Phenotypic screening of covalent compound libraries identifies chloromethyl ketone antibiotics and MiaA as a new target

The emerging antibiotic resistance requires the development of new antibiotics working on novel bacterial targets. Here, we reported an antibiotic discovery workflow by combining the cysteine-reactive compound library phenotypic screening with activity-based protein profiling, which enables the rapid identification of lead compounds as well as new druggable targets in pathogens. Compounds featuring chloromethyl ketone scaffolds exhibited a notably high hit rate against both gram-negative and gram-positive bacterial strains, but not the more commonly used warheads such as acrylamide or chloroacetamide. Target identification of the lead compound, 10-F05, revealed that its primary targets in S. flexneri are FabH Cys112 and MiaA Cys273. We validated the target relevance through biochemical and genetic interactions. Mechanistic studies revealed modification of MiaA by 10-F05 impair substrate tRNA binding, leading to decreased bacterial stress resistance and virulence. Our findings underscore chloromethyl ketone as a novel antibacterial warhead in covalent antibiotic design. The study showcases that combining covalent compound library phenotypic screening with chemoproteomics is an efficient way to identify new drug targets as well as lead compounds, with the potential to open new research directions in drug discovery and chemical biology. Graphic Abstract


Introduction
Antibiotics are essential human medicines, but antibiotics resistance is threatening the effectiveness of existing antibiotics.Expanding the targets pool is crucial to overcome antibiotics-resistance.Known antibiotics target a limited number of well-studied targets, including the biosynthesis of cell wall, proteins, and nucleic acids. 1 There is a need to explore novel strategies that can target a broader range of bacterial proteins.Inspired by the recent successful development of KRAS G12C inhibitor to target proteins that traditionally considered as 'undruggable' 2 and the existence of antibiotics that covalently target proteins (such as penicillin), we hypothesized that screening a cysteine-reactive compound library can expand the pool of druggable proteins in bacterial proteomes.By utilizing covalent interactions, this approach has the potential to engage essential proteins that are untouched in conventional drug screenings.4][5] By combining these methodologies, we can accelerate the identification and characterization of novel targets, paving the way for the development of new antibiotics.
Despite the rising interest in cysteine targeting compounds, [6][7][8] previous studies primarily used very small libraries with a limited number of cysteine warheads like chloroacetamide. 9,10The limitation of warhead diversity could hinder the identification of druggable protein targets.The diverse properties and reactivities of bacterial cysteinome may require a broader range of warheads to effectively interact with and modify target proteins.Here, we applied a larger and more diverse cysteine-targeting library (Cys-library) consisting of 3,200 fragment-like covalent ligands in antibacterial screening to discover potential new antibiotics and new targets.We discovered that chloromethyl ketone scaffolds showed promising activity in both gram-positive and gram-negative strains.One lead compound, 10-F05, displays broad-spectrum antibacterial activity.Using competitive activity-based protein profiling (ABPP), we identified and validated two proteins, FabH and MiaA, as the physiologically relevant targets of 10-F05.Importantly, MiaA is not a previously recognized antibiotics target, underscoring the advantage of employing phenotypic screening of covalent compound libraries with ABPP to identify both lead compounds and novel targets.

Results
Susceptibility screenings identify antibacterial hits with chloromethyl ketone warhead.To expand the druggable targets within bacterial cysteinome, we obtained a diverse Cys-library consisting of 3,200 fragment-like compounds from Enamine (Figure S1A-E).We first assessed the cytotoxicity of the Cys-library in HEK293T cells at a concentration of 25 μM over two days (Figure S2).Chloromethyl ketone and chloroacetamide scaffolds exhibited higher cytotoxicity in HEK293T cells compared to other warhead scaffolds.The 2-chloropropinoamide and 2-chloroethyl ketone scaffolds displayed higher tolerance in the mammalian cells.
Next, we screened the Cys-library to identify active compounds against S. aureus (MSSA476) and V. cholerae (SAD30) at 25 µM (Figure 1A and 1B).From the initial screening, we identified 48 and 17 hits that inhibited S. aureus and V. cholerae growth in liquid culture, respectively (Table S1).Interestingly, many S. aureus active hits were not active in V. cholerae, which could be due to differences of cell wall composition or cysteine targets in these two strains. 10,11We excluded 11 chloroacetamides, three αcyanoacrylamides, and a cluster of dimethyl pyrroles (DP) with chloromethyl ketone warhead due to the high cytotoxicity in HEK293T cells at 25 μM (Figure S3B, S3C, S3D, and supplementary table). 10,12Two compounds (3-J04 and 8-I07, Figure S3A) were excluded from further analysis due to the presence of a nitrofuran moiety, which is known to be toxic. 13In total,15 chloromethyl ketone compounds were found to be active against both S. aureus and V. cholerae.These hits and three additional chloromethyl ketones were selected to measure the minimum inhibitory concentrations (MIC) against an expanded collection of pathogens including S. flexneri M90T, V. cholerae SAD30, E. coli JPN15, and S. aureus MSSA476 (Table 1).Most chloromethyl ketone hits displayed MICs ranging from 5 μM to 50 μM, with slight variations observed among the four pathogens.Appropriate thiol reactivity is required for high antibacterial activity.Surprisingly, the two most frequently used cysteine covalent warheads, 14 acrylamide and chloroacetamide, were not found among V.
cholerae active hits despite that they account for the majority of the Cys-library (Table S1).Only 11 out of 751 chloroacetamide compounds were active against S. aureus and none against V. cholerae.A previous study also showed that several chloroacetamide-based MurA covalent inhibitors (47 fragment-sized chloroacetamide scaffolds) were not active against E. coli. 9In contrast, a less studied warhead, chloromethyl ketone, has the highest hit discovery rate in both V. cholerae and S. aureus screenings.
Notably, all active hits possessed a direct conjugation of the chloromethyl ketone warhead with an aromatic ring.The aromatic ring conjugated chloromethyl ketone compounds are generally more reactive than the nonaromatic ones because of the conjugation effect (Figure 2B).Both 10-C09 and 10-F05 possess electron-withdrawing groups (trifluoromethyl and fluoro, respectively, Figure 2C) on the benzene ring, which weaken the conjugation effect, leading to decreased reactivity compared to 10-L07, which has an electron-donating group on the benzene ring (Figure 2C).Interestingly, the thiol reactivity did not correlate with the antibacterial activity.For instance, 10-C09 and 10-F05 exhibited similar reactivity towards thiols, yet they displayed a 2.5~10-fold difference in their MICs across four tested pathogens, as shown in Table 1.Furthermore, the two derivatives of 10-F05, namely 10-M19 and 10-L07, which are 1.2-fold and 2.5fold more reactive, respectively, exhibited 2~10-fold worse MICs compared to 10-F05.These results emphasize that other factors contribute to the MIC values.
To better understand the structure activity relationship (SAR), we employed t-SNE distribution in Data Warrior to cluster the 210 compounds in the Cys-library that contained a chloromethyl ketone warhead (Figure 2D and Figure S5). 17Totally 30 clusters were assigned based on the structural similarity.Despite the important role of high reactivity plays in antibacterial activity, no hits were observed in several highly reactive structure clusters 1, 7, and 12, emphasizing the significance of non-covalent moieties in facilitating target engagement.We propose that the high reactivity of 10-F05 is necessary to achieve rapid target engagement, but reactivity is not the sole determining factor for its antibacterial activity.was able to eliminate ~99% of S. flexneri within a 24-hour period (Figure 3A).Furthermore, 10-F05 exhibited low cytotoxicity in HEK293T cells when compared to other chloromethyl ketone hits (Figure S6A).The IC50 value of 10-F05 is ~50 µM in both HEK293T and A549 cells after 2 days incubation.The selectivity ratio ranged from 2 to 20 in different strains (Figure S6B).
Next, we tested the development of resistance against 10-F05 in S. aureus MSSA476 and S. flexneri   Competitive ABPP identifies FabH, MiaA and PdxY as targets.Encouraged by its promising activity in both gram-positive and gram-negative bacteria, we selected 10-F05 as the lead compound for further studies.To confirm that 10-F05 inhibits bacterial growth by targeting the bacterial cysteinome, we measured the MICs of negative control compounds, in which the cysteine-reactive warheads were replaced with unreactive ones, against model pathogens (Table S3).All the negative control compounds showed no activity, suggesting that the growth inhibitory effect is through the covalent modification of bacterial proteins.
10-F05 and 10-L07 exhibited the best MICs of 2.5 and 5 µM, respectively, in S. flexneri M90T and they are quite similar in structure despite their 2.2-fold difference in thiol reactivity.We hypothesized that they engage similar cysteines to inhibit bacterial growth.We applied competitive activity-based protein profiling (ABPP) in S. flexneri M90T using both 10-F05 and 10-L07 to increase the target confidence.Live S. flexneri M90T cells suspended in PBS were treated with 10-F05, 10-L07, or DMSO and lysed, followed by treating with iodoacetamide-conjugated desthiobiotin (IA-DTB).After trypsin and Lys-C digestion, IA-DTB labeled peptides were affinity enriched and quantified using quantitative tandem mass tag (TMT) proteomics.Real-time search coupled with synchronous precursor selection and MS3 fragmentation were implemented to optimize both quantification and proteomic coverage (Figure 4A). 18 our proteomic result, we identified 1035 cysteines from 588 proteins in S. flexneri M90T.Gene ontology was assigned to the identified proteins and no obvious enrichment was observed (Figure S7).We next annotated all the identified proteins using NetGenes database to predict the essentiality of the identified proteins. 19Among 588 proteins, 103 were predicted to be essential in S. flexneri M90T (Figure S8).
We then quantified the cysteines that are engaged by the two compounds.As anticipated, 10-L07 engaged many cysteines with high ligandability due to its higher reactivity.However, 10-F05 was not as promiscuous as we expected and only engaged 5 cysteines higher than 35% ligandability in the bacterial proteome.Interestingly, FabH Cys112 was identified as the top cysteine hit in both 10-F05 and 10-L07, with a ligandability of 65% and 75%, respectively (Figure 4B and 4C).FabH [β-ketoacyl-acyl carrier protein (ACP) synthase III] converts malonyl-ACP to acetoacyl-ACP in bacterial fatty acid biosynthesis and multiple studies have reported the essentiality of FabH. 20,213][24] Among those, Oxa2 was found to bind with FabH Cys112 which is the same cysteine identified in our proteomic result.Oxa2 showed a MIC range from 0.25~0.5 µg/mL (1~2 µM) against a panel of S. aureus.However, Oxa2 is not active against gram-negative strains while 10-F05 is active against gram-negative strains.
We also follow up with other cysteine targets of 10-F05, MiaA Cys273 and PdxY Cys121.These two cysteines have slightly lower ligandability and they were also engaged by 10-L07 (Figure 4C).MiaA generates the i 6 A-37 tRNA by catalyzing the addition of an isopentenyl group onto the N6-nitrogen of Ade-37 next to the anticodon.The product is subsequently methylthiolated by the radical-S-adenosylmethionine enzyme MiaB to yield ms 2 i 6 A-37. 25This modification is known to enhance the interaction of tRNA with UNN codons (Phe, Leu, Ser, Tyr, Cys, Trp), thus promoting reading frame maintenance and translational fidelity. 26,27MiaA is necessary for bacterial fitness and virulence in diverse host niches.Intraperitoneal injection of miaA KO extraintestinal pathogenic E. coli showed significantly lower survival rates compared to the WT E. coli strain in a mouse model. 25Given its well-defined role in regulating bacterial virulence, MiaA appears to be a novel antibiotic target.Despite MiaA's conservation in prokaryotes, the identified Cys273 is only conserved in some bacteria strains and its function remains unclear (Figure S9).
The third hit PdxY, is known as a pyridoxal kinase involved in the pyridoxal 5′-phosphate (PLP) salvage pathway.However this activity is significantly lower compared to PdxK, indicating that its physiological function remains unknown. 28,29PdxY Cys121 is involved in substrate/pyridoxal binding through covalent interaction. 30To confirm the interaction between 10-F05 and the three potential target proteins identified by proteomics, we expressed and purified recombinant Sf_FabH, Sf_MiaA and Sf_PdxY proteins from E. coli BL21 strains and visualized the cysteine labeling using tetramethylrhodamine-5-iodoacetamide (5-TMRIA), with competition from 10-F05.As expected, preincubation of 10-F05 decreased the 5-TMRIA labeling in a dosedependent manner (Figure 5A).Time-dependent labeling revealed rapid covalent bond formation between 10-F05 and Sf_FabH (Figure 5B).Furthermore, 10-F05 was able to bind with S. aureus FabH (Sa_FabH, Figure 5C), which shared 39.9% sequence similarity with Sf_FabH.FabH is highly conserved among (Figure 5D) while MiaA Cys273 and PdxY are not highly conserved in different bacteria (Figure S9), suggesting that FabH is an important target for 10-F05.To test this, we obtained a reported FabH inhibitor Platencin (Figure 5E, left) and measured the MIC in the 10-F05 resistant S. aureus strain compared to WT strain. 31As expected, the 10-F05 resistant S. aureus also exhibited resistance to Platencin (Figure 5E, right).
To further confirm the identified protein targets of 10-F05 in bacteria, we used over-expression and knock-out (KO) E. coli strains (Figure 5F).We validated the ASKA strains over-expressing the selected proteins using western blot analysis (Figure S10). 32We then measured the MIC of 10-F05 in these strains compared to the parental WT strain (Figure 5H).Over-expressing FabH resulted in 2.5-fold increase the MIC.Overexpression of YiiD (also known as FabY), which compensates for the loss of FabH (Figure 5G), 33,34 fully rescued the growth inhibition by 10-F05.This difference in tolerance between FabH and FabY over-expression could be attributed to the fact that the over-expressed FabH is still inhibited by excess 10-F05, but the over-expressed YiiD is not inhibited by 10-F05.Similarly, the MIC of 10-F05 increased 2.5-fold in MiaA and PdxY over-expressing E. coli strains.Interestingly, over-expressing PdxK had no effect on 10-F05 treatment, suggesting that PdxY likely does not function in pyridoxal salvage.The MIC increase caused by FabH/MiaA/PdxY gene over-expression further supports that they are relevant target proteins of 10-F05.
We obtained the knock-out strains from Keio collection 35 and compared the MIC difference of 10-F05 in WT and fabH/miaA/pdxY KO strains.Knocking out any of the three targeted genes sensitized E. coli to 10-F05 treatment, resulting in a 2-fold MIC change (Figure 5I).Notably, the yiiD KO strain showed the largest MIC change consistent with the overexpression data.Thus, our chemical proteomics combined with the biochemical and genetic data revealed that multiple targets contribute to the observed antimicrobial activity of 10-F05, but given the strongest effect of yiiD (which bypasses FabH) overexpression or deletion on the sensitivity to 10-F05, FabH seems to be the major target underlying the antibacterial activity of 10-F05.10-F05 disrupts MiaA and tRNA substrate binding in vitro.Among the top three cysteine targets we identified, the function of MiaA Cys273 remains unclear.Thus, we aimed to investigate the impact of covalent modification of MiaA Cys273 by 10-F05 on MiaA's function.Sf_MiaA contains a total of two cysteines.To confirm that 10-F05 reacts with MiaA Cys273, we purified the MiaA C273A mutant and performed 5-TMRIA labeling.The C273A mutant had decreased 5-TMRIA labeling and the signal was not decreased by 10-F05 competition, indicating 10-F05 specifically reacts with MiaA Cys273, consistent with our proteomic data (Figure 6A).
MiaA Cys273 is located near the tRNA binding site (Figure S11A).Covalent docking was performed to visualize the interaction of 10-F05 and MiaA (Figure S11B).The predicted covalent docking structure revealed a spatial overlap between 10-F05 and the tRNA substrate, suggesting that the covalent modification by 10-F05 could interfere with tRNA binding (Figure 6B, 10-F05 is shown in orange surface).
To test this, we designed a 34-base tryptophan tRNA (tRNA Trp _34) sequence predicted to mimic the tRNA D-loop and anticodon loop, 36 and conducted a gel mobility shift assay (Figure 6C).We first incubated the purified MiaA protein with 10-F05 or DMSO, followed by the addition of the tRNA substrate.Bound and unbound tRNA substrate were separated using electrophoresis and visualized using Gel-Red staining, which were further quantified to calculate the unbound tRNA ratio.The unbound tRNA Trp _34 ratio increased from 64% to 87% after preincubation with 10-F05, in agreement with our hypothesis (Figure 6C and Figure 6D).As expected, the MiaA C273A mutant, which does not react with 10-F05, was not affected much by 10-F05 treatment in tRNA binding (Figure S12).
To examine whether the substrate blocking effect of 10-F05 could influence MiaA's activity, we performed an in vitro enzymatic activity assay. 37The reaction kinetics were measured by monitoring the i 6 A-37 tRNA product using LC-MS.As expected, preincubation of 10-F05 with MiaA decreased the initial velocity of modified tRNA forming rate by 1.7-fold using tRNA Trp _34 as the substrate (0.085 vs 0.049 µM/min, Figure 6E).We next investigated whether 10-F05 inhibits MiaA function in bacteria (Figure 7A).We utilized a dual-luciferase reporter plasmid to quantify translation fidelity change in miaA KO E. coli strains 25 .
Our results revealed an elevated translation error in the miaA KO strain compared to the WT strain, which aligns with previously reported results. 25Subsequently, we treated both the WT and miaA KO strains with 5 µM of 10-F05.The results indicated that 10-F05 increased the translation error in the WT strain but not in the miaA KO strain.This observation strongly suggests that 10-F05 reduces translation fidelity by inhibiting the activity of MiaA (Figure 7B).Interestingly, the measured translation error in 10-F05-treated WT E. coli cells was even higher than that in the miaA KO E. coli cells, likely because the miaA KO cells were somehow adapted to the increased translation error in the absence of i 6 A modified tRNA.
To further validate that 10-F05 inhibits MiaA in cells, we proceeded to purify the modified tRNA from E. coli and enzymatically digested it into individual ribonucleosides using established methods. 38,39bsequently, we quantified the presence of i 6 A and ms 2 i 6 A (from i 6 A by the action of MiaB) using LC-MS analysis (Figure 7C, top).The i 6 A modification level was extremely low and below the detectable limit of our LC-MS instrument.Nonetheless, we were able to detect a distinct peak corresponding to ms 2 i 6 A (same retention time as the synthetic standard of ms 2 i 6 A, Figure S13).Importantly, this peak was only observed in the WT strain but not in the miaA KO strain, confirming the absence of i 6 A modification in the latter.As expected, when we treated the E. coli strain (starting from ~10 8 inoculation) with 100 µM of 10-F05 for 3 hours, we observed a ~24% decrease in the level of ms 2 i 6 A compared to the DMSO control group (Figure 7C, bottom and Figure S13).This reduction in ms 2 i 6 A level further supports that 10-F05 inhibits the activity of MiaA in the bacteria.
Previous studies conducted in S. flexneri and E. coli demonstrated that the deletion of miaA led to a reduction in the expression of virulence factors and altered stress resistance (Figure 7A). 25,40,41Thus, we tested whether the pharmacological inhibition of MiaA by 10-F05 causes similar effects.The fabH, miaA, or pdxY KO strains exhibited no significant difference in growth compared to the WT strain when cultured in LB media (Figure 7D, top).However, the deletion of MiaA resulted in decreased E. coli growth rate under osmotic stress (Figure 7D, middle).The addition of sub-MIC concentration of 10-F05 completely shut down WT E. coli growth in 3% NaCl LB media but not in normal LB media, indicating that pharmacological inhibition of MiaA sensitized bacteria to osmotic stress (Figure 7D, bottom).
To evaluate the impact of MiaA inhibition on bacterial virulence, we pre-treated GFP-tagged S.  We further validated their physiological relevance in mediating the growth inhibitory activity of 10-F05 through chemical genetic interactions.
FabH has previously been indicated as an antibiotics target.Various chemical scaffolds, including 1,3,5-oxadiazin-2-one, 22 oxadiazolones, 4 platencin and its derivatives, 23,31 and benzoylaminobenzoic acid, 42 have been reported as inhibitors of FabH, through covalent or non-covalent interactions.However, despite their distinct structures, these FabH inhibitors have shown a loss of antibacterial activity in gram-negative bacterial strains due to effective drug efflux. 24,42The compound discovered here, 10-F05, is able to inhibit FabH in gram-negative bacteria.It is possible that its smaller size leads to better cell permeability, or its rapid target engagement may lead to a reduction in excretion levels through TolC, thereby retaining its potency against gram-negative bacteria.
Notably, MiaA and PdxY have not been recognized as antibiotics targets and no inhibitors targeting them have been reported to date.Therefore, 10-F05 represents a promising starting scaffold for the development of selective inhibitors for these new targets.Moreover, we validated that MiaA is relevant antibacterial target for 10-F05.Through examining published MiaA structures and experimental validation, we discovered that the covalent addition of 10-F05 to MiaA Cys273 disrupts its interactions with the tRNA substrates, inhibiting MiaA function and leading to reduced stress resistance and virulence.The 10-F05 MIC changes observed in strains overexpressing PdxY and PdxK further suggests an unidentified role for PdxY that is unrelated to PLP salvage.
Reactivity is a crucial factor to consider in covalent drug design as it influences off-target effects.
Traditional covalent drugs often employ weak electrophilic warheads to facilitate optimized covalent bond formation primarily within the desired binding complex.This minimizes off-target protein labeling.Our study here showed that many commonly used warhead scaffolds lack antibacterial activity against gramnegative bacterial strains.This could be attributed to their low affinity, as most compounds in the library are fragment-like, resulting in a low secondary second-order rate constant of inactivation (kinact/KI).In contrast, the highly reactive aromatic ring-conjugated chloromethyl ketones likely increases the chances of engagement of these fragment-like moieties.
We initially had concerns about the promiscuity of 10-F05 due to its high reactivity.However, our proteomic results demonstrated reasonable selectivity of 10-F05 in live bacterial cells.Furthermore, a previous study reported that promiscuity does not necessarily correlate with reactivity. 16Future experiments are necessary to explore the mechanism underlying the target selectivity of 10-F05.We also note that the ability of 10-F05 to target several bacterial proteins has a potential advantage -the bacteria are less likely to develop resistance to 10-F05 (Figure 3) compared to other common antibiotics tested (kanamycin and methicillin).This is because it is difficult for bacteria to accumulate mutations in all of the target proteins.
It is also important to note that the high reactivity of cysteine covalent drugs can result in a decreased stability for in vivo use.This is due to their rapid consumption by glutathione, which may explain the lower cytotoxicity of chloromethyl ketone scaffolds compared to chloroacetamide in mammalian cells.These findings highlight the complex role of reactivity that needs to be considered in covalent antibiotic design.
Further optimization of 10-F05 is required for developing clinical antibiotic candidates.Additionally, it is crucial to develop a more diversified covalent library with a broad range of reactivity to effectively target bacterial cysteines with various properties.

Conclusion
Our study demonstrates that combining covalent compound library phenotypic screening (bacterial growth) with ABPP is an efficient way to identify new protein targets (such as MiaA) as well as lead compounds.Traditional non-covalent compound library phenotypic screening is limited by the difficulty in identifying the protein targets of the hit compounds.The use of covalent compound library and ABBP chemical proteomics effectively overcomes this limitation.This approach therefore has the potential to open new research directions in drug discovery and chemical biology.

Figure 2 .
Figure 2. Suitable reactivity of warhead is important for antibacterial activity.(A) Thiol reactivity of compounds containing different warheads in the Cys-library measured using reduced DTNB.(B)

M90T (Figure
3B and 3C).S. flexneri and S. aureus were passaged daily with sub-MIC concentrations of 10-F05 or two known antibiotics (kanamycin and methicillin), and the surviving bacteria in the next day were used to measure the MIC values of the compounds.S. flexneri M90T exhibited a 2-fold increase in MIC to 10-F05 on day 5 (MIC = 5 µM, 2-fold change) and maintained less than 4-fold change of MIC after 30 days of sub-MIC daily passage.During this period, S. flexneri M90T developed higher resistance to kanamycin, with a maximum fold change in MIC of ≤ 8.In contrast, S. aureus gained resistance to methicillin with a MIC fold change higher than 200 after 5 days (MIC > 1 mg/mL).The resistance of S. aureus to 10-F05 occurred at a much slower rate, taking place on day 14 and finally reached a 25-fold change in MIC by day 16.The slow resistance development rate in both S. flexneri and S. aureus suggested a potential poly-pharmacological mechanism of 10-F05.Notably, the 10-F05 resistant S. aureus strains did not exhibit significant cross-resistance to other commonly used antibiotic classes, indicating a novel mechanism of action of 10-F05 (Figure 3D).

Figure 3 .
Figure 3. Antibacterial activity and resistance development of 10-F05.(A) Time-dependent killing of S. flexneri M90T (N=2).(B) Resistance development of S. flexneri M90T against 10-F05 and Kanamycin during daily serial passaging with sub-MIC concentrations.(C) Resistance development of S. aureus MSSA476 against 10-F05 and Methicillin during daily serial passaging with sub-MIC concentrations.(D) MIC fold change of 10-F05 resistant S. aureus MSSA476 against other classes of antibiotics.MIC values were reported in Supplementary table.

Figure 4 .
Figure 4. Competitive activity-based protein profiling identified FabH, MiaA and PdxY as targets of 10-F05.(A) Schematic of TMTpro-18plex-based workflow for mapping cysteine-drug interactions in live S. flexneri.(B) Heatmap analysis of the protein targets of 10-F05 and 10-L07.(C) Ligandability table of highlighted cysteines in B. Cysteine function is annotated based on Uniprot annotation.

Figure 5 .
Figure 5. Biochemical and chemogenetic validations suggested that FabH is the major target of 10-F05 mediated growth inhibition.(A) In-gel fluorescence labeling of three purified S. flexneri target proteins.Purified proteins were incubated with 10-F05 at indicated concentrations for 1 hour and labeled by 5-TMRIA to visualize the unbound cysteines.Uncropped gel images are shown in Supporting Information.(B) Time dependent in-gel fluorescence labeling of S. flexneri FabH protein.Purified proteins were incubated with 5 µM of 10-F05 for indicated time and labeled by 5-TMRIA to visualize the unbound

Figure 6 .
Figure 6.10-F05 disrupts MiaA and tRNA substrate binding.(A) In-gel fluorescence of purified WT and C273A MiaA with and without 10-F05 as the competitor.Uncropped gel images are shown in Supporting flexneri M90T (starting from ~ 3.2 × 10 8 inoculation) with 10-F05 for 1 h and then infected Hela cells.S. flexneri is an intracellular pathogen and its pathogenesis requires entering the host cells.We assessed changes in virulence by quantifying the number of intracellular bacteria.No growth inhibition, as indicated by the colony-forming ability, was observed after 1-hour treatment with 10 µM of 10-F05 (Figure 7E, top left).However, the average number of intracellular S. flexneri was significantly reduced following 10-F05 treatment (Figure 7E, top right and Figure S14).Similarly, an agar-based quantification demonstrated a significant decrease in the intracellular level of S. flexneri, indicating decreased virulence after 10-F05 treatment (> 100-fold change, Figure 7E, bottom).Overall, our findings revealed that the pharmacological inhibition of MiaA leads to a similar phenotype as miaA KO in bacterial pathogens, suggesting MiaA as apromising target that controls bacterial virulence and stress resistance.25,40

Figure 7 .
Figure 7. 10-F05 increases translation error through MiaA inhibition in vivo to decrease bacterial stress resistance and virulence.(A) Scheme showing MiaA's role in regulating bacteria stress resistance and virulence.(B) Translation error measured using -1 frame shift reporter plasmids transformed in WT (BW25113) and miaA KO E. coli strains.(C) Top, biosynthesis pathway of i 6 A and ms 2 i 6 A tRNA.Bottom,

Table 1 .
MIC values of hits with chloromethyl ketone warhead.

Table 1 .
The p values (unpaired two- tailed t-tests) were calculated in GraphPad Prism(version 9.3.1).****p < 0.0001.10-F05 is active against multiple ESKAPE pathogens and has slow resistance development.We carried further experiments with 10-F05 and found that 10-F05 inhibits multiple ESKAPE pathogens, including methicillin resistant S. aureus (MRSA), E. cloacae, K. pneumoniae, and E. coli.10-F05lost its activity in several multiple drug resistant Gram-negative strains (Table2, antibiotic spectrum for bacterial strains used is reported in Supplementary Data).A time-killing experiment was conducted using a starting inoculum of approximately 10 5 CFU/mL in LB medium.The results demonstrated that 2.5 µM of 10-F05

Table 2 .
MIC of 10-F05 against an expanded bacterial panel