Antibacterial sulfonimidamide-based oligopeptides as type I signal peptidase inhibitors: Synthesis and biological evaluation

Oligopeptide boronates with a lipophilic tail are known to inhibit the type I signal peptidase in E. coli , which is a promising drug target for developing novel antibiotics. Antibacterial activity depends on these oligopeptides having a cationic modi ﬁ cation to increase their permeation. Unfortunately, this modi ﬁ - cation is associated with cytotoxicity, motivating the need for novel approaches. The sulfonimidamide functionality has recently gained much interest in drug design and discovery, as a means of introducing chirality and an imine-handle, thus allowing for the incorporation of additional substituents. This in turn can tune the chemical and biological properties, which are here explored. We show that introducing the sulfonimidamide between the lipophilic tail and the peptide in a series of signal peptidase inhibitors resulted in antibacterial activity, while the sulfonamide isostere and previously known non-cationic analogs were inactive. Additionally, we show that replacing the sulfonamide with a sulfonimidamide resulted in decreased cytotoxicity, and similar results were seen by adding a cationic sidechain to the sulfonimidamide motif. This is the ﬁ rst report of incorporation of the sulfonimidamide functional group into bioactive peptides, more speci ﬁ cally into antibacterial oligopeptides, and evaluation of its biological effects. © 2021 The Authors. Published by Elsevier Masson SAS. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).


Introduction
The lack of new antibiotics, alongside the increase in resistance to existing antibiotics, is a current alarming problem that researchers all over the world are fighting to address.To discover new antibiotics, novel lead structures that are potent and specific, and possess minimal toxic effects, are needed.In medicinal chemistry, optimization of validated hits is a fundamental task involving strategies such as bioisosteric replacements of atoms or groups [1].The sulfonamide (SA) functional group, one of the first examples of a carboxamide isostere in drug design [2], has played a major role in various important biologically-active compounds such as diuretics, antivirals, and antibiotics [3].Replacing one of the SA oxygens with a nitrogen gives the isostere sulfonimidamide (SIA), which has gained much momentum over the past ten years due to its unique chemical properties and new possibilities in drug design.The extra nitrogen in the SIA scaffold acts as a weakly basic imine, providing a reactive handle for incorporating chemical modifications.This substitution furthermore introduces chirality into the molecule, in contrast to the achiral SA or carboxylic acids [4].Previous reports have shown that properties such as metabolic stability and solubility are often improved with the SIA compared to the corresponding SAs in small molecules [4,5].Until now, the major studies regarding SIAs have focused on the development and optimization of novel synthetic routes to small molecule SIAs, whereas the evaluation of SIA-based compounds in medicinal chemistry has been less explored.Examples of small molecule SIA-based analogs of known bioactive compounds are seen in Fig. 1, including oncolytic sulfonylureas [6], an SIA analog of the Alzheimer's drug Begacestat [7], antimicrobial trifluoromethylated SIAs [8], and an analog of the cancer drug Tasisulam [9].
A review by Arvidsson's group in 2016 [4] focusing on SIAs in medicinal chemistry also highlighted the fact that most emphasis has been placed on studies of the drug-related and physicochemical properties of SIAs, in contrast to the paucity of studies on biological evaluation.To the best of our knowledge, no evaluation of SIA in bioactive peptides has yet been presented.
Multidrug-resistant Gram-negative bacteria have become one of the primary concerns in the fight against antibiotic resistance, as there are currently few effective drugs available to treat such infections.The double cell membrane and special efflux mechanisms make it very challenging to develop new effective Gram-negative antibiotics [10e12].Type I signal peptidase (SPase I), commonly referred to as LepB in Escherichia coli [13], is a promising drug target as it is limited to bacteria and vital for their survival and virulence.SPase is a highly specific enzyme designed to cleave pre-proteins to release a mature protein from the bacterial cytoplasmic membrane.The active site is located at the outer surface of the cytoplasmic membrane, making it accessible to any inhibitor capable of crossing the outer membrane of Gram-negative bacteria.The peptidase has a unique Ser-Lys catalytic dyad that differs from typical eukaryotic serine proteases, which have a Ser-His-Asp catalytic triad architecture [14e16].Targeting the Ser-Lys dyad and thus blocking the SPase activity leads to the accumulation of the pre-proteins in the membrane, eventually killing the bacteria.Despite the five classes of molecules, including arylomycins [17e19], krisynomycin [20], 5S-penems [21,22], beta-aminoketone [23], and substrate-based oligopeptides [24e27], that have been identified as inhibitors of SPase I, there is to our knowledge still no candidate drug in advanced stages of development.In our laboratory, De Rosa et al., previously optimized substrate-based oligopeptides by replacing an aldehyde warhead with a boronic ester (Fig. 2A), which improved the EcLepB IC 50 from low micromolar to low nanomolar, and at the same time imparted whole-cell antibacterial activity on wild-type strains [26].Although promising starting points, these potent oligopeptide boronates were shown to be toxic to human liver cells (HepG2) and to have hemolytic effects.Therefore, improvements are necessary to reduce the toxic effects while maintaining the antibacterial activity of these compounds.Further optimization of the linear boronic ester inhibitors was recently attempted in our lab, resulting in novel boronic ester-linked macrocycles [27].These compounds were also potent inhibitors of the enzyme, but unfortunately did not solve the existing problems; when the toxic effect was avoided, the antibacterial activity was reduced.Investigation of the structure-activity relationships of the substrate-based oligopeptides has identified some key structural features important for antibacterial activity.A bulky lipophilic tail at the N-terminus, an electrophilic and highly-reactive warhead at the C-terminus, and a positive charge around the P5 position are all important for antimicrobial activity (Fig. 2).In particular, the lipophilic group is crucial for enzymatic inhibition.This tail seems to be needed for anchoring to the membrane where the LepB target is situated; it thus helps the peptide part of the inhibitor to reach the binding site.However, previous studies indicated that the lipophilic tail might also be correlated with cytotoxicity and hemolytic side effects [26,27].To develop sufficiently potent lead molecules with an acceptable toxicity profile, further exploration of the lipophilic tails is of interest.Additionally, designs for improved antibacterial activity of the oligopeptides in previous studies relied on either the addition of a cationic side chain as seen in Fig. 2A [25] or replacement of the proline with the cationic amino acid ornithine [27].The aim of the present work was to investigate the effect of the SIA functionality, introduced between the lipophilic tail and P5, on antibacterial potency via modulating permeability or efflux properties.Effects on toxicity and hemolysis of SIA-containing oligopeptides were also evaluated.Furthermore, the SIA offers an additional hydrogen bond donor or acceptor, allows for an additional substituent, introduces chirality, and alters the geometry between the peptide and the lipophilic chain compared to the amide-linked oligopeptide.We wanted to take advantage of these features and explore the differences of the SA versus SIA functionalities in the oligopeptides.We furthermore saw an opportunity to apply and challenge our previously reported synthetic protocols [28,29] for introducing the SIA group into peptides in a medicinal chemistry context.
We report herein the design, synthesis, and biological evaluation of novel SA-and SIA-linked oligopeptides targeting the E. coli type I signal peptidase.

Design and chemistry
The oligopeptide boronates with lipophilic tails designed and synthesized in this study were based on a Pro-Thr-Ala-Asn peptide sequence with an amino boronic ester as warhead at the N-terminus, a rigid biphenyl tail, and with SA or SIA as a linker functionality.Furthermore, for comparison, a SA-linked analog was designed with a flexible dodecane tail to investigate if there were differences from the more rigid tail analog.The SIA-linked oligopeptides were designed with the same sequence as the SA-linked analog for direct comparison of the linkers.As previously mentioned, the SIA group introduces chirality, therefore it was important to be able to separate the two diastereomers to investigate them for possible differences.To facilitate simple separation of the diastereomers, the SIA building blocks were designed with the tert-butyldiphenylsilyl (TBDPS)-protected SA.This bulky group has been shown to enhance differences between isomers; the fact that it is highly UV-active also makes the separation much easier [28].We have previously shown that the TBDPS group is compatible with solid-phase peptide chemistry (SPPS), and is stable when cleaving the peptide with 10e20% hexafluoroisopropanol (HFIP) in DCM from a 2-chlorotritylchloride (2-CTC) polymer resin, whereas it was removed when using standard cleavage conditions such as TFA [29].As highlighted in the initial study on the oligo-boronates [26], introducing a primary amine as a substituent on the proline in P5 was important for antibacterial activity.With this in mind, we designed an analog with an additional cationic sidechain (aminoethyl), utilizing the SIA moiety.The overall synthetic approach is described below where the SA building blocks were first prepared in solution (Scheme 1), the desired peptide sequences on solid phase (Scheme 2), then the building blocks were attached to the peptide sequence (Scheme 3), and finally the boronic ester warhead was coupled to the free C-terminus (Scheme 4).
The syntheses of the SA-based building blocks 3, 4 and 7 are outlined in Scheme 1. Compound 3 was prepared by a substitution of a bromobenzenesulfonyl chloride 1 with a proline methyl ester, followed by a Suzuki-Miyaura cross-coupling of a (4-hexylphenyl) boronic acid, then basic hydrolysis of the methyl ester to yield the free carboxylic acid at the proline.Likewise, compound 4 was prepared by substitution of an N-Boc-etylene diamine with bromobenzenesulfonyl-chloride, followed by Suzuki-Miyaura cross-coupling of the (4-hexylphenyl) boronic acid.The preparation of compound 7 started by protecting a bromobenzenesulfonamide 5 with a TBDPS group, which was followed with Suzuki-Miyaura cross-coupling of the (4-hexylphenyl) boronic acid to obtain the hexyl-biphenyl tail.The tripeptide 10 and tetrapeptide 11 were synthesized by manual SPPS on 2-CTC resin using standard protocols for N-terminal 9-fluorenylmethyloxycarbonyl-protected amino acids (Fmoc-AA-OH), using N,N-diisopropyl-ethylamine (DIPEA) and 3-[Bis(dimethylamino)methyliumyl]-3H-benzotriazol-1-oxide hexafluorophosphate (HBTU) and 20% piperidine in DMF for Fmoc removal (Scheme 2).First, the Fmoc-Asn-(Trt)-OH 8 was attached to a 2-CTC resin to yield 9, which was further reacted with the remaining Fmoc-AAs, DIPEA, and HBTU and finally the Fmoc group was removed by treating the resin with 20% piperidine solution in DMF to yield 10 and 11.To further react the peptides with the SA-based building blocks, the commercially-available sulfonyl chloride 1a was coupled to the tetramer 11 (Scheme 3-c) and the building block 3 was attached directly to the trimer 10 (Scheme 3a).Both peptides were then cleaved off the solid phase, resulting in the SA-linked oligopeptides 12e13 (Scheme 3-d).
The syntheses of the SIA-linked oligopeptides started with converting the SA building blocks 4 and 7 to a sulfonimidoylchlorides (SIC) in solution, with the chlorinating agent triphenylphosphine dichloride (Scheme 3-b).The SIC solution was then added directly to the tetramer 11 on solid-phase as seen in Scheme 3-c.Any excess of unreacted building blocks was then simply washed away from the oligopeptide-bound resins, after which the resins were treated with 20% HFIP in DCM to selectively release the oligopeptides as free C-terminal acids 14e16 (Scheme 3-d), with the sidechain protecting groups still intact.Thanks to the TBDPS group, diastereomers 14 and 15 were easily separated by preparative HPLC, in contrast to the diastereomers of the Boc-protected aminoethyl compound 16 which were inseparable under these conditions.After purification of the acids by preparative reversedphase HPLC, the final step included the attachment of the reactive boronic ester warhead via peptide coupling of the commercially-available (R)-boroAla-(þ)-pinanediol, in solution, resulting in the final compounds 17e21 (Scheme 4).It is however well known that activation of C-terminal carboxylic acids of peptides can cause racemization via oxazolone formation [30].To avoid this as much as possible, the activation was carried out at low temperature for 3 min before the addition of the amino boronate.However, as previously observed by De Rosa et al., racemization was unavoidable, resulting in partial loss of chiral integrity at the asparagine chiral side chain, and yielding mixtures of epimers in ratios ranging from 90:10 to 50:50.The epimers could not be separated in the preparative RP-HPLC, and thus the final compounds were isolated and tested as mixtures of epimers in analogy with the corresponding inhibitors evaluated by De Rosa et al. [26].

In vitro biology
The final compounds 17e21 were evaluated for in vitro inhibitory potency against the E. coli SPase in a FRET-based functional assay, reported in Table 1 as IC 50 values (mM) [26].The compounds were also evaluated for their minimal inhibitory concentration (MIC), against a panel of seven Gram-negative bacterial strains and one Gram-positive organism, S. aureus [26].The S. aureus spsB gene encodes a catalytically active and essential Type I signal peptidase enzyme with similar conserved sequence motifs to LepB in E. coli [31].Further, the oligopeptides were evaluated on the human liver cell line HepG2 for cytotoxicity, and in a hemolysis assay where the compounds (100 mM) were tested, using red blood cells from heparinized human blood, values greater than 1% in the hemolysis assay are regarded as a red flag [26].
All of the new oligopeptides (17e21) were able to inhibit the LepB protease, with IC 50 s ranging from 0.056 to 1.11 mM.As seen in Table 1, the SA-linked peptides 17 and 18 were more potent inhibitors of the enzyme (0.056e0.071 mM) compared to the SIA-linked peptides 19e21 (0.133e1.11 mM), but not as potent as the previously reported amide-linked peptides 22 and 23 (0.012e0.018 mM) [26].The difference between 22 and 23 is that the latter has a cationic sidechain in P5, which makes the compound active on bacteria, able to inhibit growth at 4 mg/mL on the efflux-defective E. coli strain and at 32 mg/mL on wild-type.Compound 22 and its new SA analog 18 do not have a cationic side chain and lack antibacterial activity even on the efflux-defective strain (>64 mg/mL).However, the SIA analogs 19e20 both inhibit bacterial growth on the efflux-defective strain at 4 mg/mL, equally potent to compound 23, which relies on the cationic side chain for this activity.The drawback of the cationic sidechain in 23 was strong (5.7%)hemolysis, whereas the hemolytic effects of the SIA-analogs 19e20 were lower, at 0.35e2.2%.The SIA analogs 19e21 were also active on the hypersensitive E. coli strain at 8e32 mg/mL, however, no activity was observed on wild-type strains, possibly because they are substrates for efflux pump activity.The cytotoxic effects of the SIA-linked peptides were 5-fold less than the SA analogs (~2.97 mM vs. 0.58 mM), however without reaching safe levels, defined as at least ten times lower than enzyme potency.Adding the cationic aminoethyl group directly on the SIA handle (diastereomers 21) resulted in enhanced inhibition of the LepB protease as compared to the unsubstituted SIA peptides (19 and 20) and 9-fold suppression of cytotoxicity compared to the SA (0.58 mM vs. 5.20 mM).However, it should be noted that the values are rough and should be analyzed with caution since the compounds are a mixture of epimers/diastereomers.Thus, it might be that the inhibition potency, as well as cytotoxicity, for individual stereoisomers could be higher.Since all compounds in Table 1 (17e23) are mixed with certain amounts of D-Asn epimer, which could not be separated, the exact influence of this epimer remains unknown.The initial set of compounds were all designed with the biphenyl lipophilic tail however, to evaluate whether the flexibility of the tail in these types of compounds would have any impact on activity or toxicity, we prepared 17, a direct analog of 18 but with a flexible tail.Notably, the more flexible tail had an impact on the antibacterial activity, resulting in activity on the efflux-defective and hypersensitive E. coli mutants, and on S. aureus, unlike epimers 18, which were inactive on all strains.The compounds showing MIC activity against S. aureus (Table 1) are believed to do so by targeting the SpsB enzyme but at present we have no direct evidence to support this.The alternative possibility is that the activity on S. aureus reflects general cytotoxicity, however, there is no correlation between MIC on S. aureus and cytotoxicity measured by hemolysis or IC 50 on HepG2 cells (Table 1).The flexibility of the tail did not result in any difference in cytotoxicity, however, the more flexible analog had negligible hemolytic effects (0.1%) compared to the more rigid one, where 14.5% hemolysis was observed.The same effect was seen by De Rosa et al. [26], i.e. the more flexible tail was moderately less hemolytic.

Conclusions
In exploring the effects of novel bioisosteric replacements, a number of sulfonamide and sulfonimidamide-based oligopeptides were designed and synthesized.The goal was to study whether such small changes in large bioactive peptides would result in marked differences in their properties.The biological evaluation of the sulfonimidamide-based oligopeptides indicated that the bioisosteric replacement was tolerated, but did not result in optimized inhibition compared to the previously-reported amide-based oligopeptides.However, in the case of the sulfonimidamide 20, hemolysis was decreased to basal levels, yet the compound was active on an efflux-defective E. coli strain.Further, replacing the sulfonamide with sulfonimidamide decreased the cytotoxicity and led to improved antibacterial properties.Additionally, utilizing the sulfonimidamide motif for adding a basic aminoethyl side chain to the "N" handle resulted in reduced cytotoxicity compared to the sulfonamide.We foresee that the synthetic approaches herein demonstrated and the awareness of SIA's ability to affect various properties of bioactive peptides will inspire and facilitate exploration of this functionality in peptide-based drug discovery.

Chemical synthesis
All chemicals and solvents were purchased from Sigma Aldrich, Fisher Scientific, FluoroChem, Enamine and Ark Pharm chemicals and were used without further purification.Analytical thin-layer chromatography (TLC) was performed using Merck aluminum sheets precoated with silica gel 60 F 254 .Column chromatography was performed on Merck silica gel 60 (40e63 mm) 1 H and 13 C NMR spectra were recorded on Varian Mercury Plus instruments; 1 H at 399.9 MHz and 13  To a solution of (S)-2-(methoxycarbonyl)pyrrolidin-1-ium chloride (12.7 mmol, 2.10 g) in dry DCM (35 mL) in a round bottom flask, commercially available 4-bromo-benzenesulfonyl chloride (19.0 mmol, 4.86 g) and triethylamine (38.0 mmol, 5.30 mL) was carefully added at 0 ᵒ C and left stirring for 30 min and then at rt overnight.The mixture was diluted with 1 M aq.HCl (35 mL) and DCM (5 mL) and the two phases were separated in a funnel.The organic phase was washed twice with 1 M HCl (30 mL).The combined organic layers were washed twice with brine (25 mL), dried over Na 2 SO 4 , filtered and the solvent was removed under reduced pressure.The title compound 2a was obtained as white solid (4.30 g, 97% yield) and used in the next step without further purification.1  Commercially available 4-bromo-benzenesulfonate chloride (2.16 mmol, 553 mg) was dissolved in 6 mL DCM, N-Boc-ethylenediamine (1.95 mmol, 308 mL) and TEA (6.49 mmol, 905 mL) was added dropwise and the mixture stirred overnight.After the completion of the reaction, excess TEA was boiled away and then the reaction mixture was diluted with 50 mL of water and extracted two times with DCM, the organic phases collected and dried over Na 2 SO 4 .The solvent was removed under reduced pressure and the crude product then purified by silica gel chromatography (isohexane/ethyl acetate 1:1) and a white solid 2b (797 mg, 96% yield) was obtained. 1H NMR (400 MHz, DMSO-d 6 ) d 7.81 (AA 0 of an AA'XX', 2H), 7.71 (XX 0 of an AA'XX', 2H), 7.33 (bs, 1H), 6.76 (t, J ¼ 5.8 Hz, 1H), 2.94 (m, 2H), 2.76 (m, 2H), 1.34 (s, 9H). 13  A 30 mL vial was charged with sulfonamide 2a (1.44 mmol, 500 mg) and hexylphenyl-boronic acid (2.87 mmol, 592 mg) followed by addition of toluene/EtOH (1:1, 11 mL), 4.0 M potassium carbonate (5.74 mmol, 1.44 mL) and at last Pd(PPh 3 ) 4 (0.0717 mmol, 82.9 mg).The vial was sealed and flushed with N 2 gas and heated to 90 C and left stirring for 3 h when it was stopped after TLC and LC-MS results confirmed that the starting materials were consumed.The reaction mixture was washed with 1 M HCl (2 Â 15 mL) and DCM (2 Â 15 mL) the organic phase was dried over Na 2 SO 4 and concentrated under reduced pressure.To the reaction mixture, LiOH (4.18 mmol, 90.6 mg) was added along with 6 mL of THF/ MeOH/H 2 O (3:2:1) and left stirring overnight.The reaction mixture was washed again with 1 M HCl (2 Â 10 mL) and DCM (2 Â 10 mL) the organic phase was dried over Na 2 SO 4 and concentrated under reduced pressure.The crude mixture was then purified by flash column chromatography (EtOAc/iso-hexane, 4:6).The pure fractions with the product were collected and concentrated affording 3 as a white solid (432 mg, 80% yield A 30 mL vial was loaded with sulfonamide 2b (1.23 mmol, 466 mg) and hexylphenyl-boronic acid (2.46 mmol, 506 mg) followed by addition of 4.0 M K 2 CO 3 (4.91 mmol, 1.23 mL), 10 mL Toluene/EtOH (1:1) and triphenyl-phosphine)palladium (0) (0.0614 mmol, 71.0 mg) was added.The vial was sealed, and the reaction mixture was stirred at 100 C for 4h.Upon completion, the reaction mixture was cooled down to rt and diluted with DCM (30 mL) and brine solution (30 mL) were added, and the two phases separated.The organic phases were collected and concentrated in vacuo and purified by column chromatography (1.5:1 iso-hexane/ ethyl acetate) yielding 515 mg of 4 as white solid (91% yield). 1

Synthesis of N 2 eO-(tert-butyl)-N-((dodecylsulfonyl)-Lprolyl)-L-allothreonyl-L-alanyl-N 4 -trityl-L-asparagine (12)
In a 12 mL disposable syringe, fitted with porous polyethylene filter, the resin bound tetramer 11 was added (0.135 mmol, 300 mg, loading 0.45 mmol/g) and swelled in DCM for 5 min.Then DCM was filtered away and solution of 20% piperidine in DMF added and reacted on rotating wheel.Into a falcon tube, commerciallyavailable dodecane sulfonyl chloride 1a was added (0.552 mmol, 149 mg) with TEA (1.11 mmol, 0.154 mL) dissolved in 12 mL anhydrous DCM.This solution was then added to the syringe containing the resin-bound peptide and left rotating on wheel for 5h.The leftover reaction mixture was then washed away and the resin washed with DCM (3 Â 4 mL), DMF (3 Â 4 mL), THF (2 Â 4 mL) and DCM (2 Â 4 mL).The peptide was then cleaved from the resin and the filtrate and two DCM washes were collected and the solvent removed in vacuo.The crude peptide was purified by preparative RP-HPLC (0.05% HCOOH in MeCN/H 2 O, 90e100% gradient, 20 min, product eluted after 10 min) and the fractions containing the desired compound were combined and lyophilized to get 88.1 mg (70% yield) of desired protected peptide as white solid. 1  4.1.9.Synthesis of N 2 eO-(tert-butyl)-N-(((4 0 -hexyl-[1,1 0 -biphenyl]-4-yl)sulfonyl)-L-prolyl)-L-allothreonyl-L-alanyl-N 4 -trityl-Lasparagine (13) The analog was prepared according to the general peptide synthesis using the trimer 10 (the Fmoc group was removed prior coupling by the general method of deprotection) (0.754 mmol, 877 mg, loading: 0.53 mmol/g), the sulfonamide building block 3 (1.04 mmol, 433 mg) in anhydrous DMF (7 mL), and adding HBTU (1.04 mmol, 394 mg) and DIPEA (2.26 mmol, 0.394 mL).The peptide was then cleaved from the resin and was directly purified by preparative RP-HPLC (0.05% HCOOH in MeCN/H 2 O, 90e100% gradient, 20 min, product eluted after 8 min) and the fractions containing the desired compound were combined and lyophilized to get 528 mg (70% yield) of desired protected peptide as white soft solid.MS (ESI): m/z 1000.4. 1   To a stirred suspension of PPh 3 Cl 2 (0.147 mmol, 49.2 mg) in dry DCM (1.1 mL) under a N 2 atmosphere was added DIPEA (0.306 mmol, 53.4 mL).The reaction mixture was stirred for 20 min at rt and thereafter cooled to 0 C. A solution of the TBDPSprotected sulfonamide building block 7 (0.123 mmol, 80.0 mg) in dry DCM (4 mL) was added, formation of a light brown-yellow solution was observed immediately.The reaction mixture was stirred for 30 min at 0 C then added to the DMF swelled tetramerbound resin 11 (0.341 mmol, 832 mg, loading: 0.41 mmol/g), in a syringe for 2h on rotating wheel.The peptide was then washed with 3 Â 5 mL of DCM and DMF and then cleaved from the resin and directly purified by preparative RP-HPLC (0.05% HCOOH in MeCN, 30 min, product eluted after 16 min) and the fractions containing the desired compound were combined and lyophilized to get 59.5 mg (65.7% yield) of desired protected peptide as white soft  13 To a stirred suspension of PPh 3 Cl 2 in dry DCM (1 mL) under a N 2 gas DIPEA (0.306 mmol, 53.4 mL) was added and the reaction mixture stirred for 30 min at rt then cooled to 0 C. A solution of the TBDPS-protected sulfonamide building block 7 (0.123 mmol, 80.0 mg) in dry DCM (4 mL) was added.The reaction mixture was stirred for 30 min then added to the DMF swelled tetramer bound resin 11 (0.341 mmol, 832 mg, loading: 0.41 mmol/g), in a filter containing syringe and left rotating for 2h.The peptide was then washed with 3 Â 5 mL of DCM and DMF and then cleaved from the resin.The crude was directly purified by preparative RP-HPLC (0.05% HCOOH in MeCN, 30 min, product eluted after 14 min) and the fractions containing the desired compound were combined and lyophilized to get 54.9 mg (63.7% yield) of desired protected peptide as white soft solid.
To a stirred suspension of PPh 3 Cl 2 in dry DCM (1 mL) under a N 2 atmosphere was added DIPEA (1.32 mmol, 230 mL).The reaction mixture was stirred for 30 min at rt then cooled to 0 C. A solution of the Boc-protected sulfonamide building block 4 (0.430 mmol, 198 mg) in dry DCM (4 mL) was added, formation of a clear light brown yellow solution was observed immediately.The reaction mixture was stirred for 30 min at 0 C then added to the DMF swelled tetramer bound resin 11 (0.341 mmol, 832 mg, loading: 0.41 mmol/g), in a filter containing syringe and left on a rotating wheel for 2h.The peptide was then washed with 3 Â 5 mL of DCM and DMF and then cleaved from the resin and directly purified by preparative RP-HPLC (0.05% HCOOH in MeCN for 30 min.The product eluted after 16 min) and the fractions containing the desired compound were combined and lyophilized to get 54.9 mg (63.7% yield) of desired protected peptide as white fluffy solid.4.1.13.General method for coupling the amino-boronic ester warhead to peptides 12e16 and deprotection of sidechain protecting groups To a solution of the peptide (1 equiv.) in anhydrous DCM (2e4 mL), HATU (2 equiv.)and DIPEA (3e10 equiv.)were added under N 2 flow and the mixture was stirred at 0 C for 5 min, after which the warhead (2 equiv.) was added and the reaction mixture was stirred for 30 min and then removed from the ice bath and stirred for another 30 min at rt till completion (monitoring via LC-MS).After completion, the solvent was removed under N 2 stream and the crude dissolved in minimum amount of DMSO for purification by RP-HPLC, the pure product was then used for the final step, deprotection of the sidechain protecting groups.The crude or pure protected peptides (1 equiv.)were dissolved in a solution of 95% TFA in H 2 O or in a solution of TFA/DCM/TES (10:9:1) and stirred at rt till completion (1 h).The solvents were removed under reduced pressure and the residues were purified by preparative RP-HPLC with TEAA/H 2 O/MeCN mobile system.

Bioassays
The E. coli signal peptidase activity and inhibition experiments were conducted in a FRET assay with the two fluorophore labeled substrates; dabcyl-VEVGGTATAYGAFSRPGLE-(EDANS)) and dabcyl-KLTFGTVKPVQAIAYGYEILE-(EDANS)-OH (arrows indicate the expected cleavage sites).The linear increase in fluorescence, corresponding to substrate cleavage, was monitored at 22 C for 2 h.For the inhibition tests, EcLepB was pre-incubated with the compound for 10 min at 22 C, then the reaction was followed at a substrate (dabcyl-VGGTATAGAFSRPGLE (EDANS)-OH) concentration of 8 mM; final EcLepB concentration was estimated to be 50 nM.Reaction rates were plotted as a function of inhibitor concentration, and half maximal inhibitory concentration (IC 50 ) values were determined by a non-linear regression analysis of the sigmoidal doseeresponse curves in GraphPad Prism (GraphPad software Inc., CA, USA).The cloning, expression and purification of the E. coli LepB protein is fully described in the supporting information by De Rosa et al. [26].
For assessing the minimal inhibitory concentration (MIC), the compounds were prepared in Mueller-Hinton II medium and dispensed into a 96-well round-bottomed plate to give final assay concentrations from 64 mg/mL down to 0.25 mg/mL (two-fold dilution series in 10 wells, with two control wells: medium control with no bacteria or compound, and growth control with bacteria added but no compound).Plates were covered and incubated without shaking for 16e20 h at 35 C ± 2 C. MIC was read visually, as complete inhibition of growth by the unaided eye, using the medium-only wells as the control [26].
The compounds were evaluated for hemolytic activity using red blood cells from heparinized human blood.Red blood cells (RBC) were washed three times in Tyrode buffer (130 mM NaCl, 4 mM KCl, 2.8 mM Na acetate, 1 mM MgCl2, 10 mM HEPES, 10 mM glucose, 1 mM CaCl2, adjusted to pH 7.4) and resuspended in the same buffer.Final concentrations in the hemolysis assay were 100 mM compound, 1% DMSO, and 50% RBC, assayed in a 200 mL vol. in a microtiter plate.The mixture was incubated at 37 C for 45 min with shaking (250 rpm).After incubation, RBCs were removed by centrifugation and clear plasma was transferred to a fresh plate, and the amount of hemoglobin measured using a spectrophotometer at 540 nm.The complete lysis control contained 2% Triton X-100 (in Tyrode buffer) instead of compound; the negative control contained Tyrode buffer but no compound.Percent hemolysis was calculated as: [Abs compound] e [Abs negative control]/[Abs complete lysis control] e [Abs negative control] x 100.Values greater than 1% hemolysis at 100 mM were regarded as a red flag [26].
In vitro cytotoxicity was determined by a fluorometric microculture cytotoxicity assay using the HepG2 cell line from ATCC, and cultured in Eagle's Minimum Essential Medium (ATCC) supplemented with 10% fetal bovine serum, penicillin/streptomycin (100 U/100 mg/mL) and L-glutamine 2 mM.HepG2 cells were passaged 2 times/week and used maximally for 20 passages.Cells were finally seeded in Nunc 384-well assay plates at a density of 1000 cells/well.Cytotoxicity was assessed after 72 h with cell survival presented as survival index (SI, %) defined as fluorescence in test wells in percent of control cultures with blank values subtracted.Criteria for a successful assay included a signal-to-noise ratio in control cultures >10, CV < 30% and a positive control (Bortezomib) SI of <5%.The half maximal inhibitory concentration (IC 50 ) was determined from

Table 1
Structures and in vitro testing results for compounds 17e23.