Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model

Abstract

Biofilm-colonized chronic wounds are difficult to treat due to a constantly evolving microbiome. In this study, a cHG augmented with antibiotics was examined for the topical treatment of biofilm-challenged wounds in vivo. Two studies were performed in series using a murine stented wound model. Mice were divided into four groups: control (wound only), infection only (IO), infection + cHG (IcHG), and infection + cHG + antibiotics (IcHG + Abx). We first examined Pseudomonas aeruginosa biofilms treated with gentamicin, and then MRSA biofilms treated with clindamycin. Wound healing was assessed using photography, immunohistochemistry, and histology. Systemic symptoms were monitored with hematological laboratory tests. Pseudomonas aeruginosa infected wounds treated with cHG + Abx healed faster and were protected from bacteremia. In the MRSA infected mice, wound treatment significantly affected the outcome, explaining 5.56% of total variance (ANOVA: F(3, 366) = 17.38, p < 0.0001). Additionally, infected wounds treated with cHG + Abx demonstrated less inflammatory tissue and accelerated closure rate on day 8 (76.53% ± 7.43% vs 48.40% ± 4.95%, p < 0.0001) and day 14 (96.00% ± 3.07% vs 82.38% ± 8.24%, p = 0.003), as compared to the infection only wounds. cHG offers a biocompatible, topical option with dual functionality: antibiotic augmentation to target biofilm pathogens, and a collagen-rich dressing to accelerate wound healing.

Introduction

The management and treatment of wounds are integral components of medical practice. In 2018, it was estimated that 8.2 million Medicare beneficiaries had wounds, with an estimated associated cost ranging from $28.1 to $96.8 billion. ¹ Today, healthcare professionals face the additional challenges of an aging population and escalating rates of comorbid diabetes, peripheral vascular disease, and obesity. ² The financial impact, in concert with the evolving healthcare demographic, emphasizes the pressing need to advance wound care strategies.

Wounds are commonly categorized as acute or chronic, with chronic wounds persisting for more than a month. ³ The treatment of chronic wounds is complicated by the colonization of biofilm-forming bacteria.^4,5 Biofilms, believed to be present in over 90% of chronic wounds, are a leading cause of amputation and morbidity.^5–7 Bacterial biofilms act offensively by hindering the host immune response, restricting surface circulation, and impeding reepithelialization. ⁸ Simultaneously, biofilms play defense by enveloping themselves in a dense extracellular polymeric substance (EPS), forming a protective shield against mechanical stimuli and antimicrobial therapy.^6,7,9–11

The standard treatment for infected chronic wounds often involves multiple courses of systemic antibiotics at high doses leading to increased adverse drug effects.^{4–7,9,12,13} During antibiotic treatment, biofilm-embedded bacteria lay dormant, allowing for increased expression and sharing of antibiotic resistance genes. ¹⁰ There are an estimated 2.8 million new antibiotic-resistant infections every year, and in 2017, an estimated $2.7 billion of healthcare expenses were attributed to the treatment of community-acquired drug resistant infections with $1.1 billion of that associated with MRSA.^14,15

Local delivery of antimicrobial agents via biomaterial carriers has emerged as a promising alternative to systemic antibiotic therapy for treatment of biofilm-colonized chronic wounds. Unlike systemic administration, local delivery enables sustained, high-concentration antibiotic release directly at the wound site, circumventing the pharmacokinetic limitations imposed by poor tissue perfusion and the physical barrier of the biofilm EPS.^6,16 This approach also reduces systemic drug exposure, minimizing adverse effects and the selective pressure that drives resistance gene expression in non-target bacterial populations. ¹⁷ Among the pathogens most commonly implicated in chronic wound biofilms, Pseudomonas aeruginosa and methicillin-resistant Staphylococcus aureus represent particularly high-priority targets due to their prevalence in both community and hospital settings, their well-characterized ability to form robust biofilms, and their association with delayed healing and increased amputation risk in diabetic foot ulcers.^18,19

A substantial portion of this cost can be attributed to wound treatment products. One promising product is the collagen hydrogel (cHG).^20–23 This adaptable polymer is capable of simultaneously promoting granulation tissue formation and neovascularization, while maintaining an appropriate level of moisture resulting in accelerated wound closure.^24–26

Previous research has shown that cHGs can carry a wide spectrum of compounds with modifiable extended drug-release profiles.^16,17,27,28 Our lab previously demonstrated that collagen-rich hydrogel (cHG) is a potential vector to improve the healing of these challenging wounds. We showed that antibiotic-loaded cHGs can disrupt biofilms in vitro. ²⁹ Building on these findings, we conducted in vivo experiments to assess the translational efficacy of antibiotic-eluting, human-derived cHG as local therapy for wounds infected with two biofilm-forming bacterial species. We used a two-step approach to increase validity while ensuring mouse safety. First, we studied wounds infected with Pseudomonas aeruginosa. This study served to test the reproducibility surgical technique, confirm that infection persisted in wounds, determine time to healing, and ensure animal safety/survival. Next, we repeated the same experiment with the more virulent and prevalent wound bacteria, MRSA, treated with clindamycin. In both arms of the study mice were divided into four categories: control, infection only (IO), infection and cHG (IcHG), or infection and cHG loaded with one of the bacteria matched antibiotics (IcHG + Abx).

We hypothesized that cHG + Abx is an effective and safe option for topical treatment, as evidenced by reduced bacterial burden, accelerated wound healing, decreased tissue inflammation, and limited systemic infection/inflammation.

Results

Pseudomonas aeruginosa cohort

Assessment of healing

A single mouse from each group was photographed over time to assess for visual changes in wound appearance (Supplemental Figure 2). Wounds treated with cHG and cHG + Abx showed less yellow exudate and re-established epithelium, particularly on days 10 and 12.

Assessment of infection

Figure 1 highlights the biofilm architecture on the surface of an IO cohort mouse on day 12 using P. aeruginosa polyclonal antibody. Figure 2 demonstrates less fluorescent intensity and depth of bacterial invasion in infected mice treated with cHG + Abx than in untreated mice.

Figure 1.

P. aeruginosa antibody labeling (RED) of an infection only mouse at post-operative day 12 against DAPI (BLUE; n = 1).

Figure 2.

P. aeruginosa antibody labeling (RED). Top row: infection only (IO) and bottom row: infection + cHG + Abx (IcHC + Abx). Left column mice harvested at day 12 and right column day 14 (n = 4).

Hematologic evaluation

Blood cultures revealed that two of the four samples from the IO group grew P. aeruginosa on days 7 and 10. In contrast, none of the antibiotic-treated mice grew P. aeruginosa. Additionally, markers of infection (white blood cell count and absolute neutrophil count) were collected on days 7, 14, and 17. Inflammatory markers were lower in the IcHG + Abx group than in the IO group at all time points (Figure 3).

Figure 3.

Lab results processed by Stanford Animal Laboratory for individual mice from both the infection + cHG + Abx (IcHG + Abx) and Infection only treatment groups at days of harvest 7, 14, and 17: (a) WBC (K/µL) with dotted line representing normal value range and (b) absolute neutrophil count with dotted lines representing normal value range (n = 6).

MRSA cohort

Assessment of healing rate

Single mice from each group were photographed throughout the experiment to assess for visual changes in the wound appearance (Supplemental Figure 3). Wounds treated with cHG and cHG + Abx showed substantially less yellow exudate and re-established epithelium at the same time points as in the Pseudomonas aeruginosa cohort (days 10 and 12).

Figure 4 summarizes the wound closure rate for all treatment groups, as measured by H&E staining. Wound closure rate was significantly increased in the IcHG + Abx group as compared to the IO group at 8 days (76.53% ± 7.43% vs 48.40% ± 4.95%, p < 0.0001, d = 3.57) and 14 days (96.00% ± 3.07% vs 82.38% ± 8.24%, p = 0.003, d = 1.75). IcHG treated wounds showed significantly improved closure compared to IO on day 8 only (73.59% ± 10.64% vs 48.40% ± 4.95%, p < 0.0001, d = 2.43). Only the control and IcHG + Abx groups achieved complete closure on histology by day 17.

Figure 4.

MRSA cohort: average wound closure presented as a percentage of total wound length − unhealed wound length for all groups as measured from H&E staining at 4×. Two tailed t-test significance (**p < 0.01) is indicated between IcHG + Abx and infection only mice (n = 406).

Figure 5 demonstrates skin wound scoring for all treatment groups at each timepoint. Using a two-way ANOVA without an interaction term, SPOT skin wound score differed by wound treatment (F(3, 366) = 17.38, p < 0.001; η ²  = 0.056) and by time (F(3, 366) = 149.38, p < 0.001; η² = 0.478). Tukey-adjusted pairwise testing showed IcHG + ABX and IcHG SPOT skin wound scores were higher than IO and control (all p ⩽ 0.0004), whereas IcHG + ABX did not differ from IcHG (p = 0.857) and IO did not differ from control (p = 0.692).

Figure 5.

MRSA cohort: average wound score for all groups as measured from H&E staining at 4×. ANOVA by treatment (F(3, 366) = 17.38, p < 0.001; η² = 0.056) and by time (F(3, 366) = 149.38, p < 0.001; η² = 0.478; n = 406).

Assessment of wound inflammation

Qualitative analysis of H&E-stained sections from wounds on days 8 and 14 showed observable differences in skin architecture (Figure 6). The IO group had more inflammatory tissue, as indicated by the asterisks, than both IcHG and IcHG + Abx groups. Additionally, IcHG + Abx wounds on day 14 showed the most robust return of adnexal structures.

Figure 6.

H&E staining of wound area sections from days 8 and 14 for mice in the control, IO, IcHG, and IcHG + Abx cohorts. The asterisk (*) indicates inflammatory tissue in the wound bed (n = 8).

Figure 7 summarizes quantitative analysis of inflammatory tissue thickness. A decreased thickness was observed on day 8 when comparing both IcHG + Abx to IO (192.78 ± 47.00 vs 401.27 ± 53.75 μm, p < 0.0001, d = 3.30) and IcHG to IO (148.36 ± 49.88 vs 221.10 ± 54.53 μm, p < 0.001, d = 3.90). Additionally, IO mice had more inflammatory tissue on day 8 compared to control mice (401.27 ± 53.75 vs 221.10 ± 54.53 μm, p = 0.001, d = 2.66). Only IcHg + Abx maintained a significant decrease in inflammatory tissue at 14 days as compared to IO (69.38 ± 46.56 vs 124.56 ± 56.19 μm, p < 0.0001, d = 0.86).

Figure 7.

MRSA cohort: average granulation tissue thickness for all groups measured from H&E staining at 4× on days 8 and 14 (n = 487). Two sample t-test.

Hematologic evaluation

Diagnostic laboratories showed normal renal and liver function as measured by blood urea nitrogen (BUN), albumin and platelets (PLT) for all IcHG + ABX mice (Supplemental Table 3).

Discussion

Chronic wounds negatively affect patients’ quality of life and impose financial strain on the U.S. healthcare system. The propensity of these wounds to become colonized and seed biofilms makes their treatment complex and is associated with substantial time and cost. Human-derived collagen-rich hydrogel, as an excellent carrier for antimicrobial agents, is a promising advancement in the treatment of infected, biofilm-challenged wounds.

Previous research has studied the additive effects of cHGs loaded with metal nanoparticles, plant extracts, growth factors, and microbes in treating clean wounds.^25,30–32 Our study builds on recent work demonstrating the healing potential of antibiotic-loaded gels for treating infected wounds, but is novel in the utilization of the stented wound to model chronic wounds. ³³ Histology revealed that the application of antibiotic-loaded hydrogel effectively reduced inflammatory tissue on days 8 and 14, compared to infection only wounds. This suggests that the treatment facilitated the progression of wounds from the inflammatory stage to the proliferative stage, thereby accelerating the overall wound healing process. Additionally, 14 days after study initiation, IcHG + Abx wounds had less inflammatory tissue than the control, suggesting that the intrinsic healing properties of cHG help decrease inflammation that occurs due to the mechanical trauma of wounding even in the absence of infection.

A substantial challenge in treating chronically infected wounds is the presence of biofilms, which defend against the effective penetration of topical and systemic antibiotics. Our study revealed that antibiotics incorporated into cHG exhibit improved biofilm penetration, as evident from the reduction of fluorescently stained bacteria in treated versus untreated wounds. Additionally, we demonstrated the absence of bacteremia in treated animals compared to the presence of positive blood cultures in infection only animals. These findings demonstrate the effectiveness of utilizing cHG to treat bacterial infections locally and prevent hematologic spread of bacteria. Infected wounds treated with cHG + Abx surprisingly did not heal faster or have less inflammatory tissue as compared to those treated with cHG alone. It is possible that in these otherwise healthy mice they can resolve cutaneous infections in the early days after infection and wounding and, therefore, the addition of topical antibiotics does not significantly improve the overall rate of healing. Future studies should be done with diabetic mice to assess whether the addition of antibiotics in these compromised mice significantly improves their healing.

Collagen-based hydrogels have been reported to exhibit a high degree of biocompatibility. ²⁰ While collagen alone is inert, one potential concern regarding the augmentation of hydrogels is that the addition of bioactive compounds may result in irritation and activation of the host immune system. ²⁵ No local tissue damage was observed on gross observation or histology of the surrounding unwounded tissue. Systemically, inflammatory marker levels were reduced in all cHG + Abx treated mice at all time points. In addition, the clindamycin treated animals in the MRSA cohort showed normal BUN and albumin/PLT levels. These results indicate preserved renal and liver function, suggesting antibiotic safety despite their topical application at high concentrations. This is in contrast with many local antimicrobial therapies, which can have harmful effects on healthy tissue or synthetic preparations of hydrogels with less biocompatibility.^20–23

The elution profile established during in vitro experiments indicated that an application reaches peak therapeutic levels after 24 h, with an additional 24 h of effective dose release, after which cHG + Abx application requires replacement. ²⁹ We anticipate that this treatment schedule will have a considerable clinical impact on both patients and providers. Patient comfort would increase, and provider workload would decrease due to the reduction in the frequency and overall number of patient encounters necessary to achieve wound closure.

Prior studies have established the compatibility of cHG with various classes and concentrations of antibiotics and antimicrobial agents.^25,34,35 This adaptability of cHG could lead to individualized preparations based on the culture results, antimicrobial sensitivity profiles, and patient allergies leading to enhanced patient safety and improved antibiotic stewardship.

Our study was limited in its approach of using only two antibiotics and two bacterial strains. However, in vitro, we have demonstrated the ability of cHG to be loaded with additional antibiotics as well as multiple antibiotics simultaneously and to treat polymicrobial biofilms. ³⁶ Two additional control groups were considered but not included. A systemic antibiotic control group was excluded because they are known to be an effective treatment in immunocompetent mouse models. ³⁷ Additionally, the target patient population for the cHG + Abx is immunocompromised individuals with poor wound vascularity, where systemic antibiotics have limited if any efficacy.^38,39 A topical liquid antibiotic group was also excluded, as liquid antibiotics without a carrier were unable to remain confined within the wound boundaries (Supplemental Video 1). As further described in the section “Methods,” the stepwise four-group design was constructed to isolate the independent contributions of infection, hydrogel, and antibiotic delivery. The IcHG group serves as the critical comparator that distinguishes the hydrogel’s intrinsic healing properties from the additive effects of antibiotic therapy. As a result, our conclusion is somewhat limited, and further investigation is needed to confidently attribute wound healing improvement to the individual treatment components versus their potentially compounding effects.

The assessment of bacterial burden was limited to qualitative data, as fluorescent bacteria were not used across enough wounds to enable quantitative evaluation of CFUs. While our data indicated that treated wounds healed more quickly, likely due to early suppression of infection, this claim is constrained by the data limitations. Accordingly, conclusions regarding bacterial clearance are based on qualitative immunofluorescence and hematologic data and further studies should incorporate quantitative CFU assessment to more precisely characterize the antimicrobial efficacy of cHG + Abx.

Finally, it is important to note the inherent limitations of the mouse model in terms of applicability to human chronic wounds. The mice used were healthy wild-type C57BL/6 animals, whereas most patients with chronic wounds present with comorbidities including diabetes mellitus, peripheral vascular disease, venous insufficiency, obesity, and poor nutrition that fundamentally alter the wound microenvironment. In diabetic wounds specifically, impaired neutrophil bactericidal function, dysregulated macrophage polarization toward a persistent pro-inflammatory M1 phenotype, and reduced growth factor signaling collectively blunt the innate immune response and prolong the inflammatory phase of healing, creating conditions far more permissive to biofilm persistence than those present in immunocompetent mice.^40,41 Additionally, the vasculopathy associated with diabetes and peripheral arterial disease reduces local tissue perfusion, limiting antibiotic delivery and making bacterial clearance substantially more difficult.^38,39 The stented wound model used here approximates the mechanical characteristics of human chronic pounds by preventing wound contraction, but does not replicate the biochemical and immunological milieu of a diabetic or ischemic wound. As a result, the accelerated healing observed in this study may overestimate the clinical benefit of cHG + Abx in the target patient population.

Future studies will focus on extending the duration of effective elution and examining additional antimicrobial agents beyond antibiotics, and utilizing an immunocompromised murine model. The utilization of cHG + Abx as a local, effective therapy holds considerable potential value for patients and the healthcare system at large.

Methods

CHG preparation

Two percent human-derived, cHG was synthesized according to a previously established protocol. ¹⁶ Flexor tendons were harvested from fresh frozen cadavers, decellularized to eliminate DNA contamination, lyophilized, and ground into fine powder. ⁴² Collagen powder (20 mg/mL) was digested with 1 mg/mL pepsin at pH 2.2. After 14 h of digestion, the pH was adjusted to 7.4 to halt the reaction. The quality of the gel was confirmed microscopically and gelation was confirmed by incubation at 37 °C for 40 min.

To synthesize cHG + Abx, the neutralized collagen hydrogel solution was mixed with a stock antibiotic (Thermo Fisher Scientific, Waltham, MA, USA) to a 100× minimum inhibitory concentration of the target bacteria (500 μg/mL gentamicin or 10 μg/mL clindamycin). The cHG + Abx mixture (100 μL) was directly applied to the wounds in liquid form, and gelation was confirmed by gross inspection of the wound 1 h after application.

Bacterial strain and growth conditions

Pseudomonas aeruginosa (ATCC 27853) and methicillin-resistant Staphylococcus aureus (MRSA252 ATCC BAA-1720; ATCC, Manassas, VA, USA) were grown in Luria–Bertani (LB) broth (Thermo Fisher Scientific) and Tryptic Soy Broth (Remel, San Diego, CA, USA), respectively, at 37 °C under aerobic conditions. These bacterial strains were chosen due to their prevalence in hospital settings and wounds, and the high likelihood of colonization to infection transition. ⁴³

Biofilms were prepared according to the protocol developed by the Center for Biofilm Engineering at Montana State University (Supplemental Figure 1). ⁴⁴ Bacteria selected from a single colony were grown to ~10⁸ CFU/mL and diluted 1:100 in broth. Polycarbonate membrane filters (0.2 um pore size; Thermo Fisher Scientific) were UV-sterilized before 2 µL of bacteria was placed on each filter. The filter was plated onto Mueller Hinton agar (Thermo Fisher Scientific) and cultured at 37 °C for 24 h under aerobic conditions.

Two cohort study design

All mice received the same standardized dorsal wound at the start of the experiment. A control group that received neither bacterial inoculation nor topical therapy was included to establish the baseline rate of wound healing in this mouse population. Three additional experimental groups were used. The infection only (IO) group received wound inoculation with one of two bacteria pathogens. The infection + collagen hydrogel (IcHG) group received bacterial inoculation followed by treatment with collagen hydrogel. The infection + antibiotic-loaded collagen hydrogel (IcHG + Abx) group received bacterial inoculation followed by treatment with collagen hydrogel loaded with the pathogen matched antibiotic.

This design was selected to isolate the independent and combined effects of infection, hydrogel treatment, and antibiotic therapy. Comparison of the control and IO groups allows assessment of the effects of infection on wound healing. Comparison of the IO and IcHG groups evaluates the effects of the hydrogel alone, while comparison of the IcHG and IcHG + Abx groups assess the additional contribution of antibiotic delivery within the hydrogel system.

We used a two-step approach to increase validity while ensuring animal safety. First, we studied wounds infected with Pseudomonas aeruginosa treated with cHG and gentamicin. We selected this organism as it is known to be an organism responsible for severe damage when present in diabetic wounds. ¹⁸ Zhao et al. described delayed wound healing in diabetic mice challenged with Pseudomonas aeruginosa as a model for studying chronic wounds. ⁴⁴ A modified version of this well-established, stented wound model using 16-week-old male wild-type C57BL/6 mice (Jackson Laboratory, Bar Harbor, ME, USA) was used. This study served to test the reproducibility of the surgical technique, confirm that infection persisted in wounds, and ensure animal safety/survival. Next we repeated the same experiment with a more virulent and prevalent wound bacteria, MRSA, treated with cHG and clindamycin. We selected MRSA due to its known high prevalence and ability to produce biofilms in diabetic ulcers, significantly contributing to healing inhibition. ¹⁹

Pseudomonas aeruginosa + gentamicin cohort

Surgical technique

Sixty-eight mice were divided into four groups: control (n = 13), infection only (IO; n = 20), infection + cHG (IcHG; n = 20), and infection + cHG + antibiotic (IcHG + Abx; n = 15; Supplemental Table 1).

Under general anesthesia (subcutaneous injection of 0.05 mg/kg buprenorphine for pain management, followed by inhaled 4% isoflurane), a 6 mm in diameter, full-thickness skin wound was created on the dorsum of each mouse after removing local hair. Wounds were stented open with an 8 mm diameter silicone ring (Thermo Fisher Scientific) anchored using six sutures and surgical adhesive around the wound circumference to limit premature wound closure. The wound was photographed and dressed with Tegaderm (3M, St. Paul, MN, USA) dressing. The animals were single-housed and allowed to recover without post-operative activity restrictions. Pain management was monitored hourly for the next 8 h with buprenorphine 0.05 mg/kg SC as needed.

The operative and treatment schedule is summarized in Figure 8. On post-operative day 2 (POD 2), the control group did not receive any treatment or inoculation. The Tegaderm was removed and the wound was photographed before replacing Tegaderm without irrigation. The IO, IcHG, and IcHG + Abx groups underwent wound biofilm inoculation. Biofilms were applied to the wound for 2 min before removal of the membrane. The wound was photographed and the Tegaderm was replaced.

Figure 8.

Surgical protocol for Pseudomonas cohort (created with BioRender.com): (a) photography of mouse after wound creation and stenting and (b) experiment schedule from wound creation to sacrifice.

On POD 4, the control and IO groups received the same treatment as on POD 2. The IcHG and IcHG + Abx groups received 100 µL of cHG or cHG + Abx treatment pipetted directly onto the wound. The wounds were imaged and Tegaderms were replaced. Every 2 days, until sacrifice, the procedures for the respective groups were identical to the POD 4 protocols.

Harvest days were 7, 10, 12, 14, and 17. All animals underwent a final round of photography before sacrifice and harvesting. Blood was collected by cardiac puncture and wound specimens were collected using a 2 mm margin around the original wound. Specimens were bisected with half embedded in optimal cutting temperature (OCT) compound (Thermo Fisher Scientific) for histology, and the other half embedded in paraffin.

Hematologic lab studies

Blood was collected via cardiac puncture and submitted to Stanford University Animal Diagnostic Laboratory. Cultures were grown from blood samples to assess for systemic bacterial infections.

Histology

Due to wound exudate and hydrogel residue, determination of wound closure via photography was limited. To overcome this, H&E-stained sections were used to quantify the rate of wound closure. The images were randomized and distributed to two independent reviewers for analysis. Wound length was determined by the distance at which the basement membrane was not intact, and the follicles had not returned. Percent healed was calculated by dividing the number of healed sections by the total number of sections in each treatment group at each time point. Histological analysis was performed using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Immunohistochemistry

Bacterial imaging was performed using a rabbit-derived whole-cell P. aeruginosa polyclonal antibody (1:800; Thermo Fisher Scientific, Waltham, MA, USA). Harvested skin samples embedded in (OCT) were treated with proteinase K for antigen retrieval (Sigma–Aldrich) for 10 min. Samples were incubated in 2% (v/v) normal donkey serum in 0.2% (v/v) Triton X-100 in PBS at room temperature. This was followed by incubation with the primary antibody at 4 °C for 24 h. Rabbit anti-Rat IgG Secondary Antibody, Texas Red conjugate, was applied for 1 h (Thermo Fisher Scientific). Finally, 4′,6-diamidino-2-phenylindole (DAPI; Vector Laboratories, Burlingame, CA, USA) was applied prior to imaging with a Keyence fluorescence microscope.

Methicillin-resistant Staphylococcus aureus (MRSA) + clindamycin cohort

Surgical technique

The same stented wound model described above was repeated for 96 mice, divided into the same four groups: control (n = 24), IO (n = 24), IcHG (n = 24), and IcHG + Abx (n = 24; Supplemental Table 2). Infected mice were inoculated with a MRSA biofilm, and the antibiotic-impregnated gels were loaded with clindamycin. The same surgical and treatment schedule shown in Figure 8 was replicated, with the exception of changes in harvest days (5, 8, 14, 17). Wounds were photographed until mice harvest days, at which point blood and tissue was obtained as described above.

Histology

An identical technique for section creation and imaging was performed for the harvested MRSA-infected wounds. Triplicate sections were created for all the wounds. Wound closure rate was calculated as the difference between measured wound length and initial wound length (6 mm), divided by initial wound length, and presented as a percentage. Additionally, local inflammation was determined by measuring the thickness of inflammatory tissue from the apical wound surface to the adipose layer at each edge of the wound. Wound score was completed for all H&E slides using the previously validated SPOT skin wound score system. ⁴⁵

Statistical analysis

The two scores of the triplicate sections for each mouse wound were averaged prior to analysis to avoid pseudoreplication and allow for statistical analysis at the individual animal level.

Statistical analyses comparing wound closure and inflammation tissue of all groups were performed using two-tailed t-tests as well as the non-parametric Mann–Whitney U test as previously described. ⁴⁶ Initially, the Jarque–Bera test was used to determine data normality. If normally distributed, the F-test was used to assess for equal variance, and the appropriate unpaired two-tailed t-test was then performed. If the sample data was non-normally distributed, the Mann–Whitney U test was used to assess for differences between the groups. All reported intervals and error bars represent 95% confidence intervals. Animals were sacrificed at each designated harvest timepoint, such that each animal contributed data at a single timepoint only. Accordingly, the study design is cross-sectional at each harvest day rather than longitudinal, and a repeated-measures or mixed-effects model was not appropriate for the pairwise wound closure and inflammatory tissue analyses. For the above outcomes, effect sizes were calculated as Cohen’s d to supplement p values and facilitate interpretation of biological significance. Effect sizes should be interpreted with caution given the modest sample size per timepoint. Analysis was done using IBM SPSS Statistics for Windows, version 29.0.1.0 (IBM Corp., Armonk, NY, USA).

We analyzed averaged (grouped) wound score data using a two-way ANOVA with fixed main effects for wound treatment and time (additive model; no treatment × time interaction term). Post hoc pairwise comparisons among treatment groups were performed using Tukey’s multiple-comparisons procedure on least-squares means, controlling the family-wise error rate at α = 0.05. All tests were two-sided. Analysis was done using GraphPad Prism version 10.0.0 for Windows (GraphPad Software, Boston, MA, USA, www.graphpad.com).

Conclusion

Human-derived collagen hydrogel is a promising carrier of antibiotics for the topical treatment of biofilm-colonized wounds. We demonstrated that human-derived collagen hydrogel disrupted Pseudomonas aeruginosa and methicillin-resistant Staphylococcus aureus biofilms in vivo, accelerated wound closure, and reduced inflammatory tissue compared to untreated infected wounds. Specifically, wounds treated with these gels progressed through the stages of wound healing significantly sooner than untreated wounds and demonstrated significantly higher SPOT skin wound scores across all timepoints, with larger effect sizes confirming the biological significance of these differences.

cHG offers a biocompatible option with dual functionality; antibiotic augmentation to reduce biofilm burden and disrupt biofilm architecture in infected wounds, and collagen-rich scaffold that intrinsically accelerates the transition from the inflammatory to the proliferative phase of healing. The biocompatibility of this human-derived construct, evidenced by preserved renal and hepatic function and reduced systemic inflammatory markers in treated animals, distinguished it from many synthetic antimicrobial alternatives and strengthens its transitional potential.

From a clinical perspective, this treatment platform holds particular promise for patients with chronic wounds refractory to standard systemic antibiotic therapy, including those with diabetes, vascular disease, or other comorbidities that impair wound vascularity and systemic drug delivery. The adaptability of cHG to carry a broad spectrum of antibiotics and antimicrobial agents enables a personalized approach. Specifically, a patient’s wounds could be cultured, sensitivities determined, and gels engineered with pathogen-matched antibiotics targeting the individual wound microbiome. This strategy would reduce unnecessary antibiotic exposure, improve stewardship and potentially lead to fewer dressing changes, wound care clinic visits, surgical procedures, and overall healthcare expenditure.

Future studies should prioritize validation of this approach in immunocompromised murine models, including the db/db diabetic mouse and streptozotocin-induced models, where impaired bacterial clearance and dysregulated healing more closely approximate the human chronic wound environment. Additional directions include extending the duration of effective antibiotic elution, evaluating combination antibiotic loading for polymicrobial infections, and ultimately progressing toward human feasibility studies.

Supplemental Material

sj-docx-1-jbf-10.1177_22808000261447657 – Supplemental material for Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model

Supplemental material, sj-docx-1-jbf-10.1177_22808000261447657 for Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model by Reid W. Smith, Evan H. Jarman, Shannon Francis, Jordan B. Burgess, Kate Hayashiganati, Ayushi Sharma, Uriel Sanchez Rangle, Amar Singh, Allen Green and Paige M. Fox in Journal of Applied Biomaterials & Functional Materials

Supplemental Material

sj-docx-2-jbf-10.1177_22808000261447657 – Supplemental material for Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model

Supplemental material, sj-docx-2-jbf-10.1177_22808000261447657 for Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model by Reid W. Smith, Evan H. Jarman, Shannon Francis, Jordan B. Burgess, Kate Hayashiganati, Ayushi Sharma, Uriel Sanchez Rangle, Amar Singh, Allen Green and Paige M. Fox in Journal of Applied Biomaterials & Functional Materials

Supplemental Material

sj-docx-3-jbf-10.1177_22808000261447657 – Supplemental material for Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model

Supplemental material, sj-docx-3-jbf-10.1177_22808000261447657 for Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model by Reid W. Smith, Evan H. Jarman, Shannon Francis, Jordan B. Burgess, Kate Hayashiganati, Ayushi Sharma, Uriel Sanchez Rangle, Amar Singh, Allen Green and Paige M. Fox in Journal of Applied Biomaterials & Functional Materials

Supplemental Material

sj-jpeg-4-jbf-10.1177_22808000261447657 – Supplemental material for Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model

Supplemental material, sj-jpeg-4-jbf-10.1177_22808000261447657 for Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model by Reid W. Smith, Evan H. Jarman, Shannon Francis, Jordan B. Burgess, Kate Hayashiganati, Ayushi Sharma, Uriel Sanchez Rangle, Amar Singh, Allen Green and Paige M. Fox in Journal of Applied Biomaterials & Functional Materials

Supplemental Material

sj-jpeg-5-jbf-10.1177_22808000261447657 – Supplemental material for Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model

Supplemental material, sj-jpeg-5-jbf-10.1177_22808000261447657 for Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model by Reid W. Smith, Evan H. Jarman, Shannon Francis, Jordan B. Burgess, Kate Hayashiganati, Ayushi Sharma, Uriel Sanchez Rangle, Amar Singh, Allen Green and Paige M. Fox in Journal of Applied Biomaterials & Functional Materials

Supplemental Material

sj-jpeg-6-jbf-10.1177_22808000261447657 – Supplemental material for Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model

Supplemental material, sj-jpeg-6-jbf-10.1177_22808000261447657 for Antibiotic eluting collagen-based hydrogel improves wound healing in a biofilm challenged murine stented wound model by Reid W. Smith, Evan H. Jarman, Shannon Francis, Jordan B. Burgess, Kate Hayashiganati, Ayushi Sharma, Uriel Sanchez Rangle, Amar Singh, Allen Green and Paige M. Fox in Journal of Applied Biomaterials & Functional Materials