Very Short and Stable Lactoferricin-Derived Antimicrobial Peptides

6 hours ago - Biography. John S. Mjøen Svendsen is a Professor of organic chemistry at UiT, The Arctic University of Norway, and Chief Scientific Off...
0 downloads 0 Views 7MB Size
Download PDF
Article Cite This: Acc. Chem. Res. XXXX, XXX, XXX−XXX

pubs.acs.org/accounts

Very Short and Stable Lactoferricin-Derived Antimicrobial Peptides: Design Principles and Potential Uses John S. Mjøen Svendsen,† Thomas M. Grant,‡ David Rennison,‡ Margaret A. Brimble,‡ and Johan Svenson*,§ †

Department of Chemistry, UiT The Arctic University of Norway, N-9037 Tromsø, Norway School of Chemical Sciences, University of Auckland, Auckland 1010, New Zealand § Department of Chemistry and Materials, RISE Research Institutes of Sweden, SE-501 15 Borås, Sweden

Acc. Chem. Res. Downloaded from pubs.acs.org by WEBSTER UNIV on 03/04/19. For personal use only.

‡

CONSPECTUS: The alarming rate at which micro-organisms are developing resistance to conventional antibiotics represents one of the global challenges of our time. There is currently ample space in the antibacterial drug pipeline, and scientists are trying to find innovative and novel strategies to target the microbial enemies. Nature has remained a source of inspiration for most of the antibiotics developed and used, and the immune molecules produced by the innate defense systems, as a first line of defense, have been heralded as the next source of antibiotics. Most living organisms produce an arsenal of antimicrobial peptides (AMPs) to rapidly fend off intruding pathogens, and several different attempts have been made to transform this versatile group of compounds into the next generation of antibiotics. However, faced with the many hurdles of using peptides as drugs, the success of these defense molecules as therapeutics remains to be realized. AMPs derived from the proteolytic degradation of the innate defense protein lactoferrin have been shown to display several favorable antimicrobial properties. In an attempt to investigate the biological and pharmacological properties of these much shorter AMPs, the sequence dependence was investigated, and it was shown, through a series of truncation experiments, that these AMPs in fact can be prepared as tripeptides, with improved antimicrobial activity, via the incorporation of unnatural hydrophobic residues and terminal cappings. In this Account, we describe how this class of promising cationic tripeptides has been developed to specifically address the main challenges limiting the general use of AMPs. This has been made possible through the identification of the antibacterial pharmacophore and via the incorporation of a range of unnatural hydrophobic and cationic amino acids. Incorporation of these residues at selected positions has allowed us to extensively establish how these compounds interact with the major proteolytic enzymes trypsin and chymotrypsin and also the two major drug-binding plasma proteins serum albumin and α-1 glycoprotein. Several of the challenges associated with using AMPs relate to their size, susceptibility to rapid proteolytic degradation, and poor oral bioavailability. Our studies have addressed these issues in detail, and the results have allowed us to effectively design and prepare active and metabolically stable AMPs that have been evaluated in a range of functional settings. The optimized short AMPs display inhibitory activities against a plethora of micro-organisms at low micromolar concentrations, and they have been shown to target resistant strains of both bacteria and fungi alike with a very rapid mode of action. Our Account further describes how these compounds behave in in vivo experiments and highlights both the challenges and possibilities of the intriguing compounds. In several areas, they have been shown to exhibit comparable or superior activity to established antibacterial, antifungal, and antifouling commercial products. This illustrates their ability to effectively target and eradicate various microbes in a variety of settings ranging from the ocean to the clinic.

■

INTRODUCTION Innate antimicrobial peptides (AMPs) are produced by most multicellular organisms as a rapid first line of defense against intruding micro-organisms, which allows them to survive in competitive environments without the aid of a sophisticated adaptive immune system.1 AMPs come in different sizes and shapes, but the majority are composed of less than 50 amino acids, often carrying a net positive charge, and readily adopt an © XXXX American Chemical Society

amphipathic structure. Negatively charged AMPs have been reported, but they represent a minority. The antibacterial mode of action of AMPs differs between the compound classes but is often associated with direct interactions with the negatively charged bacterial membrane, causing membrane depolarization Received: December 7, 2018

A

DOI: 10.1021/acs.accounts.8b00624 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 1. Selection of building blocks used and evaluated to generate short peptides with improved antibacterial and metabolic stability properties during the course of our work.

■

LACTOFERRIN/LACTOFERRICIN In mammals, the iron-binding protein lactoferrin is present in large quantities in both milk and colostrum. Lactoferrin is also found in mucosal secretions and in poly-morphonuclear leukocytes.5 Lactoferrin is a large glycosylated protein of the transferrin family composed of approximately 700 amino acids (depending on species) representing a multifunctional innate defense protein with significant antimicrobial activity (MIC = 25 μM).6 Proteolytic degradation of bovine lactoferrin in the human stomach generates a cyclic fragment (residues 17−41) known as lactoferricin B.7 Lactoferricin B is rich in cationic residues, such as lysine and arginine, in combination with

or lysis. Intracellular targets, membrane protein delocalization, and immunomodulatory effects have also been reported,2 and the exact modes of action of many AMPs have not been established. It was long thought that the many rapid mechanisms AMPs eliminate micro-organisms are highly challenging for the micro-organisms to counteract via resistance development. Recent reports on the plasmid-mediated resistance toward the membrane active peptide colistin is nevertheless challenging these assumptions.3 The nature and proposed modes of action of AMPs have been extensively reviewed elsewhere.1,4 B

DOI: 10.1021/acs.accounts.8b00624 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 2. Examples of highly active short AMPs produced by incorporating Tbt as a “superbulky” central amino acid residue.

possible to replace hydrophobic residues by including hydrophobic elements via C-terminal amides and esters. The effect of such C-terminal cappings was pronounced, and through the incorporation of C-terminal benzyl esters, it now became possible to generate highly active peptides as short as tetrapeptides, e.g., RWRWOBn with MIC values down to 5 μg/mL. It was further possible to generate active di and tripeptides with MIC values between 5 and 50 μg/mL against a selection of bacteria. These short peptides appeared to be indifferent to the order of the residues or their individual stereochemistry. It was shown that a minimum motif comprised two hydrophobic elements and at least two cationic charges to generate a “2 + 2” pharmacophore.18 Larger hydrophobic residues, for example, W vs F, generated more potent peptides, and the incorporation of more basic residues, e.g., R vs K, was similarly beneficial for good antibacterial activity.

hydrophobic residues, such as tryptophan, creating an amphiphilic distorted antiparallel beta-sheet structure.8 In the native lactoferrin protein, residues 1−13 of lactoferricin B form an α helix, illustrating that this sequence is able to adopt multiple conformations.8,9 Lactoferricin is a well-studied cyclic peptide and of particular interest is that it retains the antimicrobial capacity of the lactoferrin protein.10 The disulfide linkage is not essential for antibacterial activity as shown by Bellamy et al.,6 and Kang and co-workers later showed that shorter linear fragments also maintain a significant amount of antibacterial potency.11 This observation suggests that further sequence truncation may be feasible, providing an opportunity for the generation of much shorter peptides with increased drug-like properties.

■

THE “2 + 2” PHARMACOPHORE This hypothesis was addressed early by scientists at the Arctic University of Norway in Tromsø led by Svendsen and Rekdal,12 who designed truncated analogues of bovine lactoferricin initially focusing on the N-terminal residues. Their studies on 15-mer peptides provided insights into the structural contribution to antibacterial activity and subsequently inspired studies on shorter fragments.13,14 The more rational and radical studies that followed were based on an 11-mer (residues 4−14 of lactoferricin B, RRWQWRMKKLG), which was truncated down to pentapeptides with maintained activity.15 These studies established that only R and W residues were sufficient for making these small peptides antibacterial, and other decorative residues could be trimmed off. These types of short RW peptide analogues have also been studied and further developed by other research groups and the synAMPs, developed by Metzler-Nolte and Albada,16 and the work by Dathe and co-workers on cyclic RW peptides represent important contributions.17 In an effort to determine the “minimal antibacterial motif” employing only hydrophobic and cationic residues, it was discovered that while at least five residues appeared essential for high activity, it was

■

INCREASING THE POTENCY WITH UNNATURAL RESIDUES Having established methods to rapidly prepare short AMPs with antimicrobial efficacies nearing those of conventional antibiotics, it was decided to further probe the pharmacophore through the introduction of synthetic amino acid residues. Initial work investigated mainly the effect of introducing larger synthetic hydrophobic alternatives to W within larger peptides (15-mer peptides), revealing substantial gains in activity.14 The following studies were aimed at di and tripeptides and also at the C-terminal capping, previously demonstrated to be crucial for high potency. More recent work also focused on the cationic residues, often provided by R in native AMPs. A schematic representation of the different functionalities and scaffolds employed to yield molecular diversity is presented in Figure 1. The initial studies illustrated that it was possible to gain substantial additional antibacterial potency by further increasing the hydrophobic volume of the amino acid side chain.19 Hydrophobic side chains such as biphenylalanine (Bip) and C

DOI: 10.1021/acs.accounts.8b00624 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 3. Top image: Examples of possible hydrogen bonding interaction between a positively charged guanidinium group and the negatively charged phosphates found in the phospholipid headgroup. Bottom image: Example of short AMPs prepared with R analogues with maintained antibacterial effect.

2,5,7-tritert-butyl tryptophan (Tbt) generated the most potent peptides when combined with either C-terminal benzyl esters or amides, with MIC values as low as 2.5 μg/mL.19 In terms of selectivity, it was shown that the short AMPs display a preference for Gram-positive bacteria with 5−10-fold higher MIC values for Gram-negative bacteria. Applying these findings to a tripeptide scaffold (RXR) provided additional insights, and a variety of potent peptides (Figure 2) were subsequently designed.20 Upon realizing that Tbt appeared to induce peptides with superior efficacy, significant focus was devoted further to investigating the properties of this bulky residue. Tbt comprises both an indole and three tert-butyl groups with a total hydrophobic side chain volume (Connolly solvent excluded volume) of nearly 321.28 Å3, which compared to either a benzyl (87.04 Å3) or indole (115.06 Å3) side chain (F or W) illustrates the large enhancement in bulk. The beneficial properties generated through the introduction of unnatural hydrophobic amino acids into the peptides encouraged us to also investigate the cationic elements.21 With regards to antibacterial effect, it was shown that the length of the cationic side chain induced only marginal effects on the AMPs.22 It was interesting to observe that even AMPs incorporating only the short 2-amino-(3-guanidino)propanoic acid (Agp) as the R substituent generated highly active AMPs. The guanidine-containing side chains generally outperformed the primary amines in terms of ability to kill bacteria, which is ascribed to their ability to form additional hydrogen bonds with

the negatively charged phospholipid head groups of biological membranes. The incorporation of a hydrophobic element into the cationic side chain, through the use of 2-amino-3-(4guanidinophenyl)-propanoic acid (Gpp), was also beneficial, and Gpp-containing peptides yielded marked improvements in antibacterial potency (Figure 3).21,22

■

MODE OF ACTION STUDIES A common target for most AMPs is the negatively charged bacterial cytoplasmic membrane. Several routes toward bacterial cell death and disruption have been proposed and reported, but the initial cytoplasmic membrane association is a crucial step.1 Our peptides can be prepared with significant selectivity for bacterial cells over zwitterionic mammalian cells, but it is important to balance the hydrophobicity and charge to avoid an undesirable increase in hemolytic activity.20 To probe how the Tbt-containing AMPs interact with bacterial membranes, a range of membrane-mimicking phospholipid vesicles were prepared. The difference in membrane association between the highly active RTbtROMe and inactive RWROMe was significant in both fluorescence shift experiments and calcein release studies.20 Additional studies on how these compounds interact with real bacterial membranes were undertaken by testing the peptides on E. coli ML-35p bacteria. This genetically engineered E. coli is constitutive for cytoplasmic β-galactosidase, lacks lac permease, and expresses a plasmid-encoded periplasmic β-lactamase. Chromogenic reporter molecules enable monitorD

DOI: 10.1021/acs.accounts.8b00624 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research ing of the permeabilization of both the outer and inner membranes. It was shown that RTbtROMe induces loss of both outer and inner membrane integrity in E. coli ML-35p bacteria at 128 and 256 μg/mL, respectively.20 These values are significantly higher than the MIC values reported for E. coli (20 μg/mL) and suggest a dual mechanism for this class of AMPs, one fast membrane integrity loss mechanism operating at slightly higher peptide concentrations and one slower mechanism, probably intracellular operating at lower peptide concentrations. However, these peptides are very rapid at killing bacteria when employed in significant concentrations, and they compare favorably with commonly employed antibiotics, as shown in Figure 4. The mode of action is not only very rapid at sufficiently high concentrations, but it has also recently been shown that the postantibiotic sub-MIC effect is higher and more prolonged than that of mupirocin against Staphylococcus aureus blood isolates. These findings provide support for persistent activity for several hours after the peptide is no longer present or is below the MIC,

which is an important pharmacodynamic parameter for reduced resistance development.23 Generally, these AMPs display a higher activity toward Gram-positive bacteria, but they can be tuned to also target Gram-negative bacteria with a similar activity. The fact that the AMPs also effectively rapidly eradicate fungi24 and fouling marine organisms25 further suggest multiple modes of action given the spread in cell surface and membrane composition. It is clear that the initial “2 + 2” pharmacophore remains reasonably accurate, but as a result of its simplicity, the model fails to account for finer detail in the peptide structure and stereochemistry. The invariance of stereochemistry in the pharmacophore has its exceptions. Molecular dynamics (MD), biological, and NMR studies on the complete set (eight) of stereoisomers of RTbtREtPh revealed that they all display similar activity against S. aureus (MIC = 2−4 μg/mL).26 An increase in MIC is nevertheless observed for both E. coli (from 3 to 10 μg/mL) and P. aeruginosa (from 8 to 50 μg/mL) when going from all L- or D-stereoisomers to mixed stereoisomers. Isomers characterized by high antimicrobial activity share the same chiral sense of the central Tbt residue and the C-terminal arginine. This behavior is also strongly reflected in the retention times of the peptides that decrease from 19 to 6 min (RPHPLC), implying that the more active stereoisomers also adopt a more amphipathic structure in solution. NMR and MD experiments on the stereoisomers in a membrane environment illustrated how the highly active stereoisomers are able to form strong locked amphipathic structures, which are optimal for membrane insertion but not for water solubility (Figure 5).26 The ability of the Tbt residue to restrict the adjacent amide bond is an additional feature that may benefit efficacy beyond the high lipophilic bulk. Studies by Aldaba and co-workers have highlighted similar insights into the dependence of stereochemistry on the bioactivity of short AMPs.27

Figure 4. Rapid in vitro killing kinetics with a reduction of the bacterial density down to the detection limit (1 × 10 cfu/mL) within 120 min at 8× MIC.

■

ADME

Two major hurdles for peptide drugs are their low metabolic stability and poor oral uptake properties leading to a low oral

Figure 5. Structure of the L-L-L-RTbtREtPh peptide (left). Data from ref 26. MD snapshots of the amphipathic L-L-L (center). Nonamphipathic L-L-Dstereoisomers (right). Dashed bonds illustrate key ROESY interactions between the central Tbt unit and either the C-terminal amide phenyl group or the side chain of the C-terminal Arg. Reproduced with permission from ref 26. Copyright 2011 American Chemical Society. E

DOI: 10.1021/acs.accounts.8b00624 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research bioavailability. The serine proteases represent a major family of proteolytic enzymes in the gastrointestinal tract, and the influence of unnatural hydrophobic residues inside the peptide sequence was investigated against the endopeptidase trypsin, which effectively degrades a tripeptide like RWRNHBn (t1/2 = 1 h).28 Systematic incorporation of unnatural hydrophobic residues to probe the binding pockets (S1−S3 and S1′−S3′) surrounding the activated serine in trypsin revealed insights into peptide binding and stability. Through degradation, calorimetric, and docking studies, it was demonstrated that several unnatural residues and capping strategies could be utilized to tune the proteolytic stability, and a series of guidelines for devising small peptides with stability toward degradation by trypsin was generated:28 (1) The peptides should be kept as small as possible in order to limit interactions with as many binding sites in trypsin as possible. End-cappings at either the Nterminus or the C-terminus may act as additional residues increasing the binding to the active site. (2) The amino acid occupying the P2 site should have a side chain sterically preventing it from binding efficiently to the S2 unit of trypsin. (3) The length and stereochemistry of the C-terminal endcapping have a major influence on stability. (4) A C-terminal tertiary amide yields stable peptides in a manner analogous to that of peptide backbone N-methylation. Moving from in vitro enzyme assays to organ homogenates revealed, however, that even a short peptide such as RWR−NH2 was rapidly degraded in the stomach and duodenum (t1/2 = 5 min).29 Despite being very short peptides, the combination of cationic and hydrophobic residues appears to also make them good substrates for serine protease degradation. The rapid degradation of these peptides was thus likely a reflection of the combined complementary action of both trypsin and chymotrypsin. By also introducing unnatural cationic residues, responsible for the crucial interactions with the S1 pocket interaction in trypsin and the S1′ pocket in chymotrypsin, we further probed this complementary serine protease.22 Using metabolic and calorimetric studies, it was shown that peptide stability toward chymotryptic degradation could be effectively tuned using both sequence scrambling and through the incorporation of both hydrophobic and unnatural cationic residues.22 In particular, shorter cationic residues yielded increased stability, and peptides incorporating the short Agp residue were resistant to chymotryptic degradation. A second generation of peptides was designed incorporating an array of “optimized” cationic and hydrophobic residues to generate further optimized peptides.30 The in vitro trypsin assays and docking studies illustrated that both longer and shorter arginine analogues were sufficient to displace the sessile amide bond out of reach of the activated serine (Ser195) of the S1′-binding pocket, thereby generating more stable peptides (Figure 5). Two of these lead peptides were further analyzed in whole-organ extracts and their stabilities were compared with those of RBipRNHBn. Both peptides displayed significant increases in stomach, duodenum, and liver metabolic stability, as summarized in Table 1 and Figure 6.30 It was fascinating to observe that the arginine analogues studied also generated stability toward degradation in stomach, with a near 70-fold increase in stability. These bioactive peptides can now be prepared with sufficient activity and stability to allow for oral administration and potential uptake. Synthetically, most of the amino acid substitutions represent no major increased costs or additional synthetic steps.30

Table 1. Overview of the Metabolic Half-Lives of Three Peptides Incorporating Arginine Analogues To Increase Metabolic Stability calculated peptide half-life (min) tissue

RBipRNHBn

RBipHarNHBn

RBipAgpNHBn

stomach duodenum liver

10 6 106

401 82 64

702 324 191

Studies focusing on peptide binding to plasma proteins, which is of importance for distribution, were also performed. The focus for these experiments was the two major drug transporting proteins in the blood, serum albumin (HSA) and α-1 glycoprotein (orosomucoid, ORM). While HSA displays a preferential binding of lipophilic anionic endogenous and exogenous ligands, ORM acts as a complement by favoring cationic ligands. By employing calorimetric methods (ITC)31 in combination with computational methodologies, we studied the binding of two libraries of AMPs to these proteins (Figure 7). The binding of the peptides to HSA was stoichiometric for the included peptides with low micromolar dissociation constants.32 The free energy of binding (ΔG) varied from −26.7 to −30.6 kJ/ mol with both favorable enthalpic and entropic contributions.32 It was unexpected to see such a strong binding to albumin for a multicationic ligand, and the influence of the binding was reflected in a near 10-fold reduction in the antimicrobial potency of the peptides when albumin was included at physiological concentrations in the antimicrobial assays. Competitive ITC studies with known site I and site II ligands warfarin and dansylglycine, respectively, in conjunction with NMR experiments, placed the peptides in drug site II, which is located in subdomain IIIA of HSA.31 The peptide binding appear to be mediated solely via the bulky hydrophobic residues and not via the cationic side chains. While albumin binding was pronounced, the opposite was seen for ORM with a binding free energy between 6 and 7 kJ/mol. The peptides bound with a low stoichiometry (n = 0.1−0.3), ensuring that the peptides maintained their activity in assays incorporating ORM at physiological conditions, implying that ORM would have little effect on the performance of this family of compounds.33 Inspired by the observed increased metabolic stabilities obtained from the fine-tuning of the peptide sequences, additional studies toward uptake were initiated through studying interactions with the human intestinal peptide transporter hPEPT1 and via passive routes.34 Such a study not only answers important questions about the likely absorption via hPEPT1 but also provides insight into the mechanism behind substrate specificity and transport of the peptide transporter, as the peptides contain numerous unnatural elements. Some peptides displayed moderate affinity (0.6 and 2.7 mM) for hPEPT1, but none served as substrates for hPEPT1 expressed by Xenopus laevis oocytes, presumably because of their unnatural amino acid residues (Figure 8). Most of the included peptides were nevertheless shown to passively permeate at rates suggesting moderate to excellent human oral absorption (>70% in vivo absorption with Papp > 0.9 × 10−6 cm/s) in the applied phospholipid vesicle-based passive permeation assay. The experimental permeation rates were also compared to computational predicted uptake properties, and it was clear that the established computational model (calculated using QikProp) used for predictions does not account for the flexibility of the F

DOI: 10.1021/acs.accounts.8b00624 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 6. Examples of the metabolic profile of peptide RBipHarNHBn in duodenum (left) and an illustrative overlay of peptides interacting with the trypsin P1 pocket via different charged S1 side chains (right). Left: Data from ref 30. Right: Reproduced with permission from ref 30. Copyright 2012 American Chemical Society.

Figure 7. Representation of the reduction in the antibacterial effect of peptide RBipRNHBn upon exposure to albumin and ORM. The thermodynamics of the interactions was determined using isotherm data obtained from ITC experiments. Right image reproduced with permission from ref 31. Copyright 2014 BioMed Central.

autoradiography determination of peptide distribution was performed (Figure 9.). The results revealed that the peptide was rapidly cleared from the blood and accumulated in the liver, lymph nodes, and the renal cortex, thereby precluding an effective distribution in the body and, hence, antimicrobial efficacy. It was however interesting to observe that the animals survived the systemic tail vein injection, which is in contrast to related peptides investigated as intravenous antibiotics that yielded acute toxicity and paralysis.16 On the other hand, local and, in particular, topical applications in animal models have yielded very good antimicrobial efficacy, superior to that of clinically used topical antibiotics. The rapid mode of action makes them well-suited for several types of topical applications, for example, in a superficial skin wound model where a wound was inoculated with S. aureus, and a biofilm wound infection was established (unpublished data).

short peptides, making the predictions only suited for the smaller reference drugs.34

■

TOWARD APPLICATIONS

In Vivo Bacterial Studies

While oral bioavailability and metabolic stability in the GI system is necessary, it is not sufficient to ensure sufficient systemic exposure of the very short AMPs in vivo. Initial in vivo studies employing a RTbtR peptide with a C-terminal isopropyl amide in a peritoneal infection model gave promising results (unpublished data). The effect of the peptide at 14 mg/kg (local intraperitoneal injection) treatment showed a log reduction in the range of 0.7 to 2.0. However, when the route of administration was systemic, e.g., tail vein injection (12.5 mg/ kg), the antimicrobial efficacy was modest. To probe this discrepancy, a tritiated peptide was injected, and a whole-body G

DOI: 10.1021/acs.accounts.8b00624 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 8. Left: Representations of the biological and artificial systems used to investigate active and passive peptide permeability. Image shows oocytes from Xenopus laevis with cloned GFP coupled hPEPT1 together with fused multilamellar liposomes deposited on top of a mixed cellulose ester filter. Right: Current−voltage relations of peptide substrates in comparison with the transport of Gly−Sar, a known substrate of hPEPT1. Reproduced with permission from ref 34. Copyright 2011 American Chemical Society.

Figure 9. Longitudinal whole-body autoradiograph of a mouse 20 min after injection of a tritiated peptide. Dark areas show high radioactivity content.

RTbtREtPh either formulated in a commercial moisture cream or a hydrogel was applied three times to the wound, with a 2 h incubation time and compared to a commercial topical antibiotic cream containing 2% mupiricin, which was applied in an identical manner. The peptide treatment reduced the bacterial load 5−6-log, whereas mupiricin reduced the infection by only 1-log, illustrating the dramatic effect of the formulated peptide on topical infections (Figure 10). Several of the peptides have been shown to be efficient eradicators of established biofilms in comparison to established antibiotics.35 RTbtREtPh was investigated against 155 MRSA isolates, including strains resistant to vancomycin and strains with decreased susceptibility to daptomycin and linezolid.36 Results illustrated that the activity against MRSA activity was not influenced by resistance, as the peptide retained a low MIC of 2−4 μg/mL in all screens. The RTbtREtPh peptide has also been investigated as a nasal decolonization agent. Nasal decolonization has a proven effect on the prevention of severe infections and function to control the spread of MRSA. In the study, persistent nasal MRSA and S. aureus carriers were treated for 3 days either with a vehicle or with ranging concentrations of RTbtREtPh (1−5%). A significant effect on nasal decolonization was observed within 2 days in subjects treated with 2 or 5% compared to vehicle. No safety issues were noted during the 9 week follow-up period apart from minimal reversible epithelial lesions.37 RTbtREtPh has also been selected as a clinical candidate (“LTX-109” by Lytix Biopharma, Norway) and has undergone clinical development through to Phase 2a for impetigo. The peptide was found to be well-tolerated and safe, and subgroup

Figure 10. In vivo evaluation of the potential of peptide LTX-109 in a murine skin infection model in comparison with commercial mupiricin cream.

analysis demonstrated improvement over placebo. However, the efficacy of peptide treatment did not reach the required significance end-point (p = 0.0534, expected p = 0.0500), and thus, the peptide is currently being investigated for other antimicrobial uses. Ex Vivo Fungal Studies

The potential for other topical applications was recently investigated in a study aimed at fungal infections where we evaluated 5 short AMPs as antifungal therapeutics against 24 strains of pathogenic fungi.24 Three of the peptides displayed strong general antifungal properties at low micromolar inhibitory concentrations against strains (Candida glabrata and C. krusei) with an inherent naturally reduced sensitivity toward certain antimycotics. RTbtREtPh was selected and evaluated as an antifungal remedy for C. albicans candidiasis in a human skin model and for the treatment of Trichophyton rubrum-induced onychomycosis in an infected human nail model. The effect of the peptide was pronounced and resulted in complete fungal eradication when combined with established nail permeation enhancers such as Med Nail and urea. In fact, the peptide was shown to display antifungal properties and a H

DOI: 10.1021/acs.accounts.8b00624 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 11. Transformation of a linear short AMP to the cyclic 2,5-DKP format to yield highly antimicrobial cyclic dipeptides while still adhering to the “2 + 2” pharmacophore.

colonization, possibly by mimicking natural marine peptidic antifoulants.40 The settling events ongoing at marine interfaces are similar to those on medical surfaces, and the peptides are also currently under development as repelling coatings for applications within both the marine and the medical fields.25,41 Collectively, the results from our studies illustrate that these short cationic amphiphilic peptides can be efficiently designed to display both enhanced activity and stability. They have been shown to display comparable or even superior activity to established antibacterial, antifungal, and antifouling commercial products, suggesting that they represent a particularly promising class of compounds for the eradication of various microbes in a variety of settings, ranging from the ocean to the clinic.

rapid mode of action superior to those of the commercial comparators Loceryl and Lamisil, and it was also found to be active against a clinical isolate of C. albicans with acquired fluconazole resistance. The results suggest that further development of these peptides against different types of fungal infections is warranted and illustrates that this type of compound could hold high potential as novel antifungal treatments.24

■

CONCLUSIONS AND OUTLOOK Our Account describes our efforts aimed at investigating and optimizing short synthetic AMPs and our ongoing efforts toward finding viable applications of these compounds. During the course of these studies, valuable insights into the mode of action of these compounds and general methods to address the instability of these peptides have been revealed. We believe that many of these discoveries are of general interest and can be applied to a wide range of the emerging antimicrobial peptidomimetics currently under development.38 The initial “2 + 2” pharmacophore dictates relatively accurately the minimal functional requirements for antibacterial activity, and studies have also shown that an ability to adopt an amphiphilic structure is beneficial for antibacterial activity. One interesting question is if this is best obtained using linear peptides? This hypothesis is currently being challenged in our ongoing studies, where we have transferred the linear pharmacophore onto the cyclic dipeptidic 2,5-diketopiperazine (DKP) scaffold in an attempt to generate more potent peptidomimetics (Figure 11).39 The first generation of such mixed DKPs has been prepared, and they were shown to display high antimicrobial activities toward bacteria and fungi with selected DKPs being superior to the positive linear peptide control RTbtREtPh.39 SAR studies revealed a similar dependence on cationic charge and the volume of the hydrophobic bulk as for linear peptides. The second generation is presently under construction and will encompass more elaborate substitution patterns, as it is believed that it may influence the ability of the peptides to form stable amphipathic structures in solution. Selected DKPs have also been investigated as marine antifoulants, and initial results are promising. Studies on mixed disubstituted DKPs illustrate that already these less complex compounds, failing to fulfill to the “2 + 2” pharmacophore, are still able to effectively prevent surface

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +46 706 855075; E-mail: [email protected] (J.S.) ORCID

John S. Mjøen Svendsen: 0000-0001-5945-6123 Margaret A. Brimble: 0000-0002-7086-4096 Johan Svenson: 0000-0002-4729-9359 Notes

The authors declare the following competing financial interest(s): J.S.M.S. is chief scientific officer and shareholder of Amicoat AS. Biographies John S. Mjøen Svendsen is a Professor of organic chemistry at UiT, The Arctic University of Norway, and Chief Scientific Officer of Amicoat AS, both in Tromsø, Norway. He has pioneered the field of synthetic ultrashort antimicrobial peptides and has cofounded the company Lytix Biopharma AS to pursue the technology in the treatment of cancer and Amicoat AS for antimicrobial surface control. Outside the field of antimicrobial peptides, he also works with small molecule protein kinase inhibitors in the field of neurodegeneration. Thomas M. Grant is a PhD Candidate who completed his BSc (Hons) in 2014 at the University of Auckland, under the supervision of Prof. Brent Copp. In July 2017, he commenced a PhD under the supervision of Prof. Margaret Brimble at the University of Auckland. Main research I

DOI: 10.1021/acs.accounts.8b00624 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

(10) Vorland, L. H.; Ulvatne, H.; Andersen, J.; Haukland, H. H.; Rekdal, Ø.; Svendsen, J. S.; Gutteberg, T. J. Antibacterial Effects of Lactoferricin B. Scand. J. Infect. Dis. 1999, 31, 179−184. (11) Kang, J. H.; Lee, M. K.; Kim, K. L.; Hahm, K. S. Structure− biological Activity Relationships of 11-residue Highly Basic Peptide Segment of Bovine Lactoferrin. Int. J. Pept. Protein Res. 1996, 48, 357− 363. (12) Haug, B. E.; Strom, M. B.; Svendsen, J. S. M. The Medicinal Chemistry of Short Lactoferricin-based Antibacterial Peptides. Curr. Med. Chem. 2007, 14, 1−18. (13) Strøm, M. B.; Svendsen, J. S.; Rekdal, Ø. Antibacterial Activity of 15-residue Lactoferricin Derivatives. J. Pept. Res. 2000, 56, 265−274. (14) Strøm, M. B.; Haug, B. E.; Rekdal, Ø.; Skar, M. L.; Stensen, W.; Svendsen, J. S. Important Structural Features of 15-residue Lactoferricin Derivatives and Methods for Improvement of Antimicrobial Activity. Biochem. Cell Biol. 2002, 80, 65−74. (15) Strøm, M. B.; Rekdal, Ø.; Svendsen, J. S. Antimicrobial Activity of Short Arginine-and Tryptophan-rich peptides. J. Pept. Sci. 2002, 8, 431−437. (16) Albada, B.; Metzler-Nolte, N. Highly Potent Antibacterial Organometallic Peptide Conjugates. Acc. Chem. Res. 2017, 50, 2510− 2518. (17) Junkes, C.; Wessolowski, A.; Farnaud, S.; Evans, R. W.; Good, L.; Bienert, M.; Dathe, M. The Interaction of Arginine-and Tryptophanrich Cyclic Hexapeptides with Escherichia coli Membranes. J. Pept. Sci. 2008, 14, 535−543. (18) Strøm, M. B.; Haug, B. E.; Skar, M. L.; Stensen, W.; Stiberg, T.; Svendsen, J. S. The Pharmacophore of Short Cationic Antibacterial Peptides. J. Med. Chem. 2003, 46, 1567−1570. (19) Haug, B. E.; Stensen, W.; Stiberg, T.; Svendsen, J. S. Bulky Nonproteinogenic Amino Acids Permit the Design of Very Small and Effective Cationic Antibacterial Peptides. J. Med. Chem. 2004, 47, 4159−4162. (20) Haug, B. E.; Stensen, W.; Kalaaji, M.; Rekdal, Ø.; Svendsen, J. S. Synthetic Antimicrobial Peptidomimetics with Therapeutic Potential. J. Med. Chem. 2008, 51, 4306−4314. (21) Svenson, J.; Karstad, R.; Flaten, G. E.; Brandsdal, B. O.; Brandl, M.; Svendsen, J. S. Altered Activity and Physicochemical Properties of Short Cationic Antimicrobial Peptides by Incorporation of Arginine Analogues. Mol. Pharmaceutics 2009, 6, 996−1005. (22) Karstad, R.; Isaksen, G.; Brandsdal, B. O.; Svendsen, J. S.; Svenson, J. Unnatural Amino Acid Side Chains as S1, S1′, and S2′ Probes Yield Cationic Antimicrobial Peptides with Stability Toward Chymotryptic Degradation. J. Med. Chem. 2010, 53, 5558−5566. (23) Saravolatz, L.; Pawlak, J.; Martin, H.; Saravolatz, S.; Johnson, L.; Wold, H.; Husbyn, M.; Olsen, W. Postantibiotic Effect and Postantibiotic sub-MIC Effect of LTX-109 and Mupirocin on Staphylococcus aureus Blood Isolates. Lett. Appl. Microbiol. 2017, 65, 410−413. (24) Stensen, W.; Turner, R.; Brown, M.; Kondori, N.; Svendsen, J. S.; Svenson, J. Short Cationic Antimicrobial Peptides Display Superior Antifungal Activities Toward Candidiasis and Onychomycosis in Comparison with Terbinafine and Amorolfine. Mol. Pharmaceutics 2016, 13, 3595−3600. (25) Trepos, R.; Cervin, G.; Pile, C.; Pavia, H.; Hellio, C.; Svenson, J. Evaluation of Cationic Micropeptides Derived from the Innate Immune System as Inhibitors of Marine Biofouling. Biofouling 2015, 31, 393− 403. (26) Isaksson, J.; Brandsdal, B. O.; Engqvist, M.; Flaten, G. E.; Svendsen, J. S. M.; Stensen, W. A. Synthetic Antimicrobial Peptidomimetic (LTX 109): Stereochemical Impact on Membrane Disruption. J. Med. Chem. 2011, 54, 5786−5795. (27) Albada, H. B.; Prochnow, P.; Bobersky, S.; Bandow, J. E.; Metzler-Nolte, N. Highly Active Antibacterial Ferrocenoylated or Ruthenocenoylated Arg-Trp Peptides can be Discovered by an L-to-D Substitution Scan. Chem. Sci. 2014, 5, 4453−4459. (28) Svenson, J.; Stensen, W.; Brandsdal, B. O.; Haug, B. E.; Monrad, J.; Svendsen, J. S. Antimicrobial Peptides with Stability Toward Tryptic Degradation. Biochemistry 2008, 47, 3777−3788.

interests include green chemistry techniques and bioactive antifouling peptides. David Rennison is a Senior Research Fellow at the School of Chemical Sciences, University of Auckland, where he works in close alignment with both the pharmaceutical and agritech industries on biologically driven projects across diverse areas such as animal health, invasive pest eradication, marine biofouling, and environmental chemistry. Margaret A. Brimble is Director of Medicinal Chemistry and a Distinguished Professor at the University of Auckland, where her research focuses on the synthesis of bioactive natural products and peptide chemistry. She was elected Fellow of the Royal Society (London) and awarded the RSC Sosnovsky Award in Cancer Therapy (2018). She won the 2012 RSNZ Rutherford, MacDiarmid, and Hector Medals and was conferred the Queen’s Honour CNZM. She is PastPresident of IUPAC Division III and an Associate Editor for Organic and Biomolecular Chemistry. She discovered the first drug “trofinetide” to treat Rett Syndrome, which is in phase III clinical trials with Neuren Pharmaceuticals and cofounded the company SapVax to develop selfadjuvanting cancer vaccines. Johan Svenson is Head of Research at the Department of Chemistry and Materials at RISE Research Institutes of Sweden. He is an Associate Professor within biomaterials chemistry at the Linnaeus University (Sweden). His research focus spans bioactive peptides, natural products chemistry, and functionalized materials with the ability to repel microorganism settlement and colonization.

■

ACKNOWLEDGMENTS J.S.M.S. and J.S. wish to thank all previous and present coworkers who have made substantial and fundamental contributions to the development and understanding of these fascinating short AMPs.

■

REFERENCES

(1) Zasloff, M. Antimicrobial Peptides of Multicellular Organisms. Nature 2002, 415, 389−395. (2) Wenzel, M.; Chiriac, A. I.; Otto, A.; Zweytick, D.; May, C.; Schumacher, C.; Gust, R.; Albada, H. B.; Penkova, M.; Krämer, U.; et al. Small Cationic Antimicrobial Peptides Delocalize Peripheral Membrane Proteins. Proc. Natl. Acad. Sci. U. S. A. 2014, 111, E1409−E1418. (3) Liu, Y. Y.; Wang, Y.; Walsh, T. R.; Yi, L. X.; Zhang, R.; Spencer, J.; Doi, Y.; Tian, G.; Dong, B.; Huang, X.; et al. Emergence of PlasmidMediated Colistin Resistance Mechanism MCR-1 in Animals and Human Beings in China: A Microbiological and Molecular Biological Study. Lancet Infect. Dis. 2016, 16, 161−168. (4) Jenssen, H.; Hamill, P.; Hancock, R. E. Peptide Antimicrobial Agents. Clin. Microbiol. Rev. 2006, 19, 491−511. (5) Van Hooijdonk, A. C.; Kussendrager, K.; Steijns, J. In vivo Antimicrobial and Antiviral Activity of Components in Bovine Milk and Colostrum Involved in Non-Specific Defence. Br. J. Nutr. 2000, 84, 127−134. (6) Bellamy, W.; Takase, M.; Yamauchi, K.; Wakabayashi, H.; Kawase, K.; Tomita, M. Identification of the Bactericidal Domain of Lactoferrin. Biochim. Biophys. Acta, Protein Struct. Mol. Enzymol. 1992, 1121, 130− 136. (7) Gifford, J. L.; Hunter, H. N.; Vogel, H. Lactoferricin. Cell. Mol. Life Sci. 2005, 62, 2588−2598. (8) Hwang, P. M.; Zhou, N.; Shan, X.; Arrowsmith, C. H.; Vogel, H. J. Three-Dimensional Solution Structure of Lactoferricin B, an Antimicrobial Peptide Derived from Bovine Lactoferrin. Biochemistry 1998, 37, 4288−4298. (9) Vogel, H. J.; Schibli, D. J.; Jing, W.; Lohmeier-Vogel, E. M.; Epand, R. F.; Epand, R. M. Towards a Structure-function Analysis of Bovine Lactoferricin and Related Tryptophan- and Arginine-containing Peptides. Biochem. Cell Biol. 2002, 80, 49−63. J

DOI: 10.1021/acs.accounts.8b00624 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research (29) Svenson, J.; Vergote, V.; Karstad, R.; Burvenich, C.; Svendsen, J. S.; De Spiegeleer, B. Metabolic Fate of Lactoferricin-based Antimicrobial Peptides: Effect of Truncation and Incorporation of Amino Acid Analogs on the in vitro Metabolic Stability. J. Pharmacol. Exp. Ther. 2010, 332, 1032−1039. (30) Karstad, R.; Isaksen, G.; Wynendaele, E.; Guttormsen, Y.; De Spiegeleer, B.; Brandsdal, B. O.; Svendsen, J. S.; Svenson, J. Targeting the S1 and S3 Subsite of Trypsin with Unnatural Cationic Amino Acids Generates Antimicrobial Peptides with Potential for Oral Administration. J. Med. Chem. 2012, 55, 6294−6305. (31) Sivertsen, A.; Isaksson, J.; Leiros, H. K. S.; Svenson, J.; Svendsen, J. S.; Brandsdal, B. O. Synthetic Cationic Antimicrobial Peptides Bind with Their Hydrophobic Parts to Drug Site II of Human Serum Albumin. BMC Struct. Biol. 2014, 14, 4. (32) Svenson, J.; Brandsdal, B. O.; Stensen, W.; Svendsen, J. S. Albumin Binding of Short Cationic Antimicrobial Micropeptides and its Influence on the in vitro Bactericidal Effect. J. Med. Chem. 2007, 50, 3334−3339. (33) Sivertsen, A.; Brandsdal, B. O.; Svendsen, J. S.; Andersen, J. H.; Svenson, J. Short Cationic Antimicrobial Peptides Bind to Human Alpha-1 Acid Glycoprotein with no Implications for the in vitro Bioactivity. J. Mol. Recognit. 2013, 26, 461−469. (34) Flaten, G. E.; Kottra, G.; Stensen, W.; Isaksen, G.; Karstad, R.; Svendsen, J. S.; Daniel, H.; Svenson, J. In vitro Characterization of Human Peptide Transporter hPEPT1 Interactions and Passive Permeation Studies of Short Cationic Antimicrobial Peptides. J. Med. Chem. 2011, 54, 2422−2432. (35) Flemming, K.; Klingenberg, C.; Cavanagh, J. P.; Sletteng, M.; Stensen, W.; Svendsen, J. S.; Flægstad, T. High in vitro Antimicrobial Activity of Synthetic Antimicrobial Peptidomimetics Against Staphylococcal Biofilms. J. Antimicrob. Chemother. 2008, 63, 136−145. (36) Saravolatz, L. D.; Pawlak, J.; Johnson, L.; Bonilla, H.; Saravolatz, L. D.; Fakih, M. G.; Fugelli, A.; Olsen, W. M. In vitro Activities of LTX109, a Synthetic Antimicrobial Peptide, against Methicillin-resistant, Vancomycin-intermediate, Vancomycin-resistant, Daptomycin-nonsusceptible, and Linezolid-nonsusceptible Staphylococcus aureus. Antimicrob. Agents Chemother. 2012, 56, 4478−4482. (37) Nilsson, A. C.; Janson, H.; Wold, H.; Fugelli, A.; Andersson, K.; Håkangård, C.; Olsson, P.; Olsen, W. M. LTX-109 is a Novel Agent for Nasal Decolonization of Methicillin-resistant and -sensitive Staphylococcus aureus. Antimicrob. Agents Chemother. 2015, 59, 145−151. (38) Molchanova, N.; Hansen, P. R.; Franzyk, H. Advances in Development of Antimicrobial Peptidomimetics as Potential Drugs. Molecules 2017, 22, 1430. (39) Labrière, C.; Kondori, N.; Caous, J. S.; Boomgaren, M.; Sandholm, K.; Ekdahl, K. N.; Hansen, J. H.; Svenson, J. Development and Evaluation of Cationic Amphiphilic Antimicrobial 2,5-diketopiperazines. J. Pept. Sci. 2018, 24, e3090. (40) Trepos, R.; Cervin, G.; Hellio, C.; Pavia, H.; Stensen, W.; Stensvåg, K.; Svendsen, J. S.; Haug, T.; Svenson, J. Antifouling Compounds from the Sub-Arctic Ascidian Synoicum pulmonaria: Synoxazolidinones A and C, Pulmonarins A and B, and Synthetic Analogues. J. Nat. Prod. 2014, 77, 2105−2113. (41) Andrea, A.; Molchanova, N.; Jenssen, H. Antibiofilm Peptides and Peptidomimetics with Focus on Surface Immobilization. Biomolecules 2018, 8, 27.

K

DOI: 10.1021/acs.accounts.8b00624 Acc. Chem. Res. XXXX, XXX, XXX−XXX