surgery by visual inspection. Tendon transection. Four days later, both the partial unloaded and the fully loaded rats were anesthetized with isoflurane gas and the Achilles tendon, from the right hind leg was transected. To avoid postoperative infection, antibiotic was given once preoperatively (25 mg/ kg oxytetracycline). Analgesic was given subcutaneously pre- and postoperatively every 8-12h for 48h (0.045 mg/kg buprenorphine). Surgery was performed under aseptic conditions and the skin on the right lower leg was shaved and cleaned with chlorhexidine ethanol. The Achil- les and plantaris tendon complex was exposed through a transverse skin incision on the lateral side. The plantaris tendon was removed completely and the Achilles tendon was cut trans- versely in the middle part and was left unsutured whereas the skin was closed with 2 stitches. This procedure has been described elsewhere [3]. Mechanical testing. Eight days after surgery, the rats were anesthetized with isoflurane gas and killed with carbon dioxide. The right Achilles tendon was harvested together with the calcaneal bone and calf muscle. Sagittal and transverse diameter of the midpart of the callus tis- sue was measured with a slide calliper, and the transverse area was calculated by assuming an elliptical geometry. The distance between the old tendon stumps (gap distance) was measured, as seen through the partly transparent callus tissue. The muscle tissue was scraped off from the tendon, which was fixed in a metal clamp with sandpaper. The bone was fixed in a custom- made clamp at 30˚ dorsiflexion relative to the direction of traction in the materials-testing PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 7 / 9 PLOS ONE Microbiome, loading and tendon healing. machine (100R; DDL, Eden Prairie, MN). The machine pulled at 0.1 mm/s until failure. Peak force at failure (N) and energy uptake (N/mm), were calculated by the software of the machine. The investigator marked a linear portion of the elastic phase of the curve for automated stiff- ness (N/mm) calculation. Peak stress (MPa) and estimate of elastic modulus (MPa) were calcu- lated assuming an elliptical cylindrical shape and homogeneous mechanical properties. Estimation of elastic modulus was calculated as stiffness gap distance/transverse area [1, 3]. Characterization of gut bacterial flora. After the tendon samples were collected, SOPF rats (5 clean and 5 contaminated) from the full loading group were used to confirm changes in the microbial community. Faecal pellets were collected via laparotomy under aseptic condi- tions. The microbial community was identified by Surrey Diagnostics Ltd using the API (Bio- Merieux) and RapID (Remel) biochemical test. These data have previously been described as they were not collected solely for the present study [1]. Statistics. The mechanical results were analysed using SPSS software version 21. Only SOPF animals (clean and contaminated) were included. SPF rats were discarded. Our prede- fined primary outcome was the interaction between microbiome (contamination) and loading status for peak force, measured by a 2-way ANOVA. For post hoc, we compared groups pair- wise using Student’s t-test to describe the response to contamination among loaded and partial unloading rats separately. The presence of Staphyloccocus aureus and the tests of bacteriologi- cal status in the gut microbiota were only descriptive. Supporting information S1 Dataset. Data from mechanical evaluation and the status of the gut bacterial flora. Data are from each rat. (XLSX) Acknowledgments The authors thank Aneta Liszka for her assistance during the bacteria culture growth and Eme- lie Jansson for her help during the oral swabs collection. Dr. Per Aspenberg passed away before the submission of the final version of this manu- script. Franciele Dietrich-Zagonel accepts the responsibility for the integrity and validity of the data collected and analyzed. Author Contributions Conceptualization: Franciele Dietrich-Zagonel, Pernilla Eliasson, Per Aspenberg. Data curation: Franciele Dietrich-Zagonel, Malin Hammerman. Formal analysis: Franciele Dietrich-Zagonel, Malin Hammerman. Funding acquisition: Pernilla Eliasson, Per Aspenberg. Investigation: Franciele Dietrich-Zagonel, Malin Hammerman. Methodology: Franciele Dietrich-Zagonel, Per Aspenberg. Project administration: Franciele Dietrich-Zagonel. Supervision: Pernilla Eliasson, Per Aspenberg. Validation: Franciele Dietrich-Zagonel, Pernilla Eliasson, Per Aspenberg. Visualization: Franciele Dietrich-Zagonel, Per Aspenberg. PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 8 / 9 PLOS ONE Microbiome, loading and tendon healing. Writing – original draft: Franciele Dietrich-Zagonel, Per Aspenberg. Writing – review & editing: Franciele Dietrich-Zagonel, Malin Hammerman, Pernilla Elias- son, Per Aspenberg. References 1. Dietrich-Zagonel F, Hammerman M, Tatting L, Dietrich F, Kozak Ljunggren M, Blomgran P, et al. Stimu- lation of Tendon Healing With Delayed Dexamethasone Treatment Is Modified by the Microbiome. The American journal of sports medicine. 2018; 46(13):3281–7. https://doi.org/10.1177/0363546518799442 PMID: 30265844 2. Blomgran P, Hammerman M, Aspenberg P. Systemic corticosteroids improve tendon healing when given after the early inflammatory phase. Scientific reports. 2017; 7(1):12468. https://doi.org/10.1038/ s41598-017-12657-0 PMID: 28963482 3. Dietrich F, Hammerman M, Blomgran P, Tatting L, Bampi VF, Silva JB, et al. Effect of platelet-rich plasma on rat Achilles tendon healing is related to microbiota. Acta orthopaedica. 2017; 88(4):416–21. https://doi.org/10.1080/17453674.2017.1293447 PMID: 28296518 4. Linehan JL, Harrison OJ, Han SJ, Byrd AL, Vujkovic-Cvijin I, Villarino AV, et al. Non-classical Immunity Controls Microbiota Impact on Skin Immunity and Tissue Repair. Cell. 2018; 172(4):784–96 e18. 5. Furusawa Y, Obata Y, Fukuda S, Endo TA, Nakato G, Takahashi D, et al. Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells. Nature. 2013; 504(7480):446–50. https://doi.org/10.1038/nature12721 PMID: 24226770 6. Andersson T, Eliasson P, Hammerman M, Sandberg O, Aspenberg P. Low-level mechanical stimula- tion is sufficient to improve tendon healing in rats. J Appl Physiol (1985). 2012; 113(9):1398–402. 7. Hammerman M, Dietrich-Zagonel F, Blomgran P, Eliasson P, Aspenberg P. Different mechanisms acti- vated by mild versus strong loading in rat Achilles tendon healing. PloS one. 2018; 13(7):e0201211. https://doi.org/10.1371/journal.pone.0201211 PMID: 30044869 8. Hammerman M, Blomgran P, Dansac A, Eliasson P, Aspenberg P. Different gene response to mechani- cal loading during early and late phases of rat Achilles tendon healing. Journal of applied physiology. 2017; 123(4):800–15. https://doi.org/10.1152/japplphysiol.00323.2017 PMID: 28705996 9. Blomgran P, Blomgran R, Ernerudh J, Aspenberg P. A possible link between loading, inflammation and healing: Immune cell populations during tendon healing in the rat. Scientific reports. 2016; 6:29824. https://doi.org/10.1038/srep29824 PMID: 27405922 10. Eliasson P, Andersson T, Aspenberg P. Influence of a single loading episode on gene expression in healing rat Achilles tendons. Journal of applied physiology. 2012; 112(2):279–88. https://doi.org/10. 1152/japplphysiol.00858.2011 PMID: 21998267 11. Andersson T, Eliasson P, Aspenberg P. Tissue memory in healing tendons: short loading episodes stimulate healing. Journal of applied physiology. 2009; 107(2):417–21. https://doi.org/10.1152/ japplphysiol.00414.2009 PMID: 19541735 12. Eliasson P, Andersson T, Aspenberg P. Achilles tendon healing in rats is improved by intermittent mechanical loading during the inflammatory phase. Journal of orthopaedic research: official publication of the Orthopaedic Research Society. 2012; 30(2):274–9. 13. Eliasson P, Andersson T, Aspenberg P. Rat Achilles tendon healing: mechanical loading and gene expression. Journal of applied physiology. 2009; 107(2):399–407. https://doi.org/10.1152/japplphysiol. 91563.2008 PMID: 19541731 14. Freedman BR, Gordon JA, Bhatt PR, Pardes AM, Thomas SJ, Sarver JJ, et al. Nonsurgical treatment and early return to activity leads to improved Achilles tendon fatigue mechanics and functional out- comes during early healing in an animal model. Journal of orthopaedic research: official publication of the Orthopaedic Research Society. 2016; 34(12):2172–80. 15. Schlundt C, Schell H, Goodman SB, Vunjak-Novakovic G, Duda GN, Schmidt-Bleek K. Immune modu- lation as a therapeutic strategy in bone regeneration. Journal of experimental orthopaedics. 2015; 2 (1):1. https://doi.org/10.1186/s40634-014-0017-6 PMID: 26914869 16. Beura LK, Hamilton SE, Bi K, Schenkel JM, Odumade OA, Casey KA, et al. Normalizing the environ- ment recapitulates adult human immune traits in laboratory mice. Nature. 2016; 532(7600):512–6. https://doi.org/10.1038/nature17655 PMID: 27096360 PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 9 / 9 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png PLoS ONE Public Library of Science (PLoS) Journal

Response to mechanical loading in rat Achilles tendon healing is influenced by the microbiome

PLoS ONE, Volume 15 (3) – Mar 10, 2020

Loading next page...
 
/lp/public-library-of-science-plos-journal/response-to-mechanical-loading-in-rat-achilles-tendon-healing-is-ci2njP1J9S
Publisher
Public Library of Science (PLoS) Journal
Copyright
Copyright: © 2020 Dietrich-Zagonel et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: All relevant data are within the paper and its Supporting Information files. Funding: This study was funded by the Swedish Research Council (VR 02031-47- 5; PA). The Swedish National Centre for Research in Sports (P218-0140; PA), Linköping University and Östergötland Country Council (LIO-698351; PA) and Lions Research Foundation (no specific grant number; PE). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist.
eISSN
1932-6203
DOI
10.1371/journal.pone.0229908
Publisher site
See Article on Publisher Site

Abstract

a1111111111 a1111111111 We have previously shown that changes in the microbiome influence how the healing ten- don responds to different treatments. The aim of this study was to investigate if changes in the microbiome influence the response to mechanical loading during tendon healing. 90 Sprague-Dawley rats were used. Specific Opportunist and Pathogen Free (SOPF) rats OPENACCESS were co-housed with Specific Pathogen Free (SPF) rats, carrying Staphylococcus aureus Citation: Dietrich-Zagonel F, Hammerman M, and other opportunistic microbes. After 6 weeks of co-housing, the SOPF rats were contam- Eliasson P, Aspenberg P (2020) Response to inated which was confirmed by Staphylococcus aureus growth. Clean SOPF rats were used mechanical loading in rat Achilles tendon healing is as controls. The rats were randomized to full loading or partial unloading by Botox injections influenced by the microbiome. PLoS ONE 15(3): e0229908. https://doi.org/10.1371/journal. in their calf muscles followed by complete Achilles tendon transection. Eight days later, the pone.0229908 healing tendons were tested mechanically. The results were analysed by a 2-way ANOVA Editor: Hazel R. C. Screen, Queen Mary University with interaction between loading and contamination on peak force as the primary outcome of London, UNITED KINGDOM and there was an interaction for both peak force (p = 0.049) and stiffness (p = 0.033). Fur- Received: October 18, 2019 thermore, partial unloading had a profound effect on most outcome variables. In conclusion, the response to mechanical loading during tendon healing is influenced by changes in the Accepted: February 16, 2020 microbiome. Studies aiming for clinical relevance should therefore consider the microbiome Published: March 10, 2020 of laboratory animals. Copyright:© 2020 Dietrich-Zagonel et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and Introduction reproduction in any medium, provided the original author and source are credited. Tendon healing is dependent on the immune system [1] which has a profound effect on the Data Availability Statement: All relevant data are strength of the tendon callus [2, 3]. The immune system, and particularly T cells, changes in within the paper and its Supporting Information response to alterations in the microbial environment in the gut and skin [4, 5]. To the extent files. that T cells play a role in tendon healing, microbial changes in the gut can indirectly influence Funding: This study was funded by the Swedish this healing. Recent data show that changes in the gut microbiota can alter the levels of CD4 Research Council (VR 02031-47- 5; PA). The + and CD3+ T cells within the tendon callus [1]. Furthermore, changes in the microbiome can Swedish National Centre for Research in Sports influence the response to different immunomodulatory treatments, such as local PRP injec- (P218-0140; PA), Linko ¨ping University and tions or systemic corticosteroid treatment, during tendon healing [1, 3]. To the best of our Ostergo ¨tland Country Council (LIO-698351; PA) knowledge, these two studies are the first to confirm a connection between the microbiome and Lions Research Foundation (no specific grant number; PE). The funders had no role in study and tendon healing outcomes. PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 1 / 9 PLOS ONE Microbiome, loading and tendon healing. design, data collection and analysis, decision to Mechanical loading also profoundly influences tendon healing. In a rat Achilles tendon publish, or preparation of the manuscript. healing model, complete load protection can reduce the strength of the healing tendon by a factor of 5 [6]. The magnitude of loading appears to activate different mechanisms which will Competing interests: The authors have declared that no competing interests exist. ultimately lead to different effects on the mechanical properties [7]. The Achilles tendon heals through callus formation and the initial tissue contains weak matrix with primarily leukocytes infiltrating this matrix [8]. Full loading, when applied to this matrix, has been shown to induce microdamage and alter the immune cell composition and ultimately promote a pro-inflamma- tory response [7, 9]. Hence, in part, the response to mechanical loading seems to be closely connected to immune cell reactions and thereby possibly to changes in the microbiome [8– 10]. We therefore hypothesized that changes in the microbiome, would also influence the response to mechanical loading. Results The SPF rats were only used for co-housing as they came from a different breeder facility. Therefore, only SOPF animals (clean and contaminated) were used for the actual experiment and data collection. Co-housing model and contamination procedure 22 out of the 30 co-housed SOPF rats showed Staphyloccocus aureus, growth grade 2 or more. The remaining 8 rats with little or no growth were excluded from the analysis (Fig 1). Microbial community status 6 weeks of co-housing lead to higher levels of Staphylococcus aureus and Escherichia coli in the gut microbiota of the contaminated SOPF rats compared to the clean SOPF rats, where the lev- els of these bacteria were below the detectable limits. However, the total amount of bacteria was higher in the clean rats in comparison to the contaminated rats, including higher levels of gram-positive bacteria (Staphylococcus spp and Enterococcus spp) as previously described [1] (Table 1). Mechanical testing Partial unloading by Botox had profound effects on most outcome variables (Table 2) and lead to a reduced material and structure properties (Table 3). This was seen in both clean and con- taminated rats. The interaction between loading and contamination (by a 2-way ANOVA) was statistically significant for the predetermined primary variable peak force and also for stiffness (p = 0.049 and 0.033 respectively, Table 3). Albeit, looking at the two factors (loading and Fig 1. Bacteria growing on CHROMagar Staph Chrom plates, from oral swabs of contaminate rats. The pink colour means growth of S. aureus. The green colour means growth of other bacteria. Numbers mean: 3 dominant presence of S. aureus; 2 more than half; 1 means minimal; and 0 means no detected Staphyloccocus aureus. Only rats with growth grade 2 and 3 were used for data collection. https://doi.org/10.1371/journal.pone.0229908.g001 PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 2 / 9 PLOS ONE Microbiome, loading and tendon healing. Table 1. The gut bacterial flora of SOPF rats. Enterococcus spp /g Escherichia coli /g Staphylococcus aureus/g Staphylococcus spp/g Total amount of bacteria/g SOPF 3100000 < 100 < 100 2800 100000000 clean 1500000 < 100 < 100 380 410000000 920000 < 100 < 100 400 140000000 5700000 < 100 < 100 < 100 160000000 110000 < 100 < 100 6200 470000000 Mean 2266000 Non detectable Non detectable 2445 256000000 SOPF 140000 36000 280000 < 100 40000000 contaminated 160000 1600000 1600000 < 100 42000000 2000000 44000 320000 < 100 40000000 240000 460000 140000 < 100 90000000 3600000 28000 300000 < 100 70000000 Mean 1228000 433600 528000 Non detectable 56400000 Values are from SOPF clean (n = 5) and contaminated (n = 5) rats. Mean values are presented for each bacterial strain and the total amount of bacteria from faecal samples. https://doi.org/10.1371/journal.pone.0229908.t001 contamination) individually, it showed that loading had a pronounced response on all vari- ables while contamination had no effect in the 2-way ANOVA. A post hoc analysis showed a small reduction in stiffness (p = 0.042) after contamination in partially unloaded rats, while contamination in fully loaded rats tended to increase the peak force compared to the clean rats (p = 0.056, Table 3 and Fig 2). Discussion Our predetermined primary variable was the interaction between loading status and micro- biome on tendon callus strength and we found such an effect. There was also an interaction for stiffness. However, the two individual factors showed only a pronounced effect of loading and not by contamination. When analysing separately clean and contaminated animals, load- ing status alone (full loading vs partial unloading) had a clear effect on all material and struc- tural properties. When microbiome effects were analysed separately, contamination leads to more diverse results depending on the loading status. In a previous study using the same co-housing design as in this study, changes in the micro- biome lead to an altered response to dexamethasone treatment during tendon healing. Table 2. Mechanical results from clean and contaminated rats. Full loading X Botox p-value Clean Contaminated Structural properties Transverse area (mm ) <0.001 <0.001 Gap distance (mm) <0.001 <0.001 Peak force (N) <0.001 <0.001 Stiffness (N/mm) <0.001 <0.001 Energy uptake (N/mm) <0.001 <0.001 Material properties Peak stress (Mpa) <0.001 <0.001 Estimate of E-modulus (Mpa) <0.001 <0.001 https://doi.org/10.1371/journal.pone.0229908.t002 PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 3 / 9 PLOS ONE Microbiome, loading and tendon healing. Table 3. Full loading vs Botox in SOPF animals. Clean Contaminated Full loading Botox Full Botox Full Botox p-value 2-way ANOVA Ratio: Cont/ p-value Ratio: Cont/ p-value loading loading Interaction clean t-test clean t-test Mean - Mean - Mean - Mean - (Cont-Loading) SD SD SD SD Structural Transverse area 16.80 10.68 17.73 10.83 0.668 (0.563–0.001) 1.06 0.51 1.01 0.88 properties (mm ) (4.0) (2.4) (3.6) (2.0) Gap distance (mm) 9.90 (0.8) 3.52 (0.5) 10.28 3.96 (0.6) 0.918 (0.097–0.001) 1.04 0.32 1.13 0.06 (1.2) Peak force (N) 26.73 9.68 (2.9) 30.7 (6.7) 8.51 (2.9) 0.049 (0.278–0.001) 1.15 0.06 0.87 0.35 (3.7) Stiffness (N/mm) 3.35 (0.9) 2.18 (0.5) 3.77 (0.7) 1.71 (0.4) 0.033 (0.908–0.001) 1.12 0.18 0.78 0.04 Energy uptake 70.26 13.04 72.45 12.62 0.789 (0.856–0.001) 1.03 0.79 0.97 0.85 (N/mm) (15.2) (4.4) (27.5) (5.7) Material Peak stress (Mpa) 1.65 (0.3) 0.95 (0.4) 1.78 (0.4) 0.78 (0.2) 0.149 (0.855–0.001) 1.08 0.37 0.81 0.23 properties Estimate of 2.00 (0.5) 0.76 (0.3) 2.29 (0.8) 0.64 (0.2) 0.151 (0.571–0.001) 1.15 0.22 0.84 0.27 E-modulus (Mpa) Clean animals: Full loading (n = 15), Botox (n = 15); Contaminated: Full loading (n = 14), Botox (n = 8). Values are mean and standard deviation (SD). https://doi.org/10.1371/journal.pone.0229908.t003 Furthermore, the controls showed a different response between the clean and contaminated animals where contaminated rats had considerably decreased peak stress and estimate of elas- tic modulus [1]. One difference in the set-up between our new study and this previous one is the duration of tendon healing. The previous study was evaluated 12 days after tendon transec- tion instead of 8 days after. This could be an explanation to the less pronounced effect of con- tamination alone in this study. We have previously used 8 days of healing for several studies on unloaded and loaded heal- ing tendons [6, 7]. However, 8 days is only a few days after the inflammatory phase of healing, which last approximately 3–5 days in rats [11]. To observe a more pronounced response of the microbiota in the mechanical testing we probably need to wait further into the proliferatory and remodelling phase. We have previously seen that there is no mechanical effect 3 days after Fig 2. Mechanical data for clean and contaminated rats. Peak force and stiffness at 8 days after tendon transection with or without loading (Botox). Partial unloading by Botox reduced tendon force and stiffness in the clean and contaminated animals. Contamination decreased the stiffness on the partially unloaded animals. Boxes include median, interquartile range and total range (whiskers). ( ) Means statistically significant difference (p<0.05). https://doi.org/10.1371/journal.pone.0229908.g002 PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 4 / 9 PLOS ONE Microbiome, loading and tendon healing. one single loading episode, however there is a measurable effect 7 days after [10]. Furthermore, new unpublished data shows that loading has the most pronounced effects on tendon healing during 1–2 weeks of healing, but not at 4 weeks (unpublished data). It seems possible that a 12 days healing time in this study could have yielded clearer answers to the secondary outcome variables too. The strong effect of Botox demonstrates the paramount importance of mechanical loading for tendon healing. This has been shown in numerous previous studies using Botox and other unloading devices in this rat transection model, and does not need further discussion here [6, 11–14]. However, it is worth noticing that the response to contamination tended to diverge between Botox treated and fully loaded rats. The ratio for peak force, between clean and con- taminated rats, was 0.87 in the Botox treated animals compared to 1.15 in the fully loaded ani- mals. Previous studies have shown that full loading vs partial unloading activates different mechanisms which ultimately will lead to different mechanical response, more specifically, full loading seems to prolong the proinflammatory response [7]. Our results imply that a change in the rat microbiome using our co-housing method modi- fies the rat microbiota, as confirmed by both oral swabs tests and faecal analyses. Unpublished data, after only 3 weeks of co-housing, showed that albeit the bacteria had been transferred to the new animals, a longer time was needed to develop a change in the bacterial flora of the ani- mals. This was confirmed by a lack of change in T-cell populations between the clean and con- taminated rats in contrast to what we saw after 6 weeks of co-housing [1]. There was also a lack of effect on the peak force. It seems therefore that a longer period of co-housing is needed for the immune system to “mature” upon contact with new bacteria. Co-housing, for 6 weeks as in this study, will ultimately lead to higher levels of CD3+ and CD4+ T cells in the healing tendon callus [1]. In bone healing, CD4+ T cells are suggested to play a beneficial role in tissue regeneration [15]. This might explain the tendency to stronger tendons in fully loaded contaminated rats compared to the clean (p = 0.056). The early callus is dominated by different leukocytes, and the composition of these cells undergoes dramatic change during the early healing phases in our model [8–10]. Full loading, in contrast to partial unloading, can result in tissue microdamage followed by a pro-inflammatory response [7]. As full loading, but not partial unloading, creates microdamage, it could explain the divergent response to contamination in with different degrees of loading. Immunomodulatory changes by the microbiome have probably more profound effects on the healing of fully loaded tendons. The Botox model for partial unloading is however probably more similar to the clinical situation after an Achilles tendon rupture where the patients have the lower leg immobilized in an orthosis or cast. The response to small loads in Botox treated animals probably acts primarily through mechanotransduction while the response in fully loaded animals acts probably through a combination of a mechanotransduction and micro- damage. Partial unloading by Botox can also change the composition of T cells in the tendon callus [9] but when we compare partial unloaded clean and contaminated rats, contamination lead to a significant decrease in the stiffness of the healing tendon. However, a limitation with this study is that the fully loaded animals did not get a sham injection in the calf muscle, but we believe that it is unlikely that this minor muscle needling itself would influence the tendon healing. Another limitation of the study is that the mechanisms of how the gut microbiota influence tendon healing was not elucidate. We only used a standard commercial test for faecal analysis in this study and no extensive bacteriological analysis was performed. This specific test was chosen to confirm a change in the microbiota, and not to verify which strain of bacteria that are important for tendon healing. The objectives with this study was to understand if the response to different loading conditions were affected by the microbiome and albeit the effects PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 5 / 9 PLOS ONE Microbiome, loading and tendon healing. are minor, we did confirm a link between microbiome changes and the loading status during tendon healing. Taken together, the microbiome influences tendon healing and the response to mechanical loading. The use of different levels of clean animals could possibly lead to difficulties in repro- ducing results. Future experiments aiming for clinical relevance should probably take the microbiome into account when studying the response to loading in animal models. Conclusion The gut microflora has an effect on mechanical loading for tendon healing stimulation. Materials and methods Experimental design In total 90 rats were used for this study: 60 specific opportunist and pathogen free (SOPF) rats and 30 specific pathogen free (SPF). The SPF rats were bred under less clean conditions and were known to carry Staphylococcus aureus. 30 of the SOPF rats were co-housed one by one with an SPF rat for 6 weeks for contamination by microorganisms carried by the SPF rats. The remaining SOPF rats were kept in a separate part of the department to ensure that they were remaining clean but were also housed in pairs. Only SOPF rats (clean and contaminated) were used for data collection. These animals were randomized to either full loading or partial unloading (mild loading) before the contamination process was performed (Fig 3). Partial unloading was achieved by Botox injections in the calf muscles. Thereafter, all rats underwent Achilles tendon transection and eight days later, tendons were harvested and mechanically Fig 3. Experimental set-up. https://doi.org/10.1371/journal.pone.0229908.g003 PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 6 / 9 PLOS ONE Microbiome, loading and tendon healing. tested. Our pre-determined primary outcome variable was an interaction between loading and contamination for peak force, measured by a 2-way ANOVA. Animals and housing. Female SOPF and SPF Sprague-Dawley rats, weighing on average of 275 grams (SD 23), 11 to 12 weeks old, were purchased from Janvier, Le Genest-Saint-Isle, France, from two different breeding facilities. All experiments were approved by the local ethi- cal committee (Regional Ethics Committee for Animal Experiments in Linko ¨ping, ref. 15–15 and 592). The experiments were adhered to institutional guidelines for care and treatment of laboratory animals, and all efforts were made to minimize suffering. All animals were housed in acrylic cages, two by two, and placed on ventilated racks with humidity (55%), temperature (22˚C) and light-dark cycle (12h each). Standard food pellets and water was given ad libitum. Co-housing model and contamination procedure. The SOPF rats were, according to the breeder, free of Bordetella bronchiseptica, CAR bacillus, Clostridium piliforme (tyzzer), Coryne- bacterium kutscheri, Dermatophytes, Encephalitozoon cuniculi, Helicobacter spp, Klebsiella oxy- toca/pneumoniae, Mycoplasma pulmonis, Pasteurellaceae sp., Pneumocystis spp, Proteus spp (mirabilis, vulgaris), Pseudomonas aeruginosa, Salmonella spp., Staphylococcus aureus, Strepto- bacillus moniliformis, Streptococci ß-hemolytic, and Streptococcus pneumoniae. While all SPF rats carried Staphylococcus aureus and Pseudomonas aeruginosa at arrival. The Staphylococcus aureus was confirmed by oral swab test and bacterial culture as this was the intended bacterial strain that we wanted to contaminate the SOPF rats with. Contamination was performed by co-housing one SOPF rat with one SPF rat, as described previously [1, 16]. After 6 weeks of co- housing, contamination was confirmed in the SOPF rats by testing for Staphylococcus aureus by an oral swab test and bacterial culture. Bacteria were grown on CHROMagar Staph Chrom plates for 24 hours and the result was graded as 0, 1, 2 or 3, where 0 means no detected Staphy- loccocus aureus, 1 means minimal, 2 more than half, and 3 dominant presence. Only rats with growth grade 2 and 3 were used for analysis. Botox injections for partial unloading. 30 rats (15 clean and 15 contaminated) were anesthetized with isoflurane gas followed by botulinum toxin (Botox, Allergan, Irvine, CA) in the right hind leg. The injections were performed into the gastrocnemius lateralis, medialis and soleus muscles (dose of 1 U/muscle, in total 3 U/animal and a volume of 0.06 mL). The animals were thereafter allowed free cage activity. Botox effectiveness was confirmed prior to the surgery by visual inspection. Tendon transection. Four days later, both the partial unloaded and the fully loaded rats were anesthetized with isoflurane gas and the Achilles tendon, from the right hind leg was transected. To avoid postoperative infection, antibiotic was given once preoperatively (25 mg/ kg oxytetracycline). Analgesic was given subcutaneously pre- and postoperatively every 8-12h for 48h (0.045 mg/kg buprenorphine). Surgery was performed under aseptic conditions and the skin on the right lower leg was shaved and cleaned with chlorhexidine ethanol. The Achil- les and plantaris tendon complex was exposed through a transverse skin incision on the lateral side. The plantaris tendon was removed completely and the Achilles tendon was cut trans- versely in the middle part and was left unsutured whereas the skin was closed with 2 stitches. This procedure has been described elsewhere [3]. Mechanical testing. Eight days after surgery, the rats were anesthetized with isoflurane gas and killed with carbon dioxide. The right Achilles tendon was harvested together with the calcaneal bone and calf muscle. Sagittal and transverse diameter of the midpart of the callus tis- sue was measured with a slide calliper, and the transverse area was calculated by assuming an elliptical geometry. The distance between the old tendon stumps (gap distance) was measured, as seen through the partly transparent callus tissue. The muscle tissue was scraped off from the tendon, which was fixed in a metal clamp with sandpaper. The bone was fixed in a custom- made clamp at 30˚ dorsiflexion relative to the direction of traction in the materials-testing PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 7 / 9 PLOS ONE Microbiome, loading and tendon healing. machine (100R; DDL, Eden Prairie, MN). The machine pulled at 0.1 mm/s until failure. Peak force at failure (N) and energy uptake (N/mm), were calculated by the software of the machine. The investigator marked a linear portion of the elastic phase of the curve for automated stiff- ness (N/mm) calculation. Peak stress (MPa) and estimate of elastic modulus (MPa) were calcu- lated assuming an elliptical cylindrical shape and homogeneous mechanical properties. Estimation of elastic modulus was calculated as stiffness gap distance/transverse area [1, 3]. Characterization of gut bacterial flora. After the tendon samples were collected, SOPF rats (5 clean and 5 contaminated) from the full loading group were used to confirm changes in the microbial community. Faecal pellets were collected via laparotomy under aseptic condi- tions. The microbial community was identified by Surrey Diagnostics Ltd using the API (Bio- Merieux) and RapID (Remel) biochemical test. These data have previously been described as they were not collected solely for the present study [1]. Statistics. The mechanical results were analysed using SPSS software version 21. Only SOPF animals (clean and contaminated) were included. SPF rats were discarded. Our prede- fined primary outcome was the interaction between microbiome (contamination) and loading status for peak force, measured by a 2-way ANOVA. For post hoc, we compared groups pair- wise using Student’s t-test to describe the response to contamination among loaded and partial unloading rats separately. The presence of Staphyloccocus aureus and the tests of bacteriologi- cal status in the gut microbiota were only descriptive. Supporting information S1 Dataset. Data from mechanical evaluation and the status of the gut bacterial flora. Data are from each rat. (XLSX) Acknowledgments The authors thank Aneta Liszka for her assistance during the bacteria culture growth and Eme- lie Jansson for her help during the oral swabs collection. Dr. Per Aspenberg passed away before the submission of the final version of this manu- script. Franciele Dietrich-Zagonel accepts the responsibility for the integrity and validity of the data collected and analyzed. Author Contributions Conceptualization: Franciele Dietrich-Zagonel, Pernilla Eliasson, Per Aspenberg. Data curation: Franciele Dietrich-Zagonel, Malin Hammerman. Formal analysis: Franciele Dietrich-Zagonel, Malin Hammerman. Funding acquisition: Pernilla Eliasson, Per Aspenberg. Investigation: Franciele Dietrich-Zagonel, Malin Hammerman. Methodology: Franciele Dietrich-Zagonel, Per Aspenberg. Project administration: Franciele Dietrich-Zagonel. Supervision: Pernilla Eliasson, Per Aspenberg. Validation: Franciele Dietrich-Zagonel, Pernilla Eliasson, Per Aspenberg. Visualization: Franciele Dietrich-Zagonel, Per Aspenberg. PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 8 / 9 PLOS ONE Microbiome, loading and tendon healing. Writing – original draft: Franciele Dietrich-Zagonel, Per Aspenberg. Writing – review & editing: Franciele Dietrich-Zagonel, Malin Hammerman, Pernilla Elias- son, Per Aspenberg. References 1. Dietrich-Zagonel F, Hammerman M, Tatting L, Dietrich F, Kozak Ljunggren M, Blomgran P, et al. Stimu- lation of Tendon Healing With Delayed Dexamethasone Treatment Is Modified by the Microbiome. The American journal of sports medicine. 2018; 46(13):3281–7. https://doi.org/10.1177/0363546518799442 PMID: 30265844 2. Blomgran P, Hammerman M, Aspenberg P. Systemic corticosteroids improve tendon healing when given after the early inflammatory phase. Scientific reports. 2017; 7(1):12468. https://doi.org/10.1038/ s41598-017-12657-0 PMID: 28963482 3. Dietrich F, Hammerman M, Blomgran P, Tatting L, Bampi VF, Silva JB, et al. Effect of platelet-rich plasma on rat Achilles tendon healing is related to microbiota. Acta orthopaedica. 2017; 88(4):416–21. https://doi.org/10.1080/17453674.2017.1293447 PMID: 28296518 4. Linehan JL, Harrison OJ, Han SJ, Byrd AL, Vujkovic-Cvijin I, Villarino AV, et al. Non-classical Immunity Controls Microbiota Impact on Skin Immunity and Tissue Repair. Cell. 2018; 172(4):784–96 e18. 5. Furusawa Y, Obata Y, Fukuda S, Endo TA, Nakato G, Takahashi D, et al. Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells. Nature. 2013; 504(7480):446–50. https://doi.org/10.1038/nature12721 PMID: 24226770 6. Andersson T, Eliasson P, Hammerman M, Sandberg O, Aspenberg P. Low-level mechanical stimula- tion is sufficient to improve tendon healing in rats. J Appl Physiol (1985). 2012; 113(9):1398–402. 7. Hammerman M, Dietrich-Zagonel F, Blomgran P, Eliasson P, Aspenberg P. Different mechanisms acti- vated by mild versus strong loading in rat Achilles tendon healing. PloS one. 2018; 13(7):e0201211. https://doi.org/10.1371/journal.pone.0201211 PMID: 30044869 8. Hammerman M, Blomgran P, Dansac A, Eliasson P, Aspenberg P. Different gene response to mechani- cal loading during early and late phases of rat Achilles tendon healing. Journal of applied physiology. 2017; 123(4):800–15. https://doi.org/10.1152/japplphysiol.00323.2017 PMID: 28705996 9. Blomgran P, Blomgran R, Ernerudh J, Aspenberg P. A possible link between loading, inflammation and healing: Immune cell populations during tendon healing in the rat. Scientific reports. 2016; 6:29824. https://doi.org/10.1038/srep29824 PMID: 27405922 10. Eliasson P, Andersson T, Aspenberg P. Influence of a single loading episode on gene expression in healing rat Achilles tendons. Journal of applied physiology. 2012; 112(2):279–88. https://doi.org/10. 1152/japplphysiol.00858.2011 PMID: 21998267 11. Andersson T, Eliasson P, Aspenberg P. Tissue memory in healing tendons: short loading episodes stimulate healing. Journal of applied physiology. 2009; 107(2):417–21. https://doi.org/10.1152/ japplphysiol.00414.2009 PMID: 19541735 12. Eliasson P, Andersson T, Aspenberg P. Achilles tendon healing in rats is improved by intermittent mechanical loading during the inflammatory phase. Journal of orthopaedic research: official publication of the Orthopaedic Research Society. 2012; 30(2):274–9. 13. Eliasson P, Andersson T, Aspenberg P. Rat Achilles tendon healing: mechanical loading and gene expression. Journal of applied physiology. 2009; 107(2):399–407. https://doi.org/10.1152/japplphysiol. 91563.2008 PMID: 19541731 14. Freedman BR, Gordon JA, Bhatt PR, Pardes AM, Thomas SJ, Sarver JJ, et al. Nonsurgical treatment and early return to activity leads to improved Achilles tendon fatigue mechanics and functional out- comes during early healing in an animal model. Journal of orthopaedic research: official publication of the Orthopaedic Research Society. 2016; 34(12):2172–80. 15. Schlundt C, Schell H, Goodman SB, Vunjak-Novakovic G, Duda GN, Schmidt-Bleek K. Immune modu- lation as a therapeutic strategy in bone regeneration. Journal of experimental orthopaedics. 2015; 2 (1):1. https://doi.org/10.1186/s40634-014-0017-6 PMID: 26914869 16. Beura LK, Hamilton SE, Bi K, Schenkel JM, Odumade OA, Casey KA, et al. Normalizing the environ- ment recapitulates adult human immune traits in laboratory mice. Nature. 2016; 532(7600):512–6. https://doi.org/10.1038/nature17655 PMID: 27096360 PLOS ONE | https://doi.org/10.1371/journal.pone.0229908 March 10, 2020 9 / 9

Journal

PLoS ONEPublic Library of Science (PLoS) Journal

Published: Mar 10, 2020

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create folders to
organize your research

Export folders, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off