Pseudonajide is an antibiotic peptide derived from snake venom that alter cell wall and membrane integrity interfering on biofilm formation.

Background The increase of bacterial resistance phenotype cases is a global health problem. New strategies in scientific community must be explored in order to create new treatment alternatives. Animal venoms are a good source for antimicrobial peptides (AMPs), which are excellent candidates for new antimicrobial drug development. These molecules have highly diverse targets in prokaryotic cells, making resistance phenotype development more difficult. Results In this study we present a peptide of just 11 amino acids which has antimicrobial and antibiofilm activity against Staphyloccocus epidermidis. Named pseudonajide, it is derived from Pseudonaja textilis venom. Pseudonajide was selected based on the sequence alignments of various snake venom peptides that displayed activity against bacteria. Several concentrations of pseudonajide were tested in antibiofilm activity essay, it was detected that 25 µM was the best minimal concentration for biofilm inhibiting activity. Microscopy analysis demonstrates that pseudonajide interacts with the bacterial cell envelope, disrupting the cell wall and membrane leading to morphological defects in prokaryotes. Conclusions Our results suggest that pseudonajide’s positives charges interacts with negative charged cell wall components of S. epidermidis. Leading to cell damage and biofilm formation inhibition. up of 11 amino acids and derived from Pseudonaja textilis snake venom. It possesses antimicrobial and antibiofilm activity against S. epidermidis , and our results suggest that pseudonajide acts on the cell wall and membrane compounds of that bacteria, quite quickly and at low doses.


Sequence alignment and peptide selection
The sequences of seven snake venom peptides having related antimicrobial activities (2) deposited in the Antimicrobial Peptide Database (APD) website (http://aps.unmc.edu/AP) were aligned using Clustal X software (24). After alignment analysis of the common sequences of these seven peptides, 17 small peptides were synthesized (Fig. 1).

Peptides 1, 2, and 3 have antibiofilm activity in epidermidis
The first aim of this work was to perform a screening for antibiofilm activity in 17 small peptides derived from snake venom. For that, we chose two different species of bacteria, one Gram-negative (Pseudomonas aeruginosa PAO1), and one Gram-positive (S. epidermidis ATCCC 35984). This selection was based on their biofilm production capabilities, and both strains are known to be good models for the study of biofilm formation and structures (9,25). For screening, a crystal violet stain protocol was used with or without different concentrations of peptides. No effects were detected on biofilm formation in P. aeruginosa (Fig. S2). On the other hand, peptides 1, 2, and 3 demonstrated strong activity on the S. epidermidis biofilm. After 24 hours of exposition to different concentrations, there was a considerable reduction in biofilm mass ( Fig. 2A). At a concentration of 100 µM, the biofilm mass was reduced by 77%, 95%, and 78% for peptides 1 to 3, respectively ( Fig. 2A).
Peptide 2 demonstrated greater antibiofilm activity than peptides 1 and 3. The considerable reduction of 63% of the biofilm mass in the presence of 25 µM of peptide 2 led us to select that particular molecule at that specific concentration for the following experiments. We named the peptide "pseudonajide" after the name of the snake it was derived from, Pseudonaja textilis. In order to test its biofilm eradication activity, we precultured S. epidermidis cells for 24 h adding pseudonajide to pre-formed biofilm and incubating for another 24 h. The final quantification of biofilm mass showed a reduction of about 30% in the presence of the molecule (Fig. 2B).

Pseudonajide has antimicrobial activity against epidermidis
We decided to test the antimicrobial activity over a shorter period of time, because no difference had been observed after 24 h. Growth and colony-forming unit (CFU) tests were performed. Cells were 5 incubated in the same conditions as for the antibiofilm tests, with or without 25 µM pseudonajide.
After 1, 2, 4, and 24 h incubation, we measured the optical density at 600 nm (OD 600 ) and assessed the CFU counts. Fig. 3 shows clearly that the molecule's presence causes a huge decrease in bacterial growth as compared to the control. The same result was seen in the CFU experiments. After 1,2,or 4 h incubation with pseudonajide, the number of viable cells vastly decreases as compared to the control conditions.

Pseudonajide binds to the cell wall and membrane, causing permeabilization
To better understand pseudonajide's binding site, we synthesized peptides tagged with fluorescein isothiocyanate (FITC), and then performed confocal microscopy. Cells were incubated with 25 µM FITC-tagged pseudonajide for 1, 4, or 24 h. After incubation, confocal microscopy showed that the molecule is located around or inside the bacterial cell, but not in the biofilm matrix (Fig. 4).
Another important finding was the reduction of fluorescent cells over time, with decreased peptidetagged cell counts after 4 and 24 h incubation.
To confirm that the interaction occurs between pseudonajide and S. epidermidis cell walls and membranes, we did LIVE/DEAD experiments. Because it was demonstrated that propidium ions can enter on cell with high membrane potential (26). Cells were cultured for 4 h with or without 25 µM pseudonajide. Confocal microscopy image analysis demonstrates an increase in cell death when in the presence of pseudonajide. Moreover, statistical analysis shows that there is a significant decrease in the number of impermeable cells when the peptide is present (Fig. 5). These data suggest that pseudonajide are interfering on cell walls and membranes integrity.

Pseudonajide damages epidermidis cell walls and membranes
To check for morphological changes in S. epidermidis cells after exposure to the peptide, microscopy experiments were then performed after 1, 4, and 24 h incubation with or without 25 µM pseudonajide.
We chose to approach this in two distinct ways, using both scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The SEM experiments were performed by culturing the cells in the same conditions as before, with plastic slides added to the culture well for cell adherence. Our most notable result was that after 4 and 24 h incubation, cell adhesion was much weaker when 6 cultured with the peptide, although no difference was observed after just 1 h (Fig. 6). Another important characteristic we noted was that several cells exposed to this molecule had a shrunken morphology and were smaller than non-exposed cells (Fig. 6, white arrows). Again, this morphology was only noted after 4 and 24 h incubation. A final point that must be highlighted is that some extravasated material is present surrounding the shrunken cells, and this can be seen in the same figure in the cells that have arrows. None of these characteristics were seen in the control. Specifically, after 4 and 24 h of peptide exposition, the cell wall is not intact, and the cell sizes are completely different than those of the control. Moreover, in the peptide-exposed cells, the material inside the cytoplasm is condensed (Fig. 7).

Pseudonajide increases the expression of genes coding for teichoic acid synthesis
The results obtained from microscopy analysis led us to hypothesize that pseudonajide acts on cell walls and membranes. Indeed, cationic peptides are known to be able to interact with the cell walls of Gram-positive bacteria (27) and to influence membrane fluidity when engaging with the phospholipid bilayer (28). One of the first molecules that is supposed to interact with cationic peptides is teichoic acid, a negatively charged molecule present in Gram-positive cell walls (29). To investigate this, realtime quantitative PCR tests were done, with S. epidermidis cultured in the same conditions as the previous experiments. However, due to pseudonajide's high antimicrobial activity, we decided to use a lower concentration. We therefore tested a series of dilutions ranging from 3 to 100 µM of the molecule at 4 h incubation. We found that a concentration of 6.25 µM is enough to inhibit about 50% of growth as compared to the control (Fig. 8A). To investigate the relative expression levels of genes when bacterial cells are cultured in subtoxic concentrations of this peptide, we selected three genes that code for teichoic acid molecules. By testing these, we were able to clearly see that cells cultured 7 in the presence of 6.25 µM peptide had higher expression levels of UgtP, LtaA, and LtaS genes (Fig.   8B). These results led us to hypothesize that pseudonajide interacts with teichoic acid in the S. epidermidis cell wall, causing a strong interaction with this structure, leading to cell permeability. The same extracted RNA was used for biofilm-related gene expression analysis. We chose nine genes related to biofilm formation for expression analysis: AtleE, agrC, aap, EmbP, icaA, leuA, saeR, saeS, and sarA. No significant differences in expression were observed under control and peptide conditions for these nine genes (Fig. 8C).
One of main challenges in the development of antimicrobial peptides is their potential toxicity to human cells (30,31). We therefore performed toxicity tests using seven human cell lines: HuH7 (hepatocellular carcinoma); Caco-2 (colorectal adenocarcinoma); MDA-MB231 (breast adenocarcinoma); HCT116 (colorectal carcinoma); PC3 (prostatic adenocarcinoma); NCL-H727 (lung carcinoma); and MCF7 (breast cancer). After 24 h incubation in a concentration of 25 µM pseudonajide, there was no decrease in the living cell counts as compared to the control conditions ( Fig. 9), demonstrating that pseudonajide is not cytotoxic to human cells.

Discussion
Antimicrobial peptides are promising molecules in the fight against bacterial resistance (32). Since AMPs can interact with a large variety of cell targets, they have an advantage in the fight against the production of bacterial resistance phenotypes (33). We demonstrate here that an 11-amino-acid peptide derived from P. textilis snake venom possesses antimicrobial activity against S. epidermidis.
Our first goal during the screening was to find new molecules with antibiofilm activity. However, when we investigated the mechanism of action of peptide 2, we found that it acts directly on the bacterial cell, and not in the biofilm matrix. This led us to investigate its antimicrobial activity, the molecule's cellular binding site, as well as the bacterial molecules which might interact with this newly identified peptide.
To analyze the effects of pseudonajide on S. epidermidis cells, we performed growth curve and CFU experiments using a concentration of the peptide of 25 µM. We began by investigating the peptide's effects in the early stages of interaction. In fact, it is possible to detect a great difference in the CFU 8 counts after just 1, 2, and 4 h incubation, which is characteristic of a fast-action antibiotic. Moreover, biofilm eradication activity was detected (Fig. 2D), with around 30% lower biofilm mass as compared to the control conditions, possibly due to the ability of pseudonajide to kill the biofilm-forming bacteria. This shows that the peptide has a dual action, both antimicrobial and against biofilm formation. To discover the binding sites of pseudonajide, we produced an FITC-tagged molecule. After 1, 2, and 24 h interaction with bacterial cells, confocal microscopy demonstrated that pseudonajide interacts with the S. epidermidis cell envelope (Fig. 4). We can therefore conclude that the first bacterial cell interactions are with the cell envelope, and not with the biofilm matrix.
Based on their activities, AMPs can be divided into two main groups: they can act on the cell wall and disrupt the membrane, causing cell permeability; or they can have intracellular targets (34,35). Even though cationic peptides can have different amino acid sequences, they still have similar characteristics which permit interaction with bacterial cell membranes. As described on the literature, most of the residues in AMPs are positively charged and some are hydrophobic, ensuring the AMPs amphipathic character (27,36). In this work, structural analysis demonstrated that more than 50% of the amino acids which make up the peptide pseudonajide are positively charged (KRFKKFFMKLK). The position of methionine (M) seems to increase the antimicrobial/antibiofilm activity, and such a residue was not seen in peptides 1 (KRFKKFFKKVK) or 3 (KRFKKFFKKLK). The original peptide sequence that we based to synthesize pseudonajide was reported by Falcão's group, and belong to the vipericidins, a family of cathelicidin-related peptides derived from the venom glands of South American pit vipers.
They described these vipericidins as having antimicrobial activities against different bacteria, including S. aureus and P. aeruginosa strains (37). Their activity is probably similar to the interactions we saw between pseudonajide and the S. epidermidis cell walls and membranes. AMPs bind preferentially to the cationic bacterial membrane instead of the zwitterionic membrane in mammalian cells (36). Moreover, pseudonajide contains 36% hydrophobic amino acids, a characteristic which may explain its interactions with the bacterial cell membrane. Insertion of the peptide into the hydrophobic portion of the membrane seems to cause osmotic imbalance in the cell, which could lead to the shrunken cell morphology observed in the SEM (Fig. 6) and TEM (Fig. 7) analysis. It is important to 9 note that in TEM, the defective cells have external material surrounding them. We hypothesize that this consists of extravasated DNA and disorganized peptidoglycan, but more tests are necessary to prove it. We also surmise that the smaller cells that can be observed are the same as those seen on FITC-tagged peptides with confocal microscopy. The green fluorescent cell sizes were all smaller than those of the non-fluorescent cells. In summary, pseudonajide acts on the cell envelope, inducing an osmotic imbalance which in turn causes a reduction in cell size, leading to cell death (Fig. 5).
The cell wall in Gram-positive bacteria is a complex network of molecules in a structure composed mainly of peptidoglycan and teichoic acids. Teichoic acids are negatively charged polyglycerophosphate chains that can be linked to peptidoglycan or anchored to the cytoplasmic membrane (38,39). Moreover, D-alanylation of lipoteichoic acid is said to promote protection against cationic AMPs in Gram-positive bacteria (40). In order to test this, we assessed the expression levels of genes coding for LTA assembly molecules, namely glycosyltransferase YgfP (UgtP), flippase LtaA, and lipoteichoic acid synthase LtaS (41). In S. aureus, lipoteichoic acid synthesis starts with YgfP, encoded by the ugtP gene. This protein synthesizes the glycolipid anchor Glc2-DAG from UDP-Glc and diacylglycerol (DAG). Glc2-DAG is translocated to the outside of the membrane by LtaA (41,42), and elongation of the LTA chain is then promoted by LtaS (41,43). Based on the literature and due to the physicochemical characteristics of teichoic acids, we speculated that pseudonajide must act on teichoic acids in the S. epidermidis cell wall. We detected increases in the expressions of all three tested genes when the cells were cultured in the presence of pseudonajide (Fig. 8B). These results reinforce our theory that pseudonajide binds to S. epidermidis lipoteichoic acids, probably causing cell wall disorganization in these bacteria (Fig. 10). The increased expression may well be a compensatory mechanism to protect against the presence of the peptide or even to preserve cell viability.
It was previously suggested that cationic antimicrobial peptides kill bacterial cells. They first interact with the membrane through electrostatic interactions (44), contacts which result in membrane disruption and cell death. Other peptides can cross the bacterial lipid bilayer without causing any damage to the cell membrane, but still inhibit intracellular functions, so they also eventually lead to bacteria death. Pseudonajide contains an amino acid sequence (KRFKKFFMKLK) that is part of a peptide isolated from P. textilis venom. Of the peptides we tested, pseudonajide has the best antibiofilm formation activity, at a sub-MIC concentration of just 6.25 uM (Fig. 2B), and the best eradication of established biofilm activity in the group (Fig. 2D). Several AMPs have been described as also having antimicrobial activity against Gram-negative bacteria. In the present work, we did not observe any antibiofilm or antimicrobial activity against P. aeruginosa (Fig. S2). It is possible that the short peptides tested here suffer from P. aeruginosa protease degradation (45,46), as small peptides are typically more susceptible to proteases. The inhibition of LL-37's bactericidal activity by alginate and exopolysaccharides is another example of antimicrobial peptide protection reported in this same pathogen. The inhibition occurs through LL-37 sequestration, which diminishes AMP concentrations at the target site (47).
We have observed pseudonajide's dual activity, as it is both antimicrobial and also inhibiting S. epidermidis biofilm formation. Even though we did not see any alteration in the expression of biofilmrelated genes when the peptide was present, we did observe biofilm eradication with reduction in mass (Fig. 2B). This decrease can be explained by several elements. One is the relationship between the cell wall teichoic and lipoteichoic acids and the processes of adhesion and biofilm formation (48).
Moreover, these types of molecules have been detected in the biofilm matrix of S. epidermidis (50). If pseudonajide mainly acts on teichoic and lipoteichoic acids, the reduction in adhesion could be one of the causes of both biofilm reduction and outright eradication. It is also important to emphasize the characteristics of cationic antibiofilm peptides, described by Von Borowski et al. (51). They showed that lysine (K) and phenylalanine (F) are the most frequently found amino acids in antibiofilm peptides, and this is clearly also the case for pseudonajide (KRFKKFFMKLK).
We have showcased here the promising activity of a synthetic peptide derived from P. textilis venom.
Its dual action against S. epidermidis cells and its biofilm make pseudonajide a very promising molecule for new drug development, and this is reinforced by the fact that it has a short sequence.
Shorter sequences are advantageous both for industry and antimicrobial peptide researchers, as they are easier to synthesize and cost less. Importantly, this facilitates future research into their structures 11 and into ways to improve their efficiency.

Bacterial strains and culture conditions
S. epidermidis ATCC 35984 was used to test the antimicrobial and antibiofilm activities of the peptides. Bacteria were grown in blood agar plates and cultured overnight at 37° C. Cell suspensions were prepared in a solution of 0.9% NaCl and adjusted to OD 600 for a final concentration of ~1 0 8 cells/ml. For microscopy analysis, pre-inoculum was made in tryptic soy broth (TSB, Merck), and adjusted to OD 600 for the same concentration of cells for all tests.

Tests on antimicrobial and antibiofilm activity
Serial dilution of peptides was performed in 96-well plates, going from 100 to 3.12 µM. Cell suspensions and TSB were added to the plates and a control was made with dimethyl sulfoxide (DMSO). Antibiofilm formation tests were performed with an adapted protocol (52), which it is described that 24 h of incubation is sufficient to determinate antibiofilm activity. The OD 600 was measured, then biofilm content accessed using the crystal violet protocol (53). Biofilm eradication test was performed supplementing 24 h pre-formed biofilm with a new culture broth, containing or not 25 µM of peptide. The plates were incubated for more 24 h followed by crystal violet protocol. The antimicrobial activity of pseudonajide was analyzed using a concentration of 25 µM after 1, 2, 4, and 24 h incubation. After measuring the optical density, the supernatant was collected and diluted. A volume of 100 µl was plated in Luria broth agar plates. CFUs were counted after 24 h incubation. All experiments were performed at least three different times, each with three technical replicates.

Scanning electron microscopy
S. epidermidis was cultured in the same conditions as described previously, in the presence or absence of 25 µM pseudonajide. However, for this analysis, a plastic slide was placed inside each well, and these plates were incubated for 1, 4, and 24 h. After incubation, the plastic slides were washed three times with 0.9% NaCl solution and fixed overnight in fixation buffer (2.5% glutaraldehyde, 2% paraformaldehyde, 0.1 M sodium cacodylate) at 4° C. The adhered cells were then dehydrated with increasing concentrations of ethanol solutions. The images were obtained using a JSM-7100F scanning electron microscope (JEOL).

Transmission electron microscopy
Bacterial cells were cultured in 24-well plates in the presence or absence of 25 µM of pseudonajide.
Cells were incubated for 1, 4, and 24 h. All of the content in the well was recovered, centrifuged at 12,000 xg, then washed with saline solution. Fixation was performed for 18 h at 4° C with a buffer

Confocal microscopy
Bacterial cultures were done in the same conditions as described above, but they were incubated with pseudonajide tagged with FITC. After incubation, cells were washed with saline solution then 3 µL was added to glass slides for confocal analysis. The images were acquired using a Leica SP8 DMI 6000 CS confocal microscope, and ImageJ software was used for image analysis.  Table below. (see Table 1 in the Supplementary Files)
This featured high-throughput multiparameter image analysis, with both high-content screening and high-content analysis. The platform is equipped with an Olympus microscope and Compix SimplePCI software; a Zeiss Axio Imager M1 microscope with a Zeiss camera and AxioVision software; and imaging systems including an ArrayScan VTI Cellomics reader (Thermo Fisher), Hamilton STARlet and NIMBUS workstations, and a Scienion spotter. Cells used in the test were obtained from an alreadyexisting collection available at BIOSIT (https://biosit.univ-rennes1.fr/impaccell-imagerie-pour-analysedu-contenu-cellulaire). For the tests, seven different cell lines were used: human hepatocellular carcinoma (HuH7); colorectal adenocarcinoma (Caco-2); breast adenocarcinoma (MDA-MB231); colorectal carcinoma (HCT116); prostatic adenocarcinoma (PC3); lung carcinoma (NCL-H727); and breast cancer (MCF7). The residual cell percentages reported correspond to viable cells compared to the average viable cells in the DMSO control. Viability of 100% represents no cytotoxicity or inhibition of cell growth, while under 25-30% is considered cytotoxic and 0% represents acute cytotoxicity.

Consent for publication
Not applicable.

Availability of data and material
The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Competing interests
The authors declare no competing of interests.  Tables   Due to technical limitations Table 1

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download.