Rational Design of Amphipathic Antimicrobial Peptides with Alternating L-/D-Amino Acids That Form Helical Structures

Abstract

Antimicrobial peptides (AMPs) are promising therapeutic agents against bacteria. We have previously reported an amphipathic AMP Stripe composed of cationic L-Lys and hydrophobic L-Leu/L-Ala residues, and Stripe exhibited potent antimicrobial activity against Gram-positive and Gram-negative bacteria. Gramicidin A (GA), composed of repeating sequences of L- and D-amino acids, has a unique β^6.3-helix structure and exhibits broad antimicrobial activity. Inspired by the structural properties and antimicrobial activities of LD-alternating peptides such as GA, in this study, we designed Stripe derivatives with LD-alternating sequences. We found that simply alternating L- and D-amino acids in the Stripe sequence to give StripeLD caused a reduction in antimicrobial activity. In contrast, AltStripeLD, with cationic and hydrophobic amino acids rearranged to yield an amphipathic distribution when the peptide adopts a β^6.3-helix, displayed higher antimicrobial activity than AltStripe. These results suggest that alternating L-/D-cationic and L-/D-hydrophobic amino acids in accordance with the helical structure of an AMP may be a useful way to improve antimicrobial activity and develop new AMP drugs.

Introduction

Antimicrobial peptides (AMPs) represent alternative therapeutic agents to conventional antibiotics and a potential solution to the global health issue caused by the inappropriate use of antibiotics that has led to increasing antibiotic resistance of bacteria and the emergence of multidrug-resistant (MDR) strains.^1,2) AMPs display potent antimicrobial activity and a broad spectrum of activity against bacteria by disrupting the integrity of bacterial plasma membranes, which are essential for bacteria to sustain cellular functions.^3–8) Many AMPs are positively charged and interact with the negative charge of the bacterial cell wall/membrane to form channel-forming agents and ionophores in the bacterial membrane. Targeting bacterial cell membranes by AMPs has a low probability of leading to the emergence of bacterial resistance.^9,10) Forming stable secondary structures, such as α-helices and β-sheets, is important in ensuring the potent antimicrobial activity of AMPs. Thus, developing and optimizing the secondary structures of de novo AMPs should aid in optimizing their function.^11–13) For example, several AMPs that form stable α-helices have been developed.^14,15) We have recently reported the rationally designed amphipathic helical AMP termed Stripe. This AMP is composed of three tandem repeats of L-amino acids (L-AA), KLLKKAG, in which the hydrophobic alanine (Ala) and leucine (Leu) residues are positioned on one face of the helix and cationic lysine (Lys) residues are on the opposing face¹⁶⁾ (Fig. 1). Stripe exhibited potent antimicrobial activity against Gram-positive and Gram-negative microbes and showed no hemolytic activity.¹⁶⁾ In general, α-helical AMPs are L-peptides composed of L-amino acids and show a right-handed screw sense.⁵⁾ In contrast, LD-peptides containing L-AA and D-AA fold into α-helical structures and wide-pore β- and π-helices that are semi-extended conformations.⁷⁾ Gramicidin A (GA), composed of repeating sequences of L- and D-AA, has a unique wide-pore helical structure (i.e., β^6.3-helix) and exhibits a broad spectrum of antimicrobial activity.^17–28) Inspired by the structural properties and antimicrobial activities of LD-alternating peptides such as GA, in this study, we designed Stripe derivatives with LD-alternating sequences. We conducted the rational design of AMPs using the principles of β-helices and the Stripe sequence to produce a potent AMP (Fig. 1).

Fig. 1. Development of the Antimicrobial Peptide Based on the Stripe Sequence That Forms a β-Helix

Results and Discussion

Initially, we designed four types of peptides, Stripe, StripeLD, AltStripe and AltStripeLD. Stripe is an L-peptide containing mainly cationic L-Lys and hydrophobic L-Leu/L-Ala residues that arrange to form an amphipathic α-helical structure. StripeLD has the same AA sequence as Stripe but with alternating L-AA and D-AA to yield an LD-peptide. AltStripe is an L-peptide where the L-Lys and L-Leu/L-Ala residues have been rearranged to give an amphipathic helix when the sequence adopts a β^6.3-helix conformation, as found for GA (Fig. 2). AltStripeLD is an LD-peptide with the same composition and sequence as AltStripe with alternating L- and D-AA under the hypothesis that LD-peptide is expected to form a β^6.3-helix similar to GA. GA adopts a β^6.3-helical structure with seven amino acids forming a helix turn. Thus, we designed AltStripeLD with three tandem repeats of alternating L-AA and D-AA, i.e., LlAkKkG (where lower case letters indicate D-AA and upper case letters represent L-AA), with the three cationic AA located on one face of the helix and three hydrophobic AA found on the opposing helix face of the β^6.3-helical wheel (Fig. 2). The four designed peptides were synthesized by microwave-assisted solid-phase peptide synthesis, purified using reversed-phase high-performance liquid chromatography and identified using electrospray ionization time-of-flight mass spectrometry.

Fig. 2. Design of the Four Peptides

(a) Peptide sequences of Stripe, StripeLD, AltStripe and AltStripeLD. (b) α-Helical wheel of L-peptides and β^6.3-helical wheel of the LD-peptides. The upper-case letters indicate L-AA, and the lower-case letters indicate D-AA.

The preferred secondary structures of the four peptides were evaluated by circular dichroic (CD) spectropolarimetry in phosphate-buffered saline (PBS, pH 7.4) containing 1% sodium dodecyl sulfate (SDS) to mimic the environment of microbial membranes¹⁵⁾ (Fig. 2). The CD spectrum of the L-peptide Stripe showed negative maxima at approximately 208 and 222 nm, indicating that Stripe formed a right-handed helical structure. L-Peptide AltStripe also showed a similar spectrum, indicating the formation of a right-handed helix, but the intensity was weaker compared to the CD spectrum of Stripe (Fig. 3a). This may be caused by the loss of amphiphilicity of the AltStripe α-helix (Fig. 2b). CD spectra of StripeLD and AltStripeLD were similar with a peak around 200 nm; however, the cotton effect was weak, indicating that equal amounts of L- and D-AA formed secondary structures under the buffer conditions used.

Fig. 3. CD Spectra of (a) L-Peptides Stripe and AltStripe and (b) LD-Peptides StripeLD and AltStripeLD in Phosphate-Buffered Saline with 1% Sodium Dodecyl Sulfate (pH 7.4)

Peptide concentration = 100 µM.

We performed molecular modeling of AltStripeLD using the MOE program because defining the β^6.3-helical structure of AltStripeLD by CD was challenging. As shown in Fig. 4, molecular simulations in which the peptide was hypothesized to form β^6.3-helical structure like GA tentatively showed that AltStripeLD formed an amphipathic structure with side chains of cationic and hydrophobic AA arranged on opposite faces of the β^6.3-helix.

Fig. 4. Molecular Modeling of AltStripeLD as Viewed in (a) This Is a Top View or a View Looking Along the Helix Axis, Whereas in (b) This Is a Side View (Perpendicular to the Helix Axis) of the Helix

Cationic amino acids are shown in pink and hydrophobic amino acid shown in gray.

Next, the antimicrobial activities of Stripe, StripeLD, AltStripe and AltStripeLD against Gram-negative and Gram-positive bacteria were evaluated using the turbidimetric method and the broth micro-dilution assay.²⁹⁾ Gram-positive bacterium Staphylococcus aureus (S. aureus) NBRC13276 and Gram-negative bacteria Escherichia coli (E. coli) DH5α, Escherichia coli NBRC 3972, Proteus vulgaris (P. vulgaris) NBRC 3045, Salmonella enterica (S. enterica) subsp. enterica NBRC 100797, Klebsiella pneumoniae (K. pneumoniae) NBRC 3512 and Pseudomonas aeruginosa (P. aeruginosa) NBRC13275 were used in this assay, and the results are shown in Table 1. Stripe showed broad-spectrum activity against various bacteria except for P. vulgaris, whereas the activity of LD-peptide StripeLD was lower than that of L-peptide Stripe. This low activity may be because StripeLD does not form a secondary structure that exhibits amphipathic properties. L-Peptide AltStripe, which forms an α-helix but has lost its amphipathic property, showed much weaker antimicrobial activity than the other three peptides. In contrast, LD-peptide AltStripeLD with the same sequence as AltStripe showed higher activity than AltStripe because it would form the β^6.3-helix with the amphipathic sequence. These results suggested that alternating between L-AA and D-AA is an effective approach to increase the activity of peptides designed to form amphiphilic β^6.3-helix composed of seven amino acid repeats.

Table 1. Antimicrobial Activity (µM) of Stripe, StripeLD, AltStripe and AltStripeLD against Gram-Negative and Gram-Positive Bacteria

	Minimum inhibitory concentration (MIC) (µM)
Bacteria	Stripe	StripeLD	AltStripe	AltStripeLD
E. coli (DH5α)	3.1	25	100	25
E. coli	3.1	50	N.D.a)	25
P. vulgaris	N.D.	N.D.	N.D.	N.D.
S. enterica	1.6	6.3	N.D.	12.5
K. pneumoniae	1.6	25	100	25
P. aeruginosa	1.6	>100	N.D.	100
S. aureus	12.5	N.D.	N.D.	N.D.

a) N.D.: Not determined.

Antimicrobial peptides that are cytotoxic typically damage cellular membranes. Therefore, a hemolysis assay on human blood cells and a water-soluble tetrazolium salt (WST) assay on human cell lines were performed to evaluate the cytotoxicity of the peptides. Human blood cells were treated with Stripe, StripeLD, AltStripe and AltStripeLD at different concentrations, and the absorption at 495 nm was measured. The WST assay was performed in the presence of the peptides and TIG-3 cells or HEK293 cells. Stripe, Stripe LD, AltStripe and AltStripeLD showed no hemolytic activity or cytotoxicity up to 100 µM treatment (Fig. 5, Table 2).

Fig. 5. Effect of Stripe, StripeLD, AltStripe and AltStripeLD Peptides on Cell Growth

HEK293 (a) and TIG-3 (b) cells were incubated with the indicated concentrations of the peptides for 72 h and were subsequently subjected to the WST assay. Experiments were performed in triplicate.

Table 2. Hemolysis and Cytotoxicity of Stripe, StripeLD, AltStripe and AltStripeLD

	Stripe	StripeLD	AltStripe	AltStripeLD
Hemolysis (µM)	>100	>100	>100	>100
Cytotoxicity	N.D.a)	N.D.	N.D.	N.D.

a) N.D.: Not determined.

Conclusion

In this study, we designed and synthesized a series of antimicrobial peptides based on Stripe by replacing L-AA with D-AA and arranging cationic and hydrophobic AA to mimic the β^6.3-helix formed by GA. For elucidating the structure–activity relationship (SAR) of Stripe, the secondary structures of Stripe, StripeLD, AltStripe and AltStripeLD were analyzed, and their antimicrobial activity against Gram-positive and Gram-negative bacteria was evaluated. CD analysis of Stripe and AltStripe composed of L-AA revealed that these peptides adopted α-helical structures. StripeLD and AltStripeLD with equal numbers of L- and D-AA gave rise to a weak peak around 200 nm in the CD spectra, suggesting that these peptides form secondary conformations in PBS with 1% SDS. In addition, molecular simulations showed that AltStripeLD forms an amphipathic structure with cationic AA and hydrophobic AA arranged on opposing faces of the β^6.3-helix. Alternating between L-AA and D-AA of the Stripe sequence to give StripeLD reduced the broad antimicrobial activity (Table 1), indicating that the amphipathic nature of these peptides is crucial for activity. Rearranging the Stripe sequence using solely L-AA to give AltStripe showed weak antimicrobial activity, presumably because this peptide adopted a non-amphipathic α-helical structure (Figs. 2, 3a). AltStripeLD showed stronger and broader activity than AltStripe, indicating that alternating between L- and D-AA and reorganizing the Stripe sequence yielded an antimicrobial amphipathic β^6.3-helix. In addition, all peptides displayed no hemolytic activity and cytotoxicity, indicating LD-alternation can be performed without drastic increases in cytotoxicity. We have shown that replacing selected L-AA with D-AA can yield a peptide that adopts an amphipathic β^6.3-helix structure with antimicrobial activity. The results represent an LD-alternation approach to developing antimicrobial peptides. We anticipate that this method, which controls the secondary structure of AMPs and induces antimicrobial activity by rationally designing the selection and sequence of hydrophobic and cationic L-AA/D-AA, will be useful for developing new AMPs.

Experimental

Synthesis and Characterization of Peptides

Chemicals were purchased from Sigma-Aldrich Co. LLC (St. Louis, MO, U.S.A.), Kanto Chemicals Co. Inc. (Tokyo, Japan), Tokyo Chemical Industry Co. Ltd. (Tokyo, Japan), Wako Pure Chemical Corporation (Osaka, Japan), Watanabe Chemical Industries, Ltd. (Hiroshima, Japan), and used without further purification. Mass spectra were obtained on a Shimadzu IT-TOF MS equipped with an electrospray ionization source.

Peptide Synthesis

The designed peptides were synthesized by Fmoc-based solid-phase methods using Liberty Blue (CEM Corp., Matthews, NC, U.S.A.). A representative coupling and deprotection cycle are described as follows. Protide resin was soaked for 30 min in dichloromethane. After the resin was washed with N,N-dimethylformamide (DMF), Fmoc-amino acid (5 equivalent (equiv.)), Oxyma (10 equiv.), N,N′-diisopropylcarbodiimide (DIC) (10 equiv.) dissolved in solution of DMF were added to the resin. Fmoc protective groups were deprotected using 20% piperidine in DMF. The peptide was suspended in cleavage cocktail (95% trifluoroacetic acid (TFA), 2.5% water, 2.5% triisopropylsilane) at room temperature (r.t.) for 3 h to cleave from the resin. TFA was evaporated to a small volume under a stream of N₂ and dripped into cold ether to precipitate the peptide. The peptides were dissolved in dimethyl sulfoxide, purified using reverse-phase high performance liquid chromatography using a Discovery^® BIO Wide Pore C18 column (25 cm × 21.2 mm solvent A: 0.1% TFA/water, solvent B: 0.1% TFA/MeCN, flow rate: 10.0 mL/min, gradient: 10–90% gradient of solvent B over 30 min). After being purified, the peptide solutions were lyophilized. Peptide purity was assessed using analytical HPLC and Inertsil WP300 C18 column (25 cm × 4.6 mm; solvent A: 0.1% TFA/water, solvent B: 0.1% TFA/MeCN, flow rate: 1.0 mL/min, gradient: 10–90% gradient of solvent B over 30 min) (see the supplementary materials).

Circular Dichroism Spectrometry

Circular dichroism (CD) spectra were recorded using a 1.0 mm path length cell. Peptides were dissolved in 20 mM phosphate buffer solution (pH = 7.4) with 1% sodium dodecyl sulfate (SDS) at concentrations of 100 µM.

Peptide Modeling

PDB structure data (1MAG) for GA was loaded. Prior to modeling, a conformational search of the backbone was performed using Molecular Operating Environment (MOE) 2022.02 with Amber10:EHT as a force field. A conformational search of AltStripeLD (Rigid body set: helix backbone was constrained, RMSD gradient: 0.005, iteration limit: 10000, MM iteration limit: 500, rejection limit: 100) was performed.

Antimicrobial Activity

Selected bacterial strains were obtained from Biological Resource Center, NITE (NBRC; Tokyo, Japan) and American Type Culture Collection (ATCC; U.S.A.). E. coli DH5α was purchased from BioDynamics Laboratory Inc. (Tokyo, Japan). The antimicrobial activities of the peptides against the bacteria, including Gram-positive (Staphylococcus aureus NBRC 13276) and Gram-negative (E. coli DH5α, E. coli NBRC 3972, Proteus vulgaris NBRC 3045, Salmonella enterica ssp. enterica NBRC 100797, Klebsiella pneumoniae NBRC 3512 and Pseudomonas aeruginosa NBRC 13275) were measured using the standard broth microdilution method as previously described.²⁹⁾ Briefly, the bacteria were inoculated and grown overnight at 37°°C on the media for other organisms i (agar medium) and then collected with the media for other organisms ii (liquid medium), according to the Japanese Pharmacopoeia 18th Edition. Each peptide was 2-fold serially diluted with an initial concentration of 1000 to 2.0 µM by phosphate buffer solution for use. Then, 10 µL per well of the peptide solution was added to each well of a sterile 96-well plate. Subsequently, 90 µL per well of inoculation with 104 CFUs (colony forming units) per mL were added to each well, and the plate was incubated for 16 h at 35°C. The minimum inhibitory concentration (MIC) was defined as the lowest concentration of the peptide which completely inhibited the growth of bacteria by visual inspection at 595 nm.

Cell Culture

HEK293 human embryonic kidney cells and TIG-3 human normal fibroblasts were cultured in Dullbecco’s modified Eagle’s medium (Sigma-Aldrich) containing 10% fetal bovine serum and 100 µg/mL of kanamycin.

Cell Viability Assay

Cell viability was determined using water-soluble tetrazolium WST-8{4-[3-(2-methoxy-4-nitrophenyl)-2-(4-nitrophenyl)-2H-5-tetrazolio]-1,3-benzene disulfonate} for the spectrophotometric assay according to the manufacturer’s instructions (Dojindo, Tokyo, Japan). HEK293 or TIG-3 cells were seeded in 96-well culture plates with 3,000 cells per well and incubated at 37°C for 24 h. The cells were treated with 1–100 µM of peptides and the cells were incubated for 72 h. Cells treated with compounds were incubated with WST-8 reagent for 0.5 h at 37°C in a humidified atmosphere of 5% CO₂. The absorbance at 450 nm of the medium was measured using an EnVision Multilabel Plate Reader (PerkinElmer, Inc., Waltham, MA, U.S.A.).

Hemolysis Activity

Human red blood cells (RBCs) were kindly supplied by the Japanese Red Cross Society (Tokyo, Japan), which collects from volunteers under informed consent. The hemolysis test of peptides was performed using previously reported method.³⁰⁾ In short, the RBCs were washed three times and resuspended with 172 mM Tris–HCl buffer (pH = 7.6) and, 100 µL of RBC solution were incubated with 100 µL of each peptide for 30 min at 37°°C. The suspensions were centrifuged for 5 min at 400 rpm. The absorbance of supernatant was measured at 450 nm. M-Lycotoxin^31,32) served as the positive controls.

Acknowledgments

This study was supported in part by Grants from AMED under Grant numbers 23mk0101197, 23ae0121013, 23ak0101185, 23mk0101220, 23fk0210110 and 23fk0310506 (all to Y.D.). The study also received support from the Japan Society for the Promotion of Science (KAKENHI, Grants 21K05320 and 23H04926, both to Y.D., JP22K15257 to H.Y.), JST (ACT-X, Grant number JPMJAX222L, Japan to H.Y.) and Takeda Science Foundation (to H.Y.).

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

References

• 1) Fjell C. D., Hiss J. A., Hancock R. E., Schneider G., Nat. Rev. Drug Discov., 11, 37–51 (2011).
• 2) Spohn R., Daruka L., Lázár V., et al., Nat. Commun., 10, 4538 (2019).
• 3) Bahar A. A., Ren D., Pharmaceuticals, 6, 1543–1575 (2013).
• 4) Hancock R. E. W., Sahl H.-G., Nat. Biotechnol., 24, 1551–1557 (2006).
• 5) Zasloff M., Nature (London), 415, 389–395 (2002).
• 6) Da Cunha N. B., Cobacho N. B., Viana J. F. C., Lima L. A., Sampaio K. B. O., Dohms S. S. M., Ferreira A. C. R., de la Fuente-Núñez C., Costa F. F., Franco O. L., Dias S. C., Drug Discov. Today, 22, 234–248 (2017).
• 7) Ageitos J. M., Sańchez-Peŕez A., Calo-Mata P., Villa T. G., Biochem. Pharmacol., 133, 117–138 (2017).
• 8) Kang H.-K., Kim C., Seo C. H., Park Y., J. Microbiol., 55, 1–12 (2017).
• 9) Nikaido H., Annu. Rev. Biochem., 78, 119–146 (2009).
• 10) Chang H. H., Cohen T., Grad H. Y., Hanage P. W., O’Brien F. T., Lipsitch M., Biol. Rev. Camb. Philos. Soc., 79, 101–116 (2015).
• 11) Walensky L. D., Kung A. L., Escher I., Malla T. J., Burbuto S., Wright R. D., Wagner G., Verdine G. L., Korsmeyer S., Science, 305, 1466–1470 (2004).
• 12) Porter E. A., Wang X., Lee H. S., Weisblum B., Gellman S. H., Nature (London), 404, 565 (2000).
• 13) Elliott A. G., Huang J. X., Neve S., Zuegg J., Edwards I. A., Cain A. K., Boinett C. J., Barquist L., Lundberg C. V., Steen J., Butler M. S., Mobli M., Porter K. M., Blaskovich M. A. T., Lociuro S., Strandh M., Cooper M. A., Nat. Commun., 11, 3184 (2020).
• 14) Huang Y., Huang J., Chen Y., Protein Cell, 1, 143–152 (2010).
• 15) Yokoo H., Hirano M., Misawa T., Demizu Y., ChemMedChem, 16, 1226–1233 (2021).
• 16) Misawa T., Goto C., Shibata N., Hirano M., Kikuchi Y., Naito M., Demizu Y., MedChemComm, 10, 896–900 (2019).
• 17) Urry D. W., Goodall M. C., Glickson J. D., Mayers D. F., Proc. Natl. Acad. Sci. U.S.A., 68, 1907–1911 (1971).
• 18) Arseniev A. S., Barsukov I. L., Bystrov V. F., Lomize A. L., Ovchinnikov Y. A., FEBS Lett., 186, 168–174 (1985).
• 19) Wallace B. A., Ravikumar K., Science, 241, 182–187 (1988).
• 20) Langs D. A., Science, 241, 188–191 (1988).
• 21) Ketchem R. R., Hu W., Cross T. A., Science, 261, 1457–1460 (1993).
• 22) Doyle D. A., Wallace B. A., J. Mol. Biol., 266, 963–977 (1997).
• 23) Wallace B. A., J. Struct. Biol., 121, 123–141 (1998).
• 24) Kovacs F., Quine J., Cross T. A., Proc. Natl. Acad. Sci. U.S.A., 96, 7910–7915 (1999).
• 25) Hotchkiss R. D., Dubos R. J., J. Biol. Chem., 132, 791–792 (1940).
• 26) Sarges R., Witkop B., J. Am. Chem. Soc., 87, 2011–2020 (1965).
• 27) Herrell W. E., Heilman D., J. Clin. Invest., 20, 583–591 (1941).
• 28) Kelkar D. A., Chattopadhyay A., Biochim. Biophys. Acta, Biomembr., 1768, 2011–2025 (2007).
• 29) Goto C., Hirano M., Hayashi K., Kikuchi Y., Hara-Kudo Y., Misawa T., Demizu Y., ChemMedChem, 14, 1911–1916 (2019).
• 30) Strandberg E., Zerweck J., Horn D., Pritz G., Berditsch M., Bürck J., Wadhwani P., Ulrich A. S., J. Pept. Sci., 21, 436–445 (2015).
• 31) Akishiba M., Takeuchi T., Kawaguchi Y., Sakamoto K., Yu H. H., Nakase I., Takatani-Nakase T., Madani F., Gräslund A., Futaki S., Nat. Chem., 9, 751–761 (2017).
• 32) Yan L., Adams M. E., J. Biol. Chem., 273, 2059–2066 (1998).