Synthesis of Temporin-SHa Retro Analogs with Lysine Addition / Substitution and Antibiotic Conjugation to Enhance Antibacterial, Antifungal, and Anticancer Activities

1. Introduction

The rise of resistant pathogens has drawn attention to antimicrobial peptides (AMPs), which possess broad-spectrum efficacy. AMPs operate through unique mechanism of action that makes it harder for pathogens to develop resistance [1]. Despite this potential, many challenges like instability, protease susceptibility, hemolysis, and toxicity have limited their clinical use. Researchers have explored several ingenious strategies to address these issues. For example, one early method involved swapping the natural L-amino acids with unnatural D-amino acids to makes these peptides invisible to proteases, which typically recognize and cleave natural amino acid chains. This seemingly minor tweak enhances the stability of AMPs in the bloodstream [2]. Another effective approach that has shown promise is peptide cyclization, which not only lowers the toxicity but also improves antimicrobial activity [3]. Researchers have also tried attaching polyethylene glycol (PEG) to peptides (PEGylation), wherein PEG acts as a hydrophilic shield to improve peptide solubility in aqueous environment. PEGylation was found to enhance the renal clearance of peptides, thereby preventing the unwanted build-up to lower the potential side effects of AMPs [4,5]. Other modifications like glycosylation (attaching sugar molecules) or lipidation (conjugating fatty acids) have extended the functional activities of AMPs [6,7]. Additionally, flipping the order of amino acids to make retro analogs, and reversing the handedness of amino acids to create inverso-analogs, has shown to enhance the potency of AMPs [8]. Taking these two concepts a step further, retro-inverso peptides were also developed, which are basically mirror images of the original molecule with reversed amino acid sequence. This double twist alters the stereochemical and conformational properties of the peptides, potentially leading to new AMPs with beneficial properties [9,10]. All these innovative strategies by researchers offer a glimpse of hope in improving the activities of AMPs, which can pave the way to develop effective antimicrobial therapies. [11].

Conjugating heterocycles with small peptides is also known to improve the biological properties. Since year 2000, numerous heterocyclic-conjugated peptides with potential therapeutic benefits have been reported [12,13,14,15,16]. Likewise, arene-peptide conjugates were also developed with promising medicinal properties [17,18,19,20]. Antibiotic drugs like levofloxacin gained our attention due to broad-spectrum antimicrobial effect. Recently, its conjugation with antimicrobial peptide named indolicidin was reported by Ghaffar et al. [21]. Levofloxacin-M33 peptide conjugate did not show any major change in its ability to fight Gram-negative bacteria [22]. Interestingly, when a linear peptide (R4W4) was conjugated with levofloxacin, an improvement of antibacterial activity with low hemolytic effect was observed. [23,24]. Early research looked into retro-analogs of P53 peptides using molecular dynamics simulations and circular dichroism spectroscopy [25,26]. Temporin-SHa is a natural AMP isolated from frog skin. Recently, we used the Fmoc peptide synthesis strategy to prepare its analogs, wherein natural glycine-10 residue of the parent temporin-SHa molecule was replaced with atypical (2-naphthyl)-D-alanine, D-tyrosine and D-phenylalanine to study the effects of hydrophobic residues on antimicrobial efficacy [27]. Previously, we also synthesized [G4a]-SHa, [G7a]-SHa, and [G10a]-SHa analogs by substituting glycine residue with D-alanine at position 4, 7, and 10, respectively [28]. Among these analogs, NST-2, corresponding to [G4a]-SHa peptide, demonstrated promising activity against methicillin-resistant Staphylococcus aureus (NCTC 13277) with an MIC of 14.3 μM. It also showed low hemolysis as compared to parent temporin-SHa peptide [29]. These results make this analog a suitable choice for further investigation [30].

The peptide analog NST-2 is D-alanine modified variant of temporin-SHa [28]. In this work, we decided to synthesize NST-2 in a retro manner potentially increasing its activity [31]. Then, we further modified retro analog of NST-2 by either adding lysine residue to its C terminus or by substitution of glycine at the 4th position of this retro analog with lysine. Finally, to potentially further enhance the therapeutic potential of the peptide, we conjugated the antibiotic levofloxacin to either the side chain of lysine-3 or lysine-14 of the retro analog or to the N-terminal phenylalanine-1 via a β-alanine linker. A summary of these modifications in retro analog of NST-2 is provided in Scheme 1.

2. Materials and Methods

2.1. Reagents and Instruments

The reagents used in our experiments were 95–98% pure. Novabiochem (Hohenbrunn, Germany) and Chem-impex (Wood Dale, IL, USA) provided the Fmoc-protected amino acids, Rink amide resin, and coupling reagents. Fmoc protected Rink amide resin had a loading capacity of 0.602 mmol/g and mesh size of 200–300. HPLC grade solvents were employed. All peptides were purified with preparative HPLC (PuriFlash^®) by using PFB15C18XS-250/212 column. Purity of peptides was checked via UPLC (Agilent 1260 Infinity Diode Array, C-4 reverse-phase analytical column, 5 μm, and 150 x 4.6 mm). Electrospray ionization mass spectrometer (Q-STAR XL, Applied Biosystems, USA) with quadrupole time-of-flight analyzer (ESI-QTOF-MS) was used for molecular mass determination. NMR spectrometer of 600 MHz (Bruker, Switzerland) was used to record 1D and 2D NMR spectra. Carbon spectra were obtained by adjusting the frequency at 125 MHz.

2.2. Peptides Synthesis and Characterization

2.2.1. Synthesis of Temporin-SHa Retro-Analogs and Their Levofloxacin Conjugates

Peptides analogs and their levofloxacin conjugates were manually synthesized by using Fmoc solid phase method on Rink amide resin as shown in Scheme 1. The resin (1 g, 0.6 mmol/g) was soaked in DMF for 2 hours, followed by its treatment with 4-methylpiperidine (20%) to remove the protecting group. Fmoc-Phe-OH (6 equiv.) was then loaded on the resin with the help of 6 equivalents of N,N′-diisopropylcarbodiimide (DIC) and 2-cyano-2-(hydroxyimino) acetate (Oxyma pure) as additive to suppress racemization. After the reaction, 4-methylpiperidine in DMF (20%) was added to remove the protecting group and the next amino acid was coupled. Overall, the successive coupling was done by employing 3 equivalents each of Fmoc-protected amino acids, coupling reagent and additive. Levofloxacin was coupled with peptide sequences at different position after the removal of Alloc and Fmoc protecting groups. The peptides were then cleaved from the resin with TFA cocktail (94% TFA, 1% triisopropylsilane, 2.5% ethanedithiol, 2.5% water). The crude peptide was precipitated with diethyl ether and lyophilized.

2.2.2. Mass and NMR Spectroscopic Analysis of Peptides

For the determination of mass of peptides, electrospray ionization with quadrupole time-of-flight mass spectrometery (ESI-QIT-MS) technique was used. All spectra were recorded on Amazon Speed mass spectrometer (Bruker Daltonics). 4500 V electrospray voltage was applied at the spraying needle with 200 ᵒC. Sample was injected with gas flow at 15 psi. All the peptides were also characterized with ¹H-NMR and ¹³C-NMR spectroscopy. The 600 MHz and 800 MHz NMR spectrometer (Bruker) equipped with an Avance III HD console (5 RF channels, 2 receivers), and with a TCI (¹H/¹³C/¹⁵N/³¹P/2H) CryoProbe which is a H-optimized triple resonance NMR ‘inverse’ probe, and also a RT probe.

2.2.2.1. Synthesis of NST-2 Peptide (1)

The peptide 1 was synthesized by solid phase peptide synthesis (SPPS) method as described earlier [32]. Overall yield: 68.6%; ${\left[\alpha \right]}_{\mathrm{D}}^{24}$= -40 (c 0.22, 60% ACN, 40% H₂O, 0.082% TFA). Purity of the peptide was confirmed with UPLC (Figure S1).

2.2.2.2. Synthesis of RNST-2 Peptide (2)

Reversing the amino acid sequence of peptide 1 via SPPS gave the peptide 2. Overall yield: 24.63%; ${\left[\alpha \right]}_{\mathrm{D}}^{24}$= -202 (c 0.001, MeOH). UV-Vis. (MeOH) λ_max (log ε): 228.0 (1.349) nm. IR (KBr, cm^-1): 1054.60 (C-O stretching, CN stretching, NH bending), 1456.81 (C-H bending, Aromatic C=C stretching), 1636.50 (NHC=O stretching), 2927.3 (C-H stretching), 3334.86 (OH stretching) and 3853.15 (NH stretching). ¹H NMR (d₆-DMSO, 600 MHz): δ_H 0.72–0.81 (18H, m, (CH₃)₂–Leu^{2, 5, 12}), 0.80–0.87 (12H, m, δ-CH₃–IIe⁹, γ-CH₃–IIe⁹, (CH₃)₂–Val⁸), 1.16 (3H, d, CH₃–D-Ala¹⁰), 1.29–1.55 (18H, m, CH₂–Leu^{2, 5, 12}, δ-CH–Leu^{2, 5, 12}), 1.30–2.39 (β-CH–IIe⁹, β-CH₂–Met⁶, β-CH₂–Lys³, γ- CH₂–Lys³, δ-CH₂–Lys³), 1.45–1.50 (1H, m, γ-CH–IIe⁹), 1.92 (1H, m, β-CH–Val⁸), 1.99 (1H, m, δ- CH₃–Met⁶), 1.78–1.81 (2H, m, γ- CH₂–Met⁶), 2.72 (2H, t, Δ- CH₂–Lys³), 2.85–3.15 (4H, dd, CH₂–Phe^{1, 13}), 3.60 (2H, m, CH₂–Ser¹¹), 3.77–3.80 (4H, m, CH₂–Gly^4,7), 3.94–4.12 (6H, m, α-CH–Phe¹, α-CH–Leu², α-CH–Leu⁵, α-CH–Met⁶, α-CH–Leu¹², α-CH–Phe¹³), 4.13–4.34 (5H, m, α-CH–Lys³, α-CH–Val⁸, α-CH–IIe⁹, α-CH–D-Ala¹⁰, α-CH–Ser¹¹), 5.3 (1H, bs, OH–Ser¹¹), 7.16–4.27 (10H, m, CH_Ar–phe^1,13), 7.74–7.95 (7H, m, NH–Leu², NH–Gly⁴, NH–Leu⁵, NH–Met⁶, NH–Val⁸, NH–D-Ala¹⁰, NH–Leu¹²). 7.96–8.49 (6H, m, NH–Phe¹, NH–Lys³, NH–Gly⁷, NH–IIe⁹, NH–Ser¹¹, NH–Phe¹³). ¹³C-NMR (d₆-DMSO, 150 MHz): δ ppm 11.17, 11.19, 14.78, 15.40, 18.36, 18.45, 19.36, 21.59, 21.64, 21.73, 21.77, 22.40, 23.10, 23.22, 23.30, 23.35, 24.23, 24.32, 24.80, 26.85, 29.69, 30.63, 31.46, 31.80, 34.63, 36.39, 37.46, 40.62, 40.85, 41.06, 42.07, 42.18, 48.40, 51.33, 52.25, 52.30,52.84, 54.08, 54.84, 54.95, 57.72, 58.03, 61.90, 126.56, 127.33, 128.35, 128.73, 129.26, 129.28, 129.39, 129.74, 137.96, 168.94, 170.11, 170.89, 170.94, 171.00, 171.48, 171.72, 171.92, 171.94, 171.99, 172.36, 172.44, 172.48, 173.36. HRMS (ESI) [M+Na]⁺ m/z: calculated for [C₆₈H₁₁₁N₁₅O₁₄S+Na]⁺: 1416.8053; found: 1416.8042.

2.2.2.3. Synthesis of RSP-1 Peptide (3)

The peptide 3 was synthesized by SPPS, wherein lysine was added to position-14 of retro analog 2. Overall yield: 7.6%; ${\left[\alpha \right]}_{\mathrm{D}}^{24}$= -208.00 (c 0.001, MeOH). UV-Vis. (MeOH) λ_max (log ε): 228.0 (1.349) nm. IR (KBr, cm^-1): 1143.30, 1201.2 (C-O stretching, CN stretching, NH bending), 1472.84 (C-H bending, Aromatic C=C stretching), 1635.72 (NHC=O stretching), 2981.04 (C-H stretching), 3356.37 (OH, NH stretching). ¹H NMR (d₆-DMSO, 600 MHz): δ_H 0.72–0.87 (18H, m, (CH₃)₂–Leu^{2, 5, 12}), 0.80–0.87 (12H, m, δ-CH₃–IIe⁹, γ-CH₃–IIe⁹, (CH₃)₂–Val⁸), 1.16 (3H, d, CH₃–D-Ala¹⁰), 1.29–1.55 (18H, m, CH₂–Leu^{2, 5, 12}, δ-CH–Leu^{2, 5, 12}), 1.30–2.39 (β-CH–IIe⁹, β-CH₂–Met⁶, β-CH₂–Lys^{3, 14}, γ- CH₂–Lys^{3, 14}, δ-CH₂–Lys^{3, 14}), 1.45–1.50 (1H, m, γ-CH–IIe⁹), 1.92 (1H, m, β-CH–Val⁸), 1.99 (1H, m, δ- CH₃–Met⁶), 1.78–1.81 (2H, m, γ- CH₂–Met⁶), 2.72 (2H, t, Δ- CH₂–Lys^{3, 14}), 2.85–3.15 (4H, dd, CH₂–Phe^{1, 13}), 3.60 (2H, m, CH₂–Ser¹¹), 3.77–3.80 (4H, m, CH₂–Gly^4,7), 4.04–4.20 (4H, m, α-CH–Phe¹, α-CH–Leu⁵, α-CH–Val⁸, α-CH–IIe⁹,), 4.13–4.34 (8H, m, α-CH–Leu², α-CH–Lys³, α-CH–Met⁶, α-CH–D-Ala¹⁰, α-CH–Ser¹¹, α-CH–Leu¹², α-CH–Phe¹³, α-CH–Lys¹⁴), 5.3 (1H, bs, OH–Ser¹¹), 7.16–4.27 (10H, m, CH_Ar–phe^1,13), 7.84–8.10 (8H, m, NH–Leu⁵, NH–Gly⁷, NH–Val⁸, NH–IIe⁹, NH–D-Ala¹⁰, NH–Ser¹¹, NH–Leu¹², NH–Phe¹³). 8.11–8.60 (5H, m, NH–Leu², NH–Lys³, NH–Gly⁴, NH–Met⁶, NH–Lys¹⁴). ¹³C-NMR (d₆-DMSO, 150 MHz): δ ppm 11.05, 15.25, 15.34, 18.12, 18.20, 19.21, 19.36, 21.44, 21.46, 21.51, 21.60, 21.70, 22.21, 22.34, 22.98, 23.00, 23.04, 23.08, 23.14, 23.19, 23.93, 24.03, 24.12, 24.53, 24.57, 26.60, 26.69, 27.15, 28.12, 29.52, 30.30, 31.47, 30.52, 31.40, 34.36, 37.10, 37.88, 38.69, 40.27, 40.83, 40.94, 41.83, 48.10, 49.17, 49.26, 51.01, 51.18, 52.54, 53.27, 57.72, 61.70, 126.29, 126.42, 127.14, 127.58, 127.81, 128.06, 128.19, 128.29, 128.52, 129.12, 129.60, 134.96, 137.65, 168.57, 169.14, 170.13, 170.62, 170.77, 170.88, 171.10, 171.51, 171.69, 171.72. HRMS (ESI) [M+Na]⁺ m/z: calculated for [C₇₄H₁₂₃N₁₇O₁₅S+Na]⁺: 1544.9003; found: 1544.8978.

2.2.2.4. Synthesis of RLFP-1 (4)

The peptide 4 was synthesized by SPPS, wherein lysine was added to the retro analog 2 followed by its conjugation with levofloxacin. Overall yield: 22.5%; ${\left[\alpha \right]}_{\mathrm{D}}^{24}$= 22.5%; ${\left[\alpha \right]}_{\mathrm{D}}^{24}$= +120 (c 0.001, MeOH). UV-Vis. (MeOH) λ_max (log ε): 206.0 (1.064) nm. IR (KBr, cm^-1): 1196.40 (C-O stretching, CN stretching, NH bending), 1455.64 (C-H bending, Aromatic C=C stretching), 1635.96 (NHC=O stretching), 2981.07 (C-H stretching), 3335.69 (OH, NH stretching). ¹H NMR (d₆-DMSO, 600 MHz): δ_H 0.72–0.81 (18H, m, (CH₃)₂–Leu^{2, 5, 12}), 0.80–0.87 (12H, m, δ-CH₃–IIe⁹, γ-CH₃–IIe⁹, (CH₃)₂–Val⁸), 1.16 (3H, d, CH₃–D-Ala¹⁰), 1.29–1.55 (21H, m, 14’-CH₃–LF, α-CH₂–bAla CH₂–Leu^{2, 5, 12}, δ-CH–Leu^{2, 5, 12}), 1.30–2.39 (β-CH–IIe⁹, β-CH₂–Met⁶, β-CH₂–Lys^3,14, γ- CH₂–Lys^3,14, δ-CH₂–Lys^3,14), 1.45–1.50 (1H, m, γ-CH–IIe⁹), 1.92 (1H, m, β-CH–Val⁸), 1.99 (1H, m, δ- CH₃–Met⁶), 1.78–1.81 (2H, m, γ- CH₂–Met⁶), 2.72 (6H, t, Δ- CH₂–Lys^3,14), 2.85–3.16 (11H, dd, 15’-CH₃–LF, CH₂–Phe^{1, 13}, 3’‘ 5’‘-CH₂–LF), 3.47 (2’‘ 6’‘-CH₂–LF), 3.60 (2H, m, CH₂–Ser¹¹), 3.68–3.77 (4H, m, CH₂–Gly^4,7), 4.04–4.18 (5H, m, α-CH–Phe¹, α-CH–Lys³, α-CH–Leu⁵, α-CH–Val⁸, α-CH–IIe⁹), 4.22–4.54 (11H, m, α-CH–Leu², α-CH–Met⁶, α-CH–D-Ala¹⁰, α-CH–Ser¹¹, , α-CH–Leu¹², α-CH–Phe¹³, α-CH–Lys¹⁴, 2’-CH₂–LF), 4.84 (3’-CH–LF), 5.3 (1H, bs, OH–Ser¹¹), 7.16–7.27 (10H, m, CH_Ar–phe^1,13), 7.53 (1H, 8’-CH–LF), 7.75–8.00 (8H, m, NH–Lys³, NH–Leu⁵, NH–Met⁶, NH–Val⁸, NH–D-Ala¹⁰, NH–Ser¹¹, NH–Leu¹², NH–Phe¹³). 8.01–8.58 (6H, m, NH–Phe¹, NH–Leu², NH–Gly⁴, NH–Gly⁷, NH–IIe⁹, NH–Lys¹⁴), 8.76 (1H, 5’-CH–LF). ¹³C-NMR (d₆-DMSO, 150 MHz): δ ppm 14.7, 15.4, 15.5, 18.3, 19.3, 19.3, 21.6, 21.7, 21.7, 21.8, 22.4, 22.7, 22.8, 23.1, 23.2, 23.3, 24.1, 24.2, 24.3, 24.7, 26.8, 27.3, 29.2, 29.2, 29.7, 30.6, 30.6, 31.0, 31.5, 31.7, 34.6, 36.1, 36.4, 37.1, 38.9, 40.3, 40.8, 41.0, 42.0, 42.1, 48.8, 51.3, 51.4, 52.3, 52.8, 53.3, 57.7, 58.0, 61.8, 64.4, 73.0, 129.3, 129.3, 129.4, 129.7, 133.9, 134.8, 137.7, 156.2, 162.8, 167.9, 168.9, 171.0, 171.4, 171.7, 171.8, 171.9, 172.5, 173.8. HRMS (ESI) [M+H]⁺ m/z: calculated for [C₉₂H₁₄₁FN₂₀O₁₈S+H]⁺: 1866.0516; found: 1866.1101.

2.2.2.5. Synthesis of RLFP-2 Peptide (5)

The peptide 5 was synthesized by SPPS, wherein lysine was added to the retro analog 2 followed by levofloxacin conjugation with lysine-3. Overall yield: 17%; ${\left[\alpha \right]}_{\mathrm{D}}^{24}$= +5 (c 0.001, MeOH). UV-Vis. (MeOH) λ_max (log ε): 206 (1.064) nm. IR (KBr, cm^-1): 1145.49 (C-O stretching, CN stretching, NH bending), 1456.10 (C-H bending, Aromatic C=C stretching), 1627.49 (NHC=O stretching), 2981.05 (C-H stretching), 3285.26 (OH, NH stretching). ¹H NMR (d₆-DMSO, 600 MHz): δ_H 0.72–0.81 (18H, m, (CH₃)₂–Leu^{2, 5, 12}), 0.80–0.87 (12H, m, δ-CH₃–IIe⁹, γ-CH₃–IIe⁹, (CH₃)₂–Val⁸), 1.16 (3H, d, CH₃–D-Ala¹⁰), 1.29–1.55 (21H, m, 14’-CH₃–LF, CH₂–Leu^{2, 5, 12}, δ-CH–Leu^{2, 5, 12}), 1.30–2.39 (β-CH–IIe⁹, β-CH₂–Met⁶, β-CH₂–Lys^3,14, γ- CH₂–Lys^3,14, δ-CH₂–Lys^3,14), 1.45–1.50 (1H, m, γ-CH–IIe⁹), 1.92 (1H, m, β-CH–Val⁸), 1.99 (1H, m, δ- CH₃–Met⁶), 1.78–1.81 (2H, m, γ- CH₂–Met⁶), 2.72 (2H, t, Δ- CH₂–Lys^3,14), 2.85–3.16 (11H, dd, 15’-CH₃–LF, CH₂–Phe^{1, 13}, 3’‘ 5’‘-CH₂–LF), 3.47 (2’‘ 6’‘-CH₂–LF), 3.60 (2H, m, CH₂–Ser¹¹), 3.68–3.77 (4H, m, CH₂–Gly^4,7), 4.07–4.21 (5H, m, α-CH–Lys³, α-CH–Val⁸, α-CH–IIe⁹, α-CH–Leu¹², α-CH–Lys¹⁴), 4.22–4.54 (11H, m, α-CH–Phe¹, α-CH–Leu², α-CH–Leu⁵, α-CH–Met⁶, α-CH–D-Ala¹⁰, α-CH–Ser¹¹, α-CH–Phe¹³ 2’-CH₂–LF), 4.84 (3’-CH–LF), 5.3 (1H, bs, OH–Ser¹¹), 7.16–4.27 (10H, m, CH_Ar–phe^1,13), 7.53 (1H, 8’-CH–LF), 7.83–8.00 (7H, m, NH–Lys³, NH–Leu⁵, NH–Met⁶, NH–Val⁸, NH–IIe⁹, NH–D-Ala¹⁰, NH–Phe¹³). 8.01–8.56 (7H, m, NH–Phe¹, NH–Leu², NH–Gly⁴, NH–Gly⁷, NH–Ser¹¹, NH–Leu¹², NH–Lys¹⁴), 8.76 (1H, 5’-CH–LF). ¹³C-NMR (d₆-DMSO, 150 MHz): δ ppm 15.01, 15.64, 15.80, 18.09, 18.73, 19.55, 21.82, 21.88, 21.95, 22.04, 22.65, 22.74, 23.06, 23.35, 23.48, 2..54, 23.68, 24.40, 24.50, 24.61, 25.13, 26.91, 27.53, 29.52, 29.99, 30.66, 30.68, 31.45, 31.64, 31.85, 34.93, 36.45, 37.34, 37.39, 40.39, 40.54, 40.54, 40.78, 41.05, 42.56, 43.22, 47.80, 48.96, 51.78, 51.91, 52.33, 52.51, 52.87, 52.99, 53.62, 53.73, 54.04, 54.80, 55.65, 58.35, 58.78, 61.89, 64.84, 73.52, 116.77, 117.58, 118.25, 119.73, 127.03, 127.83, 128.73, 128.75, 129.11, 129.61, 130.05, 134.07, 134.88, 137.85, 141.16, 156.78, 159.07, 159.23, 159.38, 159.54, 168.29, 169.62, 169.73, 170.55, 171.16, 171.45, 171.59, 171.80, 172.11, 172.37, 172.87, 172.91, 173.20, 173.83, 174.38. LRMS (ESI) m/z: 1912.3 [M+2Na+H]⁺

2.2.2.6. Synthesis of RLFP-3 (6)

The peptide 6 was synthesized by SPPS, wherein lysine was added to retro analog 2 followed by levofloxacin conjugation with its phenylalanine-1 via β-alanine linker. Overall yield: 21%; ${\left[\alpha \right]}_{\mathrm{D}}^{24}$= -45 (c 0.001, MeOH). UV-Vis. (MeOH) λ_max (log ε): 224 (1.064) nm. IR (KBr, cm^-1): 1135.15 (C-O stretching, CN stretching, NH bending), 1456.85 (C-H bending, Aromatic C=C stretching), 1670.12 (NHC=O stretching), 2972.69 (C-H stretching), 3648.95 (OH, NH stretching). ¹H NMR (d₆-DMSO, 600 MHz): δ_H 0.72–0.81 (18H, m, (CH₃)₂–Leu^{2, 5, 12}), 0.80–0.87 (12H, m, δ-CH₃–IIe⁹, γ-CH₃–IIe⁹, (CH₃)₂–Val⁸), 1.16 (3H, d, CH₃–D-Ala¹⁰), 1.29–1.55 (21H, m, 14’-CH₃–LF, CH₂–Leu^{2, 5, 12}, δ-CH–Leu^{2, 5, 12}), 1.30–2.39 (β-CH–IIe⁹, β-CH₂–Met⁶, β-CH₂–Lys^3,14, γ- CH₂–Lys^3,14, δ-CH₂–Lys^3,14), 1.45–1.50 (1H, m, γ-CH–IIe⁹), 1.92 (1H, m, β-CH–Val⁸), 1.99 (1H, m, δ- CH₃–Met⁶), 1.78–1.81 (2H, m, γ- CH₂–Met⁶), 2.35 (α-CH₂–bAla), 2.72 (2H, t, Δ- CH₂–Lys^3,14), 2.85–3.16 (11H, dd, 15’-CH₃–LF, CH₂–Phe^{1, 13}, 3’‘ 5’‘-CH₂–LF), 3.37–3.44 (β-CH₂–Ala), 3.47 (2’‘ 6’‘-CH₂–LF), 3.60 (2H, m, CH₂–Ser¹¹), 3.66–3.80 (4H, m, CH₂–Gly^4,7), 4.11–4.23 (5H, m, α-CH–Lys³, α-CH–Val⁸, α-CH–IIe⁹, α-CH–Leu¹², α-CH–Lys¹⁴), 4.23–4.55 (12H, m, α-CH–Phe¹, α-CH–Leu², α-CH–Leu⁵, α-CH–Met⁶, α-CH–D-Ala¹⁰, α-CH–Ser¹¹, α-CH–Phe¹³, 2’-CH₂–LF), 4.84 (3’-CH–LF), 5.3 (1H, bs, OH–Ser¹¹), 7.16–7.27 (10H, m, CH_Ar–phe^1,13), 7.53 (1H, 8’-CH–LF), 7.85–8.00 (8H, m, NH–Leu², NH–Lys³, NH–Val⁸, NH–IIe⁹, NH–Ser¹¹, NH–Leu¹², NH–Phe¹³, NH–Lys¹⁴). 8.01–8.31 (6H, m, NH–Phe¹, NH–Gly⁴, NH–Leu⁵, NH–Met⁶, NH–Gly⁷, NH–D-Ala¹⁰), 8.76 (1H, 5’-CH–LF), 9.88 (NH–bAla). ¹³C-NMR (d₆-DMSO, 150 MHz): δ ppm 11.01, 14.58, 15.22, 17.9, 18.14, 18.39, 19.16, 21.40, 21.43, 21.55, 22.09, 22.18, 22.30, 22.96, 23.12, 23.29, 24.01, 24.07, 24.51, 26.61, 26.64, 29.48, 30.52, 31.30, 31.42, 31.81, 31.92, 35.29, 36.37, 37.11, 37.53, 38.70, 38.76, 40.19, 40.56, 40.81, 41.79, 41.89, 42.47, 48.01, 50.93, 51.11, 51.59, 51.98, 52.21, 52.30, 53.35, 53.75, 53.91, 53.95, 54.63, 57.27, 57.60, 61.68, 68.35, 126.28, 127.93, 128.01, 128.07, 129.13, 129.19, 137.59, 137.81, 168.47, 168.52, 169.29, 170.20, 170.57, 170.70, 170.98, 171.08, 171.29, 171.63, 171.91, 172.07, 173.20, 173.24. HRMS (ESI) [M+Na]⁺ m/z: calculated for [C₉₅H₁₄₆FN₂₁O₁₉S+Na+H]⁺: 1960.0785; found: 1960.0653.

2.2.2.7. Synthesis of RSP-4 Peptide (7)

The peptide 7 was prepared by SPPS, wherein glycine-4 of retro analog 2 was substituted with lysine. Overall yield: 28%; ${\left[\alpha \right]}_{\mathrm{D}}^{24}$= +88 (c 0.001, MeOH). UV-Vis. (MeOH) λ_max (log ε): 230.0 (1.979) nm. ¹H NMR (d₆-DMSO, 600 MHz): δ_H 0.72–0.87 (18H, m, (CH₃)₂–Leu^{2, 5, 12}), 0.80–0.87 (12H, m, δ-CH₃–IIe⁹, γ-CH₃–IIe⁹, (CH₃)₂–Val⁸), 1.16 (3H, d, CH₃–D-Ala¹⁰), 1.29–1.55 (18H, m, CH₂–Leu^{2, 5, 12}, δ-CH–Leu^{2, 5, 12}), 1.30–2.39 (β-CH–IIe⁹, β-CH₂–Met⁶, β-CH₂–Lys^{3, 4}, γ- CH₂–Lys^{3, 4}, δ-CH₂–Lys^{3, 4}), 1.45–1.50 (1H, m, γ-CH–IIe⁹), 1.92 (1H, m, β-CH–Val⁸), 1.99 (1H, m, δ- CH₃–Met⁶), 1.78–1.81 (2H, m, γ- CH₂–Met⁶), 2.72 (2H, t, Δ- CH₂–Lys^{3, 4}), 2.85–3.15 (4H, dd, CH₂–Phe^{1, 13}), 3.60 (2H, m, CH₂–Ser¹¹), 3.77–3.80 (4H, m, CH₂–Gly⁷), 4.04–4.22 (5H, m, α-CH–Phe¹, α-CH–Lys⁴, α-CH–Leu⁵, α-CH–Val⁸, α-CH–IIe⁹,), 4.23–4.39 (7H, m, α-CH–Leu², α-CH–Lys³, α-CH–Met⁶, α-CH–D-Ala¹⁰, α-CH–Ser¹¹, α-CH–Leu¹², α-CH–Phe¹³), 5.3 (1H, bs, OH–Ser¹¹), 7.16–4.27 (10H, m, CH_Ar–phe^1,13), 7.84–8.00 (7H, m, NH– Lys⁴, NH–Leu⁵, NH–Val⁸, NH–IIe⁹, NH–D-Ala¹⁰, NH–Leu¹², NH–Phe¹³). 8.01–8.60 (5H, m, NH–Leu², NH–Lys³, NH–Met⁶, NH–Gly⁷, NH–Ser¹¹). ¹³C-NMR (d₆-DMSO, 150 MHz): δ ppm 14.62, 15.25, 18.08, 18.13, 18.45, 19.28, 21.36, 21.45, 21.70, 22.14, 22.32, 22.94, 23.14, 24.05, 24.14, 24.55, 24.59, 26.69, 26.74, 27.17, 28.12, 29.37, 30.30, 30.58, 31.31, 31.37, 32.05, 34.37, 37.09, 37.31, 38.75, 39.07, 39.20, 39.34, 39.49, 39.62, 39.76, 39.90, 40.02, 40.48, 41.07, 41.93, 43.29, 48.16, 50.99, 51.15, 51.87, 52.04, 52.23, 52.40, 53.26, 53.86, 54.00, 57.68, 61.79, 126.27, 126.30, 126.42, 127.16, 128.11, 128.19, 128.53, 129.11, 129.16, 129.61, 134.90, 137.88, 167.97, 170.73, 171.29, 171.31, 171.51, 171.63, 171.66, 172.07. HRMS (ESI) [M+Na]⁺ m/z: calculated for [C₇₂H₁₂₀N₁₆O₁₄S+Na]⁺: 1487.8788; found: 1487.8763.

2.2.2.8. Synthesis of RLFP-4 Peptide (8)

The peptide 8 was synthesized by SPPS, wherein glycine-4 of retro analog 2 was substituted with lysine followed levofloxacin conjugation at phenylalanine-1 via β-alanine linker. Overall yield: 25%; ${\left[\alpha \right]}_{\mathrm{D}}^{24}$= +131.5 (c 0.001, MeOH). UV-Vis. (MeOH) λ_max (log ε): 225.0 (1.65) nm. ¹H NMR (d₆-DMSO, 600 MHz): δ_H 0.75–0.85 (18H, m, (CH₃)₂–Leu^{2, 5, 12}), 0.78–0.83 (12H, m, δ-CH₃–IIe⁹, γ-CH₃–IIe⁹, (CH₃)₂–Val⁸), 1.16 (3H, d, CH₃–D-Ala¹⁰), 1.26–1.61 (21H, m, 14’-CH₃–LF, CH₂–Leu^{2, 5, 12}, δ-CH–Leu^{2, 5, 12}), 1.30–2.39 (β-CH–IIe⁹, β-CH₂–Met⁶, β-CH₂–Lys^3,4, γ- CH₂–Lys^3,4, δ-CH₂–Lys^3,4), 1.45–1.50 (1H, m, γ-CH–IIe⁹), 1.92 (1H, m, β-CH–Val⁸), 2.00 (1H, m, δ- CH₃–Met⁶), 1.78–1.81 (2H, m, γ- CH₂–Met⁶), 2.35 (α-CH₂–bAla), 2.72 (2H, t, Δ- CH₂–Lys^3,4), 2.85–3.16 (11H, dd, 15’-CH₃–LF, CH₂–Phe^{1, 13}, 3’ 5’-CH₂–LF), 3.37–3.44 (β-CH₂–bAla), 3.47 (2’ 6’-CH₂–LF), 3.61–3.85 (2H, m, CH₂–Ser¹¹), 3.66–3.84 (4H, m, CH₂–Gly⁷), 4.09–4.22 (4H, m, α-CH–Lys⁴, α-CH–Leu⁵, α-CH–IIe⁹, α-CH–Leu¹²,), 4.23–4.54 (8H, m, α-CH–Phe¹, α-CH–Leu², α-CH–Lys³, α-CH–Met⁶, α-CH–Val⁸, α-CH–D-Ala¹⁰, α-CH–Ser¹¹, α-CH–Phe¹³, 2’ 6’-CH₂–LF), 4.84 (3’-CH–LF), 5.3 (1H, bs, OH–Ser¹¹), 7.16–7.27 (10H, m, CH_Ar–phe^1,13), 7.53 (1H, 8’-CH–LF), 7.80–8.03 (8H, m, NH–Lys³, NH–Lys⁴, NH–Met⁶, NH–Val⁸, NH–IIe⁹, NH–D-Ala¹⁰, NH–Ser¹¹, NH–Leu¹²). 8.04–9.88 (6H, m, NH–Phe¹, NH–Leu², NH–Leu⁵, NH–Gly⁷, NH–Phe¹³), 8.76 (1H, 5’-CH–LF), 9.88 (NH–bAla). ¹³C-NMR (d₆-DMSO, 150 MHz): δ ppm 10.98, 14.57, 15.19, 17.89, 18.05, 18.39, 19.15, 21.38, 21.41, 21.59, 22.11, 22.15, 22.17, 22.21, 22.89, 23.05, 23.09, 23.26, 23.98, 24.06, 24.08, 24.50, 26.62, 26.64, 29.30, 30.55, 31.26, 32.03, 34.92, 35.30, 36.30, 37.21, 37.29, 40.42, 40.60, 41.81, 42.45, 47.30, 47.97, 50.92, 51.13, 51.90, 52.14, 52.19, 5333, 53.71, 53.92, 54.40, 57.26, 57.52, 61.74, 68.32, 124.22, 126.09, 126.20, 127.92, 128.02, 129.02, 129.06, 137.77, 137.82, 145.01, 163.90, 168.46, 170.45, 170.53, 170.99, 171.18, 171.20, 171.25, 171.51, 171.92, 171.97, 172.88, 173.93, 173.94. HRMS (ESI) [M+Na]⁺ m/z: calculated for [C₉₃H₁₄₃FN₂₀O₁₈S+Na+H]⁺: 1903.0570; found: 1903.0484.

2.2.3. Circular Dichroism (CD) and Secondary Structure Analysis

The far-ultraviolet circular dichroism (CD) spectra were recorded using a JASCO J-810 spectropolarimeter (Jasco, Tokyo, Japan), with measurements taken in a quartz cuvette with a 10 mm path length. The temperature was kept at 22 °C, and the instrument was calibrated with D-(+)-10-camphorsulfonic acid. Peptides were dissolved in 20 mM SDS to a final concentration of 15 µM. The CD spectra were collected over a wavelength range of 190 to 260 nm with a bandwidth of 2 nm. Each spectrum was recorded with ten consecutive scans at a rate of 50 nm/min, and the baseline was obtained under identical conditions. Secondary structures, i.e., percentage of alpha-helix and other secondary structures, were quantified from CD data using Bestsel software (https://bestsel.elte.hu/index.php).

2.3. Biological Studies

2.3.1. Antibacterial Assay

The antibacterial activity of peptide analogues was tested against different bacterial strains, i.e., Staphylococcus aureus (NCTC 13277), Bacillus subtilis (ATCC 23857), Escherichia coli (ATCC 25922), Salmonella typhi (ATCC 14028), and Pseudomonas aeruginosa (ATCC 10145). These strains were obtained from the microbial bank of Dr. Panjwani Center for Molecular Medicine and Drug research (PCMD), International Center for Chemical and Biological Sciences (ICCBS), University of Karachi. Colonies of these bacterial strains were grown in respective agar media, then inoculated in Mueller Hinton (MH) broth (Oxoid, UK) and incubated at 37 °C for overnight. The peptide solutions were first diluted 1 in 100 (from stock solution at 20 mM) and then further two-fold dilution in MH broth in sterile 96-well plate were performed, resulting in 100 µL broth containing increasing concentration: of analogs. Bacteria in their exponential growth phase were diluted in MH broth, and 100 µL of this suspension was added in each well. This resulted in a final 200 µL suspension containing 0.5-1.0 X 10⁶ CFU/mL. Then, these plates were incubated at 37°C for 20–22 h. MIC (Minimum Inhibitory Concentration) values were determined as the concentrations of peptides causing >99% inhibition of bacterial growth.

2.3.2. Antifungal Assay

Candida albicans (ATCC 36082) was cultured on Sabourad Dextrose Agar (SDA). A single colony was inoculated in Sabouraud Dextrose Broth (SDB) medium and grown overnight at 37 °C while shaking. The next day, susceptibility of C. albicans was tested using broth microdilution assay. Peptide analogues were diluted in SDB media by two-fold, resulting in 100 µL medium containing increasing concentrations of analogs. The overnight culture (turbidity at OD₆₀₀ = 1) was 1000x diluted in SDB. Then 100 µL of this suspension was added to each well, resulting in a final 200 µL suspension containing 2–4 x 10⁵ CFU/mL. The plate was incubated at 37°C for 24 h. The next day, MIC (Minimum Inhibitory Concentration) values were determined as the concentrations of peptides causing >99% inhibition of yeast growth.

2.3.3. Antiproliferative Assay

Human breast cancer (MCF-7) and human cervical cancer (HeLa) cells were obtained from the cell culture bank of PCMD in ICCBS, University of Karachi. These cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum, and incubated at 37 °C and 5 % CO₂. The antiproliferative activity of samples were determined by MTT assay. Briefly, the cells were washed with PBS and trypsinized. After determination of the cell number using Malassez cell counting, 100 µL of cell suspension were seeded in 96-well plate at the density of 6,000 cells/well and incubated at 37 °C and 5% CO₂. After 24 h, the cells were treated with increasing concentrations of analogs. The next day, media was removed and 200 µL of MTT dye (0.5 mg/mL) was added in each well and incubated for 3 h at 37°C in 5% CO₂ incubator. Then media was removed and 100 µL DMSO was added to solubilize formazan crystals. After one minute of shaking, the absorbance was recorded at 540 nm in microplate reader (Multiskan GO, ThermoScientific). Finally, percent inhibition of proliferation was calculated using the following formula:

$$\%\mathrm{inhibition}\mathrm{of}\mathrm{proliferation}={\displaystyle \frac{\mathrm{O}.\mathrm{D}\mathrm{o}\mathrm{f}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{d}\mathrm{w}\mathrm{e}\mathrm{l}\mathrm{l}-\mathrm{O}.\mathrm{D}\mathrm{o}\mathrm{f}\mathrm{m}\mathrm{e}\mathrm{d}\mathrm{i}\mathrm{a}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}{\mathrm{O}.\mathrm{D}\mathrm{o}\mathrm{f}\mathrm{u}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{d}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}-\mathrm{O}.\mathrm{D}\mathrm{o}\mathrm{f}\mathrm{m}\mathrm{e}\mathrm{d}\mathrm{i}\mathrm{a}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}}x100$$

2.3.4. Hemolytic Assay

For hemolytic assay, fresh human blood (2 mL) was obtained from healthy donor in EDTA tube while following the protocol of approval from Independent Ethics Committee of ICCBS (Approval number ICCBS/IEC-047-HB-2019/Protocol/1.0). The blood was centrifuged and upper supernatant plasma was removed. The cell pellets were washed with sterile PBS for three times. The washed blood cells were then diluted 25 times to make 4% concentration of initial blood cells. Red blood cells (500 µL) were treated with increasing concentrations of analogs, triton X-100 at 0.1 % being used as positive control giving 100% hemolysis. After 1 h incubation at 37°C, tubes were centrifuged at 800 rpm for 5 min to pellet down red blood cells. After centrifugation, 200 µL of supernatant were transferred into 96 well plate and absorbance was recorded at 576 nm using microplate reader (MultiSkan Go, ThermoScientific).

Then following formula was applied to calculate percent hemolysis:

$$\%\mathrm{Hemolysis}={\displaystyle \frac{\mathrm{O}.\mathrm{D}\mathrm{o}\mathrm{f}\mathrm{t}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}-\mathrm{O}.\mathrm{D}\mathrm{o}\mathrm{f}\mathrm{P}\mathrm{B}\mathrm{S}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}{\mathrm{O}.\mathrm{D}\mathrm{o}\mathrm{f}\mathrm{T}\mathrm{r}\mathrm{i}\mathrm{t}\mathrm{o}\mathrm{x}\mathrm{X}\mathrm{p}\mathrm{o}\mathrm{s}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}-\mathrm{O}.\mathrm{D}\mathrm{o}\mathrm{f}\mathrm{P}\mathrm{B}\mathrm{S}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}}}x100$$

2.3.5. Atomic Force Microscopy (AFM) Imaging

Different bacteria (S. typhi, E. coli, and P. aeruginosa) adjusted to 2–3×10⁷ CFU/mL were treated with 2XMIC values of analogs and incubated overnight at 37°C. The next day, bacteria were washed, dispersed in sterile pure grade water, dispensed in poly-L-lysine coated mica slides, and left for air drying. Changes in the morphology of bacteria were studied with atomic force microscope (Agilent 5500, Chandler, AZ, USA). The whole analysis was done in tapping mode. Images were collected and optimized at scans velocity of 1–5 µm/s and 512 × 512-line resolution and processed through PicoView 1.2 imaging software.

3. Results

Lysine enriched peptides are known to possess strong antimicrobial activities with less toxicity towards eukaryotic cells [35,36,37]. Hence, we decided to investigate the role of lysine using two approaches — addition and substitution — in retro analog of NST-2 peptide. Briefly, L-lysine was added to retro analog 2 at position-14, which afforded RSP-1 peptide (3). In addition, glycine at position-4 in the retro analog was substituted with lysine to give RSP-4 peptide (7). In the next phase, antibacterial levofloxacin was selected for direct conjugation with the side chain of lysine-14 and lysine-3 of analog 3 to afford RLFP-1 (4) and RLPF-2 (5), respectively. Similarly, levofloxacin was also conjugated to analog 3 and analog 7 at phenylalanine-1 via β-alanine linker to get RLFP-3 (6) and RLFP-4 (8), respectively. Amino acid sequences of the peptides (1–8) are shown in Table 1. All the peptides were purified with RP-HPLC (PuriFlash^®) using PFB15C18XS-250/212 column, and eluted at a flow rate of 3 mL/min by 0.1% TFA in H₂O/ACN (40:60). Their purity was established by UPLC (Agilent 1260 Infinity Diode Array, C-4 reversed-phase analytical column, 5 μm, 150 × 4.6 mm, Santa Clara, CA, USA). Furthermore, 1D / 2D NMR, UV-Vis., FT-IR spectroscopy, polarimetry and HR-MS-ESI mass spectrometry were employed to characterize the peptides. Physiochemical parameters of the synthesized peptides showing the molecular weights, optical rotation and retention time are shown in Table 2.

3.1. Circular Dichroism (CD) and Secondary Structure Analysis

The circular dichroism (CD) spectra of temporin NST-2 (1) and its analogs in 20 mM SDS are displayed in the Figure 1. Spectra show that in lipid-like environments provided by SDS, all analogs adopt an alpha-helical conformation due to their amphipathic properties, which is crucial for their antimicrobial function by facilitating their insertion into bacterial membranes. The hydrophobic segment of the peptide interacts with the lipid’s hydrophobic portion, while the hydrophilic segment remains accessible to the aqueous environment. The percentage of alpha-helix and other secondary structures were quantified from CD data using Bestsel software (https://bestsel.elte.hu/index.php) and are shown in Table 3. NST-2 (1) exhibited an 81.5% alpha-helical content, while its retro analog RNST (2) displayed a 62% alpha-helical structure, showing that inversing amino acid sequence had major effect on secondary structure of NST-2. Addition of Lys at the C-terminus in analog 3 restaured the percentage of alpha-helix to 81.8 % whereas Lys substitution at position 4 in analog 7 decreased further helix content to 56.8 %. Conjugation with levofloxacin on Lys³ (analog 4) or Lys¹⁴ (analog 5) also restaured the helix content to 80.1 and 80.4 %, respectively. Conjugation with levofloxacin on Phe¹ (analogs 6 and 8) also increased alpha-helical content compared to analog 2 with 82.2 and 75.0%, respectively.

3.2. Antimicrobial Assay

NST-2 (1) has been already reported to possess strong antibacterial activity against Helicobacter pylori (ATCC 43504) and Staphylococcus aureus (NCTC 13277) with low hemolytic effect [28,29]. The retro analog of 1, RNST-2 (2), was remarkably active against S. aureus (NCTC 13277), B. subtilis (ATCC 23857), S. typhi (ATCC 14028), E. coli (ATCC 25922), and P. aeruginosa (ATCC 10145) as well as against Candida albicans (ATCC 36082) (Table 4) with MIC values lower than the ones of 1 in all cases. Similarly, RSP-1 peptide (3) obtained by addition of Lys at the C-terminal end of analog 2 also showed higher antimicrobial activities compared to 1. Further modifications were explored by synthesizing RSP-4 (7), an analog of 2 wherein Gly⁴ was substituted with Lys. Although supposed to increase it, this substitution did not improve the antibacterial activity of the analog and rather inhibited it as evident from higher MIC values compared to 2. The antibacterial molecule levofloxacin was then conjugated to some analogs to try to improve their activity. As expected, the addition of levofloxacin at position 1 in analog 7 via β-alanine linker generated RLFP-4 (8) with an improved activity against both Gram-positive and -negative bacteria but not against C. albicans. Similarly, RLFP-3 (6), wherein levofloxacin was again conjugated to Phe¹ of analog 3 also via β-alanine linker, demonstrated potent activities against both Gram-positive and -negative bacteria. Surprisingly and not expected due to the selective antibacteria activity of levofloxacin, it was also the most active analog against C. albicans with an MIC of 3.12 µM. In contrast to analogs 6 and 8, RLFP-1 (4) and RLFP-2 (5) obtained by conjugation with levofloxacin to the side chain of Lys¹⁴ or Lys³ of analog 3 displayed reduced antimicrobial effect compared to parent analog 3.

To further explore the mechanism of action of the analogs, three bacterial strains (S. typhi, E. coli, and P. aeruginosa) were treated with the most potent peptide analogs and analyzed through AFM technique in tapping mode. Figure 2 (A–D), Figure 3 (A–D), and Figure 4 (A–D) represents the S. typhi, E. coli, and P. aeruginosa bacteria, respectively. Untreated S. typhi (Figure 2A) appeared as short rods ranging its length between 0.8–1.5 µm, slightly irregular texture with no signs of damage or disruption. S. typhi treated with RLFP-4 (8), RLFP-3 (6), and RLFP-1 (4) at 2XMIC showed loss of morphological rod shapes. Loss of structural integrity could be seen due to leakage of cytoplasmic contents around the damaged cells demonstrating that the analogs caused membrane damages. Similarly, treatment of E. coli (Figure 3) or P. aeruginosa (Figure 4) with RLFP-3 (6), RSP-1 (3), and RLFP-4 (8) at 2XMIC values caused significant damages to the bacteria as their rod shaped morphology were lost, cells were disintegrated, and their cytoplasmic content were found around some rod shaped cells confirming again the membrane damages caused by the analogs.

3.3. Anticancer Activity and Hemolytic Effect of the Analogs

NST-2 (1) was previously reported with antibreast cancer activity (IC₅₀: 17.5 µM) in MCF-7 cells, while it was inactive against cervical cancer HeLa cells [32] with hemolysis HC₅₀ value of 90.0 µM [29]. The antiproliferative activity of the retro analog of NST-2 against cancer cells, i.e., MCF-7 and HeLa cells, as well as their hemolytic properties are presented in Figure 5. IC₅₀ and HC₅₀ values were determined from Figure 5 and are reported in Table 5. Compared to parent peptide NST-2 (1), retro analog 2 was found less active on MCF-7 (Figure 5A) (IC₅₀ of 17.9 versus 53.0 µM) but more active on HeLa (Figure 5B) (IC₅₀ of > 100 versus 60.0 µM). Compared to parent analog 2, analogs 3, 6, 7, and 8 in which Lys was added were found more active than 2 on MCF-7 cells with an efficiency order of 6 > 8 > 7 > 3 > 2. On HeLa cells, analogs 3, 7, and 8 were found more active than analog 2 with efficiency order of 8 > 3 > 7 > 2. Analogs 4 and 5 in which levofloxacin was added to side chain of Lys¹⁴ or Lys³ were found inactive with IC₅₀ > 100µM. Interestingly, analog 6 was the only analog with an IC₅₀ on MCF-7 cells lower than the one of parent peptide NST-2 (1) (i.e., 13.3 versus 17.9 µM, respectively). But like analog 1, analog 6 was found inactive on HeLa cells (IC₅₀ > 100 µM).

In term of hemolysis (Figure 5C and Table 5), all analogs were found more hemolytic than analog 1 (with HC₅₀ values ranging from 4.5 to 51.0 µM compared to 90.0 µM), except analog 4 with a similar HC₅₀ (i.e., 98.9 µM) and analog 7 that was found not hemolytic at the tested concentrations (HC₅₀ > 100 µM). When comparing with their parent molecule (i.e., the analog 2), analogs 3, 4, 5, and 7 were found less hemolytic than analog 2 (HC₅₀ values of 51.0, 98.9, 25.0, and >100 µM versus 13.8 µM) confirming that the addition of Lys to AMPs sequences reduces their hemolytic effect. Oppositely, analogs 6 and 8 in which levofloxacin was added at position 1 were found more hemolytic than 2 (HC₅₀ of 6.6 and 4.5 µM versus 13.8 µM, respectively).

3.4. Selectivity Indexes of the Analogs

The selectivity of action of the analogs against micro-organisms or cancer cells was further evaluated through calculation of their Selectivity Indexes (SI). SI were determined using either MIC on micro-organisms or IC₅₀ on cancer cells compared to HC₅₀ values, and are reported in Table 6.

Regarding SI on micro-organisms, it appears that the retro analog 2 has a lower SI compared to analog 1 (i.e., 8.8 versus 14.4). When comparing to parent analog 2, all other analogs derived from 2 were also found less selective (due to higher MIC values and/or lower HC₅₀), except analogs 3 and 7, for which SI on micro-organisms were equal (for 7) or superior (3.7-fold increase for 3) to parent analog 2. This confirmed that addition of Lys into AMPs improves their selectivity through increased antimicrobial activity and/or decreased toxicity toward human cells, this being the case for analog 3 with an decreased toxicity (HC₅₀ of 13.8 versus 51.0 µM for analogs 2 and 3) with a conserved antimicrobial activity (lowest MIC of 1.56 µM for analogs 2 and 3). Regarding levofloxacin conjugate analogs, their selectivity was reduced either due to decrease in antimicrobial effect or to increase in their toxicity. Analog 4 corresponding to levofloxacin conjugate on Lys¹⁴ displayed reduced toxicity (HC₅₀ of 98.6 µM versus 13.8 µM for analog 2) but also unfortunately reduced antimicrobial activity (lowest MIC of 25 µM for 4 versus 1.56 µM for 2) resulting in a SI of 3.9 versus 8.8. Analogs 6 and 8 corresponding to levofloxacin conjugates at position 1 were as active as analog 2 (lowest MIC of 1.56 µM) but their higher toxicity (HC₅₀ of 6.6 and 4.5 µM versus 13.8 µM) reduced their selectivity from 8.8 to 4.2 and 2.8, respectively. Regarding SI on cancer cells, all analogs were less selective than analog 1, but when compared to analog 2 from which they originated, analogs 3, 6, and 7 were found more selective, suggesting again that the presence of additional Lys in the sequence of retro NST-2 improves its selectivity also against cancer cells.

4. Discussion

In the present study, retro analogs of NST-2, a D-alanine variant of temporin-SHa were synthetized and tested in term of antimicrobial and anticancer activities. This strategy was based on the fact that retro analogs of peptides have been described in the literature as more active than their parent peptides [8,9,10,11]. Retro analogs were further modified by addition of lysine residue, a strategy also known to improve AMPs activity and to decrease their hemolytic effect [35,36,37]. Finally, retro analogs were conjugated to levofloxacin (either on side chain of Lys or at the N-terminus of the peptide), the literature describing AMPs conjugated to antibiotics displaying or not increased efficiency depending of their sequences and site of conjugation [12,13,14,15,16,17,18,19,20,21,22,23,24]. Analogs were tested in term of antimicrobial, anticancer, and hemolytic activities.

In term of antimicrobial activity, previous works have shown that NST-2 (1) possesses strong antibacterial activity with low hemolytic effect [28,29]. The retro analog of 1, RNST-2 (2), was remarkably active against both Gram-positive and Gram–negative bacteria as well as against Candida albicans with MIC values lower than the ones of 1 in all cases. This result confirmed that retro analogs possesse stronger antimicrobial activity compared to parent AMPs as described in the literature [8,9,10]. Similarly, RSP-1 peptide (3) obtained by addition of Lys at the C-terminal end of analog 2 also showed higher antimicrobial activities compared to NST-2 (1) but similar activity compared to retro NST-2 (2) except for E. coli and P. aeruginosa with an improved activity. This is in line with the literature showing that Lys addition to AMPs improves their antimicrobial activity [35,36,37]. Interestingly, RSP-4 (7), an analog of 2 wherein Gly⁴ was substituted with Lys was found less active than NST-2 (1) or retro NST-2 (2) demonstrating that the site of Lys addition/substitution is critical in enhancing the activity of AMPs. Similarly, the site/type of conjugation of AMPs with levofloxacin influences the activity of the analogs. Conjugation with levofloxacin at the N-terminus of Phe¹ through β-alanine linker in analogs 6 and 8 improved activity against both Gram-positive and Gram-negative bacteria as described in the literature for some AMPs conjugated to levofloxacin. Surprisingly, and not expected due to the selective antibacterial activity of levofloxacin, conjugation to levofloxacin on Phe¹ also improved the antifungal activity of analog 6 but not analog 8 against C. albicans. Importantly, analog 6 was also the most active analog against C. albicans with an MIC of 3.12 µM compared to 20.0 and 15.6 µM for NST-2 (1) and retro NST-2 (2), respectively. In contrast to analogs 6 and 8, RLFP-1 (4) and RLFP-2 (5) obtained by the conjugation of levofloxacin to the side chain of Lys¹⁴ or Lys³ displayed reduced antimicrobial effect compared to parent analog 3 reinforcing the idea that modifications have different effect depending of the site/amino-acid modified.

Regarding anticancer activity, retro NST-2 (2) was found less active than parent NST-2 (1) on MCF-7 but more active on HeLa cells. In accordance with the literature, the addition of Lys at the C-terminus in analog 3 and the substitution of Gly to Lys at position 4 in analog 7 both increased the anticancer activity against MCF-7 and HeLa cells. Surprinsingly and not expected due to the selective antibacterial action of levofloxacin, the conjugation with levofloxacin at Phe¹ (analogs 6 and 8) resulted in higher anticancer effect on MCF-7 cells whereas on HeLa cells, only analog 8 showed increased activity. Analogs 4 and 5 in which levofloxacin was conjugated to side chain of Lys¹⁴ or Lys³ were found inactive with IC₅₀ > 100µM showing that the site of conjugation to levofloxacin influences the anticancer activity of the analogs as observed for antimicrobial activity.

In term of hemolytic effect, retro NST-2 (2) unfortunately was found more hemolytic than the parent NST-2 (1) (HC₅₀ of 13.8 versus 90.0 µM). Regarding other analogs, when comparing with their parent molecule (i.e., the analog 2), analogs 3, 4, 5, and 7, all containing additional Lys, were found less hemolytic than analog 2 (HC₅₀ values of 51.0, 98.9, 25.0, and >100 µM versus 13.8 µM) confirming that the addition of Lys to AMPs sequences reduces their hemolytic effect. Oppositely and surprisingly as levofloxacin by it-self is not hemolytic, analogs 6 and 8 in which levofloxacin was added on Phe¹ were found more hemolytic than 2 (HC₅₀ of 6.6 and 4.5 µM versus 13.8 µM, respectively).

Although a high percentage of alpha helix seems to be important for the antimicrobial activity of different AMPs, our data did not found such dependency for NST-2 retro analogs. Indeed, analogs with the higher percentage of helix were not necessary the ones giving the lowest MIC. Although analogs 3 and 6 with 81.8 and 82.2 % helix gave good activity (MIC as low as 1.56 µM), analogs 4 and 5, with 80.1 and 80.4 % helix are the less active analogs (MIC ranging from 25 to >200 µM). Oppositely, analogs 2 and 8 with 62 and 75 % helix were found very active (MIC as low as 1.56 µM). The percentage of alpha-helix in analogs does not seems to correlate also with their anticancer activity. Indeed, the efficiency order in term of anticancer activity on MCF-7 was found as 6 (82.2% helix) > 8 (75.0% helix) > 7 (56.8% helix) > 3 (81.8% helix) > 2 (62.0% helix) with 4 (80.1% helix) and 5 (80.4% helix) being inactive. Similarly, no correlation was found between the percentage of alpha-helix and hemolytic activity of the analogs, since analogs with high percentage of alpha-helix e.g., analogs 1, 4, and 6 (81.5, 80.1, and 82.2% helix) have HC₅₀ of 90.0, 98.9, and 6.6 µM, respectively.

As the modifications of the peptides affected both the activities and hemolytic effect of the analogs, the determination of their selectivity indexes (SI) was needed to identified best analog(s). Regarding antimicrobial activity, the selectivity indexes order was 3 > 1 > 2 = 7 > 6 > 4 > 8 > 5. The retro analog of NST-2 (2) displays a lower SI compared to analog 1 (i.e., 8.8 versus 14.4) showing that although this analog is more active on micro-organisms, its parallel higher hemolytic effect reduces its selectivity. Addition of Lys in analogs 3 and 7 improves their SI by reducing their hemolytic effect compared to parent analog 2, confirming that addition of Lys into AMPs improves their selectivity through decreased toxicity. Regarding levofloxacin conjugate analogs, their selectivity was reduced either due to decrease in antimicrobial effect (analogs 4 and 5) or to increase in their hemolytic effect (analogs 6 and 8). Regarding anticancer activity, the selectivity indexes order was 1 > 7 > 3 > 4 > 6 > 2 = 8 > 5, all analogs being less selective than analog 1, but when compared to analog 2 from which they originated, analogs 3, 6, and 7 were found more selective, suggesting again that the presence of additional Lys in the sequence of retro NST-2 analogs improves their selectivity also against cancer cells.

5. Conclusion

In this study, we explored the efficacy enhancement of an antimicrobial peptide, NST-2 (1), through retro analog synthesis, lysine addition / substitution, and levofloxacin conjugation. Although analog 2 corresponding to retro analog of NST-2 was found more active of bacteria and fungi, its higher hemolytic effect compared to parent NST-2 resulted in a reduction of its selectivity index and therapeutic potential. Addition of lysine (analog 3) and lysine substitution (analog 7) reduced the hemolytic effect of the analogs resulting in safer peptides. Conjugation with levofloxacin on lysine side chain (in analogs 4 and 5) was found to decrease the hemolytic effect of the analogs but unfortunately also their antimicrobial and anticancer activities. Finally, conjugates obtained through addition of levofloxacin at the N-terminus of the peptide via β-alanine linker (analogs 6 and 8) possessed increased antimicrobial and anticancer activity but also unfortunately increased hemolytic effect, resulting in less safe/selective analogs. In conclusion, lysine addition/substitution and levofloxacin conjugation to retro analog of NST-2, at least at the N-terminal position through β-alanine linker, were found to enhance their antimicrobial/anticancer activity and/or to decrease their hemolytic effect, enhancing their therapeutic potential.