Sounding the Alarm: Multiple Functions of Host Defense Peptides

Abstract

The capacity of the skin and other organs to resist infection depends on the innate production of molecules known as antimicrobial peptides. Emerging evidence suggests that some of these peptides are important to immune defense by acting not only as natural antibiotics but also as cell-signaling molecules. In this issue Carretero et al. (2007) expand on these findings by demonstrating that expression of human cathelicidin alters multiple signaling pathways in a keratinocyte cell line and enhances wound re-epithelialization in ob/ob mice.

Microbes are a formidable adversary for all epithelial barriers and have evolved a wide variety of mechanisms to invade the skin and induce disease. These virulence factors necessitate that the host use multiple defense strategies for protection. The physical barrier provided by the epidermis is essential, but many microbes have evolved effective systems to breach the epidermis. Similarly, the recruitment of circulating immune effector cells is also an absolute requirement for adequate protection against infection, but many pathogens have evolved systems to evade the cellular immune system. In addition, all microbes proliferate too rapidly to be effectively controlled by the relatively slow inflammatory process alone. The answer to this problem is the relatively recently recognized systems for the generation of antimicrobial molecules by the epidermis. In particular, the production of peptides with antimicrobial activity has been shown to be essential for normal defense against microbial infection, providing a rapid, first-line chemical barrier to inhibit microbial growth (Zasloff, 2002).

Research such as that described by Carretero et al. (2007) is expanding our understanding of antimicrobial peptides (AMPs) to show that some of these host-defense peptides exert their effect not only through their capacity to act as an antibiotic but also by their ability to alert the host and stimulate many elements of the defense system, including barrier repair and inflammatory cell recruitment. These activities have inspired the use of the term “alarmins” (Oppenheim and Yang, 2005) to describe some AMPs.

AMPs were first described in the 1970s and 1980s as gene-encoded molecules responsible for disease resistance in plants and insects. Later, with the identification of defensins in neutrophil granules, it became clear that similar peptides might also be important to the mammalian immune system. As the more complex immune defense systems of mammals have been described, the apparent role of AMPs in host defense has continued to expand. In fact, our initial discovery of AMPs in the skin was made not because of their capacity to kill microbes but because of the ability of some of these peptides to induce the expression of syndecan-1 and -4 in fibroblasts (Gallo et al., 1994). This event is critical to the wound-repair process.

Overwhelming evidence now exists that the AMPs have many distinct and complementary functions. When these functions are disrupted by abnormal expression, human disease can result. Examples of this include atopic dermatitis, which shows decreased expression of multiple AMPs and an increased susceptibility to infections (Ong et al., 2002), and rosacea, which has excessive and abnormally processed cathelicidin peptides that will reproduce elements of the disease in mice (Yamasaki et al., 2007) How these processes are regulated is an area of ongoing investigation. However, progress has been made in understanding how specific AMPs can influence host-cell responses. The human cathelicidin peptide LL-37 has been shown to activate mitogen-activated protein kinase (MAPK) and extracellular signal-related kinase in epithelial cells (Tjabringa et al., 2003), and blocking antibodies to LL-37 hinder wound repair in human skin equivalents (Heilborn et al., 2003). In this issue Carretero et al. (2007) provide further evidence that cathelicidin participates in the process of epithelialization by demonstrating that overexpression accelerates migration of HaCat in vitro and mouse skin re-epithelialization in vivo. Adenoviral-mediated expression of this cathelicidin altered multiple signaling pathways, including both MAPK and PI-3K. Such effects are consistent with observations in many experimental systems that have shown that AMPs can trigger a set of events specific to the cell type of interest. The mechanism responsible for these responses is unclear. As illustrated in Figure 1, there are data to support several systems by which AMPs could stimulate host-cell responses. These include binding to a specific receptor, transactivation of a receptor, or specific changes to the cell membrane that result in an alteration in the overall pattern of activity of membrane-associated signaling molecules. The latter hypothesis implies that responses to AMPs are controlled in part by the composition of the cell membrane (Di Nardo et al., 2007).

Figure 1. Alternative models for cell activation by antimicrobial peptides Antimicrobial peptides such as LL-37 are typically small and cationic and can interact with hydrophobic membranes.

Three major mechanisms have been proposed to explain how LL-37 activates mammalian cells. In the transactivation model, LL-37 stimulates the release of a membrane-bound growth factor. This then binds its high-affinity receptor and activates it. In the receptor-binding model, LL-37 serves as a surrogate ligand for a specific receptor. This direct binding initiates receptor activation. In the receptor-activation model, LL-37 associates with and modifies the membrane containing the receptor. This membrane activity indirectly results in a change in receptor function such that it can signal without a ligand or becomes insensitive to binding by its specific ligand. Evidence to support the transactivation model has been reported in keratinocytes and pulmonary epithelial cells; receptor binding is supported in monocytes and endothelia, and receptor activation in dendritic cells.

Although it is clear that the AMPs have an important role to play in skin biology, progress in understanding the functions of AMPs and in developing therapies based on these naturally occurring molecules has been hindered by an unfortunate tendency to lump the many diverse peptides and proteins with antimicrobial activity into a single conceptual category. Unfortunately, because of this tendency, it is frequently and incorrectly assumed that a function seen in one peptide should be applied to another. It is inappropriate to compare these greatly diverse peptides to make conclusions regarding their relative role or to expect that a drug design based on the ability of an AMP to kill bacteria will retain its important effects on the host. In fact, confusion regarding the native peptides is common. For example, the important and excellent work by Carretero et al. in this issue is somewhat limited by their choice of an expression vector that was designed to overexpress the full-length precursor protein of cathelicidin (hCAP18). Unlike what is implied by the title of this article, it is not clear whether the peptide LL-37 is actually present and responsible for the effects observed. hCAP18 can be enzymatically processed into multiple alternative forms with different functions (Braff et al., 2005b).

We now know of more than 20 AMPs in the skin, including cathelicidins, β-defensins, Substance P, RANTES, RNase 2,3,7, S100A7, and several others (Braff et al., 2005a). Many peptides have antimicrobial action against bacteria, viruses, and fungi. Some may play a specific role against certain microbes in normal skin, whereas others act only when the skin is injured and the physical barrier disrupted. Some other peptides may play a larger role to signal host responses through chemotactic, angiogenic, growth factor, and immunosuppressive activity. The full functions of these evolutionarily ancient classes of molecules are slowly being uncovered. Stay tuned.