Alteration of the Antifungal Action Mechanism Due to Structural Changes in the Antimicrobial Peptide, HnMc

Abstract

The rapid induction of drug resistance is considered a fatal drawback of conventional antibiotics and requires the continuous development of new antibiotics. Accordingly, antibacterial peptides (AMPs) have attracted interest as next-generation antibiotics and many studies have been conducted. However, much remains unknown regarding the mechanism of AMPs and the effects of amino acid sequence changes. We compared the structures and antifungal effects of HnMc-W (F1W substitution, straight alpha-helical structure), HnMc-WP1 (S9P substitution, bending alpha-helical structure), and HnMc-WP2 (addition of the PXXP motif, helix-to-helix structure) to those of a parent hybrid AMP (HnMc) regarding their mechanism of action. The most active was HnMc-WP2, which exhibited an antifungal effect via membranolytic action on the fungal cell membrane. The others inhibited fungal growth by inducing apoptosis through reactive oxygen species production caused by mitochondrial damage. This study proposes the addition of the ‘PXXP’ motif in the design of AMPs acting on cell membranes.

1. Introduction

Recently, several researchers have attempted to identify novel antibiotics that can overcome the rapid spread of antibiotic resistance. All living organisms secrete or have defensive substances against external attacks by microorganisms, one of which are antimicrobial peptides, which are being actively studied worldwide and recognized as new next-generation antibiotics [1,2,3,4,5]. Humans and living organisms are constantly attacked by various external and internal microorganisms; however, they have an innate immune system whereby they protect themselves and sustain life through autoimmunity [6,7]. The major molecules of the innate immune system are antimicrobial peptides (AMPs). Although their amino acid sequences and structures differ, they are produced in almost all living organisms, including viruses, bacteria, protozoa, fungi, plants, and mammals [1,2,3,4,5]. AMPs act as first-line defense molecules when organisms are attacked by pathogens. Their cationic and amphipathic characteristics allow them to bind to the lipid components of the negatively charged cell membrane via electrostatic interactions and change the electrical potential of the microbial cell membrane or destroy it [8,9]. Some also enter the cell membrane, attack intracellular targets, paralyze cellular functions, and exhibit antimicrobial activity [8,9]. Because their main target is the cell or organelle membrane, pathogens can be eliminated in a very short period, making it very difficult and rare for pathogens to induce resistance to AMPs. Moreover, because most of these have excellent anti-inflammatory and antibacterial activities, they are important next-generation antimicrobial agents for overcoming drug resistance.

Because the cell types, cell membrane components, and organelles of pathogenic bacteria and fungi differ, it is difficult to determine which of the various mechanisms of AMPs is superior. Many studies have investigated the antibacterial and cytotoxic effects of various amino acid sequence changes, including deletion, substitution, hybridization, cyclization, and lipidation [10,11,12,13,14]. Because the roles of individual amino acids tend to vary depending on their sequence, there are currently no valid general design rules for AMPs. Because proline induces a “bending” in the conformational changes in peptides and proteins, one or two amino acids can cause a drastic change in the AMP structure [15,16,17,18,19]. In a previous study, it was found that one proline in the sequence of HnMc, a chimeric hybrid AMP, induced bending, and two prolines had a ‘helix-to-helix’ structure, affecting antibacterial and cytotoxic effects [20].

However, much remains unknown regarding the mechanisms of AMPs and the effects of amino acid sequence changes. Furthermore, even if an AMP is the same, its mechanism of action may change in bacteria or fungi owing to different cell types and physiochemical characteristics. Therefore, this study aimed to investigate the alterations in the antifungal effects and mechanisms caused by changes in the sequence and structure of HnMc.

2. Materials and Methods

2.1. Materials

Rink Amide ProTide resin, ethyl 2-cyano-2-(hydroxyimino)acetate (Oxyma Pure), and 9-fluorrenylmethoxycarbonyl (Fmoc) amino acids were purchased from CEM Co. (Matthews, NC, USA). Trifluoroacetic acid (TFA), triisopropylsilane (TIS), and N, N-diisopropylcarbodiimide (DIC) were purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). Alexa Fluor™ 488 NHS Ester (Succinimidyl Ester), SYTOX Green, MitoSOX™ Red, and 2′,7′-dichlorofluorescein diacetate (DCF-DA) were bought from Thermo Fisher Scientific (Waltham, MA, USA). All other chemicals were of American Chemical Society grade.

2.2. Fungal Cells and Culture Conditions

The drug-susceptible strains of Candida albicans (KCTC 7270), C. parapsilosis (KCTC 7214), and C. krusei (KCTC 7213) were obtained from the Korea Collection for Type Cultures (KCTC). Drug-resistant strains, C. albicans CCARM 14001, 14004, and 14020 (which were resistant to fluconazole) and CCARM 14007 (which was resistant to fluconazole, amphotericin B, and flucytisine), were obtained from the Culture Collection of Antimicrobial-resistant Microbes (CCARM) in South Korea and cultivated in yeast extract peptone dextrose (YPD) broth (Difco, Sparks, MD, USA) at 28 °C. Fusarium graminearum (KCTC 16656), F. oxysporum (KCTC 16909), F. solani (KCTC 6326), Colletotrichum gloeosporioides (KCTC 6169), and Trichoderma harzianum (KCTC 6043), used as mold strains, were cultured on potato dextrose (PD) (Difco) agar at 25 °C.

2.3. Peptide Synthesis

HnMc (FKRLKKLISWIKRKRQQ-CONH₂), HnMc-W (WKRLKKLISWIKRKRQQ-CONH₂), HnMc-WP1 (WKRLKKLIPWIKRKRQQ-CONH₂), HnMc-WP2 (WKRLKKLIPQQPWIKRKR-CONH₂), and melittin (GIGAVLKVLTTGLPALISWIKRKRQQ-CONH₂) peptides were synthesized using the Fmoc amino acids, coupled on a Rink Amide ProTide resin (0.58 meq/g substitution) with a DIC/Oxyma solution, and deprotected with 20% (v/v) piperidine solution in DMF under an automated peptide synthesizer (Liberty PRIME, CEM Co.). The synthesized peptides were cleaved using a TFA/diH2O/TIS (95:2.5:2.5, v/v/v) cocktail, precipitated using diethyl ether, and air-dried. Pure peptides were purified under a Shimadzu semi-preparative HPLC (Kyoto, Japan) with C₁₈ column (30 × 250 mm, 10 μm; Agilent Co.; Santa Clara, CA, USA). Matrix-assisted laser desorption ionization (MALDI) mass spectrometer (Kratos Analytical, Manchester, UK) was used to identify the molecular masses of the peptides, and the purity of peptides was over 98% [21].

2.4. Antifungal Assay

Yeast cells pre-cultured in YPD broth and mold conidia collected from 5 d old cultures grown on PD agar were suspended in appropriate buffers containing 20% YPD (yeast) or 20% PD (mold) (2 × 10⁴ cells or conidia/mL) and were added to serially diluted peptides in 96-well plates. After 24 h incubation at 25 °C or 28 °C, cell/mycelial growth was examined microscopically with an inverted light microscope and measured at the absorbance 595 nm using ELISA reader (Molecular Devices M5, Sunnyvale, CA, USA). The minimum inhibitory concentration (MIC) against each fungus was defined as the lowest concentration of a protein sample that completely inhibited visible growth. The low- and high-salt buffers used were 10 mM sodium phosphate (pH 7.2) and 10 mM sodium phosphate containing 140 mM NaCl (pH 7.2). Magnesium and calcium concentrations were adjusted to 6 mM in the high-salt buffer. All assays were performed in triplicate.

2.5. Confocal Laser Scanning Microscopy (CLSM) Analysis

AlexaFluor 488-labeled peptides were prepared by peptide synthesis. After incubating the Alexa Fluor 488-labeled peptides with C. albicans CCARM 14007 cells at 28 °C for 6 h, the fungal cells were washed three times with phosphate-buffered saline (PBS, pH 7.2), followed by observation under CLSM (A1R HD 25, Nikon, Tokyo, Japan).

2.6. SYTOX Green Uptake

C. albicans CCARM 14007 (2 × 10⁵ cells/mL) was mixed with the MIC peptides. After 2 h, the cells were incubated with SYTOX Green dye (final concentration, 0.2 μM) for 15 min in the dark and observed under a fluorescence microscope (OPTINITY KCS3-160S; Korea Lab Tech, Seongnam, Republic of Korea). Fluorescence levels were analyzed using flow cytometry (Attune NxT, ThermoFisher, Seoul, Republic of Korea).

2.7. Intercellular ROS and Superoxide Levels

Peptides of MIC value or 1 mM H₂O₂ were added to C. albicans CCARM 14007 (2 × 10⁴ cells/mL) in a 10 mM sodium phosphate buffer (pH 7.2) containing 10% YPD for 6 h at 28 °C, followed by staining of cells with DCF-DA (5 μM) for 30 min. Green fluorescence of the fungal cells was measured using flow cytometry (Attune NxT). To measure mitochondrial superoxide, MitoSOX™ Red probe (0.5 μM) was used with C. albicans CCARM 14007 with H₂O₂ or peptides for 1 h and the cells were analyzed using a fluorescence microscope and flow cytometry.

2.8. Induction Assay of Drug Resistance

C. albicans KCTC 7270 cells were incubated with fluconazole or peptides for the antifungal assays described above. After 24 h of incubation, MICs were determined using the method described above. Fungal cells that survived at half the MICs of each sample were treated with two-fold serially diluted samples and incubated for 24 h; the process was repeated 15 times.

3. Results and Discussion

3.1. Designation and Antifungal Effects of HnMc Variants

In a previous study, we reported the in vitro and in vivo antibacterial, anti-inflammatory, and antifungal activities and the mechanism of action of HnMc, a hybrid peptide connecting the N-terminal and C-terminal parts of HPA3NT3 and melittin, respectively [22,23]. We designed three model peptides, HnMc-W (F1W substitution), HnMc-WP1 (S9P substitution), and HnMc-WP2 (addition of PXXP motif), which changed the peptide structure in HnMc and demonstrated the non-toxicity and potent in vivo antibacterial effects of HnMc-WP2 in a drug-resistant Pseudomonas aeruginosa-infected mouse model [20]. In bacteria, the antibacterial mechanism appears to involve destruction of the cell membrane. However, because fungi are eukaryotic cells, this study was conducted with the expectation that the antifungal mechanisms may differ depending on their structure.

First, their antifungal effects were compared using the MIC values from microdilution assays on pathogenic fungi consisting of three drug-susceptible yeasts, four drug-resistant yeasts, and five molds. As shown in

Table 1. Antifungal activity of HnMc variants against yeast and mold species in low-ionic-strength buffer.

Fungal Cells	MIC (μM)
	HnMc	HnMc-W	HnMc-WP1	HnMc-WP2	Melittin	Fluconazole
Yeast
C. albicans	16	16	16	8	8	0.8
C. albicans CCARM 14001	16	16	16	8	4	>128.0
C. albican CCARM 14004	16	16	16	8	4	>128.0
C. albicans CCARM 14007	16	16	16	8	4	>128.0
C. albicans CCARM 14020	16	16	16	8	4	>128.0
C. krusei	8	8	8	4	4	32.0
C. parapsilosis	16	16	16	8	8	4.0
Mold
C. gloeosporioides	1	1	1	1	8	-
F. graminearum	2	2	2	1	4	-
F. oxysporum	2	2	2	2	2	-
F. solani	2	2	2	2	2	-
T. harzianum	2	2	2	1	4	-

, the MIC values of HnMc-W and Hn-Mc-WP1 did not differ from that of HnMc; however, the antifungal activity of HnMc-WP2 was enhanced against most fungi. All four peptides exhibited potent antifungal activity against drug-resistant C. albicans strains. This suggests that AMP may be a good alternative for overcoming drug resistance, although clinical studies on AMP are needed. Figure 1 shows their antifungal effects against C. gloeosporioides at the MIC and ½ MIC. Hollow hyphae, shrunken conidia, and swollen conidia were observed microscopically, indicating there was more than one mechanism acting on the fungus.

To apply AMPs clinically, it is important that their antibacterial effect is maintained under physiochemical conditions. Except for HnMc-W, the antifungal activity of the three peptides was enhanced under high-salt conditions and was maintained under divalent ions (Figure 2). Although there are issues with peptide degradation by proteolytic enzymes and side reactions in clinical applications, physicochemical conditions had little influence on antifungal activity.

3.2. Localization of AlexaFluor 488-Labeled Peptides in C. albicans Cells

To identify the site of action of the peptides in fungal cells, C. albicans CCARM 14007 cells with Alexa Fluor 488-labeled HnMc, HnMc-W, HnMc-WP1, HnMc-WP2, or melittin were observed by CLSM. Green fluorescence was regularly distributed on the surface of C. albicans cells treated with melittin and HnMc-WP2, but the fluorescence intensity of the others did not accumulate within the fungal cells (Figure 3). Alexa Fluor 488-labeled HnMc peptides emitted a strong green fluorescence from one part of the fungus. These results suggest that the antifungal mechanisms of action may differ among the model peptides.

3.3. Membrane-Permeable Effects of Peptides

We measured the membrane permeability of the peptides in fungal cells using SYTOX Green, a membrane-impermeable dye that binds to nucleic acids, emits green fluorescence when excited at 488 nm, and emits at a wavelength of 520 nm. This dye can enter the cells when the membrane is disrupted or pores are created. The membrane-permeable activity of the peptide was investigated using flow cytometry and fluorescence microscopy (Figure 4). C. albicans CCARM 14007 cells treated with melittin, a known membranolytic peptide, exhibited strong green fluorescence, indicating membrane permeability. Nevertheless, the number of green fluorescent cells in the HnMc, HnMc-W, and HnMc-WP1 cells was relatively low compared to that in the melittin cells. The number of fungi treated with HnMc-WP2 and the fluorescence intensity were measured in a manner similar to that for melittin. This suggests that the antifungal mechanism of HnMc-W may differ from that of the other three peptides.

3.4. Intracellular Reactive Oxygen Species (ROS) and Mitochondrial Superoxide Generations

Our previous study reported that HnMc induces apoptosis in fungal cells via the generation of ROS and mitochondrial superoxide [23]. HnMc has a random coil structure in an aqueous solution, but when it binds to the fungal cell membrane, it forms an alpha-helical structure, damages the cell membrane, and enters the cell itself. Once inside the cell, it affects the outer membrane of the mitochondria and causes superoxide production because the outer membrane of the mitochondria is similar in composition to the bacterial cell membrane. In conclusion, it inhibited fungal growth by inducing apoptosis. Apoptosis is characterized by a progressive activation pathway that leads to specific biochemical and morphological events, such as ROS generation, imbalance of intracellular ions, cell shrinkage, destabilization of lipid asymmetry, chromatin condensation in the early stages, activation of caspases and endonucleases, formation of apoptotic bodies, and cell fragmentation in the late stages [24,25].

Total ROS generation was detected using DCF-DA, a fluorescent probe oxidized by ROS. The green fluorescence emission of the H₂O₂-treated fungus was 85.024% of the total, confirming that ROS detection was normal. Fungi treated with melittin and HnMc-WP2 were measured at 3.956% and 0.13%, respectively. However, in HnMC, HnMc-W, and HnMc-WP1, ROS-producing fungi were detected between 22 and 40% (Figure 5). To confirm the action of the peptides that entered the cell, superoxide production in the mitochondria was confirmed using MitoSOX Red fluorescence staining. As shown in Figure 6, fungi treated with H₂O₂ were not stained with MitoSOX Red, unlike DCF-DA, which showed that this dye is selectively oxidized by superoxide rather than ROS. Melittin and HnMc-WP2 produced little mitochondrial superoxide, whereas the other peptides produced high levels of superoxide. These results suggest that HnMc, HnMc-W, and HnMc-WP1 peptides inhibit fungal growth by entering cells and producing intracellular ROS through mitochondrial damage and that HnMc-WP2 acts on the fungal cell membrane as well as melittin.

3.5. Induction of Drug Resistance in C. albicans Cells

AMPs have a different mechanism of action from conventional antibiotics, which provide the advantage of effectively inhibiting the growth of drug-resistant bacteria and take a long time to induce drug resistance. Therefore, drug-susceptible C. albicans were incubated with fluconazole or peptides, and changes in MIC over 15 passages were monitored (Figure 7). Surprisingly, the MIC of fluconazole-treated fungi increased exponentially from three passages and increased 1024-fold after 14 passages, confirming that drug resistance was easily induced. However, the growth of fungi treated with the peptides was inhibited at the same concentration range after 15 passages (Figure 7). This suggests that HnMc variant peptides do not induce drug resistance in the short term.

4. Conclusions

In summary, HnMc and HnMc-W with straight alpha-helical structures and HnMc-WP1 with bent alpha-helical structures have similar fungal growth-inhibitory capacities and mechanisms of action. They penetrate the fungal cells by forming an alpha-helical structure in the fungal membrane, causing destruction of the mitochondrial membrane, generation of excessive ROS, and apoptosis (Figure 8A). However, HnMc-WP2, which has a ‘helix-to-helix’ structure, exhibited strong antifungal activity via its membranolytic action (Figure 8B). It is pointed out that its structure causes pore formation or permeation into the cell membrane rather than cell membrane penetration. In addition, because it can kill fungi within minutes by acting directly on the cell membrane, it can inhibit the development of peptide resistance as well as overcome drug resistance. Although further research is needed, including laboratory studies, on the relationship between the various structures of AMPs and their antimicrobial activities, it is clear that AMPs are promising next-generation antibiotics.