Antimicrobial peptides: ancient molecules as modern therapeutics?

Antimicrobial peptides: ancient weapons of innate defense

We live in a world populated by an enormous quantity of potentially harmful microorganisms. Our body surface represents the first barrier against these microbiological threats and, with the ability of bacteria to double their number within 20 min under optimal conditions, it is a remarkable phenomenon that our epithelia are usually not infected. This implies the existence of a sophisticated epithelial defense system, which allows efficient resistance against harmful microbes while allowing the presence of the normal beneficial epithelial flora. Although for a long time the physical barrier was believed to represent the main component that protects body surfaces from infection, extensive research in the last 20 years has revealed the existence of a chemical defense system based on the production of antimicrobial peptides (AMPs). AMPs represent a diverse group of small, mainly cationic, endogenous proteins that exhibit antimicrobial activity against bacteria, fungi and viruses. The biological significance of AMPs is reflected by the wide distribution of these molecules throughout the plant and animal kingdom [1]. The Antimicrobial Sequences Database [101] and the Antimicrobial Peptide Database [102] indicate the abundance and diversity of AMPs. Organisms such as plants and invertebrates, which do not have an adaptive immune system, rely on the production of AMPs to fend off harmful microbes. It is likely that all multicellular organisms are capable of producing AMPs.

Approximately 20 years ago, Michael Zasloff isolated from the skin of the African clawed frog, Xenopus laevis, a broad spectrum AMP termed ‘magainin’ [2]. This was the first report indicating that the skin can function as an initial barrier against microorganisms by the production of AMPs. Extensive research in the last 20 years has also shown that the human skin and other epithelial tissues, as well as various leukocytes, are able to produce a wide variety of AMPs. The role of AMPs to protect us from infection is emerging, and a role for AMPs in innate immunity is only beginning to be elucidated. Since there is increasing evidence that AMPs, such as the defensins or cathelicidins, exhibit many immunomodulatory functions, such as recruitment and activation of antigen-presenting cells, it can be speculated that the antimicrobial activity of AMPs may not be the only function of AMPs [3]. However, several reports using mice models or overexpression experiments have provided evidence that AMPs play a crucial role in host defense and are able to control the growth of pathogens [4]. Examples are now discussed.

AMPs: key players of cutaneous defense?

There is increasing evidence that several AMPs play an important role in cutaneous defense. Examples are the human β-defensins (hBD)-2 and hBD-3, psoriasin, RNase 7, the cathelicidin LL-37 and dermcidin. Interestingly, many of them are upregulated in keratinocytes upon contact with microorganisms or microbial products. This indicates that keratinocytes do not merely function as a physical epidermal barrier, but rather are active participants in innate defense by the recognition of microorganisms and subsequent initiation of innate immune responses, such as the rapid induction of AMPs. There is also evidence that keratinocytes can distinguish between commensal and pathogenic bacteria because they activate different signaling pathways in keratinocytes [5].

The role of AMPs in cutaneous biology may be reflected by the increasing numbers of publications showing an association of AMPs with skin diseases. For example, psoriatic skin is characterized by a high expression of various AMPs, such as the β-defensins and psoriasin, which may be locally induced by microorganisms and endogenous proinflammatory cytokines [6]. The abundance of AMPs in psoriatic skin offers an explanation as to why patients with psoriasis suffer fewer cutaneous bacterial infections than expected [7]. In contrast to psoriasis, there is less induction of AMP such as the β-defensins and LL-37 in the common chronic inflammatory disease atopic dermatitis [8–10]. Possible reasons may include the lack of major AMP inducers in atopic dermatitis skin, such as the cytokines IL-1 and IL-22, as well as the suppression of AMP induction by elevated levels of Th2-cytokines, such as IL-4, IL-10 and IL-13. The reduced induction of AMPs in atopic dermatitis may contribute to the increased susceptibility of atopic dermatitis skin to Staphylococcus aureus infection [11].

An interesting hypothesis is that individuals who express lower levels of AMPs are more susceptible to infectious and inflammatory diseases. It is known that the α- and β-defensin gene cluster on chromosome 8 has gene-copy number polymorphisms [12,13]. In addition, gene-expression studies have revealed a correlation between the genomic copy number and the expression level of β-defensins, suggesting that the defensin copy-number polymorphism influences innate immunity [12]. In concordance with this hypothesis, Fellermann et al. recently reported that a low hBD-2 gene copy number predisposes to Crohn's disease of the colon [14]. Whether patients with atopic dermatitis or other recurrent inflammatory and infectious skin diseases carry a low AMP copy number will need to be investigated. A surprising observation has been reported recently by Hollox and colleagues in which they detected an association between psoriasis and an increased β-defensin genomic copy number [15]. As discussed previously, one would assume that a high β-defensin copy number should have beneficial effects. In psoriasis, the beneficial effect of a high β-defensin copy number may be reflected by a lower infection rate. However, it is known that hBD-2, hBD-3 and hBD-4 can stimulate keratinocytes to release IL-8, IL-18 and IL-20, proinflammatory cytokines that have an established role in the etiology of psoriasis. Based on this data, Hollox et al. speculated that, upon minor skin injury, infection or some other environmental trigger, either high basal levels or induced levels of β-defensins can generate an inappropriate inflammatory response, which contributes to the clinical symptoms associated with psoriasis.

An inappropriate AMP-mediated induction of an inflammatory response has also been recently suggested for the inflammatory skin disease rosacea. It has been reported that abnormally increased levels of proinflammatory cathelicidin peptides are present in the facial skin of patients suffering from rosacea [16]. In this study, evidence was provided that elevated levels of the serine protease stratum corneum tryptic enzyme (SCTE; also known as kallikrein 5) in rosacea lesions contribute to this disease because this enzyme is able to process the cathelicidin precursor hCAP18, thus generating the rosacea-specific cathelicidin peptide profile. Subcutaneous injection of these rosacea-specific cathelicidin peptides in mice induced cutaneous inflammation similar to that observed in rosacea patients. These data indicate that cathelicidin peptides may participate in the pathogenesis of rosacea and that the balance between cathelicidin and SCTE is disturbed in rosacea [16].

An important role for skin-derived AMPs in wounds and burns has also been reported. There is increasing evidence that AMPs are induced upon wounding and can promote wound healing. For example, the cutaneous expression of hBD-2 is upregulated after injury and in chronic wounds [17]. LL-37 was also shown to be upregulated after cutaneous injury due to synthesis within epidermal keratinocytes and deposition from infiltrated granulocytes [18]. Increased levels of psoriasin have been detected in keratinocytes surrounding the wound, as well as in wound exudate [19]. Sterile wounding of human skin resulted in induced hBD-3 expression through activation of the EGF receptor (EGFR). Activation of the EGFR generated antimicrobial concentrations of hBD-3 and increased the activity of organotypic epidermal cultures against S. aureus, indicating that sterile wounding initiates an innate immune response that increases resistance to infection and microbial colonization [20]. In contrast to the upregulation of AMPs upon wounding and injury, some studies suggest that decreased AMP levels found in burn wounds and burn blister fluids may promote infection [21]. These data indicate a link between burn wounds and a defect in host defense, and further suggest a potential therapeutic role for AMPs in the management of burn wounds [21]. Indeed, a transient cutaneous adenoviral gene therapy with the cathelicidin LL-37 proved to be effective for the treatment of burn wound infections [22]. In addition, decreased levels of LL-37 have been shown to be associated with chronic wounds, such as chronic leg ulcers. Using neutralizing-antibodies to LL-37, Heilborn et al. demonstrated that LL-37 promotes the re-epithelialization of human skin wounds [23]. The authors speculate that reduction of LL-37 in chronic wounds impairs re-epithelialization and may contribute to their failure to heal.

AMPs as potential drugs

Many studies using animal models have shown that an overexpression of AMPs offers an increased protection against infection. For example, mice transgenic for the gut-derived human-α-defensin-5 are resistant to infection and systemic disease from orally administered Salmonella typhimurium[24]. Overexpression of the cathelicidin LL-37 in cystic fibrosis xenografts increased the antimicrobial activity of airway surface fluid [25]. Gene-therapy studies have also shown that keratinocytes overexpressing the β-defensins hBD-2, hBD-3 or hBD-4, as well as cathelicidins, exhibit increased antibacterial activity [26–29]. These studies suggest that it can be possible to increase cutaneous defense using genetically modified keratinocytes. The use of AMPs may offer a promising alternative to conventional antibiotics in chronic and burn wounds. For example, to date, several bioengineered skin substitutes are available for wound closure. The use of such skin replacements is often associated with frequent topical administration of high doses of antibiotics to prevent microbial infection. Unfortunately, the frequent use of common antibiotics contributes to the development of resistant microbes. Furthermore, conventional antibiotics cannot be used in gene therapeutic approaches because they are not encoded by genes. The endogenous overexpression of AMPs in skin replacements may offer an alternative therapeutic approach, avoiding the use of exogenous antibiotics. In this issue of Expert Review of Dermatology, Supp and Neely provide a detailed review regarding the potential use of AMP-gene therapy in human skin replacements to combat wound infection. Another potential approach to AMP use would be direct exogenous cutaneous application. However, compared with an exogenous topical application of AMPs, the use of a cutaneous AMP-gene therapy approach has the advantage of a continuous administration and low-cost production.

However, several important issues must be addressed when considering human AMPs as novel therapeutic agents. For example, will overuse or artificial overexpression of AMPs provoke the spread of bacteria with increasing resistance against human AMPs? One could argue that acquisition of complete resistance is unlikely because AMPs have been protecting our ancestors for millions of years and are still active against many bacteria [30]. This phenomenon is surprising because AMPs are genetically encoded and the mutation rate in bacteria is much greater than the rate of adaptive mutations in mammals. This suggests that the mutations that could render resistance have already occurred and, therefore, the use of human AMPs would not elicit a new selective pressure. However, it has been shown that some bacteria have developed various strategies to decrease their susceptibility to AMPs. These strategies include reducing the negative charge of the bacterial surface, degradation of AMPs through bacterial proteases and elimination of AMPs through efflux pumps [31]. Nevertheless, these modifications do not result in complete resistance to AMPs, and even bacteria capable of using this strategy remain susceptible to high concentrations of AMPs [31].

Overexpression of AMPs in skin replacements seems to be a smart strategy because it imitates the upregulation of AMP expression in vivo at sites of infection and injury. On the other hand, genetically induced overexpression of a single AMP still represents an artificial situation that does not reflect what occurs in vivo, where a wide variety of AMPs may act in synergy to control microbial growth. The synergistic activity of AMPs in vivo together with the use of different killing mechanisms may hamper the ability of bacteria to develop resistance to AMPs. Since our understanding about the induction and regulation of AMPs at sites of infection is limited, even an overexpression of several AMPs in parallel would still represent an artificial situation since probably hundreds of molecules actually get activated in vivo upon infection and can act in synergy to fend off pathogens. As mentioned previously, rosacea is one example where an abnormal increased expression and processing of AMPs may be a triggering factor for the development of this disease. This indicates that the biology of AMPs underlies a complex regulation whose imbalance may trigger undesired effects, such as inflammation. Therefore, the administration or artificial overexpression of high doses of AMPs in their native form may lead to undesirable side effects. Indeed, in a mouse model, it has been shown that an intratracheal instillation of defensins causes acute lung inflammation and dysfunction [32].

An alternative promising strategy for the treatment or prophylaxis of cutaneous or other epithelial infections is to specifically induce endogenous AMP synthesis. Interestingly, bacteria of the commensal flora, as well as probiotic bacteria, are able to induce AMPs [5,33]. These bacteria do not normally induce inflammation, suggesting that conditions may exist that cause AMP induction without inflammation. Therefore, it would be interesting to identify microbial products that solely induce AMP expression without inducing inflammatory reactions. Such compounds, when topically applied, could initiate epithelial induction of AMPs locally in the appropriate place and within the appropriate cellular compartment, and increase resistance to infection. In addition, the development of bacterial resistance would be unlikely because this strategy resembles what occurs in vivo when microorganisms induce the local production of AMPs. Evidence that the specific induction of AMPs can help to prevent infection has been recently provided in a study reporting that oral butyrate treatment of rabbits induced the expression of the cathelicidin cell adhesion promoting (CAP)-18 in the colonic epithelium and promoted elimination of Shigella[34].

Conclusion

The high number of human AMPs and their potent antimicrobial activity suggests that these proteins play a major part in innate immunity. The constitutive, as well as inducible, production of AMPs offer a rapid response against invading microorganisms and I believe that AMPs are crucial for human survival. The discovery of AMPs provides an opportunity to develop improved antimicrobial therapies based on ‘natural antibiotics’ designed during evolution to effectively fend off infection. It is, therefore, a future challenge to understand the role of AMPs in infectious diseases and to evaluate their use as a novel class of antibiotics. However, the problem of the development of AMP resistance by pathogens must be critically addressed before AMPs can be used in clinical practice. If an uncontrolled overuse of AMPs resulted in the emergence of AMP-resistant microbes, we would destroy an important innate defense mechanism that has been protecting us for thousands of years against microbial threats. This would really be a frightening scenario.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.