Multivalent Display of Lipophilic DNA Binders for Dual-Selective Anti-Mycobacterium Peptidomimetics with Binary Mechanism of Action

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE7 Nov 2022Multivalent Display of Lipophilic DNA Binders for Dual-Selective Anti-Mycobacterium Peptidomimetics with Binary Mechanism of Action Jianxue Wang, Junfeng Song, Xianhui Chen, Rey-Ting Guo, Yingjie Wang, Guopu Huang, Nan Zheng, Peilei Hu, Xinxin Feng and Yugang Bai Jianxue Wang State Key Laboratory of Chem-/Bio-Sensing and Chemometrics, Hunan Provincial Key Laboratory of Biomacromolecular Chemical Biology, School of Chemistry and Chemical Engineering, Hunan University, Changsha, Hunan 410082 Google Scholar More articles by this author , Junfeng Song State Key Laboratory of Chem-/Bio-Sensing and Chemometrics, Hunan Provincial Key Laboratory of Biomacromolecular Chemical Biology, School of Chemistry and Chemical Engineering, Hunan University, Changsha, Hunan 410082 Google Scholar More articles by this author , Xianhui Chen State Key Laboratory of Chem-/Bio-Sensing and Chemometrics, Hunan Provincial Key Laboratory of Biomacromolecular Chemical Biology, School of Chemistry and Chemical Engineering, Hunan University, Changsha, Hunan 410082 Google Scholar More articles by this author , Rey-Ting Guo State Key Laboratory of Biocatalysis and Enzyme Engineering, School of Life Sciences, Hubei University, Wuhan, Hubei 430062 Google Scholar More articles by this author , Yingjie Wang Institute of Systems and Physical Biology, Shenzhen Bay Laboratory, Shenzhen, Guangdong 518055 Google Scholar More articles by this author , Guopu Huang State Key Laboratory of Chem-/Bio-Sensing and Chemometrics, Hunan Provincial Key Laboratory of Biomacromolecular Chemical Biology, School of Chemistry and Chemical Engineering, Hunan University, Changsha, Hunan 410082 Google Scholar More articles by this author , Nan Zheng Department of Polymer Science and Engineering, School of Chemical Engineering, Dalian University of Technology, Dalian, Liaoning 116024 Google Scholar More articles by this author , Peilei Hu *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] Hunan Institute for Tuberculosis Control, Hunan Chest Hospital, Changsha, Hunan 410013 Google Scholar More articles by this author , Xinxin Feng *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] State Key Laboratory of Chem-/Bio-Sensing and Chemometrics, Hunan Provincial Key Laboratory of Biomacromolecular Chemical Biology, School of Chemistry and Chemical Engineering, Hunan University, Changsha, Hunan 410082 Google Scholar More articles by this author and Yugang Bai *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] State Key Laboratory of Chem-/Bio-Sensing and Chemometrics, Hunan Provincial Key Laboratory of Biomacromolecular Chemical Biology, School of Chemistry and Chemical Engineering, Hunan University, Changsha, Hunan 410082 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.021.202101416 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail We made oligoamidine-based peptidomimetics highly specific for mycobacteria eradication by introducing and arraying lipophilic DNA binding motifs on macromolecular backbones. The short poly(amidino-phenylindole) (PAPI) structures feature an alternating amphiphilic structure with cationic, lipophilic DNA-binding moieties, enabling fast and selective eradication of mycobacteria through binary, membrane- and DNA-selective mechanisms of action. More importantly, PAPIs address the primary treatment challenge by combating mycobacteria in eukaryotic cells and working as a sensitizer for conventional antibiotics, in both ways promoting more thorough removal of pathogens and reducing the mycobacteria’s resistance generation rate during treatment. Structural optimization was achieved to counter specific pathogens, including Mycobacterium tuberculosis, in the Mycobacterium genus.One of the hit peptidomimetics was evaluated in a zebrafish-based aquatic infection model using Mycobacterium fortuitum and a mice tail infection model using Mycobacterium marinum, both revealing excellent in vivo performance. Download figure Download PowerPoint Introduction The Mycobacterium genus includes several pathogens responsible for serious diseases in fish, livestock, and humans,1,2 among them tuberculosis is one of the most serious human infectious diseases in the world,3 causing approximately 10 million new infection cases and 1.4 million deaths annually. As such, the responsible pathogen, Mycobacterium tuberculosis, is steadily ranked first worldwide in the cause of deaths from a single infectious agent before the COVID-19 pandemic.3 Mycobacteria infections require multidrug combination therapy and long treatment time and are prone to drug resistance,4,5 all characteristic of the Mycobacterium genus. The additional outer membrane, waxy and thick cell wall rich in mycolate,6 and slow reproduction of mycobacteria make them generally less sensitive toward many antibiotics in the “dormant state”.7 For many mycobacteria, the ability of becoming intracellular “persisters” makes them readily evade antimicrobial effects.7,8 Together these factors forge the hardiness of this genus, which significantly lengthens the treatment time of infection caused by any mycobacterium, including M. tuberculosis, Mycobacterium leprae,9Mycobacterium marinum,10Mycobacterium fortuitum,11 and others. Their hardiness also contributes to the development of antibiotic resistance that deteriorates the treatability. We have recently reported a resistance-resistant, oligoamidine-based antimicrobial peptidomimetic that kills pathogenic bacteria with dual-selective mechanisms of actions, namely membrane disruption and DNA binding.12 The compound also showed activity toward extracellular and intracellular Mycobacterium smegmatis, a non-pathogenic bacterium in the Mycobacterium genus, despite a weaker inhibitory effect against M. tuberculosis. However, designed chemical structure alteration can be used to advantageously adjust the molecular properties of peptidomimetics irrespective of the actual chemical approach utilized.13 Thus, we believe that there is room for improvement for those dual-mechanistic peptide mimics so that they can better combat pathogens in the Mycobacterium genus. Indeed, in our search for a better-performing antimicrobial oligoamidine, we noticed that the structure of the constitutional units of such peptidomimetics dramatically affected the antimicrobial performance of the final constructs. In addition, we have recently found and reported that changes in the constitutional units may lead to changes in the global amphiphilicity pattern that significantly affects antimicrobial performance and toxicity.14 Therefore, we began to seek alternative structures that could be more effective toward mycobacteria and found that the enhancement of a polymer’s lipophilicity while retaining the binary mechanism of action can be a practical and effective solution. Experimental Methods Materials and methods All reagents and organic solvents were purchased from Macklin Chemical (Shanghai, China), ADAMAS-BETA (Shanghai, China), J&K China (Beijing, China), Energy Chemical (Shanghai, China), or Merck China (Shanghai, China) and were used without further purification unless otherwise noted. Dulbecco’s modified Eagle’s medium and fetal bovine serum (FBS) were purchased from HYCLONE® (Logan, Utah, United States) and TIANHANG Biotech (Deqing, Zhejiang, China). Water was obtained on a Milli-Q purification system. Instrument information, additional synthetic procedures, and experiment protocols can be found in the Supporting Information. Special safety note All experiments involving the use of M. tuberculosis must be performed in a laboratory that meets the standards of Biosafety Level 3 or above. Representative polymerization protocol For the condensation copolymerization of monomers, AM1 (13 mg, 1 equiv) and PI2 (46 mg, 1 equiv) were mixed in a 7 mL glass vial containing dimethylformamide (DMF) (1.5 mL), and diisopropylethylamine (75 μL) was added. The mixture was heated at 45 °C with stirring for 4 days. Aqueous HCl (3.0 M, 2 mL) was added to the solution, and the resulting mixture was transferred into a dialysis bag (molecular weight cut-off = 1 kDa). The mixture was dialyzed against deionized water for 2 days, filtered, and lyophilized to obtain PAPI1-2. Other peptidomimetics were synthesized similarly using their respective monomers. Results and Discussion To enhance the peptidomimetics’ effectiveness against mycobacteria, structural adjustments should be made to counter the bacteria’s following unique defensive fortifications: (1) mycobacteria in the dormant state have a slow reproduction rate15 and low sensitivity toward many antibiotics,7 and (2) mycobacteria have thick and highly hydrophobic enclosures to protect themselves. As shown in Figure 1a, we have already demonstrated that the “membrane-plus-DNA” dual-targeting strategy was effective for those “dormant persisters”12; for the latter, a more hydrophobic antimicrobial agent may be a straightforward solution to gaining affinity toward mycobacteria. Thus, we targeted dual-mechanistic peptidomimetics with an emphasis on increased overall hydrophobicity. However, as all the antimicrobial peptidomimetics and polymers are essentially surfactant-like, incorporation of excess hydrophobicity may lead to conformational changes14 or aggregation, which are typically accompanied by eukaryotic cytotoxicity,16 serum instability, and loss of activity. The question ahead was how should highly hydrophobic motifs be introduced onto the peptidomimetics without increasing the synthetic difficulty and disrupting the beneficial amphiphilicity pattern? Figure 1 | (a) Illustration of the binary, dual-selective antimicrobial mechanism of action achieved by PAPIs. (b) Reported antimicrobial oligoamidine and oligoguanidine. (c) Preparation of peptidomimetics with enhanced lipophilicity from diamines and phenylindole bisimidates. (d) Probability distribution of the alternating hydrophilic (blue, amidinium moieties) and hydrophobic segments (orange, aromatic and alkyl moieties) on the backbone of PAPI1-2 (n = 5.5, 150 ns, isosurface N = 0.0025), predicted from the all-atom molecular dynamics simulation. The figure on the right shows the end-to-end distance fluctuation of the macromolecule, with the purple line indicating the average. Download figure Download PowerPoint The rigid and hydrophobic dsDNA-binding dye, diamidino phenylindole (DAPI), may provide an answer for the above question.17,18 DAPI analogs have already been reported to have antimicrobial and antifungal efficacy,19,20 although such an effect has a narrow-spectrum /accompanied by significant eukaryotic cytotoxicity. Importantly, DAPI is a bisamidine, which makes it possible to incorporate its analogs into our peptidomimetics using the same chemistry as our previously reported poly(phthalamidine) ( PPTA, Figure 1b) and similarly bring in a multivalent effect that can further enhance DNA binding affinity.21 Importantly, the DAPI moiety is apparently much more hydrophobic in nature than in the charged constitutional units of PPTA and poly(xylenediguanidine) ( PXDG, Figure 1b). The structure rigidity is highly beneficial in holding an alternating amphiphilic (AA) conformation for the resulting peptidomimetics,14 preventing unwanted aggregation and minimizing eukaryotic membrane disruption. In addition, the oligomerization of DAPI analogs into larger-sized molecules also restricts them from eukaryotic genetic DNA located in the envelop-protected nucleus, alleviating the associated undesired eukaryotic cytotoxicity.19 Thus, we postulated that a DAPI-incorporated peptidomimetic strategy may be a solution. Preparation, initial screening and evaluation Based on the above analysis, we prepared four different phenylindole bisimidates ( PI1– PI4, Figure 1c and Supporting Information Figures S1–S4) and polymerized them with five commercially available diamines ( AM1– AM5), yielding 20 different peptidomimetics bearing poly(amidino-phenylindole)s ( PAPIx-y as synthesized from AMx and PIy, Supporting Information Figures S5–S10). All had similar molecular weights of ca. 3–4 kDa (based on poly(ethylene glycol) calibration curve, Supporting Information Figure S11 and Table S1) and bore DAPI-like moieties for effective DNA binding. The charges of the amidinium moieties granted the peptidomimetics water solubility, and the alkyl chains on the diamines provided different degrees of hydrophobicity, so that these multivalent DNA binders also exhibited the structural feature of traditional, membrane-disruptive antimicrobial polymers,16,22,23 while making it possible to assess the effect of hydrophobicity incorporation. We screened these peptidomimetics using ESKAPE pathogens24 and M. smegmatis to evaluate their antimicrobial efficacy (Table 1) by calculating their therapeutic indices (TIs) using their minimum inhibitory concentrations (MICs), half maximal hemolytic concentrations (HC50) and half maximal inhibitory concentrations (IC50, HEK-293T). As shown in Table 1, the oligomerization had a profound effect, as all the peptidomimetics were potent and broad-spectrum against bacteria, unlike small molecular DAPI analogs that are more cytotoxic toward eukaryotic cells and are reported to be strong antifungal agents.19 Most importantly, although the PAPIs showed comparable anti-ESKAPE capability compared to our previously reported oligo(phenylenediamidine) series, most had significantly better anti-M. smegmatis MICs, supporting our hypothesis that the “membrane-plus-DNA” dual-targeting strategy may be beneficial for the anti-mycobacterium purpose. Overall, the substitution on the DAPI moiety had a smaller effect, whereas the diamine linker structure had a profound impact on the product’s performance. The two most hydrophobic amines ( AM1 and AM2) yielded PAPIs with the best performance ( PAPI1-x and PAPI2-x) on M. smegmatis, which supports our design hypothesis. The polymers’ antimicrobial effects on bacteria outside the Mycobacterium genus did not have a clear trend. Thus, we selected PAPI1-2 for most mechanistic and application studies, yet some other PAPIs also showed advantages against different mycobacteria (vide infra). Table 1 | MIC and TIs of Various PAPIs PAPI MIC (μg/mL) HC50 (μg/mL) IC50c (μg/mL) M. s TI (HC50/MIC) M. s TI (IC50/MIC) K. pneumoniae (−)a E. coli (−) A. baumannii (−) P. aeruginosa (−) S. aureus (+)b E. faecalis (+) M. smegmatis (+) 1-1 8 16 16 16 8 8 0.125 467 25 3739 204 1-2 4 2 8 8 2 2 0.0625 1063 22 17,006 354 1-3 4 2 8 8 1 2 0.0625 235 17 3762 268 1-4 4 4 16 8 2 4 0.125 566 28 4531 224 2-1 4 8 16 8 8 8 0.25 457 24 1828 96 2-2 4 8 16 8 4 8 0.125 470 26 3764 206 2-3 2 2 16 8 2 4 0.0625 <640 18 <10,240 282 2-4 4 8 16 16 8 8 0.25 <640 28 <2560 113 3-1 8 16 16 16 8 8 0.25 449 6.5 1796 26 3-2 8 8 8 8 2 8 0.25 499 7.8 1998 31 3-3 8 8 8 8 2 4 0.25 <640 6.2 <2560 25 3-4 8 16 16 16 8 8 0.25 <640 9.6 <2560 38 4-1 16 16 <32 8 8 8 16 3928 14 246 0.87 4-2 16 8 16 8 8 8 8 2910 3.7 364 0.46 4-3 16 8 16 8 8 8 4 996 3.4 249 0.85 4-4 8 16 32 8 16 16 8 1771 11 221 1.3 5-1 16 16 <32 16 16 4 8 <20,000 16 <2500 2.0 5-2 8 8 <32 8 8 8 2 ca. 20,000 13 ca. 10,000 6.5 5-3 4 8 16 8 4 2 0.5 ca. 20,000 21 ca. 40,000 42 5-4 16 4 <32 32 8 8 2 <20,000 14 <10,000 6.9 Different Gram-positive, Gram-negative bacteria, and mycobacteria strains were evaluated. aGram-negative bacteria. bGram-positive bacteria. cHEK-293T was used as the model cell. Conformation simulation of the hit peptidomimetic Molecular dynamics simulations of 150 ns characterized the conformational plasticity of PAPI1-2’s structure.14,25 As observed in the simulation movie (see Supporting Information Video S1), PAPI1-2 showed an extended conformation in water most of the time, indicating that its rigid aromatic linkages between amidinium moieties did encumber unwanted chain folding. The probability distribution of the hydrophobic moieties (aromatic and alkyl groups, orange) and the hydrophilic moiety (amidinium, blue) in Figure 1d clearly show that despite the “swaying” of the two ends, the hydrophobic and hydrophilic segments of the oligomer were alternatingly placed, forming a conformationally AA structure instead of folded structures or aggregates.26–31 The end-to-end distance was constantly fluctuating with an average of 7.44 nm (Figure 1d, right), indicating that the chain was intrinsically dynamic but extended. Consistent fluctuation of Rg was also observed in the simulation (2.50 nm on average, Supporting Information Figure S12). As such, benefiting from the rigidity of the phenylindole moieties, PAPI1-2 maintained an AA conformation despite the DNA-binding motifs, that is, the phenylindole bisamidines, that were considerably more lipophilic than the previously reported phthalamidine structure in PPTA and xylenediguanidine structure in PXDG. The importance of such enhanced lipophilicity will be discussed in a later section of this report. In vitro DNA binding of the peptidomimetic The AA conformation allows the oligomer to achieve effective multivalent display of DAPI analogs by avoiding unwanted chain collapse into random coils that have reduced net amphiphilicity32 and “wrapped” DNA binding motifs. Such a “binding moieties along a line” peptidomimetic design has already been successfully employed in pathogenic DNA/RNA binding,33 and we indeed observed enhancement of the DNA binding strength of PAPIs using propidium iodide (PI) fluorescent titration experiments (Figure 2a and Supporting Information Figure S13). The affinity of PAPI1-2 toward M. smegmatis genomic DNA (Ki = 2.8 μM against PI-DNA, per bisamidine moiety) was comparable to the oligoamidine we reported recently12 and was much higher than monomeric DAPI (Ki = 94 μM), indicating the successful enhancement in DNA binding strength through a multivalent approach. Yet, like the DAPI dye, PAPIs fluoresced upon binding to dsDNA (345 nm/455–510 nm, Supporting Information Figure S14), which facilitated the observation of their intracellular binding behavior. In addition, comparison of DNA condensation efficiency between PAPI1-2 and polyethyleneimine (2 kDa, PEI2k) unambiguously showed that PAPI1-2 had significantly higher binding strength at the same protonatable nitrogen to phosphate ratio (N/P ratio) (Figure 2b), suggesting that PAPI1-2 employed other means, presumably hydrogen bonding and hydrophobic forces, in addition to the electrostatic interactions in DNA binding solely employed by PEI2k. Consistently, PAPI1-2 prevented DNA bands from moving on gels by forming polyplexes at an oligomer to DNA (w/w) ratio greater than 1 (Figure 2c). The dynamic light scattering (DLS) study using PAPI1-2 and M. smegmatis genomic DNA confirmed the formation of such polyplexes ( Supporting Information Figure S15). Because of such DNA binding capability, PAPI1-2 partially or completely lost its antimicrobial activity when external DNA was added, as seen in Figure 2d, while non-DNA-targeting antibiotics, such as colistin, were not affected under the same condition. Figure 2 | (a) Improved DNA binding strength of PAPI1-2 compared to monomeric DAPI, as revealed in PI displacement titration experiments. (b) DNA condensation efficiency comparison between PAPI1-2 and PEI2k. (c) Plasmid DNA (pSV-pMD19-T) movement retardation effect of PAPI1-2. (d) Partial or complete loss of antimicrobial activity observed for PAPI1-2 when external M. smegmatis genomic DNA was added. NT = no treatment. (e) SEM images of M. smegmatis after being treated with various concentrations of PAPI1-2. (f) Flow cytometric analysis of M. smegmatis and RAW 264.7 cells treated with PI dye in the presence of various concentrations of PAPI1-2. PI could only enter bacteria cells, indicating a selective membrane disruption effect of PAPI 1-2. (g) Exploration of the MCF-7 cellular uptake mechanism of PAPI1-2. Chlorpromazine (CPZ), mβCD, and wortmannin are inhibitors of clathrin-mediated endocytosis, caveolae-mediated endocytosis, and macropinocytosis, respectively. ns, **, and *** indicate P < 0.05, P ≤ 0.01, and P ≤ 0.001, respectively, as determined by two-tailed Student’s t-tests. Download figure Download PowerPoint PAPI1-2 showed selective membrane disruption Based on a previous report,14 the peptidomimetic’s preference of maintaining such an AA conformation could be the key to achieving low eukaryotic cytotoxicity, and this was reflected in the high HC50 values measured for all PAPIs (Table 1). The HC10 of PAPI1-2 was <100 μg/mL, indicating that the compound had minimal impact on red blood cells at its antimicrobial concentration. Meanwhile, the oligomers’ disruptive interaction with the mycobacteria’s hydrophobic enclosure was promoted by the enhanced lipophilicity. Imaging by scanning electron microscopy (SEM) clearly revealed PAPI1-2’s disruptive effect on the M. smegmatis’s membrane (Figure 2e). The damage was more notable for samples treated with more concentrated PAPI1-2 solutions, which is consistent with the zeta potential measurements showing that the oligomer enriched on the bacterial surface ( Supporting Information Figure S16). Importantly, the potency of PAPI1-2 was not compromised by serum proteins. DLS analysis showed that PAPI1-2 did not cause aggregation of proteins at its MIC against M. smegmatis (0.0625 μg/mL) or S. aureus (2 μg/mL) in 10% or 50% FBS ( Supporting Information Figure S16), and consistently, the MIC of the compound against M. smegmatis did not show any change in 7H9 media containing 10% or 50% FBS in our assay studies. We used PI as an indicator to quantify the permeability of membranes. PI can only enter cells with a compromised membrane, thus its intracellular fluorescence would indicate a membrane disruption effect. Indeed, PAPI1-2-treated M. smegmatis (1 μg/mL and above) allowed the entry of PI (Figure 2f and Supporting Information Figure S17), indicating that the oligomer was membrane-targeting. In sharp contrast, PAPI1-2 did not show any membrane disruptive effect toward RAW 264.7 cells, even at concentrations up to 16 μg/mL (Figure 2f and Supporting Information Figure S18). Clearly, the membrane targeting effect of PAPI1-2 was highly selective toward M. smegmatis. In fact, the oligomer entered mammalian cells only via energy-dependent endocytosis pathways, mostly macropinocytosis, as reduced cell uptake was seen under various endocytic inhibitory conditions and mostly for wortmannin (Figure 2g). Thus, PAPI1-2 causes minimal damage to the mammalian cell membrane, and a large amount of it was found in lysosomes ( Supporting Information Figure S19). It should be noted that such direct and selective membrane targeting is an important feature for anti-mycobacterium agents, which targets persistent infections caused by mycobacteria that are slow-growing and not susceptible to many common antibiotics, because these antibiotics target biosynthetic pathways in growing cells.34 PAPI1-2 employed dual-selective binary mechanism against mycobacteria With its capability of selective membrane disruption and DNA binding, PAPI1-2 was expected to counter mycobacteria in a synergistic way, employing both mechanisms of action. As can be seen in Figure 3a, upon treatment with rhodamine-labeled PAPI1-2, M. smegmatis showed characteristic fluorescence under a confocal microscope. Upon channel merging, the red fluorescence from the rhodamine label was observed on the whole bacteria, whereas the cyan fluorescence, gained upon DNA binding by PAPI1-2, was only observed inside the bacteria. Such an observation strongly supports the dual mechanism of action by PAPI1-2, which was supposed to bind both membrane and DNA. In contrast, RAW 264.7 cells treated with rhodamine- PAPI1-2 showed no fluorescence from the DAPI moieties, and rhodamine’s red fluorescence could only be observed outside the nucleus ( Supporting Information Figure S20). This result clearly demonstrated that PAPI1-2 was targeting bacterial DNA in a highly selective manner, presumably because the PAPI structure is large enough to be excluded from the membrane-protected eukaryotic nucleus and chromosomal DNA.12,35 Thus, in eukaryotic cells such as RAW 264.7, gene expression levels of proteins, such as the key skeletal proteins actin and tubulin, were not significantly affected by the treatment with PAPI1-2 at much higher concentrations than its MIC against mycobacteria (Figure 3b). Figure 3 | (a) Confocal microscopy images of M. smegmatis after treatment with rhodamine-labeled PAPI1-2. (b) RT-PCR analysis showing that the transcription of two housekeeping genes in RAW 264.7 cells, actin and tubulin, was not affected by PAPI1-2. Transcription levels were normalized to non-treated cells using 18S as the standard. (c) Bacterium killing kinetics of PAPI1-2 and two common antibiotics against M. smegmatis, i.e., rifampicin (RIF) and ethambutol (EMB). (d) Flow cytometric analysis of M. smegmatis and RAW 264.7 cells treated with PAPI1-2 or EMB, using DCFH-DA as the ROS indicator. (e) Heatmap of differential expression analysis showing gene regulation changes in M. smegmatis treated with PAPI1-2, with three replicates for both control and treatment groups. (f) Volcano plot of the transcriptome results using PAPI1-2-treated M. smegmatis. −lgP: −log10 (P-value for the Log2 fold change of a certain gene), Log2 fold change: log2 (fold change of RPKM for a certain gene). (g) Comparison of resistance generation rate of M. smegmatis treated with PAPI1-2, RIF, or the combination of both. ns, *, **, *** indicate P < 0.05, P ≤ 0.05, P ≤ 0.01, and P ≤ 0.001, respectively, as determined by two-tailed Student’s t-test. Download figure Download PowerPoint With the combined efforts from two selective mechanisms, our peptidomimetic showed fast antimicrobial kinetics that was hard to achieve using traditional antibiotics and very low bactericidal concentration that was impossible for traditional antimicrobial polymers with membrane disruption as the sole working mechanism. PAPI1-2 showed a very fast antimicrobial effect on M. smegmatis at higher concentrations, killing all bacteria in 30 min at a concentration of 0.25 μg/mL and in 7 h at 0.125 μg/mL, significantly outperforming ethambutol (EMB) and rifampicin (RIF) in antimicrobial rate (Figure 3c, black and red curves). The fast kinetics is a characteristic feature of a membrane disruption-based antimicrobial mechanism. However, at low concentrations (62.5 ng/mL), PAPI1-2’s killing kinetics were similar to EMB (Figure 3c, blue and green curves). The minimal bactericidal concentration (MBC) measured for PAPI1-2 against M. smegmatis was 62 ng/mL (1.5 h) and 16 ng/mL (24 h), whereas EMB showed almost no activity in 1.5 h (MBC <32 μg/mL) and mediocre bactericidal activity in 24 h (MBC = 250 ng/mL). The compound’s strong bactericidal effect was also reflected in its stimulation of reactive oxygen species (ROS), as the production of ROS in cells is a response toward external factors that cause damage, and bactericidal antibiotics are known to stimulate the production of ROS.36,37 Using dichlorofluorescein diacetate (DCFH-DA) as the indicator, flow cytometric experiments showed that PAPI1-2 (2 μg/mL) caused significant ROS generation in M. smegmatis, whereas EMB did not at the same concentration (Figure 3d and Supporting Information Figure S18c). Importantly, PAPI1-2 did not cause ROS generation in mammalian cells (Figure 3d and Supporting Information Figure S18d), presumably because of its double-selective mechanisms of action; this observation is consistent with the oligomer’s high TI. The strong bactericidal yet eukaryote-friendly nature of PAPI1-2 is a very important feature for the treatment of mycobacteria-caused persistent infections. To better understand PAPI1-2’s mechanism of action, we performed a transcriptome study using PAPI1-2-treated M. smegmatis. The reads per kilobase of transcript per million mapped reads (RPKM) for each gene collected for the treated bacteria were compared with the data collected from the untreated bacteria ( Suppo