Employment of mastoparan-like peptides to prevent Staphylococcus aureus associated with bovine mastitis

ABSTRACT

Bovine mastitis is a frequent infection in lactating cattle, causing great economic losses. Staphylococcus aureus represents the main etiological agent, which causes recurrent and persistent intramammary infections because conventional antibiotics are ineffective against it. Mastoparan-like peptides are multifunctional molecules with broad antimicrobial potential, constituting an attractive alternative. Nevertheless, their toxicity to host cells has hindered their therapeutic application. Previously, our group engineered three mastoparan-L analogs, namely mastoparan-MO, mastoparan-R1, and [I⁵, R⁸] MP, to improve cell selectivity and potential. Here, we were interested in comparing the antibacterial efficacy of mastoparan-L and its analogs against bovine mastitis isolates of S. aureus strains, making a correlation with the physicochemical properties and structural arrangement changes promoted by the sequence modifications. As a result, the analog’s hemolytic and/or antimicrobial activity was balanced. All the peptides displayed α-helical folding in hydrophobic and membrane-mimetic environments, as determined by circular dichroism. The peptide [I⁵, R⁸] MP stood out for its enhanced selectivity and antibacterial features related to mastoparan-L and the other derivatives. Biophysical approaches revealed that [I⁵, R⁸] MP rapidly depolarizes the bacterial membrane of S. aureus, causing cell death by subsequent membrane disruption. Our results demonstrated that the [I⁵, R⁸] MP peptide could be a starting point for the development of peptide-based drugs for the treatment of bovine mastitis, with the advantage of no residue in milk, which would help reduce the use of classical antibiotics.

IMPORTANCE

Staphylococcus aureus is a leading cause of mastitis, the world’s most important dairy cattle disease. The multidrug resistance and zoonotic potential of S. aureus, besides the likelihood of antibiotic residues in milk, are of critical concern to public and animal health. Antimicrobial peptides offer a novel antimicrobial strategy. Here, we demonstrate that [I⁵, R⁸] MP is a potent and selective peptide, which acts on S. aureus by targeting the bacterial membrane. Therefore, understanding the physicochemical determinants and the modes of action of this class of antimicrobials opens novel prospects for peptide development with enhanced activities in the bovine mastitis context.

INTRODUCTION

Bovine mastitis (BM) is one of the most prevalent infectious diseases in the dairy cattle industry (1). Described as a mammary gland inflammatory condition, BM is mainly caused by teat canal bacterial colonization, resulting in substantial economic losses in dairy production systems (2, 3).

Since bacterial pathogens (e.g., staphylococci, streptococci, and coliforms) are related to the predominant cause of BM, antibiotics remain the main tool for prevention and treatment (2, 4, 5). Consequently, this leads to a wide antibiotics use in dairy production systems (6), which could cause antimicrobial resistance in food-producing animals (7) (https://www.fda.gov/media/159544/download). In this context, several studies have reported high rates of antibiotic resistance from mastitis-isolated bacteria, showing multidrug-resistant phenotypes and genotypes mainly to β-lactams, aminoglycosides, tetracyclines, and macrolides (8–12). It is also worth highlighting that antibiotic residues are commonly found in dairy products, which in turn cause a negative effect on human health.

Although numerous pathogens can induce BM, Staphylococcus aureus has been commonly reported as a leading cause of mastitis in dairy cows worldwide (13, 14). This bacterium causes persistent and antibiotic-resistant infections, which represent a challenging task in bovine mastitis control processes (15, 16). Besides, the presence of S. aureus and antibiotic residues in milk for human consumption presents an important public health concern, owing to the zoonotic potential of this bacterium and the risk of bacterial resistance dissemination among livestock and humans (13, 17). Therefore, searching for alternative approaches to treating this common and costly cattle disease is imperative.

In this scenario, antimicrobial peptides (AMPs) have been reported as an attractive alternative to conventional antibiotics in the current alarming multidrug resistance context (18, 19). Mastoparans, cationic wasp-venom isolated peptides, constitute an AMP family with multifunctional properties (20, 21). Nevertheless, the nonspecific cell toxicity of some members has made them unsuitable for clinical practice. In particular, mastoparan-L, the first isolated peptide from this family (22), is a cationic, amphipathic α-helical peptide with a wide range of reported biological activities. This peptide has been reported as a cytolytic peptide (23, 24), hindering its biotechnological and therapeutical potential. Hence, different rational design approaches have been widely employed to fine-tune the biological functions of mastoparan-like peptides (25–27).

In previous works, our group designed three mastoparan-L analogs (Fig. 1), including [I⁵, R⁸] MP (28), mastoparan-R1 (23), and mastoparan-MO (29). The design approaches applied aimed at the optimization of their hemolytic and cytotoxic activities to promote therapeutic applications to treat human pathogenic microorganisms. Therefore, to obtain the mastoparan-MO analog (Fig. 1B), a pentapeptide motif (FLPII), conserved among AMPs with high antimicrobial and immunomodulatory activities, was added at the N-terminal portion of the mastoparan-L sequence (29). Additionally, the mastoparan-R1 peptide (Fig. 1C) was computationally designed by the Joker algorithm (30) through a pattern insertion (K - [ILV] - [AL] - X - [RKD] - [ILV] - X - X - K - I) in the mastoparan-L peptide sequence, making the amino acid substitution at positions 1, 2, 5, 9, and 10 (23). Finally, the amino acid substitution strategy was used to obtain the [I⁵, R⁸] MP peptide (28) (Fig. 1D). In that study, alanine residues allocated at positions 5 and 8 in the mastoparan-L peptide sequence were replaced by isoleucine (Ile) and arginine (Arg) residues, respectively. These modifications were aimed at increasing the peptide charge (+3 to +4) by inserting Arg, without changing the polar face angle of the peptide. Besides, Arg is classified as the second residue with more propensity to form an α-helix (as per Pace-Scholtz’s α-helical propensity scale (31). Then, the insertion of an Ile residue was intended to stabilize the peptide hydrophobicity, since according to the Kyte-Doolittle hydrophobicity scale (32), Ile (4.5) and Arg (−4.5) have opposite values (32). These three peptide analogs showed improvements in their antimicrobial activities with reduced or noncytolytic activities when compared with the parental counterparts. Thus, the present work proposes the comparative investigation of the antibacterial potential of different mastoparans, including mastoparan-L, -MO, -R1, and [I⁵, R⁸] MP against S. aureus strains isolated from cattle with bovine mastitis, and the correlation of these activities with the changes in the physicochemical properties and structural arrangement promoted by the sequence modifications.

Fig 1.

Three-dimensional structures for (A) mastoparan-L (PDB:6DUL), (B) mastoparan-MO (PDB:6DUU), (C) mastoparan-R1(Theoretical PDB coordinates, (23)), and (D) [I⁵, R⁸] MP (PDB:7RUL). The electrostatic potential for each peptide was calculated using the Adaptive Poisson-Boltzmann Solver (APBS), with a potential scale ranging from −5kT/e (red) to +5 kT/e (blue). Cationic residues are shown in blue, uncharged polar in pink, and nonpolar in yellow, gray, and green on the helical wheel projections. The hydrophobic moment vector is represented by arrows starting from the center of the diagrams, whereas the estimated value of the resulting hydrophobic moment is proportional to the size of the arrows.

RESULTS

Sequence modifications in mastoparan-like peptides promote changes in physicochemical properties

The mastoparan-L analogs evaluated in the present work were obtained through different rational design strategies by our group. The modifications made to the amino acid sequence of the parental peptide led to significant changes in the physicochemical properties of the analog peptides (mastoparan-R1, mastoparan-MO, and [I⁵, R⁸] MP). For instance, the addition of five apolar residues (FLPII) to the peptide sequence of mastoparan-L resulted in an increase in the percentage of hydrophobicity (from 57.6% to 83.6%) and a decrease in the hydrophobic moment (from 0.398 to 0.325) in the mastoparan-MO derivative, which maintained the net +3 charge of the parent peptide (mastoparan-L). By contrast, the peptide mastoparan-R1 showed an increase in a net charge to +6 and a hydrophobic moment equal to 0.775 (most amphipathic peptide), as well as a decrease in the overall hydrophobicity (36.9%). Similarly, the modifications performed to obtain the [I⁵, R⁸] MP analog generated a slight increase in the values of all physicochemical parameters described here (Table 1). It is worth noting that these differences in overall hydrophobicity are intrinsically related to cell selectivity in AMPs (the higher the hydrophobicity, the lower the cell selectivity) and will be discussed in detail in the next topics.

TABLE 1.

Physicochemical properties of mastoparan-L and its derivates

	Sequencea	(z)b	<H>c	<µH>d
mastoparan-L	-----INLKALAALAKKIL-NH2	+3	57.6	0.398
mastoparan-MO	FLPIIINLKALAALAKKIL-NH2	+3	83.6	0.325
mastoparan-R1	-----KILKRLAAKIKKIL-NH2	+6	36.9	0.775
[I5, R8] MP	-----INLKILARLAKKIL-NH2	+4	58.9	0.471

^a Modified amino acid residues are in bold type.

^b (z): net charge.

^c <H>: hydrophobicity percentage.

^d <μH>: hydrophobic moment.

Mastoparan-L and its analogs undergo a coil-to-α-helix transition from aqueous solution to hydrophobic/anionic environments

To investigate secondary structural changes of the mastoparan-like peptides studied here, circular dichroism (CD) spectra were measured in hydrophilic (ultrapure water and KH₂PO₄ buffer), hydrophobic (TFE/water mixture), and membrane mimetic environments [sodium dodecyl sulfate (SDS) micelles]. As shown in Fig. 2A and B, all the peptides displayed CD spectrum characteristics of random coiled structures when in water and KH₂PO₄ buffer, with small helical fractions for mastoparan-L (18.11% in water and 20.86% in buffer), mastoparan-MO (17.58% in water and 17.54% in buffer), and [I⁵, R⁸] MP (12.65% in buffer) (Fig. 2A and B; Table 2). By contrast, in TFE 50% and SDS micelles, all the peptides exhibited strong CD signatures but, with different helicity percentages, ranging from 33.32% to 69.89% α-helix content in TFE 50% and 45.37% to 73.63% in SDS, among which mastoparan-L and mastoparan-MO peptides showed the highest helical content (Fig. 2C and D; Table 2).

Fig 2.

CD spectra for the mastoparan-like peptides studied here. CD spectra of mastoparan-L (black), mastoparan-MO (red), mastoparan-R1 (blue), and [I⁵, R⁸] MP (green), showing peptide structure transitions from random coil signatures in (A) ultrapure water and (B) 10 mmol L⁻¹ KH₂PO₄ (pH 7.4) to α-helical conformations in (C) TFE 50% in water and (D) 100 mmol L⁻¹ SDS micelles.

TABLE 2.

Calculated percentage of helicity for mastoparan-L, mastoparan-MO, mastoparan-R1, and [I⁵, R⁸] MP in hydrophilic, hydrophobic, and membrane mimetic conditions^a

Solvent	mastoparan-L (%)	mastoparan-MO (%)	mastoparan-R1 (%)	[I5, R8]MP (%)
Ultrapure water	18.11	17.58	3.31	8.90
10 mmol L−1 KH2PO4	20.86	17.54	1.48	12.65
50% TFE	69.89	55.78	36.27	33.32
100 mmol L−1 SDS	73.63	44.28	45.37	51.02

^a The percentage of helix content for all mastoparan-like peptides was calculated from the mean residue ellipticity at 222 nm ([θ]₂₂₂), as previously described (33).

Determination of resistance and sensitivity to antibiotics

The resistant profiles of S. aureus 02/18, 353/17, and Aurora strains were obtained for the commonly used antibiotics of different classes, including cefoxitin, tetracycline, gentamicin, and ciprofloxacin, as shown in Table 3. Then, isolate 02/18 showed resistance only to gentamicin, whereas isolate 353/17 was found to be resistant to most of the antibiotics tested. S. aureus Aurora isolate was susceptible to all the antibiotics tested.

TABLE 3.

Resistant profile of S. aureus 02/18, 353/17 and Aurora for cefoxitin, tetracycline, gentamicin, and ciprofloxacin, accordantly with CLSI guidelines

Antibiotics	Zone diameter (mm)a
	S. aureus 02/18	Interpretive category	S. aureus 353/17	Interpretive category	S. aureus aurora	Interpretive category
Cefoxitin	32	S	30	S	30	S
Tetracycline	25	S	0	R	25	S
Gentamicin	11	R	0	R	19	S
Ciprofloxacin	23	S	12	R	25	S

^a Inhibition zone diameter values are expressed in millimeters (mm). S: susceptible; R: resistant.

Antibacterial and time-kill kinetics assays

The minimal inhibitory concentrations (MICs) and minimal bactericidal concentrations (MBCs) for the mastoparan-L peptide and its analogs were determined against the three S. aureus strains isolated from mastitic cattle. As a result, mastoparan-L peptide was inactive against the strains 02/18 and 353/17 at 32 µmol L⁻¹ (Fig. 3A). However, the Aurora strain growth was inhibited at 32 µmol L⁻¹ of this same peptide. The MBCs for mastoparan-L were not determined against the strains 02/18 and 353/17 (MBC >32 µmol L⁻¹). However, mastoparan-L showed an MBC equal to 32 µmol L⁻¹ against the Aurora strain. By contrast, the mastoparan-MO peptide showed enhanced activity by inhibiting the growth of all S. aureus strains at 16 µmol L⁻¹ (Fig. 3A). Additionally, this peptide showed bactericidal activity at MIC against the strains 02/18 and Aurora and at 32 µmol L⁻¹ against S. aureus 353/17. Interestingly, the mastoparan-R1 peptide was incapable of inhibiting the growth of any of the S. aureus strains evaluated, even at the highest concentration tested (32 µmol L⁻¹). Finally, we observed that the [I⁵, R⁸] MP peptide exhibited the highest antibacterial potential, with MIC values of 4, 8, and 16 µmol L⁻¹ against S. aureus Aurora, 353/17, and 02/18 strains, respectively. Furthermore, [I⁵, R⁸] MP eradicated S. aureus cells with MBCs ranging from 8 to 16 µmol L⁻¹.

Fig 3.

Antibacterial, hemolytic, and killing kinetic activities of mastoparan-L and its analogs. (A) Minimal inhibitory concentrations of all peptides against the three isolated S. aureus strains. (B) Hemolytic activity of mastoparan-L peptide and its analogs against bovine red blood cells. Triton X-100 (2%) was used as a positive hemolysis control. (C) Time-kill kinetics of the [I⁵, R⁸] MP peptide against S. aureus Aurora at 8 µmol L⁻¹ (white circles). The bacterial strain growth in the absence of peptide was used as a growth control (black squares). The peptide was added at time 0, being monitored every 10 min until 90 min of incubation.

Regarding the time-kill kinetics assay, the time course of the bactericidal activity (time-kill) of the [I⁵, R⁸] MP peptide against S. aureus Aurora is shown in Fig. 3C. The time-kill curve shows that the [I⁵, R⁸] MP peptide, at 8 µmol L⁻¹ (MBC), completely reduces the initial bacterial load (4 log₁₀) after 70 min of incubation.

Hemolytic and cytotoxic assays

The hemolytic activity of the peptides was evaluated against bovine erythrocytes. As shown in Fig. 3B, the parent peptide (mastoparan-L) and the mastoparan-MO analog showed the highest percentages of hemolysis (62% and 50%, respectively) when tested at 100 µmol L⁻¹. By contrast, the [I⁵, R⁸] MP and mastoparan-R1 analogs showed less than 20% hemolytic activity at this same concentration (100 µmol L⁻¹). These differences in hemolytic potential are more likely to be related to the higher hydrophobicity (over 50%) in mastoparan-L and mastoparan-MO, impairing the cell selectivity in these two peptides.

Furthermore, the methylthiazolyldiphenyl-tetrazolium bromide (MTT) assay shows that increased concentrations of [I⁵, R⁸] MP caused a decrease in the cell viability of RAW 264.7 cells (Fig. S1). At the highest concentration tested, 128 µmol L⁻¹, we observed a significant (P > 0.05) decrease in cell viability (under 80%) when compared with the untreated control. At concentrations below or equal to 4 µmol L⁻¹, the [I⁵, R⁸] MP peptide did not compromise RAW 264.7 cells viability (cell viability >96%, Fig. S1). In addition, the LC10 value of the [I⁵, R⁸] MP peptide was approximately 4 µmol L⁻¹, whereas the LC50 value was 32 µmol L⁻¹. These results are promising because the most potent effect of the peptide against the Aurora S. aureus strain occurred at the lowest concentrations, between 4 µmol L⁻¹ (MIC) and 8 µmol L⁻¹ (MBC). Altogether, this suggests that the peptide can be used to treat infections caused by this strain of bacteria without damaging the host cells, thereby increasing the effectiveness of the treatment.

Resistance development analysis for [I⁵, R⁸] MP in S. aureus

To analyze whether [I⁵, R⁸] MP resistance could be acquired through short-term exposure, we performed a short in vitro evolutionary trajectory with S. aureus Aurora, the most susceptible strain against antibiotics and with the lowest [I⁵, R⁸] MP MIC. After challenging three lineages from this strain for 5 days at a subinhibitory concentration (2 µmol L⁻¹) of [I⁵, R⁸] MP, the MIC for lineages 1 and 2 was 4 µmol L⁻¹, whereas lineage 3 growth was inhibited at 2 µmol L⁻¹. Thus, no resistance acquisition was observed for [I⁵, R⁸] MP in a short period. This result indicates that [I⁵, R⁸] MP could be used for short treatments without induction of a fast resistance by the bacterium.

Peptide stability assay

The stability of [I⁵, R⁸] MP was tested in both milk and human serum to determine the time required to observe the complete degradation of the peptide (Fig. S2 and S3). Table 4 shows the peptide integrity percentages at different times for each treatment. In milk, the peptide signal has already decreased by about 40% after 4 h and is no longer detectable after 12 h (Fig. S2). In human serum, the degradation is much faster, and after 4 h, only about 20% of the initial amount of peptide can be detected (Fig. S3).

TABLE 4.

Percentage of [I⁵, R⁸] MP peptide remaining in cow milk and human serum after 12 h, quantified by reverse-phase high-performance liquid chromatography (RP-HPLC) using a reversed-phase Venusil ASB C18 column. In both cases, the [I⁵, R⁸] MP was diluted to a final concentration of 64 µmol L⁻¹ by mixing the peptide with human serum and milk in a 1:4 ratio^c,d

Incubation	Relative peak area (%)a
	0 h	2 h	4 h	6 h	12 h
Milk	100 ± 0.22	83.47 ± 1.89	39.16 ± 2.76	19.16 ± 3.55	∼0
Serum	100 ± 0.16	57.64 ± 3.47	18.54 ± 1.65	∼0	NDb

^a Calculated by subtracting the relative area of treatment peaks from the initial control peak area.

^b ND: not determined.

^c Results represent the mean ± standard deviation (SD) from three replicates.

^d In both cases, the [I⁵, R⁸] MP was diluted to a final concentration of 64 μmol L^-1 by mixing the peptide with human serum and milk in a 1:4 ratio.

Fluorescence microscopy indicates that the [I⁵, R⁸] MP peptide causes cell damage in bacteria

To gain more insights into the antibacterial activity of our lead peptide candidate, [I⁵, R⁸] MP, the membrane integrity of S. aureus Aurora bacteria with live/dead staining was investigated through fluorescence microscopy (Fig. 4). Untreated bacteria showed predominantly green and minor red fluorescence (Fig. 4, first column). Conversely, cells exposed to [I⁵, R⁸] MP at MIC (second column) and MBC (third column) showed higher red fluorescence, but with live cells at the MIC treatment (second column) and nonliving cells at MBC treatment (third column). These results indicate membrane cell damage at both concentrations.

Fig 4.

Fluorescence microscopy of S. aureus Aurora treated with 4 µmol L⁻¹ (MIC) (second column) and 8 µmol L⁻¹ (MBC) (third column) values of [I⁵, R⁸] MP. The bacteria were stained with the live/dead BacLight viability kit after 18 h of treatment. Untreated bacteria (first column) were used as a control. The panels show SYTO 9 dye (top row), propidium iodide (PI) dye (middle row), and a merged image of both types of staining (bottom row). Scale bars = 10 µm, 40 × objective.

Biophysical experiments suggest that the [I⁵, R⁸] MP peptide acts through a membranolytic mechanism

To monitor the effect of [I⁵, R⁸] MP treatment on S. aureus Aurora cell surface, atomic force microscopy (AFM) was also performed. Based on the antimicrobial activity results, the peptide was tested at MBC (8 µmol L⁻¹) and 10-fold higher MBC. As depicted in Fig. 5, untreated S. aureus Aurora bacteria maintained their typical spherical shape and characteristic cluster arrangement, with sharp boundaries between cells and smooth surfaces without morphological changes. Nonetheless, the peptide treatments led to greater membrane disruption and loss of cell arrangement, being more evident in the 10-fold MBC treatment (Fig. 5). As observed in Fig. 5 (bottom row), the cytoplasmic content leakage due to the membrane disruption produces a substantial cell height loss. Then, the cells that initially showed a height of 759.81 nm (untreated bacteria) exhibited a decrease in their height at 601.26 nm at MBC treatment and 391.65 nm at 10× MBC treatment of [I⁵, R⁸] MP.

Fig 5.

Atomic force microscopy imaging of S. aureus Aurora treated with 8 and 80 µmol L⁻¹ of [I⁵, R⁸] MP. The panels show the untreated bacteria (first column), used as a control, the MBC (8 µmol L⁻¹) treatment (second column), and 10× MBC (80 µmol L⁻¹) treatment (third column). The top and middle rows correspond to the deflection and height images of bacteria, respectively. The bottom rows are 3D height images. Total scanning area for image: 10 × 10 µm². Scale bar = 1 µm.

[I⁵, R⁸] MP peptide causes hydrodynamic diameter to decrease and membrane depolarization of S. aureus bacteria

To further characterize the peptide-bacteria interaction, changes in the hydrodynamic diameter and surface zeta potential of S. aureus Aurora in the presence of [I⁵, R⁸] MP peptide were evaluated. In the absence of the peptide, the size of S. aureus Aurora is around 734–1,087 nm, with an average hydrodynamic diameter of 913 nm (±150 nm) and a mean zeta potential of −31 mV (Table 5). At MIC and MBC of the [I⁵, R⁸] MP peptide, the bacteria displayed a decrease in the size range with 758 nm (±172 nm) and 744 nm (±162 nm) average hydrodynamic diameters, respectively. Moreover, upon peptide addition, the potential shifted toward more positive values, −5.84 mV at MIC and −1.51 at MBC, thus indicating that this peptide depolarizes bacterial membranes (Table 5). It is worth noting that the high polydispersity index values (Table 5) indicate a wide distribution in the size of bacteria, which may be explained by the fact that the bacterial cells are in division (midlogarithmic growing bacteria), and therefore, it is possible to find cells in different growth cycle states.

TABLE 5.

Average hydrodynamic diameter, polydispersity index, and ζ-potential of S. aureus Aurora in the absence and presence of [I⁵, R⁸] MP peptide^d

Sample	DHa (nm)	PIb	ζ-potentialc (mV)
Untreated S. aureus Aurora	913 ± 150	0.481 ± 0.09	−31 ± 2.36
S. aureus Aurora - [I5, R8]MP - [4 µmol L−1]	758 ± 172	0.621 ± 0.01	−5.84 ± 1.02
S. aureus Aurora - [I5, R8]MP - [8 µmol L−1]	744 ± 162	0.616 ± 0.05	−1.51 ± 2.36

^a D_H: hydrodynamic diameter.

^b PI: polydispersity index.

^c zeta potential.

^d Average values of three repeats are presented, together with ± standard deviation.

DISCUSSION

When it comes to the rational design of AMPs, the physicochemical determinants of their activity (e.g., charge, hydrophobicity, and hydrophobic moment) must be considered together, since there is an interdependent relationship between them (34). Here, we evaluated and compared the physicochemical properties of three mastoparan-L analogs. As previously shown, the modifications made to the mastoparan-L peptide sequence altered the physicochemical parameters of the analogs. Consequently, there was a change in the structural and bioactivity profiles of these peptides. Thus, the addition of five apolar amino acid residues to the parental peptide sequence led to an increase in the hydrophobicity of the mastoparan-MO peptide, as well as a reduction in the hydrophobic moment. This improved the activity of the analog against the strains of S. aureus tested. However, only a slight decrease in hemolytic activity was observed. Bearing this in mind, the high hydrophobicity has been commonly related to low selectivity, which can lead to not only improved antimicrobial activity but also high rates of cytotoxicity on healthy mammalian cells by AMPs (35–37).

Instead, the analog mastoparan-R1, which is less hydrophobic and more cationic and amphipathic, did not inhibit the bacterial growth of any tested strain, also showing low hemolytic activity. This may be explained by the high hydrophobic moment and charge in this peptide, which could lead to aggregation and, therefore, a molar decrease at the target site of infection (in this particular case, the bacterial cells) (23). By contrast, [I⁵, R⁸] MP peptide displayed a more balanced adjustment in the physicochemical parameters evaluated, resulting in antibacterial activities at low concentrations and low hemolytic and cytotoxicity activities. It is generally accepted that amphipathicity is directly associated with the biological activities of AMPs, being reported as an important characteristic that favors interactions with cell membranes (38). Nonetheless, some studies have also shown that the imperfect amphipathicity of α-helical peptide structures can contribute to the improvement of the activity and specificity of these molecules (39–41). Then, we proposed that in [I⁵, R⁸] MP peptide, the imperfect amphipathicity promoted by the residue substitution in the nonpolar face assisted in enhanced biological activities and cell selectivity. Therefore, this peptide stood out as our best therapeutic candidate against BM.

The ability to undergo an amphipathic α-helical conformation is commonly correlated with the biological activities of several mastoparan peptides (29, 42). Thereby, the changes in the structural conformation assessed by CD indicated that the modifications also influenced the helical fractions exhibited by the analogs. However, in general, all these peptides displayed structural switching from an unfolded conformation in an aqueous solution to an α-helix conformation when in a hydrophobic and membrane mimetic environment. Nevertheless, it is worth noting that the higher helical content is not necessarily related to improved antibacterial activities, as postulated a few decades ago and which has even been shown for other mastoparan-like peptides as well (43). For instance, mastoparan-L was the most helical peptide but with low anti-BM properties.

Here, we selected the [I⁵, R⁸] MP analog for further characterization using a suite of biophysical and biological approaches. This peptide showed an improved membrane selectivity when compared with mastoparan-L, displaying potent antibacterial activities, even against antibiotic-resistant S. aureus strains, and cytotoxicity concentrations of 2-fold to 8-fold higher than its MIC. The lower cytotoxicity displayed by [I⁵, R⁸] MP could be associated with its physicochemical properties, greater charge, and amphipathicity, when compared with the parental peptide (mastoparan-L), as proposed for analogs of mastoparan-like peptides EMP-EM1 and EMP-EM1 (44). Then, since eukaryotic membranes have greater hydrophobicity and slightly negative charge, the peptide interaction with them is expected to be weaker than with bacterial membranes (44).

The [I⁵, R⁸] MP peptide has a rapid bactericidal action and is able to depolarize and permeabilize the membrane of S. aureus Aurora strain. Similarly, in a previous work, the time required for the [I⁵, R⁸] MP peptide to cause membrane permeabilization of an S. aureus strain was longer than that required to cause membrane depolarization (28), suggesting that even when the membrane damage could represent a key step for the peptide action, this would not be the mechanism through which this molecule causes cell death (28). It is worth highlighting that in the present study, we used 6-fold to 12-fold lower concentrations of the [I⁵, R⁸] MP peptide for the killing kinetic and zeta potential assays compared with the study by Irazazabal et al (28). Then, the differences observed in our results obtained for these assays, as well as for the MIC and MBC assays, might be explained by the fact that we used C-terminal amidated form of the [I⁵, R⁸] MP peptide, contrary to Irazazabal and coworkers (-COOH) (28). Bearing this in mind, various studies have shown that amidated mastoparans show improved biological activities (36, 45).

The morphological changes on bacterial surfaces upon peptide treatment have been demonstrated by microscopy approaches, including AFM (46). In this study, the peptide-treated S. aureus Aurora bacteria showed morphological damage. At MBC and 10× MBC, the bacterium presented high loss due to the membrane cell disruption. Thus, our results of the bacterial surface changes are in agreement with those obtained by Irazazabal and coworkers (28) for this same peptide, [I⁵, R⁸] MP, through scanning electron microscopy with field emission gun (SEM-FEG). In that work, the authors reported membrane disruption in the gram-positive bacteria Listeria ivanovii. Based on the AFM images, we confirmed that bacterial death is accomplished by membrane disruption and intracellular content leakage. Moreover, these findings are supported by fluorescence microscopy data, where the bacterial cells at MBC (8 µmol L⁻¹) were all red (death) cells.

Microbial drug resistance acquisition is a challenge facing antimicrobial agents (47) and a critical pharmacodynamic parameter that determines the lifetime of new antimicrobials (48). Bearing this in mind, we evaluated whether [I⁵, R⁸] MP selected S. aureus Aurora for resistance. Our results show that in short-term exposure, S. aureus was not capable of developing resistance to the [I⁵, R⁸] MP peptide. In this sense, in previous studies, the inability of bacteria to develop resistance to AMPs was attributed to their action on the membrane and even their possible simultaneous action on multiple cellular targets (44, 47, 49).

Recently, it has been reported that the mastoparan peptide polybia-MP1 was capable of inhibiting Pseudomonas aeruginosa ATCC 27853 and multidrug-resistant P. aeruginosa isolated from mastitic cow growth at 75 µmol L⁻¹ and 450 µmol L⁻¹, respectively (50). In another study, the potential of linear and branched forms (dimers and tetramers) of peptide IRK were tested against S. aureus ATCC 29213 and methicillin-resistant Staphylococcus aureus (MRSA) (clinical isolates from mastitic milk samples). As a result, only the linear form inhibited S. aureus ATCC 29213 growth at 16 µmol L⁻¹, whereas a 4-fold higher concentration (64 µmol L⁻¹) was necessary for linear and branched peptides to inhibit MRSA strains (51). Moreover, Popitool and coworkers (52) showed that a pleurocidin-derived peptide, named Pm11, exhibits a broad antibacterial spectrum against gram-negative and gram-positive bacteria (MBC between 2.5 and 10 µmol L⁻¹). Although the antibacterial activity of Pm11 toward S. aureus was lower (5 µmol L⁻¹) than our lead peptide candidate, [I⁵, R⁸] MP (8 µmol L⁻¹), the killing kinetic is slower for Pm11 (240 min) relative to [I⁵, R⁸] MP (70 min).

In summary, although much must be done for a successful clinical translation of mastoparan-like peptides, the large body of information regarding the peptide’s functional and structural characterization constitutes an important basis to help us deepen our understanding of the physicochemical determinants that drive the biological activities of AMPs, which will assist in the design of novel AMPs with enhanced antimicrobial action and low side effects on mammalian cells. Here, we showed that the [I⁵, R⁸] MP peptide has an improved biological activity profile, when compared with the parent peptide mastoparan-L in the BM treatment scenario, exhibiting an enhanced membrane selectivity and a rapid membranolytic mechanism of action. Finally, further studies are needed to explore the therapeutic potential of [I⁵, R⁸] MP peptide in animal models to validate the in vitro results. In conclusion, this peptide could be a starting point for peptide-based drug development in bovine mastitis treatment, with a clear advantage of no residue in milk, which would help reduce the use of classical antibiotics.

MATERIALS AND METHODS

Peptide synthesis

The mastoparan-L peptide and its analogs mastoparan-R1 (23), mastoparan-MO (29), and [I⁵, R⁸] MP (28) were purchased from Peptide 2.0 Incorporated (Chantilly, VA, USA). The peptides were synthesized using the F-moc (9-fluorenylmethoxycarbonyl) solid-phase strategy, employing a rink-amide resin. All peptides were synthesized with >95% purity and amidated C-terminal. The molecular masses were confirmed by matrix-assisted laser desorption ionization-time of flight-mass spectrometry (MALDI-ToF) analysis (Ultraflex III type mass spectrometer, Bruker Daltonics) in the reflector mode. Calibration was performed using Peptide Calibration Standard II (Bruker Daltonics) as molecular mass standards.

Calculation and comparison of physicochemical properties

The theoretical physicochemical properties of mastoparan-MO, mastoparan-R1, and [I⁵, R⁸] MP derivatives, as well as their parent peptide (mastoparan-L), were calculated using the HeliQuest server (http://heliquest.impc.cnrs.fr) (53), from the amino acid sequences of the peptides. The calculated physicochemical properties included the net charges (z), hydrophobicity (<H>), and hydrophobic moment [<µH>; Eisenberg scale (54)].

Circular dichroism spectroscopy

The secondary structures of mastoparan-L and its analogs were evaluated at 50 µmol L⁻¹ in different environments, including ultrapure water, 10 mmol L⁻¹ KH₂PO₄ buffer (pH 7.4), 2,2,2 trifluoroethanol (TFE) 50% in water (vol/vol), and sodium dodecyl sulfate micelles (SDS) at 100 mmol L⁻¹. The CD spectra were recorded in a JASCO spectropolarimeter (J-1100) equipped with a Peltier temperature controller (25°C), using a quartz cuvette of 1 mm path length, wavelength from 185 to 260 nm, data pitch of 0.5 nm, and a scanning speed of 50 nm min⁻¹, as well as a step resolution of 0.1 nm with a response time of 1 s and five scan accumulations for each sample. The CD spectra of solvent solutions were recorded for background subtraction. All data were analyzed using Spectra Manager software Jasco’s Fast Fourier algorithm and corrected baseline. The α-helix content of each peptide was calculated from the ellipticity values at 222 nm, as described by Chen et al (33).

Determination of resistance and sensitivity to antibiotics

The resistance profiles of S. aureus 02/18, 353/17, and Aurora strains for cefoxitin (30 µg), tetracycline (30 µg), ciprofloxacin (30 µg), and gentamicin (30 µg) were analyzed accordingly with Clinical and Laboratory Standards Institute (CLSI) guidelines (55). The antibiotics were purchased commercially from CECON. Briefly, the strains were grown in Mueller-Hinton broth (MHB), until they reached ~1 × 10⁸ CFU mL⁻¹, swabbed on Mueller-Hinton agar (MHA) plates and incubated for 18 h at 37°C. Then, the diameter zones were measured and analyzed accordingly with CLSI interpretative categories.

Minimal inhibitory concentration and minimal bactericidal concentration assays

Three S. aureus strains (isolates 02/18, 353/17, and Aurora) were obtained from Girolando breed cows with mastitis, belonging to farms located in the municipalitiest of Rochedo (19°57'11.5''S 54°253'17.1''W), Campo Grande (20°28'09.2''S 54°37'08.7''W), and Jaraguari (20°08'22.3''S 54°24'01.0'' W) in the state of Mato Grosso do Sul and used for the antimicrobial assays. Bacterial strains were kindly provided by Professor Cassia Rejane Brito Leal, Universidade Federal do Mato Grosso do Sul (UFMS), Campo Grande, MS, Brazil. The isolates were plated in petri dishes containing MHA medium and incubated for 18 h, followed by the standardization of the CFU. MIC tests were performed according to the 96-well microplate dilution protocol established by the CLSI (56). The bacterial strains were grown in MHA plates overnight at 37°C. Afterward, three isolated colonies from each bacterium were inoculated in 5 mL of MHB (three biological replicates) and incubated at 37°C at 200 rpm overnight. After this period, 1:50 dilutions of overnight cultures were performed in MHB and grown to the midlog phase. Bacteria at 5 × 10⁵ CFU mL⁻¹ per well were exposed to different peptide concentrations ranging from 1 to 32 µmol L⁻¹. The antibiotics, chloramphenicol and ciprofloxacin, were used as a positive control (at the same concentrations as the peptides), and bacterial suspensions (5 × 10⁵ CFU mL⁻¹) in MHB were used as a negative control. All tests were performed with three biological replicates and three technical replicates. The microplates were incubated at 37°C for 18 h, and optical density (O.D.) readings were taken at 600 nm. MIC was determined as the lowest concentration of peptides or antibiotics that inhibited 100% of bacterial growth. To determine the minimal bactericidal concentration (MBC), 10 µL of bacterial cultures from each well corresponding to the MIC value (and from the preceding wells) was transferred to MHA plates, which were then incubated at 37°C for 18 h. The lowest concentration of peptides that did not allow bacterial subcultures to grow was determined as the MBC value.

Hemolytic assays

Hemolytic assays were assessed by using fresh healthy cattle blood, approved by the Ethics Committee on the Use of Animals (CEUA) of the Universidade Católica Dom Bosco (protocol number: 015/2018), as previously described (57) with some modifications. Briefly, the collected blood was centrifuged at 400 × g for 2 min at 10°C. The supernatant was discarded, and the red cells were washed three times with saline solution (0.9%) prior to plating (100 µL well⁻¹). The hemolytic activity was determined by exposing the suspensions of bovine red blood cells in NaCl (0.9%) to different peptide concentrations (from 1.5 to 100 µmol L⁻¹). Positive and negative controls were prepared by adding 2% (vol/vol) Triton X-100 and NaCl to the erythrocyte suspensions, respectively. All preparations were performed in triplicate and incubated at room temperature for 1 h. After centrifuging, 100 µL of supernatant was transferred onto a 96-well flat-bottom plate, and hemoglobin release was monitored by measuring the O.D. at 415 nm.

Cytotoxicity assay

The MTT assay was used to determine the sublethal (LC10) and lethal concentrations (LC50) of the [I⁵, R⁸] MP peptide against RAW 264.7 mammalian cells. The cell line was maintained under sterile conditions, using Dubelcco's modified eagle medium (DMEM) culture medium supplemented with 10% fetal bovine serum (FBS) and incubated at 37°C under an atmosphere of over 95% humidity and 5% CO₂, until 80% confluence. The culture cell was then harvested and resuspended at a concentration of 1 × 10⁶ cells mL⁻¹, and 100 μl of the cell suspension was plated into a 96-well plate and incubated at 37°C for 24 h. The peptide was diluted 2-fold in the range of 128 μmol L⁻¹ to 1 μmol L⁻¹, added to the wells, and incubated at 37°C for 24 h. Then, 10 µL of MTT (5 mg mL⁻¹) was added to each well, and the plate was incubated at 37°C for 4 h, with constant shaking and protection from light. To solubilize the formed formazan crystal, 100 µL of isopropanol was added to each well, and the plate was left covered in the dark at room temperature for 4 h. Finally, OD570 nm was read, and cell viability was calculated according to the following equation.

$${\displaystyle Cellviability\mathrm{\%}=\frac{opticaldensityoftreatedcells}{opticaldensityofuntreatedcells}x100\mathrm{\%}}$$

Time-kill kinetics assays

Time-kill kinetics were performed using the best candidate peptide of the present study, [I⁵, R⁸] MP, against the S. aureus Aurora strain, as previously described (28). Midlogarithmic growing S. aureus Aurora was diluted to 5 × 10⁶ CFU mL⁻¹. Afterward, bacterial cultures were exposed to MBC (8 µmol L⁻¹) determined for the [I⁵, R⁸] MP peptide (each well containing a final volume of 150 µL) against the S. aureus Aurora strain. 100 µL aliquots were removed every 10 min for 90 min, then diluted (1:10, three subsequent dilutions) in saline (0.9%), and seeded (50 µL) on MHA plates. Colony counting was performed manually after 18 h of incubation on agar plates at 37°C. Bacterial growth under the same conditions described above, but without the presence of the peptide, was evaluated to be used as a control of cell viability.

Resistance development analysis for [I⁵, R⁸] MP in S. aureus

An in vitro evolutionary trajectory was performed for 5 days as previously described (58), with modifications. Three lineages from S. aureus Aurora were challenged with 2 µmol L⁻¹ (1/2 MIC) of [I⁵, R⁸] MP in MHB for 5 days. Initially, the cultures started with ~5 × 10⁵ CFU mL⁻¹ supplemented with the peptide in 96-well plates and were grown at 37°C, 80 rpm. Every 24 h, 1 µL of each culture was propagated to 99 µL of a new fresh medium supplemented with the peptide. After five passages, a new MIC assay was made, and the results were compared with those obtained before the evolutionary trajectory.

Peptide stability assay

The stability of [I⁵, R⁸] MP toward serum and milk proteases was evaluated as previously described (59, 60), with some modifications. Fresh milk collected from healthy cows was centrifuged at 24,000 × g for 10 min, and the milk fat supernatant was removed. Then, aliquots of milk and human serum (male AB plasma, Sigma-Aldrich) were mixed with the peptide to a final concentration of 64 µmol L⁻¹. Aliquots were incubated at 37°C, centrifuged, and filtered through a 0.22 µm pore after 0, 1, 2, 6, and 12 h. The samples were analyzed by RP-HPLC, using a Shimadzu LC system (Kyoto, Japan) with LC Solution software. Chromatographic separations were performed using a gradient method and a reversed-phase Venusil ASB C18 column (250 × 4.6 mm, 5 µm, Bonna-Agela Technologies), with an injection volume of 25 µL. The segmented gradient was applied at a constant flow rate of 1.0 mL min⁻¹ and photo diode array (PDA) detection at 216 nm, using a mixture of water and acetonitrile eluents as the mobile phase. [I⁵, R⁸] MP was quantified by subtracting the relative area of the treatment peaks from the initial control peak area. All assays were performed in triplicate.

Fluorescence microscopy

S. aureus Aurora (5 × 10⁵ CFU mL⁻¹ inoculum) was incubated with 4 and 8 µmol L⁻¹ of the [I⁵, R⁸] MP peptide (MIC and MBC, respectively) on round glass coverslips (pretreated with 20% fetal bovine serum and 5% dimethyl sulfoxide for 18 h at 4°C) in a 24-well polystyrene plate at 37°C for 18 h. Bacteria without treatment were used as growth control. Subsequently, bacteria were stained with the LIVE/DEAD BacLight Viability Kit (Molecular probes—Life Technologies) containing SYTO 9 dye and propidium iodide in dark for 10 min at 37°C, following the manufacturer’s instructions. Images were acquired using a Leica DM000 LED, equipped with a Leica DFC 7000T camera. Images were made with a specific dye filter and collected in LAS V4.12 software.

Atomic force microscopy

AFM studies were performed using methods described by Domingues and coworkers (61), with some modifications. S. aureus Aurora (~10⁶ CFU mL⁻¹ per well) was incubated at 37°C for 18 h with our lead peptide candidate ([I⁵, R⁸] MP) at MBC (8 µmol L⁻¹) and 10× MBC (80 µmol L⁻¹). Then, untreated (control) and peptide-treated bacteria were centrifuged at 731 × g for 15 min and washed three times with ultrapure water. A drop of 80 µL of each sample was applied to a gelatin-coated glass slide and left at 25°C for 1 h. Images of untreated and treated bacteria were obtained in a scanning probe microscopy (SPM-9700HT, Shimadzu Corp., Kyoto, Japan) operated by dynamic contact (air) mode, using Pointprobe–silicon SPM–sensor, with a resonant frequency of 320 kHz and a force constant of 42 N/m. Height and size images were recorded with similar AFM parameters (operating point, sweep range, and gain). The scan rate was set between 0.5 and 1.0 Hz, and the operating point was close to 0.2 V. Height and deflection signals were collected, and the images were analyzed with SPM analysis software (SPM9700 analysis, Shimadzu, Japan).

Size and zeta potential measurements

The size and zeta potential measurements were performed as reported by Domingues and coworkers (61), with modifications. Midlogarithmic growing S. aureus Aurora was diluted to ~10⁶ CFU mL⁻¹. Bacteria were then centrifuged at 731 × g for 15 min and washed three times under the same conditions using ultrapure water. The size and zeta potential of S. aureus Aurora were assessed in the presence of 4 and 8 µmol L⁻¹ (MIC and MBC, respectively) of the [I⁵, R⁸] MP peptide. The measurements were carried out on a Malvern Zetasizer Advance Series Pro (Malvern Panalytical Ltda, UK) by noninvasive back scatter (NIBS) dynamic light scattering (DLS) (for size) and mixed-mode measurement phase analysis light scattering (M3-PALS) (for zeta potential) at 25°C, employing DTS1070 cells. Three independent replicates of each condition were measured, and each measurement consisted of 5 runs (with 10–100 runs). Values of viscosity and refractive index were set to 0.8872 cP and 1.330, respectively. The hydrodynamic diameter (D_H) of the samples was obtained from the highest scattered light intensity peak and calculated through the Stokes-Einstein relationship (61). The results were analyzed with ZS Xplorer software.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/jb.00071-24.

Figures S1 to S3. jb.00071-24-s0001.docx.

Cytotoxicity assay and reverse-phase HPLC chromatograms.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.