Computational Design of Cyclic Peptide Inhibitors of a Bacterial Membrane Lipoprotein Peptidase

There remains a critical need for new antibiotics against multi-drug-resistant Gram-negative bacteria, a major global threat that continues to impact mortality rates. Lipoprotein signal peptidase II is an essential enzyme in the lipoprotein biosynthetic pathway of Gram-negative bacteria, making it an attractive target for antibacterial drug discovery. Although natural inhibitors of LspA have been identified, such as the cyclic depsipeptide globomycin, poor stability and production difficulties limit their use in a clinical setting. We harness computational design to generate stable de novo cyclic peptide analogues of globomycin. Only 12 peptides needed to be synthesized and tested to yield potent inhibitors, avoiding costly preparation of large libraries and screening campaigns. The most potent analogues showed comparable or better antimicrobial activity than globomycin in microdilution assays against ESKAPE-E pathogens. This work highlights computational design as a general strategy to combat antibiotic resistance.


Materials and Methods. S3
Figure S1.Rationally designed globomycin peptide analogues with no activity against LspA from P. aeruginosa.

S7
Figure S3.Solid-phase peptide synthesis of analogues containing an 8-C lipid chain with variation in ring-size and stereochemistry at the lipid-bearing amide.

MS Spectra for Generation 1 Peptides S23 MS Spectra for Generation 2 Peptides S29
Materials & Methods Materials: All reagents were obtained from commercial sources and used without additional purification.All aqueous solutions were prepared using ultrapure laboratory grade water (deionized, filtered, and sterilized) obtained from an in-house ELGA water purification system.Reverse-phase high-performance liquid chromatography (RP-HPLC) was performed using an Agilent Technologies 1260 Series HPLC instrument with a diode array detector.For analytical analysis, a C 18 reversed-phase HPLC column was used (Higgins).Samples were eluted with a 5-95% acetonitrile/water gradient (0.1% TFA) in 45 min with a flow rate of 1 mL/min and monitored by absorbance at 214 nm.For purifications, semi-preparative C 18 reversed-phase HPLC columns were used (Higgins).Samples were eluted with a 5-65% or 25-95% acetonitrile/water gradient (0.1% TFA) in 35 min with a flow rate of 5.0 mL/min, and monitored by absorbance at 220 nm.Analytical HPLC analysis of final compounds G2a and G2d was performed on a Shimadzu Nexera HPLC instrument equipped with a diode array detector and a C 8 reversed-phase analytical column.Mass spectra were acquired on an Agilent LC-TOF or Thermo ESI direct inject mass spectrometer.
Peptide Synthesis: Peptides were synthesized using standard Fmoc solid-phase chemistry on 2-Chlorotrityl ProTide (CEM, 0.45 mmol/g) resin using a Liberty Blue peptide synthesizer from CEM.All peptides were synthesized on a 0.1 mmol scale and yielded ~20-60% product.Couplings were performed using DIC (5 equiv, Novabiochem) and Oxyma (10 equiv, Sigma) in DMF followed by Fmoc deprotection with 20% piperidine.N-alkyl amino acids were installed by coupling of the appropriate ⍺-bromo acid, followed by treatment with the appropriate amine (2 equiv.) in DMSO for 2 h.Peptides were cleaved (96.5:2.5:1DCM/triisopropylsilane/TFA) for 1 h at room temperature (20 ºC) and precipitated from cold diethyl ether.Peptides were then resuspended in DCM and cyclized using COMU coupling agent, followed by deprotection (95:2.5:2.5 TFA/H2O/triisopropylsilane) for 3.5 h at room temperature.Cyclic peptides were then purified by reverse-phase preparative HPLC.All peptides were characterized by mass analysis using ESI-MS, and the sample purity was assessed by analytical HPLC.

FRET-based peptidase assay:
The FRET-based assay is a continuous assay that monitors an increase in fluorescence at excitation and emission wavelengths of 320 and 420 nm respectively.The synthesis and purification of the FRET substrate has been described previously. [1]Inhibitor concentrations ranged from 0 to 600 µM. Assay buffer was 100 mM MES/NaOH pH 5.6, 150 mM NaCl and 0.05 %(w/v) LMNG.Final DMSO concentration in all reactions was 10 %(v/v).For PaLspA, dose-response assays were performed using 500 nM (Generation 1 compounds) or 100 nM (Generation 2 compounds) and 50 µM FRET peptide substrate.The reaction mixtures were pipetted into wells of a 96-(Thermo Scientific) or a 384-well plate (4titude®) and incubated for 10 min at 37 °C in a SpectraMax M3 (Molecular devices), prior to starting the assay.Assays were initiated by additing enzyme and were performed at 37 o C. Initial velocities (v o and v i ) were determined from the linear region (steady-state) of the reaction progress curve.Initial velocity data was fitted to the dose-response inhibition equation (Equation 1) to determine IC 50 values and to the Morrison equation (Equation 2) to determine Ki' using GraphPad Prism® (Vogeley et al, 2016; Olatunji et al, 2020). [1,2] Structure of the FRET lipopeptide substrate used for assaying the peptidase activity of LspA. [2]l shift assay: The gel-shift assay is a non-continuous SDS-PAGE based assay that monitors the formation of diacylglyceryl-ICP and signal peptide, products of the LspA catalysed reaction.Gelshift assays were set up according to published protocols (Vogeley et al, 2016; Olatunji et al, 2020), [1,2] with the following modifications.12 µM pre-proICP (ppICP) (Lgt substrate), 250 or 600 µM DOPG (Lgt substrate) (Avanti Polar Lipids, Inc.) and 1.2 µM Lgt in 50 mM Tris/HCl pH 7.5, 150 mM NaCl, 1 mM DTT, 0.02%(w/v) LMNG was mixed and incubated at 37 °C and 200 rpm for 60 min to allow for the Lgt catalyzed conversion of pre-proICP to the LspA substrate proICP (pICP).Inhibitors (globomycin, G2a or G2d) (0 -1,000 µM final concentration) in DMSO were added to the reaction mix and the LspA reaction was initiated by the addition of 100 nM PaLspA or EcLspA.The LspA reaction was performed at 37 °C and allowed to proceed for 30 min, then 20 µL aliquots were removed and the reaction was stopped by adding 10 µL 4 x SDS loading buffer (62.5 mM Tris/HCl pH 6.8, 2.5 %(w/v) SDS, 0.002%(w/v) bromophenol blue, 0.5 M β-mercaptoethanol, 10%(v/v) glycerol).10 μL of the stopped reaction mixtures were loaded on precast Mini-PROTEAN® gels (Bio-Rad) and run with Tris/glycine buffer (25 mM Tris/HCl pH 8.0, 250 mM glycine, 0.1 %(w/v) SDS).Gels were stained using InstantBlue™ (Expedeon) and imaged using a Bio-Rad Gel-Doc imager.LspA activity was determined by tracking the band intensity of the diacylglyceryl-ICP product which was quantified using Image Lab.IC 50 values were determined using GraphPad Prism by fitting the data to equation 1.

Microdilution assay:
MIC values were determined in cation-adjusted Mueller-Hinton broth using the broth microdilution method according to CLSI guidelines.Plates were incubated for 16 to 24 h at 37 °C.Six biological replicates were performed across three different 2-fold dilution series, starting at concentration of 100 µg/ml, 80 µg/ml or 64 µg/ml.The lowest concentration of antimicrobial inhibiting growth was taken as the MIC.This was established by visual inspection of the medium after the incubation period with a lack of turbidity of the growth medium indicating no growth. [3]ta analysis: All data was fitted using GraphPad Prism 9.0.0 as detailed in (Vogeley et al, 2016; Olatunji et al, 2020). [1,2] S1).
Additionally, C3 -C5 (peptides S5 -S7) and C7-Gbm(NLys) (peptide S9) amide analogues were submitted for enzyme inhibition assays to determine if these cyclic amide isosteres could still inhibit the target enzyme, LspA.Each peptide was incubated with PaLspA and the observed reduction in enzyme activity reported.Unfortunately, no reduction in activity was observed with peptides S6 and S7 (C4 and C5-Gbm(NLys) and only a moderate decrease in enzyme activity was found for peptides S5 and S9 (C3 and C7-Gbm(NLys).
An olefin isostere was also investigated as a potential replacement for the native ester linkage in globomycin.Octyl-globomycin(RCM) (isostere S13) was synthesized via side-chain immobilization of L-Ser4 and an on-resin ring-closing metathesis (RCM) reaction performed with Grubbs 2 nd generation catalyst to generate a mix of cis-and trans-isomers.Analogously, peptide S13 (RCM peptide) was void of antimicrobial activity against Gram-negative bacteria.These findings highlight the difficulty of successfully replacing the ester linkages in globomycin while retaining antimicrobial activity and native mechanism of action.
Four linear globomycin analogues were synthesized to determine the necessity of the cyclic core.
The pentapeptide core of NMeLeu-Ile-Ser-Thr-Gly-OH was synthesized as both the natural alloversion and the cheaper non-allo version in which native residues allo-L-isoleucine and allo-Lthreonine were replaced by their cheaper non-allo derivatives.Each linear pentapeptide was synthesized in both its acylated, octyl-derivative and non-acylated form and submitted for analysis of its PaLspA inhibitory activity.No PaLspA inhibition was observed, confirming the importance of globomycin's cyclic core for antimicrobial activity.Design name: Xp8

Equation 1 . 5 Figure S1 .Figure S2 .
Figure S1.Rationally designed globomycin peptide analogues with no activity againstPaLspA.Blue shading indicates the areas of variation for each analogue.

Figure S12 .Figure S13 .Figure S14 .
Figure S12.Dose-response and Morrison plots of PaLspA inhibition by compound G2d.FRET-based peptidase assays (A) Dose-response and Morrison plots.(B) Structure of compound G2d.(C) Average IC 50 and K i ' values from the data shown in (A).PaLspA concentration was 40 or 100 nM, FRET substrate concentration was 50 µM and inhibitor concentrations ranged from 0 to 600 µM.Assays were run at 37 o C.

Table S1 . Analytical and MIC (E. coli) data for synthetic analogues
, with MIC values of 100 to 200 mg/mL registered against E. coli (Table