Developing a Gram-Negative Selective Peptide–Drug Conjugate

Abstract

Resistance to fluoroquinolone antibiotics has serious implications for healthcare; here, we conjugate the widely used fluoroquinolone ciprofloxacin to a proline-rich antimicrobial peptide (PrAMP) oncocin to improve oncocin’s potency in ciprofloxacin-sensitive and ciprofloxacin-resistant strains of Escherichia coli. The conjugate molecule (oncocin-cipro-c) is ∼3× more potent than the parent oncocin, as determined by MIC, while retaining Gram-negative selectivity. We have characterized oncocin-cipro-c’s interactions with three intracellular targets, two from oncocin (DnaK and 70S ribosome) and a third from ciprofloxacin (gyrase). Oncocin-cipro-c is also able to facilitate mast cell degranulation at a lower concentration than the parent peptide. The development of multimode antibiotics like oncocin-cipro-c is essential in the coming decades of antibiotic resistance.

1. Introduction

Development of Gram-negative targeting antimicrobial agents is an urgent and significant challenge, , in part due to their highly restrictive bacterial encapsulation, limiting drug penetration. , Gram-negative pathogens are responsible for the majority of antimicrobial resistance-associated deaths, with some being resistant to all currently available antibiotics. The Gram-negative pathogens Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, Enterobacter spp., and Acinetobacter baumannii are key drug-resistant pathogens, which the WHO has classified as a critical priority for human health.

Antibiotic resistance arises due to a range of different contributing factors, but one key factor is the limited modes of action of traditional antibiotics. A single intracellular target would require single modifications to overcome. An ideal antibiotic agent would have multiple distinct modes of action. Resistance to such an agent would require multiple simultaneous mutations, which is much less likely than mutation of a single target for resistance as in traditional small-molecule antibiotics. One route to producing multiactive antibiotics is through the conjugation of antibiotics and peptides (peptide–drug conjugates, PDCs).

Here, we detail our efforts toward developing a Gram-negative-selective PDC from a proline-rich antimicrobial peptide (PrAMP). PrAMPs are particularly appealing as they are specifically active against Gram-negative bacteria, have multiple bacterial targets, − and have excellent therapeutic indexes (hemolysis and cytotoxicity vs activity).

Ciprofloxacin is a second-generation broad-spectrum fluoroquinolone antibiotic that is active against both Gram-negative and Gram-positive bacteria. It is used widely to treat a range of bacterial infections; however, widespread resistance has necessitated development of third- and fourth-generation fluoroquinolones such as levofloxacin and moxifloxacin.

Here, we explore conjugation of a PrAMP (oncocin) with a fluoroquinolone antibiotic (ciprofloxacin) to generate a single antibiotic agent with three cellular targets. We show that the conjugate is active against Gram-negative bacteria (E. coli), without affecting Gram-positive bacteria (Staphylococcus aureus). The C-terminal conjugate exhibits a 3-fold potency over the parent peptide without changes to safety profiles.

2. Results and Discussion

In this initial study, we identified the fluoroquinolone, ciprofloxacin, as a suitable antibiotic for the generation of conjugates as it contains a secondary amine and a free carboxylic acid (Figure A). This enables ciprofloxacin to be employed much like an amino acid for the generation of covalent–peptide bond conjugates of ciprofloxacin. To enable the incorporation of ciprofloxacin into oncocin via a peptide bond, we first generated Fmoc-protected ciprofloxacin (Fmoc-cipro, Figure A). Fmoc-cipro was then employed in solid-phase peptide synthesis to incorporate ciprofloxacin onto the N-terminus, or C-terminus of oncocin (n-cipro-oncocin and oncocin-cipro-c, respectively; Figures S1 and A) and potency assessed against Gram-negative ciprofloxacin-sensitive E. coli (ATCC25922; Figure B), ciprofloxacin-resistant E. coli (BAA 3051; Figure C), and S. aureus (ATCC 25923; Figure D).

1.

Generation and antimicrobial testing of oncocin and the ciprofloxacin conjugates. (A) Ciprofloxacin can be N-protected with Fmoc to enable use in SPPS to generate N-cipro-oncocin and oncocin-cipro-C. (B–D) Antimicrobial potency of oncocin, the ciprofloxacin conjugate, coadministration of oncocin and ciprofloxacin, and ciprofloxacin alone was assessed against (B) E. coli ATCC 259220, (C) BAA 3051, and (D) S. aureus ATCC 29213. The red triangle represents the activity of oncocin-cipro-C, and the star represents the activity of ciprofloxacin alone. The C-terminal acid variant of oncocin-cipro-c could not be generated.

Oncocin shows near-identical potency across the two E. coli strains tested (MIC = 1 μM), which is in line with previously reported potency. The conjugates show dramatically different potency, with the C-terminal conjugate (oncocin-cipro-C) showing improved potency over the parent (MIC = 0.3 μM), while the N-terminal conjugate (N-cipro-oncocin) shows reduced potency over the parent (MIC = 8 μM). These potencies were retained across the two strains regardless of ciprofloxacin sensitivity (Figure B,C). The conjugates showed no activity against S. aureus and neither did the parent oncocin, suggesting the specificity is retained regardless of the small-molecule conjugate. When tested for hemolysis and cytotoxicity, no change in the safety profile was observed when compared to the parent (Figure S2).

The requirement for C-terminal conjugation for ciprofloxacin to improve the potency of oncocin fits well with the known mechanism of action of ciprofloxacin whereby the carboxylic acid of ciprofloxacin is penetrant into the binding site on GyrA to elicit its antimicrobial effect (Figure A–C), which is only possible when C-terminally conjugated. While also supporting the known mechanism of action of oncocin, the N-terminus is required for the 70S ribosome inhibition (Figure D,H). The observation that oncocin-cipro-c retains activity in E. coli BAA 3051, more so than the coadministration of oncocin and ciprofloxacin, is a particularly appealing finding, suggesting conjugation in some way overcomes ciprofloxacin resistance that is not possible through coadministration.

2.

Structures of ciprofloxacin and oncocin bound to targets. (A) Structure of gyrase with ciprofloxacin bound highlighted in panel (B) showing that the amino termini of ciprofloxacin are solvent-exposed. (C) Structure of ciprofloxacin with the binding-occupied areas is indicated with a green line, and key interactions are noted. (D) Structure of DnaK with oncocin bound; highlighted in panel (E) shows the binding domain, oncocin threads through DnaK with the occupied amino acids indicated in panel (H) with the orange bar. (F) Transverse view of the structure of the 70S ribosome, with oncocin bound; highlighted in panel (G) shows the binding domain, with the occupied amino acids indicated in panel (H) with the pink bar. Together, the structural data support C-terminal conjugation of ciprofloxacin to oncocin to permit targeting of each of the enzymes. PDB IDs: (A, B) 2XCT; (D, E) 3QNJ; and (F, G) 4Z8C.

Oncocin is well-characterized to elicit inhibitory activity at DnaK and the 70S ribosome. We assessed dose–response binding at these two targets in an attempt to identify the cause of improved potency of oncocin-cipro-c (Table and Figure S3), and we identified both conjugates show slightly improved binding to DnaK; however, N-terminal conjugation dramatically reduced the potency at the 70S ribosome. Ciprofloxacin elicits its antimicrobial effect through inhibition of gyrase, and the conjugates were assessed for their capacity to inhibit gyrase (Table ). Appealingly, we observed that oncocin-cipro-c is able to inhibit gyrase, though less potently than ciprofloxacin or oncocin co-administered with ciprofloxacin. This finding suggests the improved antimicrobial activity of oncocin-cipro-c over oncocin is due to the additional enzymatic inhibition of gyrase.

1. IC₅₀ for Compounds Tested against DnaK, 70S Ribosome, and DNA Gyrase.

	compound tested: IC50
enzyme	oncocin	N-cipro-oncocin	oncocin-cipro-c	ciprofloxacin	oncocin + cipro
DnaK	1.54 ± 0.08 μM	1.01 ± 0.07 μM	0.99 ± 0.06 μM	n/a	n/a
70S ribosome	66.7 ± 6.0 nM	482.6 ± 89.2 nM	51.1 ± 8.5 nM	n/a	n/a
DNA gyrase	>10 μM	>10 μM	2.5 μM	0.3 μM	0.3 μM

Oncocin resembles most likely the sequence of a host defense peptide from Oncopeltus fasciatus. Host defense peptides are known to facilitate MRGPRX2-mediated mast cell degranulation in an immune-recruiting mechanism. Thus, we evaluated the capacity of oncocin, the ciprofloxacin conjugates, ciprofloxacin, and oncocin co-administered with ciprofloxacin (Figure ). This analysis revealed that the peptides do act via MRGPRX2 to degranulate mast cells, with the ciprofloxacin conjugates having a greater effect, suggesting the conjugates may also engage anti-microbial actions of mast cells; however, validation of this hypothesis is required in the future.

3.

Conjugate-mediated degranulation in LAD2 mast cells. (A) Mast cells were treated with compounds (40 uM) and degranulation measured through quantification of b-hexosaminidase release. responses in MRGPRX2 knock-down cells are shown in striped bars. (B) More complete concentration-degranulation relationships are shown. Responses in MRGPRX2 knock-down cells are shown as dashed lines.

We evaluated the effect of ciprofloxacin conjugation to oncocin on the conformation of oncocin via circular dichroism (CD) spectroscopy (Figure A), which showed that ciprofloxacin increased the proportion of the coil structure when conjugated to oncocin, as seen by the decrease of the 200 nm minimum intensity. However, when present as a separate molecule, ciprofloxacin appears to cause aggregation as determined by a reduction of the minimum intensity at 200 nm. We subsequently assessed the effect on the CD spectral signal in the presence of lipids with different surface charges to gain understanding of potential membrane interactions for different organisms (DPC micelles with zwitterionic charge, SDS micelles with a strong anionic charge, POPG/TOCL (9:1) phospholipid large unilamellar vesicles (LUV) to mimic Gram-positive bacterial membranes, and POPE/POPG (7:3) LUV to mimic Gram-negative bacterial membranes) (Figure B–D). In the presence of SDS micelles, the CD signal of oncocin exhibited a red shift of minimum to 215 nm, indicating some propensity for the β-sheet structure, likely due to a charge-based interaction. A similar effect was observed in the presence of the negatively charged PG/CL and PE/PG LUV, although the CD signals exhibited a lesser red shift of minimum intensity than within SDS micelles (Figure B). Oncocin-cipro-c and N-cipro-oncocin showed a similar trend to oncocin, except with the buffer and DPC results showing a greater degree of random coil structure, with a similar potential β-sheet propensity with charged membranes (SDS > PGCL > PEPG, Figure C,D). The CD spectra of oncocin in the presence of free ciprofloxacin and regardless of the membrane mimic added to the system showed severe signal loss, likely due to aggregation (Figure E). Ciprofloxacin on its own is not optically active in CD as expected (Figure F).

4.

CD spectra of 50 μM oncocin and ciprofloxacin conjugates. (A) CD spectra of oncocin, conjugates, ciprofloxacin, and nonconjugated control in buffer. (B–F) CD spectra of oncocin and conjugates in DPC micelles at L/P of 100:1 (blue line), SDS micelles at L/P of 100:1 (green line), POPG/TOCL (9:1) LUV at L/P of 100:1 (green line), and POPE/POPG LUV at L/P of 100:1 (red line). Lines are averaged of three repeats performed at 298 K.

Oncocin has been well-characterized to not kill via membrane disruption. , However, the propensity for polyproline helices, in the presence of charged lipid membranes, indicates interactions at the membrane interface. Thus, the possibility of membrane permeabilization was investigated using dye-leakage assays in various membrane systems (Figure ): neutral mammalian membrane (POPC), Gram-positive bacterial membrane (POPG/TOCL 9:1), and Gram-negative bacterial membrane (POPE/POPG 7:3). Against POPC LUV, ciprofloxacin causes permeabilization in a dose-dependent manner starting from ∼5 μM with oncocin, and the conjugates show dose-dependent leakage above ∼20 μM (Figure A). Against PEPG, the conjugates (N-cipro-oncocin and oncocin-cipro-C) show an increase in permeabilization starting at ∼0.1 μM with an initial maximum at 2 μM, which does not increase as the concentration increases for N-cipro-oncocin, but at 10 μM they begin to increase again to ∼60% leakage (Figure B). Oncocin shows a slow increase in leakage from 0.1 to 10 μM, which then does not increase further. Ciprofloxacin alone and oncocin co-administered with ciprofloxacin showed a similar trend to POPC. Against PGCL, the Gram-negative model membrane, the conjugates showed a much higher maximal leakage, reaching the maxima at 10 μM, and oncocin-cipro-c then decreases leakage as the concentration increases. Oncocin shows a similar profile against PGCL and PEPG below 10 μM, above which, against PGCL, a dramatic shift in permeabilization was observed. While ciprofloxacin and oncocin with free ciprofloxacin showed a similar trend as in POPC (Figure C). Based on the membrane leakage profiles observed here, it is likely that the leakage here is due to vesicle fusion as has been observed for other cationic AMPs.

5.

Calcein-leakage assays of 100 μM LUV in the presence of oncocin and conjugates at increasing concentrations: (A) POPC LUV, (B) POPE/POPG (7:3) LUV, and (C) POPG/TOCL LUV incubated with oncocin (circles), Oncocin and ciprofloxacin (triangles), ciprofloxacin (squares), oncocin-cipro-c (filled squares), and N-cipro-oncocin (filled triangles). Error bars are SD from three replicates performed at 298 K.

With the observation that model membrane permeabilization varies depending on the lipid composition and the potential that the leakage observed is due to vesicle fusion, we moved to investigate membrane permeabilization of E. coli ATCC25922 at a range of MIC relative concentrations (Figures and S4). At the concentrations tested here, we observed no differences in membrane permeabilization, suggesting the improvement in activity of oncocin-cipro-c is due to enzymatic inhibition. Interestingly, for both oncocin and oncocin-cipro-c, we observed that sub-MIC increases in membrane permeabilization with 1/4 MIC concentrations for both, showing an approximate 4-fold increase in permeabilization relative to the MIC concentration. The exact mechanism here is not clear but is quite curious and warrants further detailed investigations.

6.

Membrane permeabilization of E. coli ATCC25922 in the presence of oncocin, cipro, and the conjugates. Each compound was tested at 2×, 1×, 5×, 25×, and 125× MIC to assess permeabilization. Oncocin and oncocin-cipro-C showed similar profiles with 25× MIC showing the greatest permeabilization, while Oncocin and free cipro showed no dose-dependent effect, while N-cipro-oncocin showed reduced permeabilization in a dose-dependent effect.

3. Conclusions

Here, we have detailed our development and evaluation of oncocin-cirpo-c, a c-terminal conjugation of ciprofloxacin, to the potent PrAMP, oncocin. The conjugate has retained Gram-negative selectivity and gained potency over the parent peptide (Figure ), which is likely to assist in reduced bystander selection and further improve the time to resistance. The improved capacity at lower concentrations to induce mast cell degranulation suggests a potential synergism with the innate immune system, which may further improve the in vivo activity.

The proposed mode of action for the improved activity of oncocin-cipro-c over oncocin is due to the acquisition of a third enzymatic target (DNA gyrase; Table ), and an improvement in DnaK binding (IC₅₀ 990 nM (K_i 273 nM) for oncocin-cipro-c vs IC₅₀ 1540 nM (K _i 390 nM) for oncocin; Table ). This conjugate, to the best of our knowledge, represents the first known antibiotic peptide–drug conjugate with at least 3 distinct binding partners. Resistance to oncocin-cipro-c is proposed to require simultaneous resistance mutations at DnaK, 70S ribosome, and DNA gyrase, which may come with a substantial fitness cost to the organism. At present, no report exists detailing resistance to oncocin.

4. Materials and Methods

4.1. Generation of Fmoc-Ciprofloxacin

Ciprofloxacin (Sigma-Aldrich), was suspended in 1:1 acetone:H₂O at approximately 100 mg/mL concentration, before the pH was adjusted to ∼11 with saturated sodium carbonate to clarify the solution. To this, 1.5 equiv of Fmoc-Osu (Auspep, Australia) dissolved in acetone was added dropwise. The reaction was allowed to continue at room temperature for 1 h with stirring before the precipitate was collected and washed with acetone over vacuum before drying via lyophilization. The product was confirmed by MALDI-TOF-MS (Shimadzu, Japan), where we observed m/z = 553, consistent with the expected mass of our product, at a yield of up to 84% at 100 mg scale synthesis.

4.2. Peptide Synthesis and Purification

All peptides were synthesized by Fmoc/ ^t Bu solid-phase synthesis methods using a Biotage Initiator + Alstra microwave-assisted synthesizer on TentaGel MB Rink amide resin with a 0.43 mmol/g loading (RAPP Polymere, Germany). Standard Fmoc-peptide synthesis protocols were used with 4-fold molar excess of the Fmoc-protected amino acid at 0.1 M concentration in the presence of 4-fold HCTU and 8-fold N,N-diisopropylethylamine (DIPEA) in dimethylformamide (DMF) for 10 min at 70 °C, with double couplings for arginine residues and residues following arginine (all at 70 °C). 20% piperidine in DMF was used for deprotection steps, which was achieved with 2× 5 min deprotection at room temperature. Between coupling and deprotection steps, the resin was washed 3 times with 5 mL of DMF. After final deprotection, the resin was washed 3× with 5 mL of DMF, followed by 2× with 5 mL of DCM and dried under vacuum.

After synthesis, the peptides were cleaved from the solid support using a cleavage cocktail of 95% TFA: 2.5% TIPS: 2.5% H₂O, for 2 h at room temperature. After cleavage, the resin was removed through a cotton filter, and the eluate was transferred to a fresh vessel before blowing off the TFA under a continuous flow of N₂. The scavengers were removed with 2 successive washes with ice-cold diethyl ether where the ether precipitated the peptide that was collected by centrifugation at 4000g for 5 min before decantation of diethyl ether. The peptides were then purified by reverse-phase high-performance liquid chromatography (RP-HPLC) on a Phenomenex Gemini C-18 column (150 × 21.2 mm, pore size 110 Å, 5 μm particle size on a Shimadzu (Japan) Nexera Prep-RP-HPLC with a SPD-M40 UV detector at a flow rate of 10 mL/min) in water and acetonitrile containing 0.1% TFA using a gradient of 10–60% acetonitrile in 60 min. The final products were characterized by RP-HPLC (on a Phenomenex Gemini C-18 column (250 × 4.6 mm, pore size 110 Å, 5 μm particle size) on a Shimadzu (Japan) Nexera analytical RP-HPLC with a SPD-M40 UV detector at 1.5 mL/min flow rate) over 20 min from 0 to 100% acetonitrile (214 nm wavelength) as well as by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). After achieving >95% purity as determined by RP-HPLC, all peptides were exchanged to HCl salt through 4 rounds of resuspension in 0.1 M HCl followed by freeze-drying.

4.3. MIC Assay

Peptides were assessed for their MIC against various bacterial strains as reported previously. Peptides were serially diluted at 2× final concentration across a 96-well plate, with the final well containing no peptide and control lanes of no antibacterial agent and lanes of no bacteria or antibacterial agent. All dilutions were in sterile MHB. Each bacterial strain was grown overnight before inoculation of a fresh culture in the morning of the assay, the culture was allowed to grow until OD₆₀₀ = ∼1.0 before dilution to achieve 2 × 10⁶ CFU/mL based on eqs 2e + 7e^{((2.9218*OD₆₀₀))} for E. coli and 2e + 7e^{((2.5537*OD₆₀₀))}. Bacteria were added to wells containing the serially diluted peptide(s) and were allowed to grow for 6 h in E. coli, or 24 h in S. aureus, before measuring OD₆₀₀ on a plate reader. Bacterial growth per well was determined using the following equation

$$\mathrm{b}\mathrm{acterial\; growth}\%=(({\mathrm{OD}}_{600}\mathrm{sample}-{\mathrm{OD}}_{600}\mathrm{negative\; control}))/(({\mathrm{OD}}_{600}\mathrm{positive\; control}-{\mathrm{OD}}_{600}\mathrm{negative\; control}))\times 100$$

The MIC breakpoint was determined using GraphPad PRISM observing for the concentration at which growth surpassed 10%.

4.4. DnaK Binding Assay

Peptides were tested for DnaK inhibition as previously reported. Black 384-well plates (flat bottom, Greiner Bio-One GmbH, Frickenhausen, Germany) were blocked with 0.5% (w/v) casein in phosphate buffered saline (PBS) containing 0.05% (w/v) Tween 20 (PBST) at 4 °C overnight and washed three times with PBST. IC₅₀ values were determined using a 2-fold dilution series of unlabeled peptide (20 μL/well) and DnaK (10 μL/well, 5 μM) diluted in FP-buffer (20 mM Tris-HCl, 150 mM KCl, 5 mM MgCl₂, 1 mM NaN₃, 2 mM DTT, pH 7.5). The plate was incubated at 28 °C in the dark for 90 min before the Cf-labeled oncocin was added (10 μL/well, 80 nM). After a second incubation (90 min, 28 °C), the fluorescence polarization was recorded (λ_ex = 485 nm, λ_em = 535 nm). Inhibitory constants were calculated by fitting the data with a variable slope parameter [y = min + (max – min)/(1 + (x/K _d) – Hill slope)] using SigmaPlot 13 (Systat Software Inc., San Jose, CA).

4.5. 70S Ribosome Binding Assay

Peptides were tested for 70S ribosome inhibition, as previously reported. Briefly, E. coli BW25113 was cultivated in Luria-Bertani (LB) medium and cells were harvested after reaching an optical density of ∼4 at 600 nm by centrifugation (5000g, 15 min, 4 °C, Rotor JLA 8.100, Avanti J-20 XP, Beckman Coulter, Krefeld, Germany). The cell pellets were frozen and stored at −80 °C. Cells were resuspended in ribosome buffer (2 mL/g cells; 20 mM HEPES-KOH, 6 mM MgCl₂, 30 mM NH₄Cl, and 4 mM 2-mercaptoethanol, pH 7.6). Lysozyme (0.25 g/L cell suspension) was added, and the mixture was incubated on ice for 30 min. Cells were disrupted using the bead mill homogenizer FastPrep-24 5G (40 s, 4 m/s, 6 cycles, MP Biomedicals Germany GmbH, Eschwege, Germany) and zirconia/silica beads (0.1 mm diameter). The lysate was centrifuged (1500g, 5 min, 4 °C, Rotor S4180, Allegra 21R, Beckman Coulter) and the supernatant was incubated with DNase (5 U/mL) on ice for 60 min. The cell debris was removed by two centrifugation steps (16,000g, 30 min, 4 °C followed by 32,000g, 60 min, 4 °C, Rotor JA 30.50 Ti, Avanti J-30I, Beckman Coulter). The ribosome was pelleted by ultracentrifugation (165,000g, 17 h, 4 °C, Rotor 70 Ti, Optima LE-80K, Beckman Coulter), resuspended in ribosome buffer (0.1 mL/g pellet), and stored at −80 °C. The ribosome concentration was determined by recording the absorbance of RNA at 260 nm (NanoPhotometer NP80, Implen GmbH, München, Germany), assuming that 1 AU corresponds to a ribosome concentration of 28 nM. The molecular weight of the E. coli 70S ribosome was assumed to be 2.3 MDa. IC₅₀ values were determined using a 2-fold serial dilution series of the unlabeled peptide from 150 μM to 70 pM in ribosome buffer (20 μL). Ribosome solution was added (10 μL; 10 μM) and the plate was incubated after centrifugation (2 min, 500g, Rotor S2096, Allegra 21R, Beckman Coulter) at 28 °C for 90 min. Cf-labeled peptide was added (10 μL; final concentration of 20 nM; final ribosome concentration of 0.5 μM) and the plate was centrifuged and incubated again (90 min, 28 °C, dark). Fluorescence polarization was recorded (λ_ex = 485 nm, λ_em = 535 nm) on a PARADIGM microplate reader. IC₅₀ values were calculated by fitting the data with a variable slope parameter [y = min + (max – min)/(1 + (x/K _d) – Hill slope)] using SigmaPlot 13 (Systat Software Inc., San Jose, CA).

4.6. DNA Gyrase Inhibition

Peptides were tested for DNA gyrase inhibition as previously described. Briefly, a total of 500 ng of relaxed pBR322 plasmid was used as a substrate for each 30 μL reaction. The amount of DNA gyrase (A₂B₂) added was normalized by testing various dilutions of the stock (A₂B₂ containing 0.5 mg/mL of each subunit) without compound. The amount sufficient to supercoil 50% of the substrate was then used for the testing of compounds. This ensured a limiting amount of enzyme for every mutant tested. The individual gyrase subunits were either prepared in the lab as described previously (25) or purchased from Inspiralis. IC₅₀ (compound concentration giving only 50% of the supercoiled substrate obtained with the uninhibited enzyme) was determined by plotting the quantified (using ImageJ) proportion of supercoiled DNA to the total of the lane against the compound concentration and fitting it to a four-parameter binding curve (y = min + ((max – min)/(1 + (x/IC₅₀)∧Hill Slope))) with Scipy. The measured value was the best fit for the IC₅₀ parameter. Comparing our measurement with ciprofloxacin to the published results showed overall consistency and validated our methodology (for instance, compare our ciprofloxacin data with those reported in ref ).

4.7. Mast Cell Degranulation Assay

4.7.1. Culture of LAD2 Cells

Laboratory of allergic disease 2 (LAD2) cells were provided by Dr Arnold Kirshenbaum (National Institute of Health, Bethesda). LAD2 cells were cultured in StemPro-34 medium (ThermoFisher) supplemented with human stem cell factor (100 ng/mL, Peprotech), GlutaMAX (2 mM), penicillin (50 units/mL), and streptomycin (50 μg/mL) (all from ThermoFisher) and maintained in a humidified incubator (5% CO₂, 37 °C). MRGPRX2 knockdown (KD) LAD2 cells were generated from wild-type (WT) LAD2 cells with CRISPR-Cas9 technology (Fernandopulle et al.) and cultured as above.

4.7.2. Quantification of LAD2 Mast Cell Degranulation

LAD2 cell degranulation is commonly quantified by measurement of β-hexosaminidase release. WT and MRGPRX2 KD LAD2 cells were sensitized with human IgE (1 μg/mL; conditioned media from JW8/5/13 cells) for 48 h prior to degranulation assays. The cells were harvested, washed, and resuspended in Hank’s buffered saline solution (HBSS) (1× Hank’s buffered salts, 0.14% sodium bicarbonate, 10 mM HEPES, 5.5 mM glucose, 0.1% BSA, 0.7 mM MgSO₄, and 1.8 mM CaCl₂, pH 7.3). Cells were then seeded (35,000 cells/well) into 96-well plates and stimulated with ciprofloxacin peptides, compound 48/80 for the MRGPRX2 pathway, and NIP-conjugated BSA (4-hydroxy-3-iodo-5-nitrophenylacetyl bovine serum albumin; NIP-BSA; Biosearch Technologies) for the IgE-dependent pathway for 30 min in a 37 °C oscillating incubator (70 rpm). Triton-X-100 (0.1%; Sigma-Aldrich) was used to quantify the total β-hexosaminidase release in the cell. The cell plate was centrifuged (120g, 5 min). The cell supernatant (25 μL) was then transferred to a separate 96-well plate and p-nitrophenyl N-acetyl-β-d-glucosaminide substrate (75 μL; 2 mM PNAG; pH 4.5) was added for an incubation of 2 h at 37 °C in an oscillating incubator (70 rpm). The reaction was stopped by addition of glycine (100 μL, 0.4 mM; pH 10.7), and the absorbance readings were measured at 405 nm in a microplate reader (Multiskan Ascent, Thermo Fischer). β-hexosaminidase release was calculated as the percentage of the total cellular β-hexosaminidase release (quantified using cellular lysis by 0.1% Triton-X-100) and the percentage of spontaneous cellular β-hexosaminidase release was subtracted from all responses.

4.8. Lactate Dehydrogenase (LDH) Release Assay

To test for drug-induced cytotoxicity, the supernatant (25 μL) of LAD2 cells treated with stimuli or 0.1% Triton-X-100 from the degranulation assay was collected and incubated with the reaction mixture (25 μL) in the LDH cytotoxicity assay kit (Pierce, Thermo Scientific) for 30 min in the dark. Reactions were stopped by adding the stop solution, and the color was measured at absorbance of 492 and 650 nm using a microplate spectrophotometer (Multiskan Ascent, ThermoFisher). LDH activity was determined by subtracting the 650 nm background absorbance value from the reading at 492 nm, and the percentage of cytotoxicity was calculated as the percentage of total cellular LDH release (quantified using cellular lysis by 0.1% Triton-X-100).

4.9. Hemolysis Assay

Sheep red blood cells (RBCs) were diluted at a 1:20 ratio in PBS buffer at pH 7.4 after washing with 10 mL of PBS 2 times (1000 g, 10 min). The number of RBCs was counted using a cell counter (Coulter particle counter Z series, Beckman Coulter), and they were then diluted to 2 × 10⁷ cells/mL. A stock solution of peptides (100 μL) was serially diluted in a V-bottomed 96-well plate from 32 μM. Diluted RBCs (100 μL, 2 × 10⁷ cells/mL in PBS) were then added to the V-bottomed 96-well plate. The cells were incubated at 37 °C, 5% CO₂ for 2 h before centrifugation at 1000g for 10 min. Aliquots (100 μL) of the supernatant were transferred into a new flat-bottomed 96-well plate. The amount of hemoglobin released from RBCs was determined from absorbance values, which were measured at 405 nm with a PerkinElmer1420 Multilabel Counter VICTOR plate reader and Wallac software. Percentage hemolysis was determined with the following equation. Wells with only RBCs in PBS buffer were used as a negative control, while cells that contained 0.5% v/v Triton-X 100 were used as a positive control. % Hemolysis = ((OD₄₀₅ sample – OD₄₀₅ blank)/(OD₄₀₅ 0.5% Triton-X 100 – OD₄₀₅ blank)) × 100.

4.10. Circular Dichroism Spectroscopy Assessment

The secondary structure of peptides was estimated using CD spectroscopy. Peptides and ciprofloxacin were assessed at 50 μM (when both ciprofloxacin and oncocin are present, each at 50 μM) in aqueous solutions (100 mM NaF, 10 mM KH₂PO₄, at pH 7.5) and supplemented with lipid vesicles (PGCL (9:1to model Gram-positive membranes) or PEPG (7:3to model Gram-negative membranes)) at 100 μM or micelles at 100 μM DPC and SDS micelles. PG, CL, and PE lipids were dissolved in chloroform (Merck) at 10 mg/mL, the suspension had the chloroform evaporated and resuspended in cyclohexane. The suspension was subjected to 10 rounds of freeze–thaw by subjecting the vessel to an acetone–dry ice bath before thawing in room temperature water. Following the freeze–thaw cycle, the lipid mixture had the chloroform evaporated under a flow of N₂ gas and the lipids resuspended in 100 mM NaF, 10 mM KH₂PO₄, pH 7.5 at 2 mM final concentration, and extruded through a 100 nm lipid extrusion manifold (610000-1EA, Avanti polar lipids), with 11 passes through the manifold, ensuring the starting side is not the final collection side. Lipids were diluted to 200 μM for CD assessment with the peptides. All samples were assessed in a Jasco J810 spectropolarimeter at 0.5 nm increments from 185 to 260 nm. Appropriate blanks were subtracted before calculating the mean residual ellipticity (θ) with the formula: θ = mdeg/(c·l·N _res), where mdeg is the baseline and blank adjusted data from the CD, c is the molar concentration, l is the light path length (mm), and N _res is the number of amide bonds in the sequence. ,

4.11. Dye-Leakage Assay

Preparation of Carboxyfluorescein Solution

5­(6)-Carboxyfluorescein (CF) was dissolved in aqueous KOH, to which imidazole and ethylenediaminetetraacetic acid (EDTA) were added to achieve a final solution of 55 mM CF, 40 mM imidazole, and 1 mM EDTA pH 7.4.

Lipids were prepared as in the circular dichroism method except that the buffer in which the lipids were resuspended prior to extrusion was supplemented with the CF solution above. The extruded membranes containing CF were separated from free CF with a PD-10 desalting column (2.5 mL volume). The vesicles were eluted using a buffer of 40 mM imidazole, 1 mM EDTA, and 110 mM KCl, pH 7.4. CF vesicles were used as 50-Membrane leakage detected with a spectrofluorometric excitation and emission with 492 and 517 nm, respectively.

4.12. Membrane Permeabilization Assay

To observe membrane permeabilization/disruption, 100 μL aliquots of E. coli (2 × 10⁶ CFU/mL) were incubated with 100 μL of peptide at 2×, 1×, 5×, 25×, and 125× MIC equivalent concentrations, after the 90 min incubation, a 50 μL aliquot of the bacteria/peptide was mixed with 50 μL of saline (0.9% w/v NaCl) containing 0.07% v/v of SYTO 9 green-fluorescent nucleic acid stain (5 mM stock solution) and 0.04% v/v of propidium iodide (PI) (1.5 mM stock solution). Then, the bacteria/peptide samples were analyzed using a CytoFLEX LX flow cytometer (Beckman Coulter). The fluorescence from SYTO 9 was measured through a 525/40 nm band-pass filter, the red emission of PI was measured with a 610/20 nm band-pass filter, and the fluorescence from AF647 was measured through a 660/10 nm band-pass filter. A minimum of 20,000 events or events collected over 60 s were recorded. The level of membrane disruption is expressed as % membrane disruption (PI⁺ cells).

Data is made available upon request to the corresponding author.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.4c08000.

• Peptide purity and mass (Figure S1); hemolysis and cytotoxicity data (Figure S2); competitive binding data for oncocin, ciprofloxacin and conjugates (Figure S3) (PDF)

#.

Department of Radiopharmaceutical Sciences, Cancer Imaging, The Peter MacCallum Cancer Centre, Victoria 3000, Australia; Sir Peter MacCallum Department of Oncology, The University of Melbourne, Victoria 3010, Australia

Conceptualization, T.N.G.H.; methodology and experimentation, all authors; writing-original draft preparation, T.N.G.H.; writing-review and editing, all authors; funding acquisition, J.D.W., M.A.H.. All authors have read and agreed to the published version of the manuscript.

Disclaimer/Publisher’s Note The statements, opinions, and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions, or products referred to in the content.

The authors declare no competing financial interest.