ORIGINAL RESEARCH
Developing and evaluating the effectiveness of wound-healing compounds based on cationic peptides and fullerene
1 National Research Center — Institute of Immunology of the Federal Medical Biological Agency, Moscow, Russia
2 Federal Research Centre "Fundamentals of Biotechnology" of the Russian Academy of Sciences, Moscow, Russia
3 Pirogov Russian National Research Medical University, Moscow, Russia
Correspondence should be addressed: Anastasiia А. Galkina
Kashirskoe sh., 24, Moscow, 115522, Russia; moc.liamg@aniklag.a.aisatsana
Funding: the study was conducted as part of the State Assignment of FMBA “Fullerene-21” (code 612.017.1:616.9).
Author contribution: Galkina AA — setting up methodology, laboratory tests, manuscript writing; Bolyakina DK — laboratory tests; Shatilova AV, Shatilov AA, Babikhina MO — laboratory tests; Golomidova AK — setting up methodology; Shershakova NN — setting up methodology, study concept, manuscript writing, research supervision; Andreev SM — setting up methodology, study concept, research supervision; Khaitov MR — study concept, research supervision.
Compliance with ethical standards: the study was approved by the Bioethics Commission of the Institute of Immunology of FMBA of Russia (order № 102 of November 2015) and conducted in accordance with the Directive 2010/63/EU for animal experiments and the Regulations Regarding Research Involving Laboratory Animals in the National Research Center — Institute of Immunology of FMBA of Russia.
Skin wound healing is a complex process involving cells of various types and multiple regulatory factors, and abnormal wound healing can result in scarring and the transition from acute to chronic inflammation. Scars can restrict movements, cause pain and itching; scars can be the cause of physiological stress in case they remain visible and cannot be hidden by clothing or makeup, thereby seriously affecting the person’s self-confidence and quality of life. Removal and treatment of scars remain a pressing issue, since, despite the diversity of available treatment methods, the methods’ efficacy is quite low. The total volume of the global market in this field was 19.6 billion dollars in 2019. It is expected that it will grow by 11.5% in the next decade [1].
Comorbidities, such as diabetes mellitus, hypertension, and other vascular and autoimmune disorders, can hinder effective treatment of the wound healing process complications [2]. Bacterial superinfections play an important role in chronification of inflammation.
All wounds are to some extent contaminated with microorganisms being a part of saprophytic skin microflora. The type and abundance of such microorganisms vary depending on the wound type [3]. Pseudomonas aeruginosa, Staphylococcus aureus, Klebsiella pneumoniae, Enterococcus faecalis, and Acinetobacter baumannii are the most common bacterial species causing wound infection [4]. It is well-known that wound infections constitute a third of cases of nosocomial infections in surgical patients and cause 70–80% of deaths from wounds [5]. Chronic wounds affect the patients’ quality of life, along with the increased morbidity and mortality, and represent a huge financial burden on health systems all over the world, since such wounds are associated with the expenses for long-term hospital stay, diagnostic tests, antibiotics, and sometimes invasive surgery [6, 7]. That is why wound healing is a serious health problem that requires the development of safe and effective therapeutic agents.
The drugs for wound healing are based on adsorbents, anti-inflammatory components, antibiotics or dexpanthenol that stimulates regeneration processes. The search for new approaches to healing wounds complicated by bacterial infection has become a pressing issue due to the alarming increase in resistance to conventional antimicrobial drugs. Among alternative antimicrobial agents, special attention is paid to cationic antimicrobial peptides (CAMPs) [8].
Cationic peptides (CPs) attract much attention as transporters and biologically active substances due to high affinity for cell membranes and specific structure enabling mass spectrometry analysis and wide possibilities to design the variety of such molecules. CPs are widespread in nature and found in all mammals, especially in the skin, where they play a protective role against pathogenic microorganisms. Cationic antimicrobial peptides, or host defense peptides, are a heterogeneous group of short positively charged peptides, mostly amphiphilic. These peptides are secreted by immune (for example, neutrophils and macrophages [9]) and epithelial cells of vertebrates and invertebrates to ensure protection against microbial invasion [10]. The shortcomings of using peptides as potential antimicrobial drugs for treatment include a very complex structure that hampers synthesis and the CPs’ proteolytic instability. It has been assumed, that the CAMP conjugation with other biologically active molecules, such as other peptides, polypeptides, proteins, and antibiotics in general, can contribute to improvement of antimicrobial properties and provide the basis for the development of drugs possessing multiple biological activities. In particular, the development of drugs with low toxicity combining antibacterial and anti-inflammatory effects and showing low probability of developing resistance would contribute significantly to both fundamental research and practical healthcare.
Since CAMPs show proteolytic instability, we believe that it seems promising to use such molecules, as fullerene C60, as carriers for peptides.
Fullerene C60 is a carbon-based molecule that has a shape of truncated icosahedron and possesses strong antioxidant activity. It is well-known that the fullerene C60 water-soluble forms have multiple biological effects, including antiviral, anti-inflammatory, anti-allergy, and regenerative effects [11, 12]. An in vivo model of wound healing process has shown that some covalent fullerene C60 derivatives accelerate wound healing and prevent inflammatory cell infiltration [13].
It is known that fullerene C60 is insoluble in aqueous media, which is clearly a significant obstacle to its extensive use in medicine. We have earlier developed a unique scalable technology to obtain stable aqueous dispersion of fullerene C60, enabling assessment of its biological activity [14]. It should be noted that this technology does not involve the use of organic solvents, ultrasonic processing or heating, which ensures bio-compatibility and safety of the resulting solution. Our method makes it possible to generate highly concentrated stable AFD with the concentration of at least 1 g/l.
The main purpose of the study was to assess the woundhealing activity of the AFD-based ointment composition in the experimental model of wound inflammation, as well as to estimate the effects of AFD on the expression of marker genes involved in regeneration, analyze the CP antibacterial activity, and assess the prospects of creating the CP- and fullerene C60-based complexes for treatment of wounds.
METHODS
AFD-based composition
The AFD-containing ointment (composition: AFD, vaseline, sucrose palmitate in the ratio of 40 : 36 : 24 (by weight)) for in vivo treatment of wounds was prepared using the IKA 25 digital homogenizer IKA 25 as a mixing device (“AFD ointment”). The fullerene C60 aqueous dispersion was acquired by the dialysis method [14]. This bio-compatible method does not involve the use of toxic organic solvents, ultrasonic processing or heating. At the same time, it ensures high yield of fullerene C60 during transition from crystals into solution (the concentration of sterile fullerene C60 solution was 1 mg/mL). The hydrodynamic particle size determined by the dynamic light scattering method was 100–200 nm.
In vivo model of wound inflammation
Female BALB/c mice aged 4–6 weeks (Stolbovaya breeding nursery; Moscow, Russia) were used to model the wound healing process [15]. The following housing conditions were used for animals: ambient temperature 18–26 °С; automatic 12 h ligh/dark cycle; relative humidity 30–70%. All animals had unlimited access to drinking water and food. The animals were anesthetized with 4% isoflurane solution for 2 min via airways and topically administered 0.5% lidocaine solution prior to making the incised wounds. To model the surgical wound, a skin fragment (1 × 1 cm) was excised from the back of the BALB/c mouse. The AFD ointment was applied to the wound surface (40 µg of С60/mouse) 24 h after the surgical procedure (“AFD ointment” group). The widely used therapeutic agent, cream for treatment of surgical wounds, was used as positive control (“C+” group). The ointment containing phosphate buffered saline (PBS) instead of AFD was used as negative control (“PBS” group). The “intact” group (with no skin wounds) was also used as negative control. The listed above preparations were used once a day for 11 days. On day 12 mice were euthanized by cervical dislocation, and skin samples were collected for quantitative RT-PCR.
Assessment of wound healing efficiency
The skin wound healing rate was assessed by measuring the wound area in the longitudinal and transverse directions (mm) every day and calculating the wound area using the following formula:
Sel = πab,
where Sel — area of an ellipse, a — semi-major axis (half of the longer diameter or transverse dimension), b — semiaxis (half of the shorter diameter or longitudinal dimension).
The wound healing efficiency (Х) was calculated as percentage using the following formula:
X = (1 – Sf/SI) × 100%,
где Sf — final wound area, SI — initial wound area [16].
Real-time polymerase chain reaction
Total RNA was extracted from skin samples using the RNeasy Mini Kit (Qiagen, Courtaboeuf; France) in accordance with the manufacturer’s instructions; cDNA was synthesized using the Reverta-L kit (InterLabService; Russia). The reverse transcription product was amplified by the real-time polymerase chain reaction (RT-PCR) using the iCycler iQ real-time PCR detection system (Bio-Rad Laboratories; USA) and the PCR Mix kit (Syntol; Russia).
The calculations to determine relative gene expression were performed using the comparative Ct method (ΔCt) against mHPRT.
Relative quantification by RT-PCR was used to detect changes in the expression of target genes relative to the reference gene represented by murine hprt gene. Quantitative PCR results for mRNA expression were compared as ΔCt values calculated using the following formula: ratio (reference/ target) = 2Ct(hprt)–Ct(target gene) [17].
Assessment of antibacterial effects of AFD and cationic peptides
Antibacterial activity of AFD and the synthesized peptides was assessed in vitro on the example of the E. coli Dh5α strain by colony counting relative to the well-known antibiotic ampicillin selected as positive control. In the method the bacterial suspension was incubated with various CAMP concentrations in the LB liquid medium for 4 h at 37 °С, then it was applied dropwise to the surface of the dried agar medium. The culture was incubated overnight at 37 °С.
It is important to note that the abovementioned E. сoli strain is not pathogenic and shows no resistance to antibiotics. Activity of peptides against the selected strain was assessed based on the determined minimum bactericidal concentration.
Cationic peptide synthesis
Peptides were synthesized by the solid-phase method in the automated PS3 Peptide Synthesizer (Gyros Protein Technologies Inc.; USA) according to the Fmoc-chemistry protocol using the N-hydroxybenzotriazole and diisopropylcarbodiimide (HOBt/DIC) mixture as a condensing agent. The starting Fmoc-aminoacyl polymers and the Rink Amide СhemMatrix gel-type resin were used for synthesis. The side carboxyl and hydroxyl groups of amino acids were protected by the tert-butyl group (t-Bu), the lysine Ɛ-amino group by Boc, the cysteine SH-group by Trt, the arginine guanidinium functional group by Pbf, and the carboxyl and hydroxyl groups of amino acids by tertbutyl ethers. The standard cycle included washing (DMF), removal of Fmoc protection (20% 4-methylpiperidine in DMF), preliminary Fmoc amino acid (DIC/HOBt) activation and condensation on the DMF/N-methylpyrrolidone medium with the twofold carboxyl component excess (~0.5–1 h). The extent of the reaction was controlled using the Kaiser Test (ninhydrin test), the condensation reaction was repeated when necessary (0.5 h). The terminal peptides were cleaved from the polymer with trifluoroacetic acid in the presence of scavengers (triisopropylsilane, ethanediol, water, dimethyl sulfide). The raw product was precipitated with the dry methyl tert-butyl ether, then the peptide was extracted with the acetic acid aqueous solution, and the extract was lyophilized (VirTis AdVantage 2.0 EL freeze dryer; SP Scientific, USA). The peptides were purified by preparative HPLC chromatography (LC-20 Shimadzu; Japan) on the reversed phase column (С18) using acetonitrile — 0.1% trifluoroacetic acid aqueous solution as a mobile phase (gradient elution). The resulting peptides were tested for homogeneity by capillary zone electrophoresis in the Kapel105М system (Lumex; Russia) with photometric detection at 226 nm. Molecular weight was analyzed using the Microflex™ LT MALDI-TOF mass spectrometer (Bruker Daltonic; USA).
Statistical analysis
Statistical analysis was performed using the Statistica 8.0 software (StatSoft Inc.; USA). Significance was determined based on Student's t-test. The differences were considered significant at р < 0.05. The data were presented as mean ± standard error.
RESULTS
In vivo assessment of the AFD regenerative effect in the model of wound inflammation
Analysis of the AFD-based ointment regenerative activity relative to commercial drug (“C+”, positive control) was performed using the model of wound healing. Visual assessment of the wound healing process by measuring the lesion area was carried out individually for each mouse. The baseline average lesion area was 143.5 ± 6.1 mm2 (“before treatment” group). On the last day of the experiment the lesion area by groups was as follows: “no treatment” — 44.4 ± 6.5 mm2, “AFD ointment” — 14.8 ± 2.7 mm2, “C+” — 26.0 ± 2.6 mm2. These values show that healing of wounds treated with AFD was effective, and the healing rate was comparable with that of the positive control group and even slightly superior to the latter. Since visual assessment of healing was a subjective parameter, statistical analysis of wound area was performed in the groups that revealed significant differences between the values of the “AFD ointment” and “C+” groups, which, in turn, differed from the values of the “PBS” group. It was found that the residual wound area was the least when the AFD-based ointment was used for surgical wound treatment.
Assessment of the expression of pathogenetically significant genes
Expression of a number of genes was analyzed to assess the AFD capability of affecting the pathogenetically significant factors of regenerative process. The expression of genes in murine skin with wounds treated/not treated with AFD was determined by real-time PCR. It was shown that the expression of such proinflammatory factor, as tnfα produced in response to pathogen entry and tissue damage that stimulates local inflammatory response, was significantly lower in all experimental groups, where animals received AFD, than in the group of animals that received no therapy (“PBS”). The expression of genes encoding other pro-inflammatory cytokines, such as il6 and il1a, was also significantly lower in mice with wounds treated with AFD than in animals that received no therapy. Furthermore, we have revealed the fullerene C60 capability of enhancing the expression of HMGB1 factor that violates the collagen synthesis and can ensure scarless tissue healing observed when treating the wound with AFD.
Thus, our findings suggest that fullerene С60 can inhibit the expression of genes encoding pro-inflammatory cytokines, which results in the pro-inflammatory effect of this substance that is likely to contribute to the healing process acceleration.
Cationic peptide design
CPs are widespread in nature and produced by almost all organisms as part of the nonspecific immune system. These compounds were initially considered as potential substitutes for antibiotics, however, it was found out that the compounds had a broader spectrum of therapeutic effects, including the effects on viruses, bacteria, and microbial biofilms. The naturally occurring CPs are linear molecules consisting of up to 50 amino acids having a high share of hydrophobic and cationic residues. This makes the molecules to fold into amphipathic structures to form α-helices and β-sheets. Such peptides form specific loop conformations due to high cysteine content and disulfide bond formation. The charges of the vast majority of natural antimicrobial CPs vary between +3 and +9. The mechanism underlying their effects is associated primarily with the cell membrane damage [18]. Today, technology makes it possible to build the structures which are quite different from natural constructs in terms of topology, including the nonnaturally occurring dendrimeric structures. The amino acid sequence construction included building the construct with low toxicity, which was stable in the serum medium. Furthermore, the structure had to demonstrate high efficiency of transfection stimulation. The plan also involved building modular constructs. One of such modules is the N-terminal supercationic fragment represented by the arginine and/or lysine residues, which is essential for interaction with the nucleic acid and the cell surface. The central module is represented by the hydrophobic core consisting of the lysine residues and short hydrophobic/ amphiphilic inserts. The С-terminal module also forms a hydrophobic fragment ensuring additional affinity for cell membrane, it contains the cysteine residue with a free thiol group intended for reporter label attachment.
Hydrophobic interactions between the aliphatic chains of lipid membranes and the peptide hydrophobic residues play an important role in the mechanism underlying membrane damage, which contributes to its incorporation in the membrane bilayer through various interactions, such as pore formation. In the carpet model, the cationic peptide is directed in parallel to the cell to cover it and saturate via interaction with the outer phospholipid layer of the membrane. After the threshold value is achieved, the peptides start spinning and embed in the membrane causing its permeabilization. The branched structures, dendrimeric CPs, are of special interest. It should be noted that these show higher resistance to proteolytic enzymes along with lower toxicity compared to the linear peptides with similar amino acid composition. At the same time these show stronger binding to cells due to cooperative effects resulting from the fact that the molecule has several chains. Our early experiments demonstrated their high permeability through cell membranes enabling using such peptides as carriers for cell transfection, transfer of genes and other biologically active compounds [19].
Assessment of antibacterial effects of AFD and cationic peptides
When assessing antibacterial activity of AFD and CPs, we showed that AFD possessed no bactericidal activity and was unable to inhibit bacterial growth. Then we analyzed a number of CPs, which were expected to possess potential antibacterial activity based on their structural features. Thus, among 35 linear and dendrimeric CAMPs with the molecular weight not exceeding 4.5 kDa we had synthesized, four cationic peptides, specifically linear peptides АB-1, AB-3 and dendrimeric peptides AB-4, ST-10, to various extent inhibited growth of the E. coli Dh5a microbial culture and showed some bactericidal activity against this strain (table).
Minimum bactericidal concentrations (MBC) of the cationic peptides, i.e. the lowest concentrations killing all bacteria under standard experimental conditions, were determined by colony counting.
Thus, MBC of the AB-1 peptide was 1.15 mM. It should be noted that the AB-1 peptide concentration of 0.23 mM (five times lower compared to bactericidal concentration) showed no significant activity against this bacterial culture, which was indicative of the extremely narrow operative range of peptide concentrations (fig. 1).
The AB-3 peptide has higher bactericidal activity against the E. coli Dh5a strain than АВ-1 (fig. 2).
MBC of this peptide was 0.11 mM. Meanwhile, activity of appropriate ampicillin doses was about 6 times lower. MBC of ampicillin was 0.74 mM.
The studied AB-4 peptide exerted considerable bactericidal activity that was comparable with that of control antibiotic (ampicillin). MBC of this peptide was 0.74 mM (fig. 3).
MBC of the ST-10 dendrimeric peptide was 0.74 mM (fig. 4).
The above concentration was the least concentration that killed almost 100% of cells. It should be noted that the level of activity exerted by the ST-10 peptide was slightly higher compared to the control sample of ampicillin antibiotic. Thus, with comparable concentrations of 0.15 mM and the same E. coli dilution (1 : 10) it is evident that the number of bacterial colonies detected in the culture treated with the ST-10 peptide is considerably lower than in the culture treated with ampicillin (16 and 95, respectively). Therefore, the effect of bacterial culture growth inhibition under exposure to 0.15 mM of peptide and antibiotic was more prominent in ST-10.
DISCUSSION
In the majority of cases wounds are associated with bleeding from the damaged blood vessels and the release of inflammatory mediators, such as serotonin, histamine, vasoactive substances, and cytokines, into the surrounding tissues. Normal wound healing includes the following phases: inflammation, proliferation, maturation, and remodeling. To determine the AFD regenerative effects, we assessed the expression of a number of marker genes, such as tnfα, il-6 and il-1α, involved in regenerative process by RT-PCR. Elevated expression of these cytokines is observed during the inflammatory phase of wound healing. It is well-known that TNFα stimulates production of not only IL1, IL6, but also other pro-inflammatory cytokines [20, 21]. IL6 is one of the most important mediators of the acute phase of inflammation. It is known that delayed wound epithelization is observed in mice with IL6 deficiency. However, excess IL6 levels serve as a signal for fibroblast proliferation suppression during the late wound healing phase and lead to scar formation [22]. As for IL1a, it has been earlier shown that it stimulates collagenase production, and overexpression of this cytokine can be associated with abnormal wound healing due to collagen breakdown. The moderate il1a expression increase later mediates keratinocyte proliferation in the wound area [23]. Low levels of il1a are observed in the wound fluid from acute wounds, while fluid from surgical wounds shows elevated levels of il1a.
Thus, suppression of the tnfα, il6, and il1a expression under exposure to fullerene С60 suggests that it not only possesses anti-inflammatory effect, but is also capable of preventing wound chronization [15]. We have earlier shown antiallergy effects of AFD with inhibition of Th1 cytokines in the model of atopic dermatitis, along with the elevated expression of genes Foxp3 and FLG (filaggrin) [24]. Therefore, AFD has shown its ability to suppress inflammation associated with not only wounds, but also allergy. This makes AFD a promising compound for treatment of inflammatory skin disorders.
Wound healing is often accompanied by accession of secondary bacterial infection. It is well-known that CAMPs have a broad range of antimicrobial and immunomodulatory effects against Gram positive and Gram negative bacteria, biofilms, viruses, fungi, and parasites; CAMPs are also effective against multidrug-resistant strains. It is important to note that the likelihood of developing resistance to cationic peptides is extremely low and requires multiple mutations, including that mediating changes in the cell wall structure, due to fast bactericidal action of CAMPs and the diversity of mechanisms of action and targets [25–27]. The above makes CAMPs promising compounds for the development of the CAMP-based antibacterial drugs. Understanding of how the antimicrobial peptide properties depend on the amino acid sequence will make it possible to timely respond to the emergence of new antibiotic-resistant bacterial strains in the future due to targeted reconstruction of peptide sequences [28].
We have created the library of CAMPs supposed to possess high antimicrobial activity and have low toxicity within the framework of current research. When developing the CP panel, we relied on the databases of already known peptides and the literature data on the CP biological (antibacterial) activity. Thus, the peptide sequences were constructed considering the content of positively charged amino acids, hydrophobic amino acids. Furthermore, both linear and dendrimeric molecules were obtained when constructing the sequences.
Thus, by defining some rules for creating cationic peptides showing antibacterial activity we look forward to creating more active CPs. The AFD anti-inflammatory activity makes the idea of developing hybrid molecules based on CPs and AFD perspective.
CONCLUSIONS
As result of the research, we have developed and synthesized CPs possessing antibacterial effects. Thus, we believe that the AB-1, AB-2, AB-3, and ST-10 peptides are promising in terms of developing antimicrobial drugs on their basis. During further studies we plan to develop hybrid compounds based on the CPs and fullerene C60 to combine the anti-inflammatory and wound-healing effects with antibacterial activity. Fullerene can play a role of carrier platform for CPs. Since the area of the spherical fullerene molecule is rather large, up to 4–8 peptide molecules can be attached to it. Such multivalent structure is less prone to biodegradation, and antimicrobial activity can be increased due to cooperative effect, simultaneous attachment of several CP chains to the bacterial cell membrane. It should also be considered that fullerene C60 itself and its amino acid adducts can permeate through biological membranes.