2.1 Materials

2-Chlorotrityl chloride resin, N,N,N′,N′-Tetramethyl-O-(1H-benzotriazol-1-yl) uronium hexafluorophosphate (HBTU), Fluorenylmethyloxycarbonyl chloride (Fmoc) amino acids, Triisopropylsilane (TIS), and Hydroxybenzotriazole (HOBt) were purchased from GL Biochem Ltd., Shanghai, China. N,N’−Dimethylformamide (DMF) Dichloromethane (DCM), Diisopropylethylamine (DIPEA), Acetonitrile, and Dimethyl sulfoxide (DMSO) were purchased from Radchem (PTY) LTD., Alberton, South Africa. Palmitic acid (Pal), 1-Adamantanecaboxylic acid (Ada), Trifluoroacetic acid (TFA), Thionyl chloride, Formic acid, Iodine, Dulbecco's modified Eagle Medium (DMEM), phosphate, Trypsin, Penicillin-Streptomycin, Fetal bovine serum (FBS), and trypan blue stain dye were purchased from Sigma Aldrich (St Louis, MO, USA). Human embryonic kidney (HEK)-293 and Colorectal adenocarcinoma (Caco-2) were purchased from Cellonex (Randburg, Gauteng Province, South Africa).

2.2 Solid phase synthesis of ADEP1 analogues

The ADEP1 analogues were synthesized via a modified solid phase peptide synthesis (SPPS) using Fmoc-amino acids, following a previously reported method as depicted in Scheme 1 (Merrifield, 1963). The 2-chlorotrityl chloride resin was used as a solid structure and activated for 2 h by 20% thionyl chloride in distilled (dry) DCM solution. The first amino acids in all ADEP1 analogues were loaded at four equivalence to the activated resin with five equivalence of the DIPEA. The concentration of reagents used during coupling cycles of the peptides were 0.5 M DIPEA, 0.19 M HBTU, 0.095 M HOBt and 0.2 M amino acids. Each amino acid was double-coupled while Pal or Ada were triple-coupled. The peptide sequence, stereochemistry and net charge are summarized in Table 1.

Close

Scheme 1

Schematic representation of a solid phase peptide synthesis of ADEP1 analogues.

Export to PPT

2.3 Cyclization of ADEP1 analogues

2.4 ADEP1 analogues purification by HPLC

The peptides were purified using Preparative High-pressure liquid chromatography (prep-HPLC) (Shimadzu, Kyoto, Japan) with a prep-HPLC c-18 column with 30 × 250 mm, 10 µm dimensions (Phenomenex, California, United States) with UV detection at 215 and 245 nm. The purification method was a 20 min running time with 20 mL/min flow rate. The gradient elution was 5–95% B (0.1% FA in acetonitrile) in A (0.1% FA in water) for 12 min and 95% B in A from 12 to 20 min. Molecular weight of the peptides were confirmed using Thermo Scientific Ultimate 3000 (Bruker, Massachusetts, United States). The HR-LCMS also was also performed on a c-18 column (5 µm, 100 Å, 4.60 mm × 150 mm) with a two-buffer solvent system, solvent A and B at a flow rate of 0.3 mL/min. The expected peptide molecular weight confirmed by HR-LCMS eluted between 13 and 14 min on the prep-HPLC. The corresponding prep-HPLC peaks were therefore collected and solvent removed using rotary evaporator at 60-65 ℃. The residuals were lyophilized using a freeze dryer for further biochemical analysis.

2.5 Antibacterial activity tests of the ADEP1 analogues

The antibacterial activity of the ADEP1 analogues was tested against three Gram-positive (Bacillus subtilis ATCC 11774, Staphylococcus aureus ATCC 25923 and Methicillin-resistant Staphylococcus aureus (MRSA) ATCC 43300) and two Gram-negative (Escherichia coli ATCC 33876 and Pseudomonas aeruginosa ATCC 15442) bacterial strains using agar well diffusion and microdilution assays according to previously reported protocols (Adeyemi et al., 2018). The bacterial strains were purchased from the American Type Culture Collection (Manassas, Virginia, USA).

2.6 Cell viability studies

The cytotoxicity of the ADEP1 analogues was investigated on HEK-293 and Caco-2 cell lines using the MTS cell proliferation assay. The cells were purchased from Cellonex (South Africa) and cultured in DMEM containing 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin at 37 °C in a 5% CO₂ humidified incubator. For the cell viability assay, the cells were seeded at a density of 1x10⁵ cells/mL in 96 well plates (100 µL/well). The plates were incubated at 37 °C overnight, after which the ADEP1 analogues at 15.6 µM to 2 mM were added. The plates were incubated at 37 °C for an additional 48 h. Auranofin (0.78125 - 100 µM) was used as a positive control and 2% DMSO was used as a negative control. The MTS dye reagent (10%) was added to the wells and incubated for 4 h. The absorbance was read at 490 nm wavelength using SpectraMax Plus 384 Microplate Reader. The percentage cell viability was calculated using the equation below. $$\%Cellviability=\left(\frac{\mathit{OD}\mathit{of}\mathit{test}\mathit{sample}}{\mathit{OD}\mathit{of}\mathit{control}}\right)x100\%$$

2.7 Statistical analysis

Differences between groups were assessed by one-way analysis of variance (ANOVA) followed by two-tailed t-test analysis of variance using the GraphPad Prism 9.510. Results with p < 0.05 were considered statistically significant and were expressed as the mean ± SEM of n = 6.

3.1 Synthesis and characterization of ADEP1 analogues

ADEPs have potential to overcome microbial resistance to antimicrobial agents as they follow a different mode of action, however, their bioactivity is limited by metabolic instability and side effects, among others. There were several strategies that were previously used to overcome these challenges. In the current study, the modifications introduced to the ADEP1 structure (Scheme 2A) were motivated by: 1. The need to produce low cost drug candidates in high yield by developing a shorter synthetic route and use of natural peptides. Therefore, 4-methyl proline and N-methylated alanine were substituted with natural amino acids (Pro2 and Ala4) (Scheme 1B and C). 2. To improve lipophilicity and biocompatability by replacing OTEA with Ada in SC006 and SC007 (Scheme 2B) or Pal in SC005 and SC008 (Scheme 2C). 3. Incorporation of D-amino acids to improve metabolic stability and prevent biodegradation. SC005 and SC006 contained D-amino acids while SC007 and SC008 were made of L-amino acids.

Close

Scheme 2

ADEP1 analogues in relation to ADEP1 parent structure (A). The ADEP analogues were modified with either Ada (SC006 and SC007) (B) or Pal (SC005 and SC008) (C).

Export to PPT

The structure of ADEP1 is cyclized by an ester bond between the carboxylic end of Pro1 and the hydroxyl functional side chain of Ser5. The ester bond linkage was replaced by a disulphide bond by adding Cys1 and substituting Ser5 with Cys6. Substitution of amino acid residues with Cys produce disulphide bridges and increase stability of the peptides. For example, mutations in subtilisin E that resulted in replacement of Gly61 and Ser98 with Cys residues increased its activity, thermal stability and shelf-life (Liu et al., 2016; Takagi et al., 1990; Zabetakis et al., 2014).

The synthesis of ADEP analogues via SPPS generally pose difficulties in cyclization with the ester bond (Wen et al., 2008; White & Yudin, 2011). The natural ADEPs synthesis in biological systems easily form the ester bond linkage through bacterial depsipeptide synthases (Alonzo & Schmeing, 2020). However, chemical synthesis make use of esterification techniques, which result in poor (∼ 6%) yields (Eyermann et al., 2020). Although fragment convergence synthesis have better yields, this synthesis route does not follow the normal peptide synthesis. It uses different protecting groups, prone to intermolecular reaction, and it also comes with numerous coupling and purification steps (Goodreid et al., 2014, 2015; Li et al., 2017). This synthesis approach disassembles the ADEPs into three fragments namely tripeptide, linkage, and Phe-heptenamide (Li et al., 2017). The incorporation of disulphide cyclization allows for a simple SPPS using one protecting group (Fmoc), with single cleavage and purification steps.

HOBt was used as a coupling additive to suppress racemization. It is also a nucleophilic catalyst, which helps accelerate the rate of the coupling reaction time (Al-Warhi et al., 2012). HOBt considerably reduced coupling reaction times from 1 h to 45 min while reducing the presence of impurities. The expected molecular mass (MM) of the analogues was confirmed by MS. Figs. A1, A3, A5, and A7 shows MM [M + 1] of the peptides before cyclization. After cyclization with iodine or 20% DMSO, the MM observed on the MS (Figs. A4, A6, and A8) suggested the loss of the two hydrogen molecules during the formation of the disulphide bond. This confirmed the synthesis of cyclic analogues.

With the analysis on both prep-HPLC and LCMS, the peptides presented a higher retention time (Figs. A9-A12). This was attributed to the hydrophobic and nonpolar nature of the peptides. Out of seven amino acids in their sequence, five of those (Pro, Ala, and Phe) possess nonpolar and hydrophobic side chains. In addition, the formation of the disulphide bond reduced the polarity and hydrophilicity of the Cys residues. The peptides containing Pal were analysed with a longer LC method (16 min) compared to the Ada-containing peptides (14 min). The LCMS chromatogram and spectra of the four ADEP1 analogues are presented in Table 2. The analogues presented high purity (≥96%) with improved yields.

The anionic nature of all the analogues is due to the ionisable -COOH of Cys at neutral pH range to give -COO^–. This was confirmed by their negative zeta potential values and the net charges of the ADEP1 analogues reported in Table 2. As a consequence of using the Cys1 and Cys6 thionyl groups to cyclize the peptides, the -COO^– of Cys1 was exposed and therefore giving the analogues a negative net charge. The addition of the two Cys also enlarged the ring, which can allow for slight conformational freedom.

3.2 Antibacterial activity of the ADEP analogues

The potential antibacterial activity of the ADEP1 analogues was screened against the benign (B. subtilis and E. coli) and the pathogenic (P. aeruginosa, S. aureus and MRSA) strains. Table 3 shows that the Pal-based ADEP1 analogues had higher activity compared to Ada-based ADEP1 analogues, in the order SC008 > 5 > 6 > 7. The Pal-based ADEP1 analogues were active against all the tested bacterial strains with SC008 showing more potency compared to its D-enantiomer, SC005. The Ada-based ADEP1 analogues showed selective activity, SC006 did not show any activity on P. aeruginosa while SC007 showed no activity on S. aureus and E.coli. Pal is a long chain of hydrocarbons (saturated fatty acid) that better mimics the OTEA than the Ada. Pal has been used to increase lipophilicity of drug-like compounds (Carta et al., 2017; Irby et al., 2017), and its conjugation to ADEP1 analogues increased their lipophilicity, which enhanced their interaction with the bacterial cell and increased cellular uptake. Both Ada and Pal are lipophilic molecules and incorporated in the ADEP1 analogues to increase their cell permeation ability. Fatty acids (linoleic, oleic acid, Pal, and Ada) are exploited in pharmaceuticals to serve as chemical cell permeability enhancers (Cizinauskas et al., 2017). They are widely used as drug delivery systems and targeted therapies in biomedical applications (Spilovska et al., 2016; Schleyer, 1957; Stimac et al., 2017). However, agar diffusion methods have been reported to lack sensitivity for non-polar molecules (Eloff, 2019) and highly dependent on their ability to diffuse in the agar. The microdilution assay was thus used to further evaluate the activity of the peptides against all the bacterial strains, their MIC and MBC values are presented in Table 4.

All the ADEP1 analogues showed a broad-spectrum antibacterial activity on the selected bacterial species at concentration range of 125 – 500 µM (MIC) and 125 – 750 µM (MBC). The mode of action of these ADEP1 analogues has not been determined yet. However, previous reports have highlighted the specific mechanism of ADEP by introducing ClpP mutations in B. subtilis, S. aureus, and E. coli. The bacterial strains with ClpP mutation showed high resistance to ADEP treatment (Brotz-Oesterhelt & Sass, 2014; Malik et al., 2020). This suggested that it is unlikely for the ADEPs to target any bacterial component than ClpP, thus making their mode of action target specific. The ADEPs binds to the ClpP, which then degrade proteins necessary for whole cell metabolism and survival (Frees et al., 2014; Sass et al., 2011). Therefore, the ADEP1 analogues might follow mode of action similar to the one for ADEPs due to their structural similarities.

All the ADEP1 analogues were able to inhibit growth of the Gram-positive (MIC = 63–125 µM) and Gram-negative (MIC = 250 µM) bacterial strains. The Gram-negative bacteria showed less susceptibility towards the ADEP1 analogues. E. coli presented the highest resistance with an MIC of 500 µM for all the analogues while P. aeruginosa was inhibited by an MIC of 250 µM for SC0006 – 8 and more susceptible to SC005 (MIC = 125 µM). This is in line with what has been reported about ADEPs or anionic AMPs which have higher potency on Gram-positive than Gram-negative bacteria (Culp & Wright, 2017). Overall, the peptides showed similar antibacterial trends despite different antimicrobial activities previously discussed between the D- and L-enantiomers. In literature, D-enantiomer peptides possess higher antimicrobial activity compared to the L-enantiomer (Ye & Aparicio, 2022), this might be due to their (D-enantiomer) proteolytic stability leading to increased bioavailability in vivo. Natural peptides, ADEPs included, consist of L-amino acids, which are susceptible to enzymatic degradation (Vlieghe et al., 2010), while peptides composed of D-amino acids may escape protease degradation thereby increasing their bioavailability (Garton et al., 2018; Uppalapati et al., 2016). Similarly, disulphide bonds also play a major role in peptide/protein structure stability by increasing resistance to proteolysis while also increasing activity (Siddiqui et al., 2005; Trivedi et al., 2009). All four ADEP1 analogues were cyclized with a disulphide bond to replace the ester bond linkage that has been reported to be susceptible to enzymatic degradation (Finer & Santerre, 2004; Yourtee et al., 2001).

The naturally occurring ester bond has not yet been reported to play a vital role in molecular conformation, interaction with ClpP, and biological activity of the ADEPs. However, a study by Li and colleagues (Li et al., 2017) made strides in comparing the effect of using different linkage motifs on the conformational behavior and bioactivity of the ADEP analogues. Their study showed a significant decline in in vitro biochemical activity when the ester bond was substituted with amide bond or N-methyl amide bond (Li et al., 2017). The same phenomenon was observed in this study, where the MIC values for ADEP1 analogues were multiple folds higher than that reported for ADEP1. Another study presented challenges of incorporating 4-methylated Pro and N-methyl-Ala during synthesis of ADEPs, and removing the methyl groups in the amino acids resulted in complete loss of antibacterial activity (Hinzen et al., 2006). This then highlighted the importance of methyl amino acids in the antimicrobial potency of ADEPs.

Although the ADEP1 analogues had antibacterial activity in the micromolar range, they showed great bactericidal effects. Bactericidal effects (MBC) of potential drugs are vital as they represent a concentration that completely eradicates bacteria rather than inhibiting their growth. For example, ADEPs has shown inhibition of cell division in B. subtilis, however, bacterial cells recover upon incubation on a fresh media (Mayer et al., 2019). This shows that even though the MICs of most ADEPs are low, they lack the bactericidal effect. In some cases, their bactericidal effect is multiple folds higher than the MIC (Sass et al., 2011). In this study, all four tested analogues had the MIC that was at the same concentration as the MBC against B. subtilis. Usually, the MBC of antimicrobial drugs is greater or equals to two fold of the MIC (Levison & Levison, 2009), as observed in the other three bacterial strains where MBC was twice the concentration of MIC.

3.3 Cytotoxicity

It is imperative that antimicrobial agents are selective in their actions to prevent or reduce off-target effects. However, the biocompatibility of natural ADEPs is challenged by their non-selectivity and toxicity to mammalian cells. Similar to microorganisms, human cells also express ClpP (hClpP) which is structurally similar to its bacterial counterparts and might respond the same way to the ADEPs. Human proteostasis is also coordinated by a network of chaperones and proteases including hClpP (Kang et al., 2005). The activity of hClpP is regulated by ClpX, an ATP-dependent protease, and dysregulated by ADEPs (Cole et al., 2015). Thus, the effects of ADEP1 analogues were evaluated on HEK-293 and Caco-2 cells at ≤ 2 mM using MTS assay. There was no significant reduction in cell viability of HEK-293 (Fig. 1A) and Caco-2 (Fig. 1B) cells in response to increasing concentrations of ADEP1 analogues. Fig. 1C shows that Auranofin, a food and drug administration (FDA) approved antirheumatic drug (Boullosa et al., 2021), was cytotoxic to all the cells at ≥ 0.125 mM.

Close

Fig. 1

Effect of ADEP analogues on the viability of HEK-293 (A) Caco-2 (B) cells, and Auranofin treatment on HEK-293 and Cac-2 (C). Cell viability was assessed by MTS assay, where p < 0.05 was considered statistically significant. *=p < 0.05, **=p < 0.01, and***=p < 0.001.

Export to PPT

The mode of action involved in the dysregulation of hClpP and the physiological consequences have not been thoroughly investigated. However, in 2018, Wong and colleagues reported that the dysregulation of hClpP by ADEPs induced cytotoxicity to various human cell types including HEK-293 cells (Wong et al., 2018). The concentration dependent toxicity effect of ADEPs on hClpP was demonstrated in vitro and in vivo using different ADEPs and ADEP analogues (Wong et al., 2018). Toxicity has been the major hindrance in the development of ADEPs as active pharmaceutical ingredient. The ADEP1 analogues under study showed no significant toxicity at a dose multiple folds higher (2 mM) than their MICs. The ADEP1 analogues in this study showed an essential improvement in their biocompatibility compared to the cytotoxicity of ADEP analogues in literature. Generally, ADEPs start showing toxicity at concentrations from ≥ 0.215 µM (Wong et al., 2018). The ADEP1 analogues showed insignificant activity up to 2 mM, and implied that these synthetic peptides are not cytotoxic at their MICs and therefore has potential for clinical applications.