Bioactivity of synthetic peptides from Ecuadorian frog skin secretions against Leishmania mexicana, Plasmodium falciparum, and Trypanosoma cruzi

ABSTRACT Chagas disease, leishmaniasis, and malaria are major parasitic diseases disproportionately affecting the underprivileged population in developing nations. Finding new, alternative anti-parasitic compounds to treat these diseases is crucial because of the limited number of options currently available, the side effects they cause, the need for long treatment courses, and the emergence of drug-resistant parasites. Anti-microbial peptides (AMPs) derived from amphibian skin secretions are small bioactive molecules capable of lysing the cell membrane of pathogens while having low toxicity against human cells. Here, we report the anti-parasitic activity of five AMPs derived from skin secretions of three Ecuadorian frogs: cruzioseptin-1, cruzioseptin-4 (CZS-4), and cruzioseptin-16 from Cruziohyla calcarifer; dermaseptin-SP2 from Agalychnis spurrelli; and pictuseptin-1 from Boana picturata. These five AMPs were chemically synthesized. Initially, the hemolytic activity of CZS-4 and its minimal inhibitory concentration against Escherichia coli, Staphylococcus aureus, and Candida albicans were determined. Subsequently, the cytotoxicity of the synthetic AMPs against mammalian cells and their anti-parasitic activity against Leishmania mexicana promastigotes, erythrocytic stages of Plasmodium falciparum and mammalian stages of Trypanosoma cruzi were evaluated in vitro. The five AMPs displayed activity against the pathogens studied, with different levels of cytotoxicity against mammalian cells. In silico molecular docking analysis suggests this bioactivity may occur via pore formation in the plasma membrane, resulting in microbial lysis. CZS-4 displayed anti-bacterial, anti-fungal, and anti-parasitic activities with low cytotoxicity against mammalian cells. Further studies about this promising AMP are required to gain a better understanding of its activity. IMPORTANCE Chagas disease, malaria, and leishmaniasis are major tropical diseases that cause extensive morbidity and mortality, for which available treatment options are unsatisfactory because of limited efficacy and side effects. Frog skin secretions contain molecules with anti-microbial properties known as anti-microbial peptides. We synthesized five peptides derived from the skin secretions of different species of tropical frogs and tested them against cultures of the causative agents of these three diseases, parasites known as Trypanosoma cruzi, Plasmodium falciparum, and Leishmania mexicana. All the different synthetic peptides studied showed activity against one of more of the parasites. Peptide cruzioseptin-4 is of special interest since it displayed intense activity against parasites while being innocuous against cultured mammalian cells, which indicates it does not simply hold general toxic properties; rather, its activity is specific against the parasites.

KEYWORDS anti-microbial peptide, malaria, Chagas, leishmaniasis, anti-parasitic, frog C hagas disease, leishmaniasis, and malaria, caused by the protozoan parasites Trypanosoma cruzi, Leishmania spp., and Plasmodium spp., respectively, are among the most important parasitic infections in Latin America and globally (1)(2)(3).However, special challenges exist for the development of anti-microbials to treat these tropical infections.In the case of Chagas disease and leishmaniasis, the pharmaceutical industry has traditionally shown little interest (4,5).
Chagas disease is the most important parasitic disease in Latin America (6), affecting an estimated 7-8 million people (7).In recent years, human migration has caused the disease to extend to non-endemic regions (7,8).Chronic infection with T. cruzi causes irreversible heart or digestive damages, which may lead to disability and even death in ~30% of those infected (9).Only two drugs, benznidazole and nifurtimox, are currently approved for Chagas disease treatment (10), and neither of them is satisfactory because of lack of effectiveness, especially in chronic infections, in addition to their toxicity and side effects (11).
Leishmaniasis is a neglected tropical disease that mainly affects the economically disadvantaged population in different countries in Asia, Africa, the Caribbean, and Latin America (12)(13)(14)(15).Current treatment is based on drugs such as miltefosine, amphotericin B, and pentavalent antimonials; the latter two require parenteral administration.These drugs cause significant side effects and must be administered under close medical supervision, which complicates adherence to treatment (12).
Malaria kills more people in the world than any other parasitic disease, and its control and elimination are greatly dependent on effective anti-parasitic drugs (16).Nevertheless, Plasmodium falciparum, the most aggressive among the species of the genus capable of infecting humans, has developed resistance to the majority of the anti-malarial drugs in use (16).The current treatment based on artemisinin derivatives is losing efficacy in Southeast Asia, making the search for new anti-malarial drugs urgent.
Peptides were partially purified by Sepacore Flash chromatography system X50 (BUCHI).Peptide aliquots (10 mg/mL) were injected several times using a Reveleris C18 Flash Cartridge (4 g, 12.3 × 6 mm).The elution gradient was 5%-100% solvent B (ACN/ 0.1% TFA) for 35 min.Detection was set at 214 and 280 nm, and manual collection was performed to obtain 100 mg of each purified peptide, at >95% purity.Due to the difficult separation of some peptides by Flash chromatography, further purification was achieved by RP-HPLC.

Anti-microbial activity and hemolytic assay of CZS-4
The minimum inhibitory concentration (MIC) of CZS-4 over Escherichia coli American Type Culture Collection (ATCC) 25922, Staphylococcus aureus ATCC 29213, and Candida albicans ATCC 10231 was determined, as previously described (28,30,31).In brief, overnight cultures of each microorganism were subcultured in Muller-Hinton Broth (MHB) until reaching 1 × 10 6 CFU/mL for bacteria and 1 × 10 5 CFU/mL for yeast.Peptide serial dilutions in dimethylsulfoxide (DMSO), ranging from 0.4 to 209.4 µM were prepared, and 2 µL of each peptide dilution was added to 198-µL diluted bacterial or yeast culture in a 96-well sterile plate (with five replicates).Sterile MHB and microbial culture with DMSO were negative controls.Plates were incubated for 16 h at 37°C.Microorganism growth was measured at 600 nm.
Hemolytic activity was determined using 200 µL of 4% red blood cell solution incubated with 200 µL of serial dilutions peptide (0.4-209.4 µM) in phosphate-buffered saline (PBS) 1× (five replicates).Negative controls contained PBS instead of peptide; positive controls contained 2% vol/vol Triton X-100 to yield complete hemolysis.Assays were incubated at 37°C for 2 h, and samples were centrifuged at 1,00 × g for 5 min.Two-hundred microliters of supernatant was transferred to a 96-well plate and absorb ance was measured at 550 nm.

Parasite culture
L. mexicana promastigote stock cultures were maintained in USHMARU biphasic medium [blood agar slant overlayed with 3 mL of Schneider's Drosophila medium (SDM), containing 10% fetal bovine serum (FBS)] at 25°C.Promastigotes were collected by centrifugation and washed with PBS, and culture medium was replaced every 2 days.Every 4 days, the parasites were transferred to a new tube and to a 25-cm 2 tissue culture flask containing 10 mL of monophasic medium (SDM + 10% FBS + 1% penicillin/strepto mycin).If present, rosettes were disrupted by passing the culture through 10-mL syringes with 27G needles.
P. falciparum was cultured in human O + erythrocytes in Roswell Park Memorial Institute medium 1640, supplemented with 25-mM HEPES buffer, 10-mM glucose, 2-mM glutamine, and O + human plasma.Parasites were cultured under low-oxygen atmosphere (5% O 2 , 5% CO 2 , and 90% N 2 ) (34) and maintained in fresh human erythro cytes suspended at 4% hematocrit in complete medium at 37°C.Stock cultures were subpassaged every 3-4 days by transfer of infected red cells to a flask containing complete medium and uninfected erythrocytes.

In vitro anti-P. falciparum activity assays
In vitro sensitivity to peptides was tested using a previously described SYBR green I fluorescence-based method (39,40).Assays were set up in 96-well plates with twofold peptide dilutions in 150-μL total volume and 1.5% (vol/vol) final red blood cell concentra tion.Stock solutions of each peptide were prepared in DMSO.Experiments were started at an initial parasitemia of 0.5% (80% rings) synchronous parasite-infected red blood cells.Plates were incubated for 72 h at 37°C in an atmosphere of 5% CO 2 , 5% O 2 , and 90% N 2 .The SYBR green I dye-lysis mixture (l:100) was added to the parasites in black plates that were incubated at room temperature for an hour in the dark.The plates were then read using a fluorescence plate reader at excitation and emission wavelengths of 480 and 535 nm, respectively.Experiments were performed in duplicate wells.Three independent assays were performed for each peptide.

Activity against T. cruzi trypomastigotes
Trypomastigotes were rinsed, suspended in phenol red-free and FBS-free Dulbecco's Modified Eagle's Medium, placed in 96-well plates (1 × 10 6 /well), and incubated with 100-µM chlorophenol-β-D-galactopyranoside red (CPRG) (41) in the presence of serial dilutions of AMPs (100.0-to0.195-µM concentration range).Wells containing parasites not exposed to peptides and treated with 0.1% Triton served as 100% lysis reference.Plates were incubated in the dark at 37°C and 5% CO 2 for 4 h, and absorbance was read at 590 nm.Experiments were performed in duplicate wells.Three independent assays were performed for each peptide.

Activity against intracellular T. cruzi amastigotes
LLC-MK2 cells, 2 × 10 4 /well, were seeded and allowed to attach overnight to 96-well plates.he medium was removed and cells were infected with 1 × 10 5 parasites in DMEM-2 for 24 h.Cell monolayers were washed four times with PBS, and serial dilu tions of AMPs (100.0-to0.195-µM range) in DMEM-2 without phenol red were added to duplicate wells.Wells containing infected cells not exposed to peptides served as parasite growth reference.The infection was allowed to proceed for 96 h.Subsequently, lysis solution was added to each well to a final concentration of 100-µM CPRG and 0.1% Triton and incubated for 2 h.Absorbance was measured at 590 nm (41).Three independent assays were performed for each peptide.

Cytotoxicity over mammalian cells
The toxicity of the studied AMPs over LLC-MK2 and RAW 264 cells was determined via a resazurin (RZN) reduction assay (42).Briefly, 2 × 10 4 cells/well were seeded in DMEM-10 in 96-well plates.Twenty-four hours later, culture medium was removed, and 10 twofold serial dilutions (100.0-0.195µM) of each AMP in 200-µL volume were placed in duplicate wells.The plate was placed on the incubator in the dark at 37°C and 5% CO 2. Ten microliters of 3-mM RZN sodium salt in PBS per well was added, and the plate was incubated for 24 h.Finally, the fluorescence was measured (530-to 560-nm excitation and 590-nm emission wavelengths).Three independent assays were performed for each peptide.

Statistical data analysis
Half-maximal inhibitory concentration (IC 50 ) and half-maximal cytotoxic concentration (CC 50 ) with 95% confidence intervals in GraphPad Prism software (v.9.2.0., non-linear regression with curve fitting [model: log (inhibitor) vs response (three parameters)].The selectivity index (SI) was determined by dividing the CC 50 value of RAW 264.7 cells by the IC 50 of the parasites.

Peptide bioinformatics analysis
Sequence similarity between studied peptides and previous entries into the National Centre for Biotechnology Information database was explored with PSI-Blast (43,44).Furthermore, studied peptide sequences were compared among each other using T-Coffee tool (45).Complete physicochemical and biochemical characterization for each peptide was performed using ExPasy (46), HeliQuest (47), and Peptide Mass Calculator (v.3.2).Peptide secondary structure was predicted using JPred (48), PSIPred (49), and SOPMA (50).With the information obtained from the prediction, the five peptides were modeled using Pymol and optimized using ChemBioDraw and Gaussian software.Optimized structures were employed for docking studies.

Docking
Molecular docking simulations were performed using a phosphatidylcholine model containing 128 lipids and 2,460 water molecules simulated for 1.6 ns, which resemble eukaryotic cell membranes like those present in the studied parasites (51).Autodock tools were used to prepare the membrane model and peptide structures for calculations, which were performed using Autodock VINA (52), applying a 1-Å spacing, and a box size of 25 in X, 25 in Y, and 55 in Z. Exhaustiveness was set to 8 and full flexibility of the side chains was allowed.

Anti-microbial and hemolytic activity of CZS-4
Except for CZS-4, we have previously reported on the activity of all peptides included in this study against bacteria and Candida albicans, as well as their hemolytic activity (28)(29)(30)(31).Here, we show that CZS-4 displayed anti-microbial activity against bacteria (E. coli and S. aureus) and yeast (Candida albicans).The lowest MIC value obtained was against E. coli (13.09 µM) followed by S. aureus (26.18 µM).The highest MIC corresponded to the yeast C. albicans with a value of 52.36 µM (Fig. 2A).On the other hand, CZS-4 induced hemolysis of 13% at 128 µM.While in anti-bacterial concentration, this peptide showed hemolysis of less than 9.1% (Fig. 2B).

Cytotoxicity of peptides over mammalian cells
All studied AMPs displayed some degree of cytotoxicity against mammalian cells (Table 3; Fig. 3A and B).CZS-1 displayed IC 50 values of 3.17 and 2.38 µM against LL-MCK2 and RAW macrophages, respectively.Interestingly, CZS-4 caused very low cytotoxicity, with IC 50 values of 70.6 µM for LL-MCK-2 cells and 47.96 µM for RAW macrophages.Mean while, CZS-16 and DRS-2 displayed cytotoxic activity against both types of mammalian cells, ranging from 3.14 to 5.59 µM.In a case where mammalian cell types were affected

L. mexicana
Promastigotes displayed the greatest AMP sensitivity among the studied parasites.Promastigotes were incubated with AMPs for 48 h at different concentrations.L. mexicana was susceptible to all studied peptides, with IC 50 values ranging from 0.09 to 10.0 µM (Table 3).CZS-4 appeared to be the most effective of all the tested AMPs, while CZS-16 showed a reduced effect.

P. falciparum
The studied peptides had anti-plasmodial activity against the drug-sensitive NF54 and the multiple drug-resistant C2B strains of P. falciparum.IC 50 values ranged from 4.87 to 52.56 µM, where DRS-SP2 and CZS-4 showed higher activity in both strains (Table 3).

SI
As depicted in Table 4, CZS-4 showed the highest SI against L. mexicana and T. cruzi trypomastigotes at 532.89 and 30.94, respectively.On the other hand, CZS-16 is selectively active against mammalian cells.PST-1 also displayed low selectivity against T. cruzi and P. falciparum, with SI values from 0.05 and 1.775.PST-1's selectivity for L. mexicana was much greater, with an SI of 25.20.

Structural model prediction
Secondary structure analysis predicts an alpha-helical conformation for all peptides (Fig. 4), in agreement with the secondary structures reported in the literature for other members of their families and for other anti-microbial peptides found in different species of frogs.The five peptides were modeled in an alpha-helical secondary structure and optimized.Optimized three-dimensional peptide structures are shown in Fig. 5.

Molecular docking
A membrane model containing 128 phosphatidylcholine molecules and 2,460 water molecules was chosen because it adequately represents eukaryotic pathogen cell membranes.Our analysis predicts favorable interaction between this membrane model and optimized peptide structures, as indicated by the negative docking scores obtained, which ranged from −5.4 kcal/mol in CZS-16 to −7.6 kcal/mol in CZS-1.Additionally, the most favorable interaction conformation for all studied peptides would be to locate along the lipid bilayer (Fig. 6), which would in turn be predicted to cause membrane destabilization, ultimately resulting in parasite lysis.

DISCUSSION
Amphibian skin AMPs have been extensively studied and proposed as therapeu tic alternatives for the treatment of infectious diseases, including those caused by multidrug-resistant microorganisms.To date, more than 1,000 amphibian AMPs with broad structural and biological diversity have been reported (53).Anti-parasitic activity has been described in the dermaseptin peptide family, but the recently described cruzioseptin and pictuseptin peptide families are anti-bacterial and anti-fungal peptides with unexplored anti-parasitic activity (28,30,31).
In this study, we report findings on the bioactivity of CZS-4 against bacteria and yeast, and CZS-1, CZS-4, CZS-16, DRS-SP2, and PTS-1 against mammalian cells and protozoan parasites.CZS-4 exhibited anti-microbial activity against S. aureus, C. albicans, and E. coli, similar to what was previously reported for other cruzioseptin family members (29,30).Interestingly, C. albicans is much less sensitive to the peptide compared to the bacteria, likely because of the resistance conferred by the fungal cell wall structure.MICs differ among cruzioseptins, with CZS-1 being the most potent one.It has been proposed that the biological activity of AMPs is correlated with certain physicochemical characteristics, such as hydrophobicity, high net charge, and helicity (20).Therefore, an exhaustive study must be performed to determine the physicochemical properties and structural determinants required to induce a potent effect against Gram-negative bacteria, Gram-positive bacteria, and yeasts.
As expected, when the studied AMPs were tested against the mammalian-specific life cycle stages of T. cruzi, they displayed greater activity against trypomastigotes than intracellular amastigotes.These findings were similar to those from other AMPs with trypanocidal activity, including melittin (54), NK-lysin (55), and several dermasep tins (24,56).Conversely, the effect on intracellular amastigotes is probably reduced because peptides must first cross the host cell's plasma membrane in order to reach the cytoplasm, where amastigotes multiply (55).Persistence of intracellular amastigote nests in different host tissues, including cardiac tissue, is crucial for long-term parasite survival during chronic Chagas disease (57).In this sense, the elimination of amastigotes is crucial, and therefore, peptide intracellular delivery and associated cytotoxicity are important challenges to be overcome for a peptide-based treatment.
While there was anti-malarial activity against the erythrocytic stages of both resistant and sensitive strains of P. falciparum, this activity was generally modest for most peptides and lower than some previously related peptides such as dermaseptin S4 derivatives  (23).DRS-SP2 and CZS-4 had the highest activity against P. falciparum, which was comparable to derivatives of DRS S3 and DRS S4 (22), suggesting that further research of these peptides can result in more promising activities.To reach the intracellular malaria parasite membrane, the peptides have to cross the erythrocytic membrane and enter the parasitophorous membrane, which may explain the lower activity when compared to other parasites (55).
As for L. mexicana, CZS-1, CZS-4, CZS-16, DRS-SP2, and PTS-1 demonstrated great leishmanicidal activity against promastigotes.These peptides were more active compared to other anti-microbial peptides isolated from amphibians (58) and marine sources (59).Although L. mexicana promastigotes were more sensitive to these AMPs than T. cruzi trypomastigotes, a limitation of our study is the lack of data for intracellular Leishmania amastigotes.Testing the studied peptides against both intracellular and extracellular forms of T. cruzi was possible because of the availability of the recombinant, beta-galactosidase expressing strain (Tula-beta gal).However, the colorimetric method employed to quantify the viability of L. mexicana parasites (MTT reduction) is not useful with intracellular parasites because the signal produced by the parasites is indistinguish able from that caused by the host cell, confounding the results.Therefore, a direct comparison of the sensitivity of the (clinically relevant) intracellular amastigotes of both parasite species is not possible from our data.Likely, intracellular amastigotes of L. mexicana will be less sensitive to the peptides, as is the case for T. cruzi.
Most AMPs interact with the plasma membrane of microorganisms as an initial mechanism of action.Several members of the cruzioseptin and dermaseptin families have shown a membranolytic effect on S. aureus, T. cruzi, and Leishmania spp.(32,56,60).Based on this observation and on our research, we believe that CZS-1, CZS-4, CZS-16, FIG 4 Secondary structure prediction for the five studied peptides obtained with JPred, PSIPred, and SOPMA bioinformatic tools.DRS-SP2, and PTS-1 act on parasites via membrane destabilization.Our hypothesis is supported by the results of molecular docking, which indicate a peptide-membrane interaction as evidenced by negative docking values.Previous computational studies have also shown that there is a favorable interaction between anti-microbial peptides and phospholipid bilayers (61,62).
Toxicity is an important concern when studying peptides for drug development (63).In our study, this challenge was evaluated exposing murine macrophages and LLC-MK2 cells to the peptides.All studied peptides displayed activity against RAW 264.7 macrophages; however, the activity was quite low for CZS-4.These data are relevant for anti-parasitic drugs, given that macrophages are one of the cells that parasites invade in the vertebrate host.Indeed, CZS-4 has great specificity for target ing L. mexicana promastigotes (SI = 532.89)and T. cruzi trypomastigotes (SI = 30.94).Conversely, CZS-4 and PTS-1 induced low cytotoxicity toward LL-MCK 2 cells.(CC 50 = 70.66 and 80.07 µM, respectively), while CZS-1 and CZS-16 were highly cytotoxic (CC 50 = 3.17 and 5,59 µM, respectively).Currently, several strategies have been used with the aim of reducing bioactive peptide toxicity, including cyclization, incorporation of D-amino acids, peptides, and computational techniques (63).
Although in vitro studies are a useful starting point for characterization of AMP biological properties, their anti-microbial and anti-parasitic activities as well as cytotoxic ity may differ greatly in vivo.Therefore, additional studies are warranted to clarify the therapeutic potential of the studied peptides.
In conclusion, we have shown that five amphibian AMPs, with anti-bacterial and anti-fungal properties previously reported, also display differing degrees of activity against the protozoan parasites L. mexicana, P. falciparum, and T. cruzi at micromolar concentrations.As expected, peptides were more active against extracellular parasite forms.Among the studied peptides, CZS-4 is the most promising due to its low toxicity and high efficacy in eliminating parasites in in vitro assays.In addition, bioinformatic analysis suggest that these peptides act through a membranolytic effect.

FIG 1
FIG1 Multiple sequence alignment (A) and identity matrix (B) of the studied peptides.Three conserved residues (glycine, leucine, and lysine) are present in all studied peptides.Glycine and leucine are hydrophobic amino acids, while lysine is basic and contributes to the positive charge of the peptides.The physicochemical characterization of the peptides is depicted in Table2.Peptides are composed of 300-400 atoms and 21-28 amino acids, corresponding to molecular weights between 2,118 and 2,990 Da.All of them are basic and positively charged at physiological pH.Their isoelectric point is higher than 7, and the number of their basic residues doubles that of negative ones.Hydrophobic amino acids constitute around 50% of the structure of each peptide.

FIG 2
FIG 2 Anti-microbial and hemolytic activities of the CZS-4 peptide.(A) Inhibitory effect of CZS-4 against E. coli, S. aureus, and C. albicans.(B) Hemolysis caused by CZS-4.One hundred percent of hemolysis was determined using Triton X-100.

FIG 3
FIG 3 Dose-response curves for the activity of the studied synthetic peptides against mammalian cells (A and B) and protozoan parasites (C-G).Data are shown as mean values of independent assays ± standard deviation.

FIG 5
FIG5 Optimized three-dimensional structure of the studied peptides obtained using ChemBioDraw and Gaussian software.

TABLE 1
Amino acid sequences and purity of the studied peptides a MALDI-TOF MS, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry.

TABLE 2
Physicochemical characterization of the five studied peptides

TABLE 3
Anti-parasitic activity and cytotoxicity of AMPs a Data are reported as IC 50 values (µM) and 95% confidence intervals.ND, not done. a

TABLE 4
SI of studied AMPsa,b SI denotes RAW 264.7 CC 50 /parasite IC 50 .b Larger values indicate greater peptide specificity for parasites over host cells. a