Effect of charge on the antimicrobial activity of alpha-helical amphibian antimicrobial peptide

Highlights

• •The cationic alpha-helical amphibian AMPs (CαAMPs) from different amphibian species have been classified into five categories from net charge +1 to +5, and the antibacterial activity of each category has been summarized using MIC values.
• •Changing the charge of the peptides in an optimum way will provide CαAMPs with the highest antibacterial and lowest hemolytic activity.
• •Establishing a validated standard approach to assess AMPs' antibacterial and hemolytic activities is essential.

Abstract

Antibiotic resistance is a severe threat to the world's public health, which has increased the need to discover novel antibacterial molecules. In this context, an emerging class of naturally occurring short peptide molecules called antimicrobial peptides (AMPs) has been considered potent antibacterial agents. Amphibians are one of the significant sources of AMPs, which have been extensively studied for the last few decades. Most amphibian AMPs are cationic, and several of these cationic AMPs adopt a well-defined alpha-helical structure in the presence of bacterial membranes. These cationic alpha-helical amphibian AMPs (CαAMPs) can selectively and preferentially bind with the negatively charged surfaces of Gram-positive and Gram-negative bacteria through electrostatic interaction, considered the main reason for their antibacterial activities. Here, we categorized these CαAMPs according to their charge, and to calculate the charge density; we divided the charge of each peptide by its corresponding length. To investigate the effect of charge among these categories, charge or charge density under each charge category was plotted against their corresponding minimum inhibitory concentration (MIC). Moreover, the effect of charge modification of some CαAMPs under specific charge categories in the context of MIC and hemolysis was also discussed. The information in this review will help us understand the antibacterial activity of accessible CαAMPs depending on each charge category across species. Additionally, this study suggests that designing novel functional antibacterial agents requires charge modification optimally.

Graphical abstract

1. Introduction

With the increasing usage and administration of antibiotics, the efficacy of existing antimicrobial agents is becoming resisted by most pathogens at an alarming rate (Michael et al., 2014). Besides, mutation (Read and Woods, 2014) and the impetuous evolving nature of the bacterial genome (San Millan, 2018; Rozwandowicz et al., 2018) and gaining the antibacterial resistant genome via lateral gene transfer (Sun et al., 2019) are also prominent factors of antibacterial resistance. Scientists and researchers are trying to find alternative options, especially broad-spectrum antibiotics, with complete devotion and sincerity to overcome this threatening situation (Levy and Marshall, 2004). Antimicrobial peptides (AMPs) can be an excellent option as effective antimicrobial agents against resistant variants.

AMPs are a varied class of molecules consisting of short polypeptide sequences that are synthesized naturally in almost every multicellular organism as their first line of defense (Zasloff, 2002). They are the immunoregulatory molecules of the organisms, regulated and produced for their inherited immunity (Zhang, 2016). Despite their differences in structural properties, AMPs share several qualities. Because most AMPs are cationic, with a net positive charge range from +3 to +11, they can preferentially connect to the negatively charged surface of bacteria via electrostatic binding (Hancock and Diamond, 2000). Gram-negative bacteria (e.g., Escherichia coli, Pseudomonas aeruginosa) have an extra outer membrane enriched with anionic lipopolysaccharide, whereas Gram-positive bacteria (e.g., Staphylococcus aureus, Streptococcus pneumoniae, and Bacillus subtilis) have a thick peptidoglycan layer enriched with anionic teichoic acids surrounding their cytoplasmic membrane. Furthermore, both Gram-positive and Gram-negative bacteria have the anionic cytoplasmic membranes due to the presence of phosphatidylglycerol (PG) and cardiolipin. Unlike bacteria, human RBCs have a cell membrane of neutral phospholipids such as phosphatidylcholine, sphingomyelin, and phosphatidylethanolamine (Otvos, 2005). As a result, these peptides have weak interaction with this neutral membrane. Due to these structural differences, they have lesser toxicity to mammalian cells (lower hemolytic activity) but higher toxicity to both Gram-positive and Gram-negative bacteria. The peptides' hydrophobic residues cause the molecule to bind tightly to the hydrophobic portion of lipid membranes, while their cationic residues are responsible for electrostatic interactions with negatively charged bacterial surfaces (Ebenhan et al., 2014). These electrostatic and hydrophobic interactions disrupt the bacterial membrane. Several models have been proposed to explain the mechanism of disruption, such as the barrel-stave, toroidal pore, and carpet model. The structural features of the peptide and the lipid membrane affect the mode of membrane disruption (Sani and Separovic, 2016).

Cationic AMPs are obtained from various sources, and amphibians are one of the major sources of AMPs. Many amphibian, cationic AMPs remain unstructured in an aqueous solvent. However, in the presence of a lipid membrane, they are converted into a well-defined secondary structure (e.g., alpha-helix or beta-sheet), which plays an important role in antibacterial activity (Takahashi et al., 2010; Hasan et al., 2022). These amphibian, cationic AMPs having alpha-helical structures can be called as CαAMPs. In the alpha-helical structure, a rod-like structure is formed by a tight right-handed twist in the amino acid chain. This shape is caused by hydrogen bonding between the hydrogen in an amino group and the oxygen in a carboxyl group. Most alpha helices are amphiphilic, which means they contain both hydrophilic and hydrophobic regions. Usually, the hydrophilic residues are located opposite to the hydrophobic residues. The secondary structure and helical wheel projection of two well-studied CαAMPs are shown in Fig. 1.

Fig. 1.

Helical wheel representation and alpha-helix structure of Magainin 2 (A) and Hylaseptin 4 (B). The helical wheel representations were obtained from https://heliquest.ipmc.cnrs.fr/cgi-bin/ComputParams.py. The alpha-helix structure was drawn by PyMOL (version 2.4.1.). In both A & B, cationic and hydrophobic residues are represented by blue and yellow color, respectively. The letters 'N' and 'C' stand for the N- and C-terminal, respectively.

These CαAMPs have received much attention in recent years due to their powerful broad-spectrum antibacterial properties, and some notable CαAMPs are summarized in Table 1. Here, we classified these CαAMPs based on their charge, and to calculate the charge density, we divided the charge of each peptide by its length. To investigate the effect of charge among these categories, charge or charge density under each charge category was plotted against the corresponding minimum inhibitory concentration (MIC). Furthermore, the effect of charge modification of some CαAMPs in specific charge categories on MIC and hemolysis has been discussed. It is challenging to compare the antibacterial activities of these CαAMPs because numerous researchers employed various microbial strains to investigate the MIC values to assess the antimicrobial activity. For ease of comparison, we have selected a particular strain for a single plot or graph from different literature. The information in this review will be helpful for understanding the antibacterial activity of accessible CαAMPs depending on each charge category across species. In addition, this study suggests that designing novel functional antibacterial agents requires charge modification in an optimal way.

Table 1.

Some notable amphibian alpha-helical peptides with their sequences, sources, and major bioactivities.

Name	Length	Sequence	Source	Major bioactivities	Reference
Aurein 2.5	16	GLFDIVKKVVGAFGSL	Ranoidea aurea	Antibacterial, antifungal, anticancer	(Rozek, 2000)
Brevinin-1E	24	FLPLLAGLAANFLPKIFCKITRKC	Rana esculenta	Antibacterial	(Simmaco et al., 1994)
Caerin-1.1	25	GLLSVLGSVAKHVLPHVVPVIAEHL	Litoria splendida	Antibacterial, antiviral	(Wong et al., 1997)
Citropin-1.1	16	GLFDVIKKVASVIGGL	Litoria citropa	Antibacterial	(Wegener et al., 1999)
Distinctin	25	NLVSGLIEARKYLEQLHRKLKNCKV	Phyllomedusa distincta	Antibacterial	(Batista et al., 2001)
Esculentin-1	46	GIFSKLGRKKIKNLLISGLKNVGKEVGMDVVRTGIDIAGCKIKGEC	Rana esculenta	Antibacterial	(Rozek, 2000)
Gaegurin-1	33	SLFSLIKAGAKFLGKNLLKQGACYAACKASKQC	Rana rugosa	Antibacterial, antifungal, antiprotozoal	(Park et al., 1994)
Hylaseptin P1	14	GILDAIKAIAKAAG	Hyla punctata	Antibacterial	(Prates et al., 2004a)
Kassinatuerin-1	21	GFMKYIGPLIPHAVKAISDLI	Kassina senegalensis	Antibacterial, antifungal	(Mattute et al., 2000)
Magainin 1	23	GIGKFLHSAGKFGKAFVGEIMKS	Xenopus laevis	Antibacterial, antifungal, antiviral, antiprotozoal	(Zasloff, 1987)
Magainin 2	23	GIGKFLHSAKKFGKAFVGEIMNS	Xenopus laevis	Antibacterial, antifungal, antiviral, antiprotozoal	(Zasloff, 1987)
Nigrocin-1	33	GLLDSIKGMAISAGKGALQNLLKVASCKLDKTC	Rana nigromaculata	Antibacterial, antifungal	(Park et al., 2001)
Ocellatin-1	25	GVVDILKGAGKDLLAHLVGKISEKV	Leptodactylus ocellatus	Antibacterial	(Nascimento et al., 2004)
PGLa	21	GMASKAGAIAGKIAKVALKAL	Xenopus laevis	Antibacterial, antifungal	(Andreu et al., 1985)
Phylloseptin-1	19	FLSLIPHIVSGVASIAKHF	Phyllomedusa sauvagei	Antibacterial, antifungal	(Zhang et al., 2010)
Pleurain-A1	26	SIITMTKEAKLPQLWKQIACRLYNTC	Babina pleuraden	Antibacterial, antifungal	(Wang et al., 2007)
Ranatuerin-1	25	SMLSVLKNLGKVGLGFVACKINKQC	Rana catesbeiana	Antibacterial, antifungal	(Goraya et al., 1998)
Rugosin-A	33	GLLNTFKDWAISIAKGAGKGVLTTLSCKLDKSC	Rana rugosa	Antibacterial	(Suzuki et al., 1995)
Temporin-A	13	FLPLIGRVLSGIL	Rana temporaria	Antibacterial, antifungal, antiviral	(Simmaco et al., 1996)

2. Effect of charge on CαAMPs from different sources

The net charge of CαAMPs is one of the most critical parameters for antibacterial activities. However, other structural parameters such as peptide length, charge, amino acid composition, helicity, amphiphilicity, etc., also contribute in their activities. Investigating the effects of charge or charge density on the antibacterial activity of these CαAMPs derived from various amphibian sources is crucial.

The net charge of a peptide can be defined as the sum of all the charges of its polar groups of the peptide, which is calculated by the following expression:

$${\mathrm{Q}}_{\text{peptide}}=\Sigma {\mathrm{Q}}^{-}+\Sigma {\mathrm{Q}}^{+}$$

Here, ƩQ⁻, the sum of the negative charge for functional groups (e.g., -SH, -PhOH, -COOH), and ƩQ⁺, the sum of the positive charge for functional groups (e.g., -NH3+, =NH2+, NH+) of the certain peptide (Lee et al., 2016). For example, magainin 2 has the sequence of GIGKFLHSAKKFGKAFVGEIMNS and contains 4 Lys-residues, each having +1 charge and 1 Glu-residue having −1 charge. As a result, Magainin 2 has a net charge of +3. Then, the charge density of each CαAMP is calculated by net charge divided by the AMP length (i.e., the total number of amino acids in the sequence). As magainin 2 contains 23 amino acids, the charge density of this CαAMP is 0.1304.

At first, the CαAMPs from different amphibian species have been classified into five categories from net charge +1 to +5. Table 2 summarizes the length, charge density, MIC, and source of several CαAMPs of these five categories. For ease of comparison, we used the MIC values obtained from E. coli ATCC 25922.

Table 2.

Charge density and minimum inhibitory concentration of several AMPs against Escherichia coli ATCC 25922.

Peptide	Sequence	Length	Charge density	MIC (μM)	Species	Reference
			Charge +1
Aurein-1.2	GLFDIIKKIAESF	13	0.076	16	Litoria aurea, Litoria raniformis	(Rozek et al., 2000)
Caerin-1.1	GLLSVLGSVAKHVLPHVVPVIAEHL	25	0.04	8.45	Litoria splendida	(Wong et al., 1997)
Nigrocin-2GRa	GLLSGILGAGKHIVCGLSGLC	21	0.047	25	Rana grahami	(Conlon et al., 2006)
Ocellatin-4	GLLDFVTGVGKDIFAQLIKQI	21	0.047	64	Leptodactylus ocellatus	(Nascimento et al., 2007)
Phylloseptin-2	FLSLIPHAINAVSAIAKHS	19	0.052	15	Phyllomedusa genus	(Leite et al., 2005)
			Charge +2
Citropin-1.1	GLFDVIKKVASVIGGL	16	0.125	10	Litoria citropa	(Wegener et al., 1999)
Hylaseptin-P1, HSP1	GILDAIKAIAKAAG	14	0.142	24.4	Hypsiboas punctatus	(Prates et al., 2004b)
Japonicin-1	FFPIGVFCKIFKTC	14	0.142	30	Rana japonica	(Isaacson et al., 2002)
Kassinatuerin-1	GFMKYIGPLIPHAVKAISDLI	21	0.095	6.25	Kassina senegalensis	(Conlon et al., 2005b)
Temporin-A	FLPLIGRVLSGIL	13	0.153	100	Rana temporaria	(Simmaco et al., 1996)
			Charge +3
Alyteserin-1c	GLKEIFKAGLGSLVKGIAAHVAS	23	0.13	25	Alytes obstetricans	(Michael Conlon et al., 2009)
Brevinin-1SPb	FLPIIAGMAAKVICAITKKC	20	0.15	50	Rana septentrionalis	(Bevier et al., 2004)
Brevinin-2LTa	GAFGDLLKGVAKEAGMKLLNMAQCKLSGKC	30	0.1	32.5	Hylarana latouchii	(Wang et al., 2012)
Hylin a1	IFGAILPLALGALKNLIK	18	0.166	32	Hypsiboas albopunctatus	(Castro et al., 2009)
Magainin 2	GIGKFLHSAKKFGKAFVGEIMNS	23	0.13	12.5	Xenopus laevis	(Zasloff, 1987)
Ocellatin-F1	GVVDILKGAAKDIAGHLASKVMNKL	25	0.12	100	Leptodactylus fallax	(Rollins-Smith et al., 2005)
Palustrin-2AJ1	GFMDTAKNVAKNVAVTLIDKLRCKVTGGC	29	0.103	13	Amolops jingdongensis	(Chen et al., 2012b)
Palustrin-2c	GFLSTVKNLATNVAGTVIDTLKCKVTGGCRS	31	0.096	120	Rana palustris	(Basir et al., 2000)
Pseudin-2	GLNALKKVFQGIHEAIKLINNHVQ	24	0.124	8	Pseudis paradoxa	(Park et al., 2011)
Ranalexin-1Ca	FLGGLMKAFPALICAVTKKC	20	0.15	4	Rana clamitans	(Halverson et al., 2000)
Raniseptin-1	AWLDKLKSLGKVVGKVALGVAQNYLNPQQ	29	0.103	5	Hypsiboas raniceps	(Magalhães et al., 2008)
XPF-St1	GVWSTVLGGLKKFAKGGLEAIVNPK	25	0.12	8	Xenopus tropicalis	(Ali et al., 2001)
			Charge +4
Ascaphin-1	GFRDVLKGAAKAFVKTVAGHIAN	23	0.173	3	Ascaphus truei	(Conlon et al., 2004)
Brevinin-1AUa	FLPILAGLAAKLVPKVFCSITKKC	24	0.166	13	Rana aurora	(Conlon et al., 2005a)
Brevinin-1VLa	FLGAIAGVAAKFLPKVFCFITKKC	24	0.166	25	Lithobates vaillanti	(Conlon et al., 2009)
Brevinin-2LTb	SILDKIKNVALGVARGAGTGILKALLCKLDKSC	33	0.121	30	Hylarana latouchii	(Wang et al., 2012)
Brevinin-2PRb	GLMSLFRGVLKTAGKHIFKNVGGSLLDQAKCKITGEC	37	0.108	3	Rana pirica	(Conlon et al., 2004)
Japonicin-2	FGLPMLSILPKALCILLKRKC	21	0.19	12	Rana japonica	(Isaacson et al., 2002)
Palustrin-2CG1	GLWNTIKEAGKKFAINVLDKIRCGIAGGCKT	31	0.129	150	Amolops chunganensis	(Yang et al., 2012)
Palustrin-3a	GIFPKIIGKGIKTGIVNGIKSLVKGVGMKVFKAGLNNIGNTGCNEDEC	48	0.083	1	Rana palustris	(Basir et al., 2000)
Ranatuerin-1	SMLSVLKNLGKVGLGFVACKINKQC	25	0.16	20	R. catesbeiana	(Goraya et al., 1998)
XT-1	GFLGPLLKLAAKGVAKVIPHLIPSRQQ	27	0.148	6	Xenopus tropicalis	(Ali et al., 2001)
			Charge +5
Brevinin-2PRd	GLMSVLKGVLKTAGKHIFKNVGGSLLDQAKCKITGQC	37	0.135	3	Rana pirica	(Conlon et al., 2004)
Brevinin-2PRe	GLLSVLKGVLKTTGKHIFKNVGGSLLDQAKCKISGQC	37	0.135	3	Rana pirica	(Conlon et al., 2004)
Esculentin-2B	GLFSILRGAAKFASKGLGKDLTKLGVDLVACKISKQC	37	0.135	1	Rana berlandieri	(Goraya et al., 2000)
Palustrin-1c	ALSILRGLEKLAKMGIALTNCKATKKC	27	0.185	28	Rana palustris	(Basir et al., 2000)
Palustrin-1d	ALSILKGLEKLAKMGIALTNCKATKKC	27	0.185	28	Rana palustris	(Basir et al., 2000)
Palustrin-3AR	GIFPKIIGKGIVNGIKSLAKGVGMKVFKAGLNNIGNTGCNNRDEC	45	0.111	6	Rana areolata	(Ali et al., 2002)

The MIC values of each peptide category were plotted against their corresponding net charges to compare their antibacterial activity (Fig. 2).

Fig. 2.

MIC values of different CαAMPs based on charges.

Fig. 2 shows that for most of the CαAMPs (discussed in Table 2), the MIC values are less than 50 µM (E. coli ATCC 25922). Moreover, the antibacterial activity is not proportional to the net charge of CαAMPs obtained from different sources. In this study, we used only one bacterial strain's MIC values. Further investigations may be performed using other strains, although it's difficult to find similar data from published works.

In the same charge group, MIC values may also vary significantly. For example, brevinin-1SPb and ranalexin-1Ca have the same net charge (+3) and peptide length but different MIC values (Bevier et al., 2004; Halverson et al., 2000). This variation may be due to differences in other structural parameters such as hydrophobicity. XT-1 shows 25 times more antibacterial activity than palustrin-2CG1, although they have the same net charge (+4) (Ali et al., 2001; Yang et al., 2012). This may be due to the difference in peptide length, interfacial hydrophobicity, amphiphilicity, and angle subtended by the hydrophobic residues (Hasan et al., 2022). In the case of temporin-PF, the highest helicity-forming peptide showed the strongest antibacterial and hemolytic activities than the other analogs though their charge was the same (+2). Therefore, peptides' helix content and aggregation pattern are directly related to the antimicrobial and hemolytic activities (Zai et al., 2021).

In the next step, we plot charge density and the MIC values to observe the relationship between charge density and the antibacterial activity of CαAMPs for charges from +1 to +5.

According to Fig. 3, most of the MIC values of CαAMPs lie between the charge density of 0.10 to 0.17. Using this bacterial strain, no specific relationship has been found between the charge density and the antimicrobial activity of different categories of CαAMPs. A similar plot should be made using other bacterial species to get a comprehensive relationship between the charge density and the antimicrobial activity.

Fig. 3.

Relationship among charge densities of CαAMPs of different categories and minimal inhibitory concentration (MIC) values against E. coli ATCC 25922.

3. Effect of charge modification on single CαAMPs

To improve the antibacterial activities of CαAMPs, several analogs have been synthesized by altering their net charges. Positively charged amino acids such as Arg, Lys, and His-have been introduced or deducted to modify the net charge of the alpha-helices. This modification can be done in both the helix or the N- and C-terminal. For example, several analogs of amphibian antimicrobial peptides medusin-PT have been synthesized by adding lysine to its nonpolar face to increase the peptide's net charge (Gao et al., 2017). In the case of phylloseptin-PT, net charge was modified by replacing Asn-and His, with Lys-and d-lysin (Gao et al., 2016). On the other hand, positively charged residue Lys-has been added to the N- and C-terminal of magainin 2 to increase the net charge (Bessalle et al., 1992).

Modification of net charge affects both the antibacterial and hemolytic activities of CαAMPs. Increased net charge of CαAMPs will increase their antibacterial activity against Gram-negative and Gram-positive bacteria. An excessively high net charge will lead to increased hemolytic propensity and decreased antimicrobial activity. For example, magainin 2 analogs showed the maximum antimicrobial activity at +5 charge with low hemolytic activity. When the charge is greater than +5, the hemolytic activity rapidly increases while the antibacterial activity decreases (Dathe et al., 2001). In the case of brevinin-1OS, the antibacterial activity decreases with the decrease of net charge. There was a complete loss in antibacterial activity after reducing the net charge from +3 to +1 (Zhou et al., 2020).

The net charge also influences the antibacterial and hemolytic activities of Brevinin-2-related peptide (B2RP: GIWDTIKSMGKVFAGKILQNL) analogs (Conlon et al., 2009). The relationship is shown in Fig. 4.

Fig. 4.

Relationship among the charge of different analogs of Brevinin-2 related peptide (B2RP) with the Minimum Inhibitory Concentration (MIC) against E. coli (ATCC 25726) and hemolysis (LC₅₀) of the human erythrocyte. The LC₅₀ is the mean concentration of peptides producing 50% hemolysis. ★ Marked points indicate greater than the value of that point (> 200).

In Fig. 4, the maximum antibacterial activity (less MIC) is observed at charges +4 and +5, further increase in the charge will decrease the activity. Peptides with a net charge of +5 and +6 exhibited almost no hemolytic activity (LC₅₀ = >200 μM). Therefore, the net charge +5 can be considered the optimum charge of B2RP, which provides higher antibacterial activity (low MIC) and lower hemolytic activity (high LC₅₀). However, the MIC values can be changed due to the variation in the bacterial strain. Changing the net charge of the peptides significantly influences the antibacterial and hemolytic activity of the peptide. During the modification of the net charge, one should find out the optimum charge for that particular AMPs by plotting MIC and LC₅₀ values against the corresponding charge.

4. Conclusion

The CαAMPs can be used as a functional antibacterial agent for the current antibiotic resistance situation. The net charge of the peptides greatly regulates the antibacterial activity of these peptides. The CαAMPs from different amphibian species have been classified into five categories from net charge +1 to +5, and the antibacterial activity of each category has been observed using MIC values of E. coli ATCC 25922. Most of the CαAMPs of different charge groups have MIC values less than 50 µM, which lie between the charge density of 0.10 to 0.17. A similar plot should be made using other bacterial species to get a comprehensive relationship between the charge density and the antimicrobial activity. Changing the charge of the peptides influences the antibacterial and hemolytic activity. During the modification of net charge, one should find the optimum charge for that particular AMPs by plotting antibacterial and hemolysis values against the corresponding charge. The antibacterial activity of CαAMPs can also be increased by modulating other structural parameters such as peptide length, charge, amino acid composition, helicity, amphiphilicity, etc. For each modification, however, the optimum changes must be determined using a similar approach. As there is no universal standard for observing the antimicrobial activity of CαAMPs, researchers use varied ways for their study. There is a significant variation in the process of peptide preparation, the selection of microorganisms, the measurement of the MIC values, etc. That's why it's not easy to compare all the available data. There should be a validated standard procedure to determine the antibacterial and hemolytic activities of CαAMPs. The information in this review will be useful in understanding the antibacterial activity of available CαAMPs across species and charge categories. Furthermore, our research implies that creating novel effective antibacterial agents necessitates an appropriate charge adjustment.

Funding

This work didn't get any funding.

CRediT authorship contribution statement

Md. Monirul Islam: Conceptualization, Methodology, Software. Fahim Asif: Data curation, Writing – original draft. Sabbir Uz Zaman: Writing – original draft, Investigation. Md. Kamrul Hasan Arnab: Visualization, Software. Md. Mostafizur Rahman: Validation. Moynul Hasan: Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

• No data was used for the research described in the article.