Research ArticleWound Healing

A dual-action peptide-containing hydrogel targets wound infection and inflammation

See allHide authors and affiliations

Science Translational Medicine  01 Jan 2020:
Vol. 12, Issue 524, eaax6601
DOI: 10.1126/scitranslmed.aax6601

Double defense

Wound infections can increase risk of systemic complications, such as sepsis, and can be difficult to treat due to growing antimicrobial resistance. Puthia et al. developed a hydrogel containing a thrombin-derived peptide, TCP-25, that kills bacteria and reduces inflammation. The hydrogel was effective against Staphylococcus aureus, Pseudomonas aeruginosa, and clinical bacterial isolates in vitro, and treated murine models of subcutaneous infections and porcine partial thickness wound infections. Bioactive cleavage fragments of the peptide were similar to those found in human wound fluid. Results suggest that this dual-action anti-inflammatory and antibacterial peptide-functionalized hydrogel is a promising approach for wound healing.

Abstract

There is a clinical need for improved wound treatments that prevent both infection and excessive inflammation. TCP-25, a thrombin-derived peptide, is antibacterial and scavenges pathogen-associated molecular patterns (PAMPs), such as lipopolysaccharide, thereby preventing CD14 interaction and Toll-like receptor dimerization, leading to reduced downstream immune activation. Here, we describe the development of a hydrogel formulation that was functionalized with TCP-25 to target bacteria and associated PAMP-induced inflammation. In vitro studies determined the polymer prerequisites for such TCP-25–mediated dual action, favoring the use of noncharged hydrophilic hydrogels, which enabled peptide conformational changes and LPS binding. The TCP-25–functionalized hydrogels killed Gram-positive Staphylococcus aureus and Gram-negative Pseudomonas aeruginosa bacteria in vitro, as well as in experimental mouse models of subcutaneous infection. The TCP-25 hydrogel also mediated reduction of LPS-induced local inflammatory responses, as demonstrated by analysis of local cytokine production and in vivo bioimaging using nuclear factor κB (NF-κB) reporter mice. In porcine partial thickness wound models, TCP-25 prevented infection with S. aureus and reduced concentrations of proinflammatory cytokines. Proteolytic fragmentation of TCP-25 in vitro yielded a series of bioactive TCP fragments that were identical or similar to those present in wounds in vivo. Together, the results demonstrate the therapeutic potential of TCP-25 hydrogel, a wound treatment based on the body’s peptide defense, for prevention of both bacterial infection and the accompanying inflammation.

INTRODUCTION

Wounds of various types have an immense impact on patients, health care, and society (1). Types of wounds include acute postsurgical wounds, burns, and nonhealing ulcers resulting from diabetes or circulatory disturbances (1). With a point prevalence of around 2 per 1000 (2), costs for chronic wounds are substantial and account for 1 to 3% of the total health system costs in developed countries (1). Considering burns, 67 million injuries were reported in 2015, resulting in about 2.9 million hospitalizations and 176,000 deaths (3, 4). In a study published in 2014, the mean total cost for burn care in high-income countries was estimated to around 88,000 USD per patient (5).

From a physiological perspective, wound healing is an evolutionarily conserved sequence of biologically interlinked events. An initial phase of hemostasis is followed by phases of inflammation, proliferation, and tissue remodeling (6, 7). Initial surveillance mediated by human innate immunity is instrumental in the control of bacteria during wounding, and lipopolysaccharide (LPS) sensing by Toll-like receptors (TLRs) is crucial in early responses to infection (8). However, an excessive TLR response causes localized and sometimes disproportionate inflammation, as observed in postoperative infections, infected burn wounds, or nonhealing ulcers. These wound complications delay proper healing, increasing the risk of severe infections and potentially leading to scar formation (1, 911). Prophylactic use of systemic antibiotics can reduce the incidence of wound and surgical infections (12). However, this use of antibiotics drives the development of resistance (13) and infections caused by antibiotic-resistant strains of Staphylococcus aureus and Pseudomonas aeruginosa, which are bacteria that cause postoperative infections and infections in chronic wounds and burns, present a major challenge (1). For example, in European hospitals, the overall rates of surgical site infection (SSI) range between 3 and 4% of patients undergoing surgery; depending on the type of surgery, the incidence of SSI ranges between <1 and >10%. In the future, as the population ages, the incidence of SSI is expected to sharply increase because the incidence is associated with age, with a doubling of the rate in patients older than 64 years (1). Besides antibiotic treatment, current strategies to counteract wound infection involve functionalization of gels, dressings, or biomaterials with various anti-infective components (12, 14, 15). Commonly used additives in the clinic include silver and polyhexanide [polyhexamethylene biguanide (PHMB)], which is used on acute wounds and burns and on nonhealing ulcers (1618). Although such treatments can kill the bacteria, they do not address the associated inflammatory component. Conversely, treatments addressing inflammation mainly aim to inactivate and scavenge proteases, such as collagen-based wound dressings (19). Thus, today’s wound care only addresses one issue (infection or protease action), and there are no therapeutic modalities currently available that both control bacteria and target the origins and causes of excessive infection-inflammation in wounds or surgical settings.

Because wound healing is important for survival (9), it is expected that multiple natural host defense systems are activated during injury involving initial hemostasis and clot formation and that proteins and peptides are activated in our innate immune system (2022). In humans, examples of such host defense systems include neutrophil-derived α-defensins and the cathelicidin LL-37 (22, 23) and proteolytic products of plasma proteins such as thrombin (2427). Thrombin, which is initially formed by selective proteolysis by coagulation factor X, mediates fibrinogen degradation and clot formation during the acute wounding phase. However, subsequent proteolysis leads to formation of fragments of about 11 kDa, which mediate aggregation of LPS and bacteria, facilitating endotoxin clearance and microbial killing (28). Further proteolysis leads to formation of smaller thrombin-derived C-terminal peptides (TCPs) of roughly 2 kDa, such as FYT21 (FYTHVFRLKKWIQKVIDQFGE) and HVF18 (HVFRLKKWIQKVIDQFGE), which are present in human wound fluids (29, 30) and have been demonstrated to exert anti-endotoxic functions in vitro and in vivo (27, 30, 31). The peptide TCP-25 (GKYGFYTHVFRLKKWIQKVIDQFGE), which encompasses these endogenous sequences, is antimicrobial and binds to and neutralizes bacterial LPS and protects against P. aeruginosa–induced sepsis and LPS-mediated shock in experimental animal models, mainly via reduction of systemic cytokine responses (27, 32). Moreover, the peptide interacts directly with monocytes and macrophages and inhibits TLR4- and TLR2-induced NF-κB activation in response to several microbe-derived agonists (27, 3133). We recently showed that such TCPs, apart from their interactions with bacterial membranes and LPS, also bind to the LPS-binding hydrophobic pocket of CD14 (33). This interference with TLR signaling provides a molecular explanation for the previously observed therapeutic effects of TCP-25 in experimental models of bacterial sepsis and endotoxin shock (27, 32, 33). In addition, the peptide reduces inflammatory responses to intact bacteria during phagocytosis (34) and inhibits neutrophil responses to LPS in vitro and in vivo (35).

We hypothesized that treatment concepts based on such endogenous innate defense strategies in wounds could have therapeutic potential. Here, we developed a TCP-25–based hydrogel, providing a local delivery scaffold, which mimics the endogenous actions of wound-derived host defense peptides (HDPs) that are found in biological matrices such as fibrin (27). We demonstrate that the TCP-25 hydrogel can act as a dual-function local therapeutic, targeting both bacteria and the accompanying inflammatory response in experimental wound models, and that TCP-25 can be cleaved into biologically active fragments that are similar to those found in wounds in vivo. Moreover, TCP-25 reduces the inflammation induced by wound fluids from patients with nonhealing venous ulcers that are infected by S. aureus and P. aeruginosa.

RESULTS

Evaluation of peptide effects and structure in the presence of different formulation components

Given that the action of TCP-25 involves structural transitions upon LPS binding, such as formation of a C-formed turn and a helical structure, and that it requires the ability for both bacterial membrane and CD14 interactions (33), it was important to develop a topical gel formulation with preserved TCP-25 functions. Therefore, TCP-25 activity in formulation components pharmaceutically and clinically compatible with use in surgery and wounding was evaluated. First, antibacterial activity was determined for the peptide alone or in the presence of hydroxypropyl cellulose (HPC), carboxymethyl cellulose (CMC), or pluronic F-127 (hereafter called pluronic). Radial diffusion analysis (RDA) is an agar diffusion–based method measuring bacteriostatic/bactericidal effects. Using RDA against the Gram-negative Escherichia coli and P. aeruginosa and the Gram-positive S. aureus, we demonstrated retention of peptide activity for TCP-25 against Gram-negative bacteria E. coli and P. aeruginosa. Addition of CMC, however, inhibited peptide activity in S. aureus (Fig. 1A). Analysis using viable count assay (VCA), measuring bactericidal effects in solution, demonstrated that both the anionic CMC polymer and the micelle-forming pluronic interfered with the antibacterial action of TCP-25, whereas the peptide’s antibacterial activity was preserved in HPC (Fig. 1B). Clinical and regulatory considerations also prompted a comparison with the related neutral polymer hydroxyethyl cellulose (HEC), and the results were similar to those obtained with HPC (fig. S1, A and B).

Fig. 1 Antibacterial and anti-endotoxic effects of TCP-25 in various formulations.

(A) Peptide activity and the release profile of TCP-25 formulations. The activity of TCP-25 in various formulations (HPC, CMC, and pluronic) was determined by evaluating the antimicrobial activity against E. coli, P. aeruginosa, and S. aureus using RDA. The heatmap illustrates measurements of the zones of clearance obtained, corresponding to the inhibitory effect of released peptide. Color and values in each box represent mean values (n = 3). (B) Heatmap showing antimicrobial effects of TCP-25 in various formulations (HPC, CMC, and pluronic) as assessed by a viable count assay (VCA). E. coli, P. aeruginosa, and S. aureus were incubated with formulation substances with or without TCP-25. Color and values in each box represent mean values (n = 3). (C) To investigate whether TCP-25 formulations block endotoxin-induced proinflammatory responses, THP-1–XBlue-CD14 cells were stimulated with E. coli LPS in the presence of various formulations (HPC, CMC, and pluronic) with and without TCP-25. The heatmap indicates NF-κB activation, as determined by measuring the production of secreted embryonic alkaline phosphatase (SEAP). Color and values in each box represent mean values (n = 3). OD600, optical density at 600 nm. (D) To assess cell viability of the TCP-25 formulations, an MTT assay was used. A bar chart shows the percentage of viable cells, quantified using the MTT assay. Values are shown in comparison to the untreated live cells (100%, dotted line). Data are presented as the means ± SEM (n = 3). P values were determined using a Kruskal-Wallis test followed by Dunn’s post hoc test. *P ≤ 0.05; NS, not significant.

Because the endotoxin-blocking effects of TCP-25 depend on specific interactions with both LPS and cells (33), it is possible that the structural prerequisites for these anti-inflammatory activities may be separated from those required for the antibacterial action in a specific formulation. We therefore evaluated the anti-endotoxic activity of the peptide in the presence of the different formulation components in vitro using LPS-stimulated THP-1–XBlue-CD14 cells. Cells were incubated with E. coli LPS (10 ng/ml) and with TCP-25 in the presence or absence of HPC, CMC, and pluronic. After 18 to 24 hours of incubation, NF-κB and AP-1 activation was assessed. The results showed that CMC in particular and, to a lesser extent, pluronic interfered with TCP-25 anti-endotoxic action. However, HPC did not exert any inhibitory effects on TCP-25 (Fig. 1C). As above, a comparison with the related polymer HEC yielded results that were similar to those obtained with HPC (fig. S1C). Simultaneous analyses of toxic effects of formulation components alone and in combination with TCP-25 were performed, and the results showed that the formulation combinations did not affect cell viability, as assessed using an [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide] (MTT) assay (Fig. 1D and fig. S1D).

To gain an understanding of the structure-function relationship explaining the observed differences for the different gel formulations, we investigated structural features and possible structural transitions of TCP-25 upon LPS binding in the presence of the gel formulation components using circular dichroism (CD) analysis (Fig. 2). Overall, the results showed that, in contrast to the neutral HPC/HEC, the negatively charged CMC induced a strong structural change in TCP-25, which was consistent with an increase in its α-helical content (Fig. 2, A and B). The α-helical content of TCP-25 was calculated from molar ellipsometry at 222 nm in the presence of tris buffer, LPS, and polymers (at the ratio of 1:5). In the presence of LPS, TCP-25 showed a significant (P ≤ 0.05) increase in α-helical content in the HEC gel formulation (Fig. 2B), which was similar to its conformational transition induced by LPS in buffer only. Together, these structural studies corresponded well with the functional studies, demonstrating that TCP-25 action is facilitated by formulations showing few TCP-25 interactions per se, and thus, enabling the peptide’s LPS and cell interactions. Figure 2C depicts the schematic description of TCP-25 peptide and LPS interaction in the presence of our studied gel components.

Fig. 2 Secondary structural changes of TCP-25 determined by CD spectroscopy.

(A) CD spectra of TCP-25, measured after incubation with tris buffer, LPS, HPC, HEC, CMC, or pluronic (TCP-25–to–polymer ratios of 5:1, 1:1, and 1:5). (B) α-Helical content of TCP-25 calculated from molar ellipsometry at 222 nm in the presence of tris buffer, LPS, and polymers (ratio of 1:5). Data are presented as the means ± SEM (n = 3). P values were determined using a Mann-Whitney U test. *P ≤ 0.05. (C) Schematic description of TCP-25 peptide and LPS interaction in the presence of various formulation components.

On the basis of the above data, we selected a gel base consisting of HEC polymer, 1.3% glycerol, and 10 mM tris (pH 7.4) for further studies. Initial dose-response studies using S. aureus and P. aeruginosa showed that doses of 0.01 to 0.05% TCP-25 (0.1 to 0.5 mg/ml, corresponding to 0.03 to 0.15 mM TCP-25) in HEC gel exhibited antibacterial effects (fig. S2, A and B). Thus, to achieve a safe margin for bacterial killing, a dose of 0.1% TCP-25 (0.3 mM) in HEC gel was selected for further studies (hereafter denoted TCP-25 gel). As illustrated using bioimaging (Fig. 3A) and quantified by use of luminometry (Fig. 3B), the peptide-functionalized gel yielded a rapid reduction in P. aeruginosa PAO1 and S. aureus bioluminescence after only 5 min of incubation. The antibacterial effect was mediated by bacterial permeabilization, as demonstrated by the use of a live-dead assay, which uses propidium iodide (red color) to detect loss of membrane integrity (Fig. 3C). Using the VCA, the bactericidal effect of the TCP-25 gel on S. aureus and P. aeruginosa PAO1 was further demonstrated, yielding more than three log reductions for the two bacteria (Fig. 3D). To extend the findings, we next analyzed the TCP-25 hydrogel effects on a series of human wound isolates. As demonstrated, the functionalized gel yielded more than three log reductions of all clinically derived isolates of S. aureus and P. aeruginosa as well as additional wound isolates and reference strains (Fig. 3E and fig. S3). The results were further substantiated using TCP-25 in standard minimum inhibitory concentration (MIC) assay (table S1) (36). The peptide was also active against a series of multidrug-resistant isolates (table S2). Note that the MIC values in all cases were below the set TCP-25 dose of 0.1% (0.3 mM) in the gel formulation. Together, these results demonstrate that TCP-25 retains its antibacterial activity in neutral polymers such as HPC and HEC, is active against multiple bacterial Gram-negative and Gram-positive bacterial isolates, and that the formulated 0.1% TCP-25 gel exerts a rapid killing effect that is mediated by bacterial permeabilization.

Fig. 3 In vitro antibacterial effects of TCP-25 formulated in a HEC gel.

(A) Bioluminescent S. aureus (S. a.) or P. aeruginosa (P. a.) bacteria were incubated with TCP-25 HEC gel for various time points, and the bioluminescent signal was analyzed using bioimaging (IVIS Spectrum). Representative images are shown (n = 3). (B) Bacterial bioluminescence measurement after treatment with TCP-25 HEC gel. Bioluminescent S. aureus or P. aeruginosa (107/ml CFU) were treated with TCP-25 formulation, and bioluminescence was measured using a luminescence plate reader. Line chart shows total bioluminescence count at the indicated time points. Data are presented as the means ± SEM (n = 3). P values were determined using a two-way ANOVA with Tukey’s posttest. (C) A bacterial live-dead assay demonstrating the antibacterial effects of the TCP-25 gel. S. aureus and P. aeruginosa bacteria were treated with TCP-25 formulation for 30 min. Green, live bacteria; red, dead bacteria. Representative images are shown (n = 3). (D) VCA showing antimicrobial effects of TCP-25 HEC formulation against S. aureus or P. aeruginosa. Data are presented as the means ± SEM (n = 3). P values were determined using a Mann-Whitney U test. (E) Antimicrobial activity of TCP-25 HEC formulation against clinical isolates of P. aeruginosa, S. aureus, Staphylococcus epidermidis, or Enterococcus faecalis. The number of CFU was determined using VCA. Data are presented as the means ± SEM (n = 3). P values were determined using an unpaired t tests. Comparisons were made with respective gel controls. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Effects of TCP-25 wound gel on bacteria and endotoxin responses in experimental mouse models

In an experimental animal model, we wanted to simulate a situation of surgical contamination with bacteria. Two different models were used, both mimicking situations of relevance for surgery and wounding. In the first model, mimicking a direct and immediate contamination, TCP-25 gel was inoculated with bioluminescent S. aureus or P. aeruginosa bacteria [106 colony-forming units (CFU) per animal] and immediately injected subcutaneously in SKH1 mice. The results showed that the TCP-25 gel reduced the bacterial load as assessed by in vivo bioimaging and analyses of CFU (Fig. 4A and fig. S4). When compared with the initial reduction of bioluminescence after 15 min, a higher signal was recorded after 24 hours, possibly due to bacterial regrowth due to the single dosage regimen (fig. S4). The reduction in CFU achieved using the gel was, however, >2 log for the two bacteria after 24 hours (fig. S4). To visualize the tissue distribution of TCP-25, we spiked the formulation with 1% Cy5-labeled TCP-25, and the peptide was observed to localize to the site of gel administration during the time period studied (Fig. 4A). Histological analyses of the infected tissue areas corresponding to the bacterial analyses showed an abrogated inflammatory response in the TCP-25–treated animals (Fig. 4B). In the second prevention model, the TCP-25 gel was first injected subcutaneously in BALB/c mice. Thirty minutes later, bioluminescent S. aureus and P. aeruginosa bacteria (106 CFU per animal) were injected into the site of gel deposition. As above, the results showed reductions of bacteria as assessed by in vivo bioimaging (fig. S5).

Fig. 4 Antibacterial and anti-inflammatory effects of TCP-25 HEC gel formulation in a mouse model of subcutaneous infection and inflammation.

(A) In vivo infection imaging by IVIS in the mouse model of subcutaneous infection. HEC gel with and without TCP-25 (spiked with Cy5) was deposited subcutaneously in the dorsum of SKH1 mice after inoculation with 106 CFU of bioluminescent P. aeruginosa or S. aureus bacteria. Bacterial bioluminescence intensity and TCP-25 Cy5 fluorescence were noninvasively analyzed using the IVIS bioimaging system. Representative images show bacterial luminescence (lum) and TCP-25 Cy5 fluorescence (flu) at 6 hours after infection. The bar chart shows measured bioluminescence intensity emitted by the bacteria at 6 hours after infection. Data are presented as the means ± SEM (n = 7 mice for gel group and 7 mice for TCP-25 gel group for each bacterial infection). P values were determined using Mann-Whitney U test. (B) Representative images show hematoxylin and eosin (H&E) staining of mouse skin tissue from the site of gel deposition. Arrows show tissue destruction and the hyperinflammatory condition of the tissue. (C) In vivo inflammation imaging by IVIS in NF-κB reporter mice. LPS in HEC gel or in TCP-25 HEC formulation (spiked with Cy5-labeled TCP-25) was subcutaneously deposited on the back of transgenic BALB/c Tg(NF-κB-RE-luc)-Xen reporter mice. In vivo bioimaging of NF-κB reporter gene expression was performed using the IVIS Spectrum system. Representative images show bioluminescence (lum) and TCP-25 Cy5 fluorescence (flu) at 6 hours. A bar chart shows measured light intensity emitted from these reporter mice. Data are presented as the means ± SEM (n = 7 mice for gel group, 5 mice for TCP-25 gel group). P values were determined using Mann-Whitney U test. (D) Cytokine analysis from the wound fluid extracted from implanted polyurethane discs. Data are presented as the means ± SEM (n = 4 gels, n = 4 TCP-25 gels). P values were determined using Mann-Whitney U test. **P ≤ 0.01, ***P ≤ 0.001.

We next explored whether TCP-25 gel could suppress LPS-triggered local inflammation in vivo. Control gel or TCP-25 gel was injected subcutaneously, with simultaneous addition of LPS. Using mice reporting NF-κB activation, we found that the addition of LPS to the formulation yielded a local inflammatory response, which was abrogated by the gel containing TCP-25 (Fig. 4C). Using gels spiked with Cy5-labeled TCP-25, we observed that the peptide localized to the site of gel administration during the time period studied (Fig. 4C). In a separate experiment, we used BALB/c mice that had 6-mm polyurethane discs implanted subcutaneously to collect local wound exudates as a model of a surgical implant. Application of LPS yielded an increase of interleukin-6 (IL-6) and tumor necrosis factor–α (TNF-α), which were reduced upon addition of the TCP-25 gel (Fig. 4D). This result is comparable to the results of the bioimaging studies (Fig. 4C). Together, the results demonstrated that the TCP-25 peptide also retains its dual anti-infective and anti-inflammatory function in vivo in subcutaneous models of infection and endotoxin-driven inflammation.

Effects of TCP-25 hydrogel in a porcine partial thickness wound model

We next wanted to explore the effects of TCP-25 gel in a model that is translatable to the human wounding situation. For this purpose, we used a partial thickness wound model in Göttingen minipigs (study outlines are presented in Fig. 5A). In the initial 4-day study, mimicking a clinical situation in which treatment is applied to injury and bacterial contamination (here denoted “contaminated wound model”), wounds were inoculated with S. aureus, followed by application of control or TCP-25 gel after a 30-min incubation time and subsequent daily gel treatments at dressing changes. On day 4, nontreated control wounds showed visible signs of inflammation and infection (Fig. 5B). TCP-25 gel treatment abrogated the infection, leading to an improved clinical score (Fig. 5, B and C), reduced bacterial counts (Fig. 5D), lower IL-6 and TNF-α (Fig. 5E), and reduced inflammatory signs at the tissue level (Fig. 5F). Clinical scoring of minipig wounds was accomplished as described in table S3.

Fig. 5 Effects of TCP-25 gel in a porcine partial thickness wound model.

(A) Schematic illustrating the wounding and wound dressing plan in minipigs. Twelve partial thickness wounds, six on each side, were created using an electric dermatome on the backs of Göttingen minipigs and infected with S. aureus (107 CFU per wound). After infection and application of gel, wounds were covered with a primary polyurethane (PU) dressing, followed by a transparent breathable fixation dressing secured with skin staples. Two layers of sterile cotton gauze were secured with adhesive tape, topped with a layer of flexible self-adhesive bandage. (B) Representative photographic images of minipig wounds after the short-term treatment regimen. Wounds with either S. aureus or having a mixed infection (S. aureus and superinfection with P. aeruginosa) were treated every day with gel with or without TCP-25. Uninfected control wounds were treated with gel without TCP-25. Scale bar, 1 cm. (C) Clinical scoring of wounds after the short-term treatment regimen. Data are presented as the medians with 95% confidence intervals (For S. aureus infection group, n = 10 wounds for gel, n = 9 wounds for TCP-25 gel, and n = 3 wounds for uninfected controls from four pigs. For mixed infection group, n = 4 wounds for gel, n = 5 wounds for TCP-25 gel, and n = 3 wounds for uninfected controls from 2 pigs). P values were determined using a Kruskal-Wallis test followed by Dunn’s posttest. (D) Microbiological analysis of wounds from days 2, 3, and 4. Data are presented as the means ± SEM (For S. aureus infection group, n = 10 wounds for gel, n = 9 wounds for TCP-25 gel, and n = 3 wounds for uninfected controls from four pigs. For mixed infection group, n = 4 wounds for gel, n = 5 wounds for TCP-25 gel, and n = 3 wounds for uninfected controls from two pigs). P values were determined using a Kruskal-Wallis test followed by Dunn’s posttest. (E) Analysis of wound fluid cytokines collected on days 2, 3, and 4. Data are presented as the means ± SEM (For the S. aureus infection group, n = 8 to 10 wounds for gel, n = 7 to 9 wounds for TCP-25 gel, and n = 3 wounds for uninfected controls from four pigs. For the mixed infection group, n = 4 wounds for gel, n = 5 wounds for TCP-25 gel, and n = 3 wounds for uninfected controls from two pigs). P values were determined using a Kruskal-Wallis test followed by Dunn’s posttest. (F) Representative images showing H&E staining of wound biopsies after 4 days of treatment. Arrows show severe tissue destruction and hyperinflammatory condition of the wound. Arrowheads show wound re-epithelization. The bar chart shows histological analysis of wound tissues. Data are presented as the means ± SEM (n = 12 wounds for gel, n = 12 wounds for TCP-25 gel). P values were determined using a Mann-Whitney U test. (G) Representative photographic images of minipig wounds after the long-term treatment regimen. Wounds were infected with S. aureus and treated on days 1, 2, 3, 5, 7, and 9 with TCP-25 gel. Bottom: Images show H&E staining of wound biopsies. Dot plot shows microbiological analysis of wounds from days 2, 5, and 7. Data are presented as the means ± SEM (n = 10 wounds for gel, n = 10 wounds for TCP-25 gel from four pigs). P values were determined using a Mann-Whitney U test. (H) Effect of TCP-25 gel treatment on minipig wound healing (noninfected wounds). Partial thickness wounds were created on minipigs and treated with TCP-25 gel. Representative photographic images of wounds and H&E-stained wound biopsies are shown. The bar chart shows histological analysis of the wound tissues. Data are presented as the means ± SEM (n = 10 wounds for gel, n = 9 wounds for TCP-25 gel from four pigs). P values were determined using Mann-Whitney U test. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

In two separate experiments, we observed that two animals developed a superinfection on day 2, leading to a mixed infection and prompting us to analyze and present these wound data separately. Upon isolation and identification, the bacterium responsible for superinfection was found to be P. aeruginosa. TCP-25 gel also prevented this secondary, mixed infection (Fig. 5, B to E). These two P. aeruginosa isolates showed sensitivity to TCP-25 in RDA and the peptide abrogated proinflammatory effects of bacterial supernatants on THP-1 cells (fig. S6). To reproduce the spontaneous superinfection with P. aeruginosa, we repeated the bacterial contamination model using S. aureus, followed by inoculation of the same wounds with one P. aeruginosa strain collected from the mixed infection above (fig. S7A). As above, TCP-25 gel treatment prevented the development of infection, leading to an improved clinical score (fig. S7B), reduced bacterial counts (fig. S7C), and lower TNF-α and IL-1β (fig. S7E). These experiments were followed by a 10-day study in which wounds were inoculated with S. aureus as above, followed by daily treatments for 3 days and thereafter every second day. In addition, here, TCP-25 gel treatment prevented S. aureus infection and improved wound status (Fig. 5G). Histological analysis showed that peptide-treated wounds were completely re-epithelialized, which was not observed in the infected untreated wounds (Fig. 5G). Last, to evaluate effects on normal healing of uninfected wounds, TCP-25 gel was similarly applied for 10 days on the partial thickness wounds in a separate experiment. Both control gel and TCP-25 gel–treated wounds showed normal wound healing with no signs of tissue toxicity (Fig. 5H).

TCP-25 gel pharmacokinetics in vitro and in vivo and rheological analysis

Next, we investigated the pharmacokinetics of TCP-25 in the gel formulation in vitro and in an in vivo model. In vitro, we analyzed the diffusion rate of the tetramethylrhodamine-labeled TCP-25 from the gel to a buffer solution (fig. S8A, inset). In the two-compartment system used, the peptide was eluted gradually, with no observed initial burst, from the gel phase as determined by the fluorescence readings. The peptide was detected in the buffer compartment after 2 hours, and about half of the peptide amount was released from the hydrogel after 6 hours, data compatible with the weak peptide-polymer interactions detected by CD (Fig. 2).

To assess the distribution of TCP-25 in vivo, we subcutaneously deposited the TCP-25 gel (spiked with Cy5-labeled TCP-25) in the dorsum of SKH1 mice. Longitudinal IVIS bioimaging was used to track the diffusion of the peptide from the gel into the surrounding tissues, and these results showed that TCP-25 was largely retained at the injection site (fig. S8B). Thus, the results were compatible with the slow diffusion observed in the in vitro model (fig. S8A). The presence of LPS did not influence the distribution of TCP-25 in this in vivo model (fig. S8B). To assess possible peptide uptake through the skin and wounds, we applied TCP-25 gel, spiked with Cy5-labeled TCP-25 as above, either onto intact porcine skin or onto wounds ex vivo and in vivo. The total peptide concentration was varied between 0.1%, 1% (in vivo), and 2% (ex vivo). The images showed that the fluorescent peptide remained locally at the application site, and no visible uptake was observed through skin or wounds into the underlying tissues (fig. S8, C and D). To assess possible systemic uptake of the peptide after topical treatment, we analyzed plasma from the partial thickness wound models using mass spectrometry. No TCP-25 peptide was detected in the porcine plasma after a 24-hour application of TCP-25 gel on partial thickness wounds; however, intact TCP-25 was detected in wound fluids from dressings obtained from infected and uninfected wounds after a 24-hour treatment period (fig. S8E). Histological analysis confirmed the absence of the epidermal layer after dermatome wounding (fig. S9). The rheology measurements of flow point indicate that the peptide does not modify the gel characteristics (fig. S10A). These results are compatible with those presented in Fig. 2, indicating that the peptide does not interact significantly with the HEC polymer. Furthermore, as G′ < G′, at low strain, all formulations display gel-like characteristics (fig. S10B).

Degradation of TCP-25 by neutrophil elastase in vitro and comparison with proteolytic TCP fragments generated in vitro and in vivo

We have previously shown that multiple TCPs are generated by proteolytic digestion of thrombin in vitro by the major protease human neutrophil elastase (HNE) (27, 29), a dominant enzyme active during wound healing and inflammation (37). The corresponding C-terminal peptide sequences were identified in wound fluids from acute and nonhealing ulcers, and among these were the TCP fragments FYTHVFRLKKWIQKVIDQFGE and HVFRLKKWIQKVIDQFGE (29, 30). Therefore, we hypothesized that similar fragments may be generated from synthetically produced TCP-25, and our goal was to determine the digestion pattern of TCP-25 after it was subjected to HNE. Enzyme digestion was performed for different time periods, and the fragmentation was evaluated by liquid chromatography–tandem mass spectrometry. Figure 6A shows the major peptides that were obtained after digestion for different time periods. All the TCP-25 fragments that were identified are presented graphically in Fig. 6B, and TCP fragments were compared to fragments found after thrombin digestion with HNE. TCPs detected in human wounds in vivo are also shown in Fig. 6B. The results show that multiple peptides detected in wound fluid, as well as after digestion of thrombin with HNE, overlap structurally with those identified after HNE digestion of TCP-25 (Fig. 6B). For example, the peptide HVFRLKKWIQKVIDQFGE (HVF18) was detected after proteolysis of TCP-25 by HNE, as well as the major fragment GKYGFYTHVFRLKKWIQKVI (GKY20) (Fig. 6A). The digestion patterns were also similar in the buffer and in the HEC formulation (Fig. 6C). The generated peptide fragments retained their antibacterial activities for digestion periods of up to 6 hours in the RDA, although longer digestion times led to a reduction of peptide activity particularly when RDA was performed at physiological salt conditions (0.15 M NaCl) (Fig. 6D). In summary, the results show that HNE degradation of TCP-25 resulted in the generation of a multitude of bioactive TCP-fragments, of which several overlapped with identical peptides generated from human thrombin and were also present in human wounds in vivo.

Fig. 6 Degradation of TCP-25 by human neutrophil elastase in vitro and comparison with proteolytic thrombin fragments generated in vitro and in vivo.

(A) Table of digestion pattern of TCP-25 after treatment with HNE. Sequences of major peptides and the number of successful identifications by mass spectrometry at 10, 30, 60, and 180 min are shown. (B) Graphical representation of major peptides obtained after digestion and comparison with peptides found after digestion of thrombin (29) and those detected in human wounds in vivo (29, 30). Asterisk (*) indicates peptides reported to show antibacterial effects. (C) Representative high-resolution MALDI mass spectra of HNE digested TCP-25. The same peptide fragments were detected in the buffer solution and the gel. After 180 min, no intact TCP-25 could be detected from the solution or gel sample. Identified peptide sequences are shown in the bottom. (D) The release and activity of TCP-25 degradation products were determined by evaluating the antimicrobial activity against E. coli by RDA in 10 mM tris (pH 7.4) with or without 0.15 M NaCl. The bar chart illustrates measurements of the zones of clearance obtained. Data are presented as the means ± SEM (n = 6). P values were determined using Mann-Whitney U test. Comparisons were made with buffer control. **P ≤ 0.01.

Stability of TCP-25 in vitro

In contrast to the rapid degradation of TCP-25 when subjected to HNE, the peptide did not show changes when stored either in buffer or in the HEC formulation for extended periods of time at 4° or 20°C. Mass spectrometry analysis using matrix-assisted laser desorption/ionization (MALDI)–mass spectrometry found no indication of degradation or oxidation/deamination for up to 180 days at these temperatures. However, storage for 180 days at 37°C resulted in mass changes. Activity assays corresponded well with the mass analyses and showed that the peptides’ antibacterial and immunomodulatory effects were preserved after storage for up to 180 days (fig. S11). It has been previously shown for other peptides that their half-life in plasma depends on the species used for plasma collection, possibly due to differences in endoproteinases or other factors affecting stability (3840). Hence, the stability of TCP-25 was studied in human, porcine, and mouse plasma using mass spectrometry. TCP-25, when incubated at 37°C, was stable when incubated in phosphate-buffered saline and mouse plasma. In human and minipig plasma, the half-lives were calculated as 8.1 and 2.5 hours (fig. S12).

Comparison of TCP-25 hydrogel with clinically used wound treatments and effects on established infection

Various silver dressings and products containing PHMB are commonly used to prevent infections on burns or surgical wounds or as treatments for chronic leg ulcers (16, 17, 41, 42). Thus, in the next study, using the previous treatment scheme for the contaminated wound model, we compared TCP-25 gel with the common wound treatments Mepilex Ag and Prontosan gel, which contain silver and PHMB, respectively. As a second objective, we wanted to study a higher dose of TCP-25 to detect any negative side effects of TCP-25. Thus, wounds were inoculated with S. aureus and then treated with either of the three wound treatments. As demonstrated previously for 0.1% TCP-25 gel, the 1% TCP-25 gel treatment used in this study also prevented S. aureus infection (Fig. 7, A and B) without any observed negative effects in this 4-day contaminated wound model, as assessed by clinical scoring (Fig. 7C) and histological analysis (Fig. 7D). We observed, however, that Mepilex Ag was not effective in preventing S. aureus infection because the clinical status and wound bacterial numbers were similar to those for the infected control (Fig. 7, A to D). Although a reduction of bacteria in the Mepilex Ag dressing extracts was observed, this change was not statistically significant (P > 0.05) (fig. S13A). In contrast, TCP-25 gel was able to reduce (>5 log) bacterial numbers in the dressing extracts (fig. S13A). In agreement with this observation, in vitro, we found that the Mepilex Ag dressing, while showing antibacterial effects against S. aureus in buffer conditions, was not effective in the presence of plasma and wound fluid (fig. S13B). In this infection model, which was mainly aimed at evaluating antibacterial effects, Prontosan treatment yielded similar results to TCP-25 gel on clinical wound scoring, reduction of bacterial infection, and TNF-α (Fig. 7, A to D, and fig. S14).

Fig. 7 Comparison of TCP-25 gel with wound treatment benchmarks.

TCP-25 gel was compared with Mepilex Ag and Prontosan, two current standard benchmarks in wound care. (A) Representative photographic images of minipig wounds after the short-term treatment regimen. Wounds were infected with 107 CFU of S. aureus and treated once daily with TCP-25 gel, Mepilex Ag, or Prontosan. (B) Microbiological analysis of wounds from days 2, 3, and 4. Swab samples were collected from wounds, and appropriate dilutions were plated on Todd Hewitt (TH) broth agar, and the number of CFU was determined. Data are presented as the means ± SEM (n = 6 wounds for gel, n = 6 wounds for TCP-25 gel, n = 6 wounds for Mepilex Ag, and n = 6 wounds for Prontosan from three pigs). Comparisons are shown against “gel” group, and P values were determined using a Kruskal-Wallis test followed by Dunn’s posttest. (C) Clinical scoring of wounds after the short-term treatment regimen. Data are presented as the medians with 95% confidence intervals (n = 6 wounds for gel, n = 6 wounds for TCP-25 gel, n = 6 wounds for Mepilex Ag, and n = 6 wounds for Prontosan from three pigs). P values were determined using a Kruskal-Wallis test followed by Dunn’s posttest. (D) Representative images showing H&E staining of wound biopsies. Arrows show severe tissue destruction and inflammatory infiltrates in the wound. Arrowhead indicates areas of re-epithelization of the wound. (E) Timeline of established infection model experimental plan in minipigs. (F) Representative photographic images of minipig wounds at days 2 and 10 of established infection treatment regimen. Wounds were infected with S. aureus and after establishment of infection, treated on days 2, 3, 5, 7, and 9 with control gel, TCP-25 gel, or Prontosan. Scale bar, 1 cm. (G) Microbiological analysis of wounds (from established infection model) from days 2, 3, 5, 9, and 10. Data are presented as the means ± SEM (n = 7 wounds for gel, n = 7 wounds for TCP-25 gel, and n = 7 wounds for Prontosan from two pigs). P values were determined using a Kruskal-Wallis test followed by Dunn’s posttest. (H) Analysis of TNF-α in wound fluid collected on days 2, 3, and 5. Data are presented as the means ± SEM (n = 7 wounds for gel, n = 7 wounds for TCP-25 gel, and n = 7 wounds for Prontosan from two pigs). P values were determined using a Kruskal-Wallis test followed by Dunn’s posttest. (I) In vivo infection imaging by IVIS in a mouse model of subcutaneous infection. TCP-25 gel formulations were deposited subcutaneously on the dorsum of SKH1 mice after adding bioluminescent S. aureus or P. aeruginosa bacteria. Representative images show bacterial luminescence at 6 hours after infection (n = 6 for each group). (J) In vivo inflammation imaging by IVIS in NF-κB reporter mice. Prontosan or TCP-25 gel was mixed with LPS and subcutaneously deposited on the left and right sides, respectively, on the back of transgenic BALB/c Tg(NF-κB-RE-luc)-Xen reporter mice. In vivo imaging of NF-κB reporter gene expression was achieved using an IVIS Spectrum bioimaging system. Representative images show bioluminescence 6 hours after subcutaneous deposition. Bar chart shows the measured bioluminescence intensity emitted from these mice. Data are presented as the means ± SEM (n = 5 each group). P values were determined using a Mann-Whitney U test. (K) Comparison of anti-inflammatory ability of TCP-25 and PHMB, the antiseptic ingredient of Prontosan. THP-1–XBlue-CD14 reporter cells were stimulated with E. coli LPS, in the presence of PHMB and TCP-25. Data are presented as the means ± SEM (n = 6). P values were determined using a one-way ANOVA with Tukey’s posttest. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Last, to evaluate the effect of TCP-25 in a model of established infection, wounds were inoculated with S. aureus, infections allowed to establish for 24 hours, and subsequently treated with 1% TCP-25 gel or Prontosan (Fig. 7E). Mepilex Ag was omitted because it was ineffective in the contaminated wound model. On day 2, before initiation of treatment, all wounds showed clinically visible signs of infection and similar bacterial numbers (Fig. 7, F and G). Treatments were applied daily for two consecutive days and then every second day until day 10. The results showed that 1% TCP-25 gel treatment indeed reduced S. aureus infection and related inflammation as assessed by bacterial numbers (Fig. 7G) and lowered concentrations of cytokines TNF-α (Fig. 7H) and IL-1β (fig. S15). Reductions in cytokine concentrations preceded the observed antibacterial effects in the TCP-25 gel–treated wounds (Fig. 7H and fig. S15). Thus, results obtained at day 3 (after 24 hours of treatment with TCP-25 gel) showed reductions of TNF-α (Fig. 7H) and IL-1β (fig. S15) despite bacterial numbers in the wounds remaining similar to those of the control. Although we observed an initial effect on bacterial numbers by Prontosan at day 5, it was overall not effective in reducing S. aureus infection and inflammation in the model of established infection (Fig. 7, F to H, and fig. S15). Together, these results demonstrate that TCP-25 gel is effective in targeting S. aureus in models of contaminated and infected wounds and that the treatment has a capacity of reducing cytokines independently of bacterial presence.

Although the porcine wound models above indicated that TCP-25 can target bacteria and inflammation, we wanted to separately evaluate the latter function of the TCP-25 gel further. In a mouse model of subcutaneous infection, Prontosan gel was equally as effective as TCP-25 gel in reducing S. aureus and P. aeruginosa (Fig. 7I), which was consistent with the observed preventive effects of both treatments in the S. aureus–contaminated wound model (Fig. 7, A and B). To specifically assess possible anti-endotoxic effects, we next injected LPS subcutaneously with addition of either Prontosan gel or 0.1% TCP-25 gel, on each side of the mouse dorsum. In this model, TCP-25 gel exhibited a significant (P < 0.01) anti-inflammatory activity when compared to Prontosan gel (Fig. 7J). In agreement with these in vivo data, in vitro experiments showed that TCP-25 exerted higher LPS-quenching effect compared to PHMB (Fig. 7K). Last, a single-dose toxicity study was performed. Mice were subcutaneously injected with 5 mg of TCP-25, and organs were collected after 24 hours. Histology of the lung, kidney, liver, spleen, and skin did not show any signs of toxicity (fig. S16).

TCP-25 targets inflammation in wounds

Having demonstrated the anti-infective and anti-endotoxic properties of TCP-25 gel, we next wanted to explore whether the treatment concept could target TLR-mediated inflammation related to wound infection in general. Initial experiments demonstrated that wound fluid derived from the mixed infection wounds described above (Fig. 5D) induced NF-κB activation in the THP-1 cell model system, whereas such induction was not observed for either TCP-25 gel–treated wounds or uninfected wounds (Fig. 8A). However, because this absence of NF-κB induction could be ascribed to the anti-infective effects of the treatment, we subsequently added TCP-25 to wound fluids derived from the animals infected with both S. aureus and P. aeruginosa. In this case, TCP-25 reduced the wound fluid–induced inflammation (Fig. 8B). Patients with nonhealing venous ulcers are commonly colonized or infected by S. aureus and P. aeruginosa (43). Wound fluids derived from five patients with wounds infected by these bacteria were selected and found to activate THP-1 cells to varying degrees (Fig. 8C). In this complex environment, adding TCP-25 reduced the NF-κB activation, as noted above (Fig. 8B). Together, the results indicate that TCP-25 also has the potential to attenuate inflammation in complex wound environments containing endotoxins and other TLR agonists and cytokines.

Fig. 8 TCP-25 targets inflammation in wounds.

(A) NF-κB activation in THP-1–XBlue-CD14 reporter cells in response to stimulation with wound fluid from infected and TCP-25 gel–treated minipigs wounds. Data are presented as the means ± SEM (n = 6). P values were determined using a Kruskal-Wallis test followed by Dunn’s posttest. (B) NF-κB/AP-1 activity as a marker of inflammation in THP-1–XBlue-CD14 reporter cells stimulated by wound fluid from minipigs days 1, 2, and 3 in the presence of TCP-25. Data are presented as the means ± SEM (n = 4). P values were determined using a one-way ANOVA with Tukey’s posttest. (C) NF-κB/AP-1 activity as a marker of inflammation in THP-1–XBlue-CD14 reporter cells stimulated by chronic wound fluid (CWF) from infected wounds from patients, in the presence of TCP-25. CWF1 to CWF5 represent five human patients. Data are presented as the means ± SEM (n = 6). P values were determined using a one-way ANOVA with Tukey’s posttest. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

DISCUSSION

Infectious diseases account for millions of deaths worldwide each year. In the wound and surgical areas, infected burn wounds and postoperative infections cause significant morbidity and are associated with a risk of deadly systemic complications such as sepsis. The decreasing effectiveness of antibiotics and other antimicrobial agents because of resistance development is an increasing problem. We have reached a point for certain infections where therapeutic agents are no longer available (44). There is, therefore, an important and unmet need for new treatments that will improve healing and reduce infection and inflammation in various types of wounds. Current therapies using systemic antibiotics or local antiseptics only target the bacterial burden without any effects on the associated inflammation. Conversely, therapies reducing the excessive inflammatory responses in wounds mainly target matrix metalloproteinases (MMPs) (19) or locally produced cytokines, and the latter is currently at the preclinical testing stage (45). The data presented in this study demonstrate an alternative approach based on a hydrogel incorporating an HDP targeting bacteria and the proinflammatory products that are released. As illustrated in fig. S17, such a “dual-action” gel has antiseptic functions and targets the proinflammatory responses by blocking bacterial products such as endotoxins. TCP-25 acting upstream of NF-κB also distinguishes it from previous concepts that target MMPs and cytokines, which are downstream of the NF-κB–mediated response.

As proof of concept, we demonstrated that the TCP-25 hydrogel effectively kills the pathogens S. aureus and P. aeruginosa in vitro and in vivo. Staphylococci are a major cause of underlying postoperative surgical infections (12, 46), and emerging multidrug-resistant strains complicate treatment possibilities (44). Because the TCP-25 mode of action is different from existing antibiotics, its ability to target MRSA in vitro is of interest: It could potentially reduce the risk of infections with resistant staphylococci. The MIC analyses for TCP-25 showed that the peptide had comparable activity to other antibacterial peptides such as LL-37 (47) and omiganan (48), which are in clinical development. Similarly, as shown here, TCP-25 also killed a series of P. aeruginosa and S. aureus isolates in vitro when formulated in a gel. The data from the in vivo bioimaging studies using mouse models mimicking situations of acute wounding and bacterial contamination demonstrate that the TCP-25 gel also has the ability to prevent subcutaneous S. aureus and P. aeruginosa infection in vivo. Of particular importance was the finding that the TCP-25 gel also reduced local responses to subcutaneously injected endotoxins in the experimental mouse models.

Porcine wound healing studies are considered to have a good translation to the human wound healing situation (49). In the short-term partial thickness model of a contaminated wound, we observed that the TCP-25 hydrogel was efficient in preventing infection after inoculation of the wounds with S. aureus. In two separate experiments, we serendipitously observed that animals acquired a superinfection on day 2 with P. aeruginosa, and the TCP-25 gel treatment prevented such a mixed infection. After isolation and characterization of the P. aeruginosa strain, we successfully reproduced this model by contaminating the wounds with such P. aeruginosa bacteria 24 hours after S. aureus inoculation.

In these above experiments, a reduction in the cytokine concentrations was regularly observed after TCP-25 gel treatment. It could be argued that these effects largely resulted from the observed concomitant reductions in bacterial numbers, leading to reduced toxin release into the wound areas and, hence, causing less inflammation. However, the observation that the proinflammatory effects of the wound exudates derived from infected partial thickness wounds could also be blocked by exogenously added TCP-25 shows that the peptide has the capacity to also reduce bacteria-induced inflammation during wounding in vivo. These findings are compatible with the initial proof-of-principle experiments demonstrating that TCP-25 gel reduces endotoxin-induced tissue inflammation in experimental mouse models. TCP-25 gel was also effective in a model of established S. aureus infection. In this model, we observed that cytokine concentrations were reduced before bacterial numbers were affected, lending further support for TCP-25 gel’s anti-inflammatory action in vivo.

All these observations are of clinical relevance because large patient groups with nonhealing ulcers of various etiologies, such as diabetes and arterial or venous insufficiency, have an inhibited wound healing process. The latter group is the largest and is particularly characterized by chronic and dysfunctional inflammation with high concentrations of proinflammatory factors, proteases, and bacteria (43, 50). The observation that TCP-25 also reduces proinflammatory monocyte responses to wound fluids derived from patients with nonhealing venous ulcers that are infected with S. aureus and P. aeruginosa is therefore highly relevant from a clinical perspective, as it indicates that the peptide has the potential to target excessive inflammation in wounds that contain a mixture of various TLR agonists. This observation is also consistent with the observation that TCP-25 targets endotoxins and other TLR agonists such as lipoteichoic acid, peptidoglycan, and cystosine-phosphate-guanine (CpG) DNA (31).

To further illustrate the clinical and translational potential of the dual-action gel concept, we compared the efficacy of TCP-25 gel with two commonly used wound treatments, Mepilex Ag and Prontosan wound gel. The intended use for both treatments is to treat wounds such as burns and nonhealing ulcers (41, 51). The silver-containing dressing did not prevent S. aureus infection, which was unexpected given the widespread use of silver as an antiseptic in various wound indications. However, despite its long history and common usage, there is little clinical evidence demonstrating that silver-containing dressings or creams improve wound healing or prevent infection (41). Both TCP-25 gel and Prontosan were antibacterial in the mouse model of subcutaneous bacterial infection and in the porcine partial thickness S. aureus–contaminated wound model. In contrast to the contaminated wound model, Prontosan overall neither reduced bacterial infection nor cytokines in the model of established S. aureus infection. This observation corresponds to recent findings on surgical wounds demonstrating that dressings soaked with Prontosan solution were not effective in reducing bacterial numbers and infection (52). Last, studies using the NF-κB reporter mouse model demonstrated that the endotoxin scavenging effects were unique for the TCP-25 hydrogel, further illustrating the functional differences between PHMB and TCP-25.

Previous studies have addressed structure function relationships of TCP-25 and its bioactive epitopes. For example, HVF18 (HVFRLKKWIQKVIDQFGE) is present in wound fluids in vivo (29, 30). The antibacterial and LPS binding epitope of TCP-25 is also present in GKY20 (GKYGFYTHVFRLKKWIQKVI), which contains the first 20 amino acids of TCP-25 (53). Similar to TCP-25, both of these peptides were shown to exert dual antibacterial and anti-inflammatory effects, and they also reduce mortality in mouse models of endotoxic shock (32, 53). Using a screening-based approach, GKY20 was found to display an improved therapeutic index because this peptide retained its anti-infective capacity while showing less hemolysis in human blood (53). The findings that fragments such as HVF18 and related truncations of TCP-25 are present in vivo in wound fluids illustrate a concept of redundancy, with multiple bioactive peptide fragments that are simultaneously present. It has been increasingly appreciated that a multitude of transient, biological interactions (dissociation constant in the micromolar range) occur frequently in biological systems. Sharing many characteristics with “transient drugs,” the TCP family, with its multiple interactions and affinities in the micromolar range, therefore represents an elegant example of such an endogenous biological system that modulates the host responses to infection (33).

From a pharmaceutical perspective, it was therefore of interest to explore whether similar TCP fragments as those that are present in wounds in vivo could be generated from synthetically produced TCP-25. TCP-25 was cleaved by HNE, a major enzyme that is active during normal wound healing (11), and notably, HVF18 was identified as a major bioactive peptide metabolite. One of the other major fragments was identical to the previously described GKY20 peptide (53). Mass spectrometry analyses also identified a series of other truncated TCP-25–derived fragments that were previously described in wounds in vivo (29, 30), of which several have been shown to retain both antibacterial and anti-endotoxic effects in vitro (53). In addition, RDA assays demonstrated that cleaved TCP-25 retains antibacterial activity, and a reduction in activity was particularly noted in the presence of physiological salt conditions, which is compatible with previous observations that shorter TCP fragments exhibit reduced salt resistance (53). However, the data indicate that the activity of the TCP fragments is also retained after digestion periods for up to 3 to 6 hours. A comparison between the degradation profiles of pure TCP-25 and TCP-25 in hydrogel identified similar peptide fragments, indicating that the formulation polymer did not interfere with the degradation patterns obtained. Thus, these data demonstrate that upon proteolysis, TCP-25 may release several bioactive fragments with retained transient interactions and dual-action functionalities in vivo, motivating the use of TCP-25 as an active drug, which, in turn, can act as a precursor for bioactive peptides such as the previously defined N- and C-terminally truncated TCP-25 variants HVF18 and GKY20, respectively (33, 53). In contrast to the rapid and specific degradation by neutrophil elastase, TCP-25 was highly stable for up to 180 days at room temperature in both the buffer and the hydrogel formulation, which is relevant for translation into therapy and clinical use. In addition, the TCP-25 gel was retained to a high degree at the site of injection in the mouse models and at the wound and skin surface in the porcine models with no systemic absorption of TCP-25.

From a drug delivery perspective, the selection of clinically and pharmaceutically acceptable formulations that enable the dual function of TCP-25 is not trivial because this peptide and the fragments HVF18, GKY20, and FYT21 all require a flexible conformation that enables critical bacterial and host cell interactions (33, 53, 54). To study the influence of formulations on TCP-25, we used a combination of both bioassays and structural analysis using CD. The analyses showed that neutral hydrogels enabled TCP-25 action, whereas anionic CMC and the micelle-forming F127 were inhibitory to variable degrees. In addition, the endotoxin-scavenging capacity was particularly sensitive to inhibition by CMC. Structural clues about this observation were obtained using CD, which showed that the peptide assumed an ordered conformation, involving helical structure induction, particularly in the presence of CMC, whereas it was unordered in the presence of HEC. The observed conformational change in the presence of CMC is compatible with interactions between TCP-25 and the anionic polymer. As demonstrated by the bioassays, the peptide scavenging effect by this polymer, in turn, reduces TCP-25 binding to bacteria and LPS. A similar reasoning may be applied on the poloxamer F127, which has hydrophobic polyoxypropylene units between hydrophilic units of polyoxyethylene (55), likely enabling binding to the amphipathic TCP-25. Thus, on the basis of these structural and functional analyses, it was concluded that TCP-25 requires a noninteracting carrier formulation, such as HEC/HPC, enabling the peptide’s direct interactions with target bacteria and host cells.

Limitations of this study include that the work has been performed in murine and porcine models with selected types of bacteria. Although, in particular, porcine models have a recognized translational potential, studies on human wounds are needed to evaluate the clinical potential of TCP-25 gel. Such future studies will also address the efficacy of the drug concept in relation to factors such as comorbidities, wound type, and the diverse bacterial wound flora often present. It is generally believed, and also supported by recent evidence (56), that bacterial resistance to HDPs is more difficult to acquire than for antibiotics. However, such concerns should be addressed in future preclinical and clinical studies.

In conclusion, we have developed a TCP-25–containing hydrogel with a dual action that targets both bacterial infection and the associated TLR-driven inflammatory component in vitro and in experimental wound models. Proteolytic degradation of the TCP-25 peptide generates bioactive TCP fragments that are similar to those that normally occur in wounds. The TCP-25 hydrogel mimics the endogenous generation of redundant and transiently acting HDPs in wounds and may, therefore, be of potential interest in the clinical development of new wound treatments for the prevention and reduction of infection and its harmful consequences.

MATERIALS AND METHODS

Study design

The goal of this study was to develop a hydrogel formulation functionalized with TCP-25 as a wound healing treatment to target bacterial load and the associated PAMP-induced inflammation. Initial in vitro studies using pharmacologically approved formulation components determined the prerequisites for TCP-25–mediated action, which depends on simultaneous interactions with bacterial membranes, LPS, and CD14 (33). Using various in vitro assays, the selected TCP-25 gel was rigorously tested for efficacy against the Gram-positive S. aureus, Gram-negative P. aeruginosa, and various other clinical bacterial isolates. The dual antimicrobial and anti-inflammatory action of the TCP-25 gel was demonstrated in experimental mouse models of subcutaneous S. aureus and P. aeruginosa infection and in NF-κB reporter mouse models of endotoxin-induced inflammation. Efficacy of the TCP-25 gel was shown in preclinical porcine partial thickness wound infection models. Pharmacokinetics of TCP-25 in the hydrogel was investigated in vitro, ex vivo, and in vivo using fluorescence spectrometry, IVIS bioimaging, and mass spectrometry analyses. To study the fate of active compound in the hydrogel, degradation of TCP-25 was analyzed by mass spectrometry. Bioactivity of major TCP-25 fragments was demonstrated by in vitro assays. In addition, stability of TCP-25 in gel after long-term storage and in plasma was analyzed by mass spectrometry. Last, efficacy of TCP-25 gel treatment was compared with clinically used wound treatments in a preclinical porcine partial thickness wound model. To further demonstrate the clinical translation, the effect of TCP-25 on the proinflammatory actions of wound fluids from the above porcine infected wounds, as well from patients with nonhealing wounds colonized by S. aureus and P. aeruginosa, was evaluated using monocyte models.

Ethics statement

All animal experiments were performed according to Swedish Animal Welfare Act SFS 1988:534 and were approved by the Animal Ethics Committee of Malmö/Lund, Sweden (permit numbers M252-11, M131-16, M88-91/14, M5934-19, and 8871-19). The use of human wound materials was approved by the Ethics Committee at Lund University (LU 708-01 and LU 509-01).

Statistical analysis

All microbiological and cell culture–based assays show biological replicates and were repeated at least three times. Data are presented as means ± SEM. Clinical scoring of wounds is presented as medians. Differences in the mean between two groups were analyzed using Student’s t test for normally distributed data and Mann-Whitney U test otherwise. To compare means between more than two groups, a one-way analysis of variance (ANOVA) with post hoc (Tukey) for normally distributed data or Kruskal-Wallis test with post hoc (Dunn’s) was used otherwise. Statistical analysis, as indicated in each figure legend, was performed using GraphPad Prism software v8. P < 0.05 was considered to be statistically significant. Individual subject-level data are reported in data file S1.

SUPPLEMENTARY MATERIALS

stm.sciencemag.org/cgi/content/full/12/524/eaax6601/DC1

Materials and Methods

Fig. S1. Comparison of TCP-25 formulation in HPC with the related polymer HEC.

Fig. S2. Dose-response studies of TCP-25 gel formulations.

Fig. S3. Antimicrobial activity of TCP-25 HEC formulation against clinical isolates and reference strains.

Fig. S4. In vivo antibacterial effects of TCP-25 HEC gel in a mouse model of subcutaneous infection.

Fig. S5. In vivo antibacterial effects of TCP-25 HEC gel in a mouse model of prevention of infection.

Fig. S6. In vitro activity of TCP-25 against P. aeruginosa isolated from minipig wounds with secondary infection.

Fig. S7. Effects of TCP-25 gel in a porcine partial thickness wound model of superinfection.

Fig. S8. In vitro release and in vivo pharmacokinetics of TCP-25 gel.

Fig. S9. Partial thickness wound on minipig skin.

Fig. S10. Rheological properties of TCP-25 gel.

Fig. S11. Stability, activity, and release profile of the TCP-25 HEC formulation.

Fig. S12. Stability of TCP-25 in human, minipig, and mouse plasma.

Fig. S13. Activity of the silver-containing dressing Mepilex Ag against S. aureus.

Fig. S14. Cytokine analysis of wound fluid collected on days 2 and 3.

Fig. S15. IL-1β analysis of the wound fluid collected on days 2, 3, and 5 from the animal model of established infection.

Fig. S16. Single-dose toxicity study in mice.

Fig. S17. Schematic figure illustrating the dual function of TCP-25 gel in the wound environment.

Table S1. MIC values for TCP-25 against clinical isolates and reference strains.

Table S2. MIC values of TCP-25 against multidrug-resistant bacteria.

Table S3. Clinical scoring of minipig wounds.

Data file S1. Individual subject-level data.

REFERENCES AND NOTES

Acknowledgments: We acknowledge L. Nilsson (Associate Professor, Food Technology, Lund University) for rheological analyses, K. Riesbeck (Professor, Clinical Microbiology, Lund University) for identification of porcine P. aeruginosa bacteria, and A. Ström (Medical Statistics and Epidemiology, Skane University Hospital) for a thorough review of statistical methods. We thank T. Gustavsson, Å. Andersson, and E. Eriksson (BMC, Lund University) for assistance in the animal studies and S. Strömblad (Bioimaging Center, Lund University) for histological support. We thank L. B. Madsen, K. Erneholm, and J. C. Bendel Møller (Timeline Bioresearch, Lund) for excellent support during the minipig studies. Funding: This work was supported by grants from Alfred Österlunds Foundation, Edvard Welanders Stiftelse and Finsenstiftelsen, The Knut and Alice Wallenberg Foundation, Thelma Zoégas Foundation, the Royal Physiographic Society, the Swedish Strategic Research Foundation, Vinnova, the Swedish Government Funds for Clinical Research (ALF), the Novo Nordisk Foundation (Novo Seeds NNF17OC0030158), and the Swedish Research Council (2012-1883, 2017-02341, and 2018-05916). Author contributions: M.P. and A.S. conceptualized and designed the study. M.P. conducted the in vivo experiments and analyzed the data. M.P., M.B., J.P., A-C.S., M.Å.A., and S.K. performed the in vitro experiments and analyzed the data. M.P. and A.S. wrote the manuscript. All co-authors revised the manuscript. Competing interests: A.S. is a shareholder and board member of in2cure AB, a company developing anti-inflammatory peptides for therapeutic applications. M.P. has received consultancy fees from in2cure AB. M.Å.A. is an employee of in2cure AB. M.B. has worked for in2cure AB in 2018. The peptide TCP-25, and variants thereof, is patent protected. EP1987056B1, Novel antimicrobial peptides and use thereof, A.S.; US8076286B2, Antimicrobial peptides and use thereof, A.S.; EP2480567B1, Thrombin C-terminal polypeptides and uses thereof for treating inflammation or coagulation disorders, A.S.; US8735353B2, Polypeptides and uses thereof, A.S. All other authors declare that they have no competing interests. Data and materials availability: All data associated with this study are present in the paper or the Supplementary Materials.

Stay Connected to Science Translational Medicine

Navigate This Article