1. Introduction
Since the discovery of the first antibiotic, penicillin [
1], public health has been significantly improved by the subsequent development of a variety of antibiotics. The discovery of streptomycin, the first antimicrobial agent effective against tuberculosis, is one of the greatest milestones in medical history [
2]. Antibiotics have significantly improved the living environment of human beings, and have cured previously incurable infections. Currently, the leading causes of death in developed countries are non-infectious diseases such as cancer, heart diseases, and cerebrovascular diseases [
3]. The development of various antibiotics has inspired the concept of “disease eradication” in the human world; however, overuse of antibiotics has caused the emergence of multidrug-resistant microorganisms. Indeed, strains of important human pathogens, such as
Mycobacterium tuberculosis [
4,
5],
Pseudomonas aeruginosa [
6],
Staphylococcus aureus [
7], and
Acinetobacter baumannii [
8], now exhibit increased resistance to almost all conventional antibiotics and the elimination of these adapted strains has become increasingly difficult. To overcome this problem, the synthetic antibiotic linezolid, which belongs to a novel oxazolidinone class of antibiotics, was developed and has been found to be effective against multidrug-resistant
M. tuberculosis [
9] and
S. aureus [
10]. Linezolid binds to the 23S subunit of the ribosome and prevents the formation of the initiation complex [
11]. It was hoped that linezolid would be a panacea for infections caused by antibiotic-resistant strains; however, an outbreak of linezolid-resistant
S. aureus has already been reported [
12]. Drug-resistant microorganisms have become a global concern, leading to ceaseless demands for novel antibiotics.
Antimicrobial peptides (AMPs) have received much attention as a novel class of antibiotics. AMPs are peptide antibiotics characterized by an amphipathic nature derived from their positive charges and hydrophobic amino acid residues [
13,
14]. Since the isolation of the first AMPs, the magainins, from the skin of the African clawed frog
Xenopus laevis by Zasloff
et al. [
15,
16,
17], AMPs have been shown to function as an essential component of innate immunity against pathogenic organisms and have evolved in most living organisms over 2.6 billion years [
18,
19]. AMPs exhibit surprisingly diverse mechanisms of action that are different from those of conventional antibiotics. AMPs disrupt membrane structure, inhibit protein and DNA synthesis, and repress cellular processes, including protein folding, cell wall synthesis, and metabolic turnover [
20,
21]. Due to these diverse mechanisms of action, AMPs have strong antimicrobial activity in the nanomolar or micromolar range against a broad spectrum of microorganisms, including Gram-positive and Gram-negative bacteria, fungi, and viruses [
22,
23]. In addition, they are also effective against pathogenic organisms that are resistant to conventional drugs [
24]. Therefore, AMPs have been considered as potential future antibiotics.
Despite their great potentials, AMPs have several drawbacks that severely limit their clinical utility, including hemolytic activity [
25], broad spectrum of activity [
26], rapid turnover in the human body [
27], deactivation by high salt concentrations [
28], and high cost of production [
13]. For example, AMPs can directly interact with host cells and lyse them [
29,
30]. Furthermore, their broad spectrum of activity can also cause severe problems. Administration of broad-spectrum antibiotics can disrupt the indigenous microflora that provides protective colonization against pathogenic organisms, thereby increasing the risks of diarrhea and other fatal infections [
31]. Improvement of the above drawbacks will be necessary for clinical application of AMPs. This article reviews the basic properties of AMPs and the progress toward their clinical application. The peptides reviewed in this article are listed in
Table 1.
Table 1.
Antimicrobial peptides reviewed in this article.
Table 1.
Antimicrobial peptides reviewed in this article.
Antimicrobial peptide | Sequence | Origin | Description |
---|
Magainin 2 | GIGKFLHSAKKFGKAFVGEIMNS | X. laevis | First AMP isolated from X. laevis |
Lactoferricin | GRRRRSVQWCA | Homo sapiens | AMP derived from lactoferrin |
Buforin II | TRSSRAGLQFPVGRVHRLLRK | Bufo gargarizans | AMP derived from histone H2A |
Drosocin | GKPRPYSPRPTSHPRPIRV | Drosophila melanogaster | The Thr residue is O-glycosylated. |
Pyrrhocoricin | VDKGSYLPRPTPPRPIYNRN | Pyrrhocoris apterus | Inducible AMP of a sap-sucking insect |
Apidaecin | GNNRPVYIPQPRPPHPRL | Apis mellifera | Isolated from the lymph fluid of honeybees |
Lasioglossin-III | VNWKKILGKIIKVVK | Lasioglossum laticeps | AMP derived from bee venom |
HNP1 | ACYCRIPACIAGERRYGTCIYQGRLWAFCC | Neutrophils | Human defensins stored in azurophil granules |
HNP2 | CYCRIPACIAGERRYGTCIYQGRLWAFCC | Neutrophils |
HNP3 | DCYCRIPACIAGERRYGTCIYQGRLWAFCC | Neutrophils |
HNP4 | VCSCRLVFCRRTELRVGNCLIGGVSFTYCCTRV | Neutrophils |
HBD1 | DHYNCVSSGGQCLYSACPIFTKIQGTCYRGKAKCCK | Epithelial cells | Human defensins secreted by epithelial cells |
HBD2 | TCLKSGAICHPVFCPRRYKQIGTCGLPGTKCCKKP | Epithelial cells |
HBD3 | GIINTLQKYYCRVRGGRCAVLSCLPKEEQIGKCSTRGRKCCRRKK | Epithelial cells |
HBD4 | ELDRICGYGTARCRKKCRSQEYRIGRCPNTYACCLRK | Epithelial cells |
RTD1 | GFCRCLCRRGVCRCICTR | Primate | Premature stop codons in the human θ-defensin sequence |
Melittin | GIGAVLKVLTTGLPALISWIKRKRQQ | A. mellifera | Peptide antibiotic with toxicity to human cells |
Gramicidin S | VOrnLdFPVOrnLdFP | Bacillus brevis | Peptide antibiotic with toxicity to human cells |
Adepantin 1 | GIGKHVGKALKGLKGLLKGLGES | Artificial | Predicted by AMPad to have low hemolytic activity |
R5L | PLCRCRVRPYRCRCVG | Artificial | Designed to mimic the LPS-binding sites of LBP, cyclic |
Oncocin | VDKPPYLPRPRPPRRIYNR | Artificial | Proline-rich, Gram-selective AMP |
M8G2 | TFFRLFNRGGGKNLRIIRKGIHIIKKY | Artificial | Designed using STAMP technology to target Streptococcus mutans |
Clavanin A | VFQFLGKIIHHVGNFVHGFSHVF | Styela clava | Histidine-rich, pH-dependent AMP |
AAP2 | FHFFHHFFHFFHHF | Artificial | Acid-activated AMP based on clavanin A |
Protease-activated AMP | DDAEAVGPEAFADEDLDEGFIKAFPKRRWQWRMKKLG | Artificial | Protease-activated AMP based on lactoferricin |
2. Mechanism of Action of AMPs
Cationic AMPs are amphipathic peptides characterized by a significant proportion of hydrophobic amino acid residues and an overall positive charge. In this section, we have reviewed the mechanism underlying the action of AMPs by using the example of lactoferricin, which is one of the most extensively studied AMPs [
32]. Lactoferricin (GRRRRSVQWCA) is naturally produced through proteolysis of lactoferrin by pepsin under acidic conditions [
33]. Lactoferricin is rich in arginine and hydrophobic valine and tryptophan residues and possesses strong antimicrobial activity against multidrug-resistant pathogens, including
S. aureus [
34],
A. baumannii [
35], and
Candida albicans [
36,
37].
Lactoferricin and other linear AMPs disrupt bacterial cell membranes by drastically changing their tertiary structure depending on the surrounding environment [
38,
39]. Nuclear magnetic resonance spectroscopy has shown that in aqueous solution, AMPs adapt a partially folded structure (
Figure 1A) [
40,
41]. By contrast, the membrane-mimetic environment in detergents induces a significantly amphipathic helix structure that separates all the hydrophobic residues to one side, and the positively charged residues to the other (
Figure 1B) [
33,
41,
42].
Figure 1.
Structure of lactoferricin. (A) Representative structure and electrostatic surface display of lactoferricin in aqueous solvent (PDB accession, 1Z6W). (B) Representative structure and electrostatic surface display of lactoferricin with a distinct α-helix in membrane-mimetic solvent (PDB accession, 1Z6V). The negatively charged surface is shown in red, and the positively charged surface in blue. The molecular surface corresponding to the hydrophobic residues was also indicated by addition of yellow color to the electrostatic potential color; hence, orange color indicates hydrophobic and negatively charged surface; yellow color shows hydrophobic and neutral surface; green color indicates hydrophobic and positively charged surface.
Figure 1.
Structure of lactoferricin. (A) Representative structure and electrostatic surface display of lactoferricin in aqueous solvent (PDB accession, 1Z6W). (B) Representative structure and electrostatic surface display of lactoferricin with a distinct α-helix in membrane-mimetic solvent (PDB accession, 1Z6V). The negatively charged surface is shown in red, and the positively charged surface in blue. The molecular surface corresponding to the hydrophobic residues was also indicated by addition of yellow color to the electrostatic potential color; hence, orange color indicates hydrophobic and negatively charged surface; yellow color shows hydrophobic and neutral surface; green color indicates hydrophobic and positively charged surface.
AMPs with positive charges are electrostatically attracted to the negatively charged surface of the bacterial cytoplasmic membrane. Subsequent to their interaction with the bacterial membrane, AMPs spontaneously oligomerize and form transmembrane pores that cause leakage of the cellular contents (
Figure 2) [
20]. Indeed, replacement of positively charged arginine or hydrophobic tryptophan residues with alanine causes significant reduction in antibacterial activity. This result suggests importance of positively charged amino acids and hydrophobic amino acids to interact with bacterial membrane and to form transmembrane pores, supporting the above mechanism of action [
43].
Figure 2.
Mechanism of action of antimicrobial peptides (AMPs). (A) Comparison of human and bacterial plasma membranes. (B) Disruption of bacterial membrane by AMPs. AMPs preferentially interact with bacterial plasma membrane due to their electrical charge. When AMPs interact with the negatively charged bacterial plasma membrane, they spontaneously form pores and disrupt membrane integrity.
Figure 2.
Mechanism of action of antimicrobial peptides (AMPs). (A) Comparison of human and bacterial plasma membranes. (B) Disruption of bacterial membrane by AMPs. AMPs preferentially interact with bacterial plasma membrane due to their electrical charge. When AMPs interact with the negatively charged bacterial plasma membrane, they spontaneously form pores and disrupt membrane integrity.
Although AMPs possess antimicrobial activity that disrupts bacterial membrane integrity, other modes of action targeting key cellular processes, including DNA and protein synthesis, protein folding, cell wall synthesis, and metabolic turnover, have been characterized [
20,
44]. Thus, transmembrane pore formation is not the only mechanism, making it necessary to carefully determine how each AMP kills microorganisms. For example, buforin II (
Table 1), an AMP isolated from the stomach tissue of the Asian toad
Bufo gargarizans, penetrates the cell membrane and strongly inhibits the functions of DNA and RNA in cells [
45,
46]. Drosocin, pyrrhocoricin, and apidaecin, originally isolated from insects, act on heat shock proteins (DnaK and GroEL) and repress the stress response of cells [
47]. Recent studies have revealed novel alternative functions for AMPs, including neutralization of endotoxins [
48], wound healing [
49], cytotoxicity against neoplastic cells [
50], and immunomodulation [
51]. Some peptides can even use multiple antimicrobial mechanisms [
44]. This multiple-hit strategy may be effective in increasing antimicrobial efficiency and evading potential resistance mechanisms.
The acquisition of resistance to AMPs is very rare, compared to conventional antibiotics, because microorganisms are killed by direct disruption of cellular components, including the microbial membrane and DNA [
16,
52]. The appearance of AMP-resistant strains is less likely because development of microbial resistance by gene mutation to such a microbicidal mechanism of action is difficult [
53], although microorganisms can coordinate countermeasures to circumvent the attack by AMPs to some extent [
54].
For example,
S. aureus, a Gram-positive bacterium, exhibits a higher minimum inhibitory concentration (MIC) for AMPs by modifying its cell surface teichoic acid with
d-alanine [
55,
56]. Though teichoic acids are polyanionic, incorporation of
d-alanine reduces the negative charges on the cell wall. Furthermore,
S. aureus can modify its membrane lipid phosphatidylglycerol via enzymatic transfer of a lysine [
57]. These modifications weaken the interaction between cationic AMPs and the cell wall, leading to acquisition of resistance.
P. aeruginosa produces elastase, a serine protease that hydrolyzes amides and esters, in a wound fluid environment and degrades the AMP LL-37 [
58]. The production of bacterial elastase degrades AMPs and enhances the survival of
P. aeruginosa. Another resistance mechanism is external trapping of AMPs through the actions of bacterial surface-associated or secreted proteins. These interfering molecules neutralize AMPs and reduce bacterial killing.
S. aureus produces staphylokinase, which directly binds to defensins produced by host immune cells [
59]. Staphylokinase neutralizes AMPs, and the formation of the kinase-AMP complex results in complete inhibition of bactericidal effects.
Recent studies have reported an interesting resistance mechanism of a dysentery bacillus.
Shigella flexneri, which causes dysentery, lowers host innate immunity by targeting the expression of host defensins [
60]. Upon infection,
S. flexneri produces MxiE, a transcriptional activator, which modulates the expression of host defensins. Another study reported the possibility that
S. flexneri might use host defense molecules to enhance virulence and subvert innate immunity [
61]. Neutrophils are the first line of defense and an essential component of innate immunity; they kill pathogenic microorganisms via phagocytosis, neutrophil extracellular traps [
62], and degranulation [
63]. During the process of degranulation, the
Shigella cell surface binds to cationic granular antimicrobial proteins, causing increased adhesion and hyperinvasion [
61]. This effect is considered to be caused by surface negative charges because a lipopolysaccharide (LPS) mutant has been found to show enhanced hyperinvasion.
4. Temporal and Spacial Regulation of AMPs in Nature
AMPs have several disadvantages and their activity is limited under physiological conditions; however, the human body efficiently uses AMPs as innate immunity, a front line defense against pathogenic organisms. It is evident that AMPs play important roles in protecting against fatal diseases, given that many types of diseases are caused by gene mutations that inactivate AMPs [
96]. Investigation of the mechanisms of action
in vivo will provide valuable information to develop effective AMPs for clinical use. In this section, we have reviewed our current understanding of the temporal and spatial regulation of AMPs. We have focused on the mechanisms used by multicellular organisms to avoid adverse effects and to effectively kill pathogenic organisms using AMPs.
The human body has a symbiotic relationship with commensal microflora (microbiota), an integral part of complex mucosal surfaces. The human distal gut microbiota is composed of 10
13–10
14 microorganism cells, including 72 bacterial phylotypes and one archaeal phylotype [
97]. The microbiota exhibits various traits (e.g., methanogenesis) that humans have not been required to evolve on their own [
98,
99]. Thus, humans are recognized as super-organisms whose metabolic processes represent a mixture of microbial and human attributes. This situation leads to the question of how human bodies can rapidly eliminate pathogenic organisms while maintaining the commensal microbiota.
The innate immune system, including AMPs and pattern recognition receptors such as toll-like receptors (TLRs), plays an important role in determining the microbial population balance of mucosal surfaces. The importance of innate immunity is demonstrated by the existence of severe diseases such as cystic fibrosis [
100], Kostmann syndrome [
101], and Papillon-Lefèvre syndrome [
102], which are caused by local defects in AMP activities. For example, cystic fibrosis can be lethal because of progressive destruction of the airways by recurrent infections and inflammation caused by
P. aeruginosa [
103]. Cystic fibrosis is an autosomal recessive genetic disorder caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR). Mutations in CFTR lead to aberrant high salt concentrations in body fluids and deactivation of AMPs, resulting in fatal infections [
102,
104]. These diseases prove the essential roles of AMPs in normal immune responses.
The oral cavity and the airways are major barriers to pathogenic organisms because most external microorganisms enter the human body via these surfaces [
105]. Epithelial cells, neutrophils, and salivary glands in the oral cavity secrete over 45 types of AMPs and antimicrobial proteins; however, the concentrations of these antimicrobial agents do not exceed the MIC for most microbes [
106]. Thus, basal AMPs in the oral cavity can be regarded as modulators that maintain the microbiota and prevent outgrowth, rather than weapons that eliminate individual microorganisms. The oral epithelium specifically recognizes a subset of bacteria and TLR ligands and produces stronger AMPs to maintain the balance between health and disease [
102].
Defensins, the most studied AMPs in vertebrates, are abundant in nature. The defensin family is evolutionarily conserved in vertebrates, and its members are characterized by the presence of similar structures of β-sheet-rich folds and three disulfide bonds [
107]. Primates have three defensin subfamilies, namely, α-defensins, β-defensins, and θ-defensins (retrocyclins such as rhesus theta-defensin, RTD1 shown in
Table 1). Each family has different secondary structures. Among them, α-defensins and β-defensins are the two main defensin subfamilies in humans because the human θ-defensin gene has premature stop codons in its genome sequence and has been recognized as a pseudogene [
108,
109]. In this section, we use the defensin family as an example to describe how human tissues temporally and spatially regulate the activity of AMPs.
4.1. Neutrophils
α-Defensins (human neutrophil peptides, HNP1-4 shown in
Table 1) are predominant in neutrophils [
110,
111], a type of white blood cell (leukocyte), and account for 30%–50% of the total protein content of the azurophil granule, which is an arsenal-storage organelle in leukocytes [
112]. Synthesis of α-defensins occurs in the bone marrow where promyelocytes reside [
113,
114], and the translated α-defensin precursors are proteolytically processed to mature α-defensins and packaged into azurophil granules [
115] (
Figure 4A). The concentration of intravacuolar α-defensins reaches several grams per liter, which exceeds the MICs for most bacteria [
116,
117]. When mature neutrophils encounter and ingest pathogenic organisms in phagocytic vacuoles, the azurophil granules fuse with the phagosome and release α-defensins [
118,
119] (
Figure 4B). Because phagosomes have little space, neutrophils are able to attack the pathogens by using the minimally diluted AMPs at a high concentration. Neutrophils use AMPs against phagocytosed organisms in the limited space, leading to high antimicrobial activity and preservation of the good microbiota, with minimal adverse effects. This observation suggests that it is important to specifically concentrate AMPs at sites where pathogens cause diseases.
Figure 4.
Phagocytosis of pathogenic cells by neutrophils. (A) Synthesis of defensins in neutrophils. α-Defensins are proteolytically processed to the mature form and then packaged into azurophil granules. (B) Phagocytosis of pathogenic cells. When a mature neutrophil ingests a pathogenic cell, it simultaneously evokes the fusion of azurophil granules and the phagosome, and the α-defensins then exert antimicrobial activity in the limited space.
Figure 4.
Phagocytosis of pathogenic cells by neutrophils. (A) Synthesis of defensins in neutrophils. α-Defensins are proteolytically processed to the mature form and then packaged into azurophil granules. (B) Phagocytosis of pathogenic cells. When a mature neutrophil ingests a pathogenic cell, it simultaneously evokes the fusion of azurophil granules and the phagosome, and the α-defensins then exert antimicrobial activity in the limited space.
4.2. Epithelial Cells
β-Defensins (human beta-defensins, HBD1-4 shown in
Table 1) are mainly produced in epithelial cells. Among them, HBD1 is constitutively produced in epithelial tissues, and it exhibits mild antimicrobial activity compared to HBD2 and HBD3. In addition, its concentration in the mucosal fluid is not high and is less than the MIC for most microbes [
106,
120]. Thus, HBD1 may be recognized by modulators of the human microbiome. HBD2 is constitutively produced in gingival tissues; in addition, HBD2 is an inducible AMP that is upregulated only in inflamed skin [
121,
122]. Upregulation of HBD2 is caused by pathogenic organisms via activation of a pathway that requires interleukin-1 and nuclear factor (NF)-κB [
123,
124]. HBD3 and HBD4 are also recognized as inducible defensins; however, they are regulated by NF-κB-independent mechanisms [
125,
126]. Based on these observations, it is proposed that HBD1 plays a role in the maintenance of steady-state microflora in epithelial tissues, whereas HBD2-4 function as effective antibiotics against pathogens.
6. Conclusions
In recent years, the appearance of drug-resistant pathogens has provoked the need to develop novel antibiotics. Although AMPs have several promising properties, their clinical application has not been successful because of toxic side effects, rapid turnover, and low activity under physiological conditions. Several studies have dramatically improved the pharmacokinetics of AMPs in vivo; further improvements are still necessary.
Recently, designer AMPs have enabled temporal and spatial regulation of AMP activity; these designer AMPs possess ideal pharmacokinetic properties. Such drugs could increase the compliance of patients and improve their quality of life. Until now, few peptides have been used in the clinical setting so far, because of their unpredictable kinetics in mice and humans [
166]. Therefore, further investigation with a focus on clinical trials to demonstrate effectiveness
in vivo is required. Smart polymer technologies may improve the pharmacokinetics of designer AMPs. Smart polymers are now available to control the release of incorporated chemical compounds [
167,
168] and could also provide high stability against serum, proteolytic enzymes, and harsh acidic environments [
169]. We believe that these efforts will produce innovative technologies and bear fruit that will benefit public health in the near future.