Investigation of siderophore–platinum(IV) conjugates reveals differing antibacterial activity and DNA damage depending on the platinum cargo

Abstract

The growing threat of bacterial infections coupled with the dwindling arsenal of effective antibiotics has heightened the urgency for innovative strategies to combat bacterial pathogens, particularly Gram-negative strains, which pose a significant challenge due to their outer membrane permeability barrier. In this study, we repurpose clinically approved anticancer agents as targeted antibacterials. We report two new siderophore–platinum(IV) conjugates, both of which consist of an oxaliplatin-based Pt(IV) prodrug (oxPt(IV)) conjugated to enterobactin (Ent), a triscatecholate siderophore employed by Enterobacteriaceae for iron acquisition. We demonstrate that L/D-Ent-oxPt(IV) (L/D-EOP) are selectively delivered into the Escherichia coli cytoplasm, achieving targeted antibacterial activity, causing filamentous morphology, and leading to enhanced Pt uptake by bacterial cells but reduced Pt uptake by human cells. D-EOP exhibits enhanced potency compared to oxaliplatin and L-EOP, primarily attributed to the intrinsic antibacterial activity of its non-native siderophore moiety. To further elucidate the antibacterial activity of Ent–Pt(IV) conjugates, we probed DNA damage caused by L/D-EOP and the previously reported cisplatin-based conjugates L/D-Ent-Pt(IV) (L/D-EP). A comparative analysis of these four conjugates reveals a correlation between antibacterial activity and ability to induce DNA damage. This work expands the scope of Pt cargos targeted to the cytoplasm of Gram-negative bacteria via Ent conjugation, provides insight into the cellular consequences of Ent–Pt(IV) conjugates in E. coli, and furthers our understanding of the potential of Pt-based therapeutics for antibacterial applications.

Graphical Abstract

Infectious diseases are among the top ten causes of death globally.¹ Recently, increased attention has been given to developing new antibiotics against bacterial pathogens, especially Gram-negative species, which are intrinsically difficult to treat due to the presence of an outer membrane (OM).² To overcome the permeability barrier of the OM, research efforts focused on understanding OM structure and function have revealed innovative strategies for drug delivery.³ For example, characterization and exploitation of uptake systems that are critical for the establishment and progression of infection have provided new approaches to antibiotic development.³

The siderophore-based “Trojan-horse” strategy involves leveraging bacterial acquisition pathways for the essential nutrient iron (Fe) to deliver drugs into bacterial cells.⁴ Bacteria utilize siderophores, small molecules that chelate Fe(III), to obtain this essential transition metal nutrient under Fe-limiting conditions, such as during infection.^5–7 The molecular recognition of siderophores by their cognate membrane-associated uptake machinery can be hijacked for efficient and targeted drug delivery, which has been demonstrated by the natural occurrence of sideromycins and class IIb microcins, as well as synthetic siderophore–antibiotic conjugates (SACs).^{4, 8–15} Inspired by natural SACs, there have been decades of studies on developing synthetic SACs as new antibacterial agents with improved potency and narrow-spectrum antibacterial activity, including imparting Gram-negative antibacterial activity to traditional Gram-positive antibiotics and repurposing other therapeutic agents for antibacterial purposes (e.g. using the anti-malarial drug artemisinin to target Mycobacterium tuberculosis).¹⁶ In 2019, cefiderocol, a catecholate cephalosporin developed by Shionogi, received FDA approval for the treatment of complicated urinary tract infections.¹⁷ Our laboratory has designed SACs using enterobactin (Ent), one of the best-studied bacterial siderophores with a remarkably high affinity for Fe (K_a ~ 10⁴⁹ M⁻1),¹⁸ as the delivery vector.¹⁹ Previously, we conjugated Ent with β-lactam antibiotics (Ent-β-lactams) and a fluoroquinolone (Ent-ciprofloxacin) and demonstrated that Ent can efficiently deliver drug cargos into the periplasm and cytoplasm of Gram-negative bacteria expressing the Ent transport machinery.^20–23 Notably, Ent-β-lactams achieve 1000-fold enhanced antibacterial activity against E. coli and Salmonella compared to the parent antibiotics, and Ent-ciprofloxacin selectively targets pathogenic E. coli with comparable activity to the unmodified drug.

Beyond traditional antibiotics, we recently evaluated the Ent-based SAC approach for drug repurposing.²⁴ We focused on the FDA-approved Pt-based anticancer agent cisplatin because studies of Pt therapeutics originated from the antibacterial activity and mode of action of cisplatin in bacterial cells.^25–28 In recent years, there has been a renewed interest regarding the antibacterial effects of Pt agents.^{27, 29–32} We took inspiration from the strategy of Pt(IV) prodrugs, which have been widely studied for the development of safe and targeted Pt anticancer therapy.^33–35 Pt(IV) complexes are known for their kinetic inertness compared to Pt(II) compounds, making them good candidates for masking Pt(II) toxicity prior to cell entry.³³ In a proof-of-concept study, we designed and evaluated L/D-EP 1,2 (Figure 1A), a pair of conjugates consisting of a cisplatin-based Pt(IV) prodrug attached to either the natural siderophore L-Ent or to its enantiomer D-Ent. L/D-EP are delivered via Ent transport machinery into the E. coli cytoplasm, where the Pt(IV) prodrug is presumably activated by the reducing environment, leading to the release of cisplatin that causes growth inhibition and cell filamentation (Figure 1B).²⁴ We found that D-EP showed greater antibacterial activity than L-EP. We attributed this increased potency to the D-Ent moiety directing conjugate delivery through the Ent transport machinery at levels similar to L-EP while preventing Fe release for bacterial metabolic use, as D-Ent is not a substrate for the cytoplasmic L-Ent esterases (Figure 1B).³⁶ Lastly, we showed that Ent conjugation results in markedly reduced Pt uptake by human cells and markedly enhanced Pt uptake by bacterial cells as compared to cisplatin. Overall, Ent modification repurposed the generally toxic anticancer agent cisplatin as a targeted antibiotic and set the stage for further investigations into Ent-based Pt antibacterials.²⁴

Figure 1.

(A) Chemical structures of previously reported cisplatin-based L/D-Ent-Pt(IV) conjugates (L/D-EP) 1,2²⁴ and oxaliplatin-based L/D-Ent-oxPt(IV) conjugates (L/D-EOP) 3,4 described in this work. Chemical structures of cisplatin and oxaliplatin are highlighted in blue. (B) Cartoon depiction of the working model for repurposing Pt-based anticancer agents as antibiotics that selectively target Gram-negative bacteria via Ent transport machinery, based on prior work of L/D-EP.²⁴ Ent transport and processing machinery is shown for E. coli K12. The Fe reductase that facilitates Fe release is not shown. DHBS, 2,3-dihydroxybenzoyl serine; G, a guanine base.

Siderophore-based drug repurposing for antibacterial purposes is gaining considerable attention as illustrated by an independent report on repurposing the anticancer agent methotrexate as an antibiotic against Gram-positive and Gram-negative bacteria.³⁷ Along these lines, we are motivated to generalize Ent-based drug repurposing by expanding this application to other metal-based anticancer agents, including the globally approved Pt agents carboplatin and oxaliplatin.³⁸ Carboplatin shares a similar spectrum of activity with cisplatin but has overall lower toxicity,^39–40 and oxaliplatin is used for treating cancers where cisplatin has minimal efficacy due to its distinct spectrum of antitumor activity and no reported cross-resistance with cisplatin and carboplatin.^41–42 Intrigued by the unique features of oxaliplatin, we selected to evaluate oxaliplatin as a new cargo for Ent–Pt(IV) conjugates.

In this study, we report two new Ent–Pt(IV) conjugates harboring an oxaliplatin(IV) prodrug cargo (L/D-EOP 3,4, Figure 1A) and investigate their effects on the growth and morphology of E. coli. Our work demonstrates that L/D-EOP are selectively imported into the E. coli cytoplasm via the Ent transport machinery, enhancing Pt uptake by bacterial cells relative to oxaliplatin and inducing cell filamentation. Compared to the previously reported cisplatin-based conjugates L/D-EP, the antibacterial activities of L/D-EOP are lower. To further understand the differing antibacterial activities of Ent–Pt(IV) conjugates, we performed a comparative analysis of the DNA damage caused by L/D-EOP and L/D-EP and found that the antibacterial activity of Ent-Pt(IV) conjugates largely depends on the intrinsic potency of the Pt cargo and correlates with the extent of DNA damage. Taken together, this work expands the scope of the SAC approach for repurposing anticancer agents as antibiotics and demonstrates the dependence of cellular consequences on the molecular identity of the Pt cargo. These findings offer valuable insights for guiding the future design of Ent–Pt(IV) conjugates for the purpose of antibacterial therapeutics.

Results

Design and synthesis of L/D-EOP.

Based on our prior studies of Ent–antibiotic conjugates and L/D-EP 1,2, the oxaliplatin-based conjugates L/D-EOP 3,4 were designed with an oxaliplatin-based Pt(IV) prodrug attached to the Ent scaffold through a poly(ethylene glycol) (PEG3) linker at the C5 position of one catechol moiety (Figure 1A and Scheme 1).^19–24 We chose dicarboxylato Pt(IV) complexes because they have been widely studied for targeted chemotherapy and shown to exhibit favorable stability in biological fluids and effective reductive activation upon cellular entry.^43–44

The syntheses were carried out via copper-catalyzed azide-alkyne cycloaddition (CuAAC) from L/D-Ent-PEG₃-N₃ 6,7¹⁹⁻²⁰ and the alkyne-functionalized oxaliplatin(IV) precursor oxPt(IV)-alkyne 5. The CuAAC reaction was first performed based on the procedure for L/D-EP.²⁴ Specifically, the Cu(I) salt Cu(MeCN)₄PF₆ was used as the catalyst, tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA) was used as the Cu-binding ligand to accelerate the reaction, and a mixture of water and DMF instead of DMSO (used in syntheses of Ent-β-lactam conjugates²⁰) was used to avoid potential ligand exchange of Pt center with the nucleophilic sulfur atom of DMSO. Following HPLC purification, L/D-EOP 3,4 were obtained with a yield of ~16%, which was slightly higher than that of L/D-EP.²⁴

We routinely synthesize Ent-β-lactams from L-Ent-PEG₃-N₃ 6 and alkyne-modified β-lactams using Cu(II) and sodium ascorbate (NaAsc) to generate Cu(I) in situ.^{20, 22, 45} Although this approach affords Ent-β-lactams in moderate to high yields (40–80%),^{20, 22} we did not employ it in the synthesis of L/D-EP due to the concern of potential Pt(IV) reduction caused by NaAsc.^{40, 46–49} During attempts to optimize the CuAAC reaction described above that afforded L/D-EOP 3,4, we reconsidered in situ reduction of Cu(II) and evaluated the stability of oxPt(IV)-alkyne 5 in the presence of NaAsc in water and DMF. We found that oxPt(IV)-alkyne 5 remained stable throughout the 8 h incubation with NaAsc. Consequently, we coupled oxPt(IV)-alkyne 5 to L/D-Ent-PEG₃-N₃ 6,7 using NaAsc and CuSO₄ in a mixture of water and DMF with the addition of TBTA. We detected negligible quantities of the dissociated Ent-containing axial ligand during HPLC purification, suggesting that a small portion of L/D-EOP 3,4 was reduced under these reaction conditions. Nevertheless, the reaction reached completion within 20 min, affording L/D-EOP 3,4 with an improved yield of 41% and high purity following HPLC purification (Figure S1).

D-EOP exhibits enhanced antibacterial activity against E. coli that depends on Ent transport machinery.

We first evaluated the stability of D-EOP in a modified M9 growth medium supplemented with thiamine, which was employed in prior studies of Ent-based SACs including L/D-EP.^{21–24, 45} The low Fe content of this medium (600–700 nM Fe by ICP-MS) induces bacterial siderophore biosynthesis and transport machineries, and thiamine is added to support the growth of E. coli CFT073 mutants.²⁴ We found that D-EOP was adequately stable in this medium for microbiology experiments, showing stability similar to what we previously observed for L-EP (Supporting Discussion). Given the low Fe content of the medium and micromolar concentration range of L/D-EOP used throughout the biological studies, L/D-EOP were preloaded with 0.9 equivalents of Fe(III) prior to each experiment as we have done previously.²⁴

In contrast to cisplatin, which was discovered for its ability to inhibit cell division and induce filamentation in E. coli,^25–26 there is limited information about how oxaliplatin impacts bacterial growth. One recent report published during the course of this investigation showed that oxaliplatin has lower antibacterial activity than cisplatin against E. coli MG1655, a close K12 derivative.⁵⁰ We also evaluated the antibacterial activity of oxaliplatin against two E. coli strains by monitoring culture turbidity, including the laboratory strain BW25113 (hereafter K12)⁵¹ and the uropathogenic clinical isolate CFT073.⁵² Our results were consistent with the recent report.⁵⁰ We found that oxaliplatin exhibited negligible growth inhibition against both E. coli K12 and CFT073 at concentrations up to 60 μM (Figure 2). By contrast, cisplatin significantly reduced the turbidity of K12 and CFT073 cultures at ≥30 μM.²⁴

Figure 2.

Antibacterial activities of oxaliplatin and L/D-EOP against (A) E. coli K12 and (B) E. coli CFT073. All assays were performed in modified M9 medium (11 h, 30 °C with shaking; mean ± standard deviation, n = 8).

We assayed the antibacterial activity of L/D-EOP against K12 and CFT073. We found that L-EOP treatment resulted in no antibacterial activity and a slight growth promotion in both strains (Figure 2). This observation could be attributed to the combined effects resulting from co-delivery of a Pt payload with limited potency and nutrient Fe following hydrolysis of the L-Ent moiety by cytoplasmic esterases. The result is consistent with our prior study where L-EP was less potent than cisplatin and also our early observation that Ent conjugates with non-toxic cargos provided growth recovery of E. coli.^{19, 24} By contrast, the enantiomer D-EOP showed modest growth inhibitory activity, and treatment of E. coli K12 and CFT073 with 7.5 μM D-EOP reduced culture turbidity by 50% compared to the untreated control (Figure 2). This result indicates that conjugation of oxaliplatin to D-Ent provides more antibacterial activity than oxaliplatin alone.

Based on our design, L/D-EOP are actively imported into the E. coli cytoplasm by the Ent transport machinery expressed by all E. coli strains, including the OM receptor FepA, the periplasmic binding protein FepB, and the inner membrane (IM) ATP-binding cassette transporter FepCDG.²⁴ E. coli CFT073 harbors the pathogen-associated iroA gene cluster which encodes an OM receptor named IroN that also transports Ent.⁵³ To examine the uptake pathway, we evaluated whether key components of the Ent transport machinery are required for D-EOP antibacterial activity against E. coli.

We first investigated the role of the OM receptor FepA by treating K12 and its ΔfepA mutant (JW5086 from the Keio collection)⁵¹ with 7.5–60 μM D-EOP. In contrast to the growth inhibition of D-EOP against the parent strain, the ΔfepA mutant was not inhibited by D-EOP at all tested concentrations (Figure 3A), indicating that FepA is required for the uptake of D-EOP into E. coli K12. We note that the K12 ΔfepA mutant exhibits a growth defect in the modified M9 medium and appears to be very sensitive to any slight change in Fe content. Given the similar stability of L/D-EOP to that of L/D-EP and our previous studies of L/D-EP,²⁴ the growth promotion observed for K12 ΔfepA treated with D-EOP was attributed to partial decomposition of D-EOP, which releases Fe into the growth medium providing this nutrient to the bacteria. We also evaluated the antibacterial activity of D-EOP against the CFT073 ΔfepA ΔiroN mutant. D-EOP inhibited the growth of this mutant (Figure 3B), which was expected based on prior studies with Ent-β-lactams and L/D-EP.^{22, 24} This result may reflect the fact that CFT073 has additional receptors for Fe(III)-Ent (e.g., IhA)^54–55 or possibly unknown OM receptors for Ent and its hydrolysis products.^{22, 24} As before, we speculate that one or more of these OM receptors also transports D-EOP and other Ent-based conjugates (vide infra).

Figure 3.

Antibacterial activity of D-EOP against (A) E. coli K12 and its ΔfepA mutant; (B) E. coli CFT073 and its ΔfepA ΔiroN, and ΔfepC mutants. All assays were performed in modified M9 medium (11 h, 30 °C with shaking; mean ± standard deviation, n ≥ 5).

We then studied whether the IM transporter FepCDG is necessary for the antibacterial activity of D-EOP. We focused these studies on the CFT073 ΔfepC mutant because the K12 ΔfepC, ΔfepD, and ΔfepG mutants showed severe growth defects in the modified M9 medium. In contrast to the parent strain, no growth inhibition was observed for the CFT073 ΔfepC mutant after treatment with up to 60 μM D-EOP (Figure 3B), indicating that the Ent IM transporter is required for the antibacterial activity of D-EOP.

Considering that oxaliplatin treatment had a negligible effect on E. coli growth (Figure 2), we decided to investigate the effect of the D-Ent moiety on E. coli growth and treated K12 and CFT073 with D-Ent, D-EOP, and D-EP pre-loaded with 0.9 equivalents of Fe(III) (Figure S2A,B). We found that over the concentration range of 7.5–60 μM, D-EP exhibits the highest potency with ≥7.5 μM D-EP fully reducing the turbidity of E. coli K12 and CFT073 cultures to the baseline, which was consistent with our prior report.²⁴ When comparing the growth inhibitory activities of D-EOP and D-Ent, the potency of 7.5 μM D-Ent was comparable to or somewhat lower than that of 7.5 μM D-EOP against CFT073 or K12, respectively; however, higher concentrations of D-Ent (15–60 μM) resulted in substantially greater growth inhibition of both strains. We also examined a lower concentration range of 0.001–10 μM (Figure S2C,D). As expected, D-EP exhibited the greatest potency, with 1 μM D-EP almost fully reducing the culture turbidity of both E. coli strains to the baseline (OD₆₀₀ = 0.07 compared to 0.2 for the untreated control). Treatment with ≤ 1 μM D-EOP caused comparable or greater growth inhibition against CFT073 or K12 relative to that caused by ≤ 1 μM D-Ent, respectively. Both E. coli strains were inhibited to a greater extent in the presence of 10 μM D-Ent compared to 10 μM D-EOP. These observations suggest that oxaliplatin is not highly potent against E. coli and that the intrinsic activity of the D-Ent moiety contributes to the growth inhibitory activity of high concentrations of D-EOP (>7.5 μM).

We then questioned whether and how the D-Ent moiety contributed to the growth inhibitory activity of D-EOP, in addition to mediating active transport of the oxaliplatin warhead. At high D-Ent concentrations, pre-loading with 0.9 equivalents of Fe(III) would result in a non-negligible (micromolar) quantity of apo chelator in the culture medium that we reasoned could scavenge Fe(III) in the medium and cause growth inhibition. We tested this notion by preloading D-Ent and D-EOP with 1.0 equivalent of Fe and found that their antibacterial activities were attenuated but some activity of D-Ent remained (Figure S3). This analysis provides a cautionary tale regarding potential effects of Fe starvation when high concentrations of D-Ent are employed in growth studies as well as the possibility of antibacterial activity from a non-native siderophore moiety. With these caveats in mind, we then proceeded to investigate the cellular effects of the oxaliplatin warhead.

Bacterial morphologies indicate cytoplasmic delivery of L/D-EOP and activity of the Pt warhead.

Many Pt compounds slow bacterial cell division and produce a filamentous phenotype, an indicator of Pt-induced DNA damage.^{27, 56} We observed this phenomenon during studies of L/D-EP and concluded that turbidity measurements are insufficient to describe Pt-induced growth defects and microscopy serves as a better method for investigating the cellular uptake of Pt-containing conjugates.²⁴ To gain further evidence for cytoplasmic delivery by the Ent uptake system and Pt-based activity, we investigated bacterial morphological changes following treatment with oxaliplatin and L/D-EOP by bright-field microscopy. We also examined cell viability by fluorescence microscopy after LIVE/DEAD staining, which distinguishes cells with intact or compromised OM using the fluorescent dyes SYTO 9 (green) and propidium iodide (red), respectively.⁵⁷

To our knowledge, the effects of oxaliplatin treatment on bacterial morphology have not been described. We first treated E. coli with 7.5–60 μM oxaliplatin and observed that oxaliplatin induced filamentation (cell perimeter 20–50 μm) in both K12 and CFT073 cells. Filamentation occurred starting at 15 μM oxaliplatin and LIVE/DEAD staining showed that most elongated cells were live at this concentration (Figure S4). At higher concentrations of oxaliplatin (30 and 60 μM), long filaments (cell perimeter >50 μm) formed, and LIVE-DEAD staining revealed a mixture of live and dead filamentous cells (Figures 4A and 4C). We previously reported that cisplatin induced filamentation in both K12 and CFT073 starting at a lower concentration (7.5 μM) and that filaments induced by cisplatin were generally longer than those induced by an equivalent concentration of oxaliplatin within the range of 7.5–60 μM (Figure S4).²⁴ Overall, oxaliplatin induces filamentous morphologies like cisplatin and other Pt compounds,^{24, 27} but appears to be less potent than cisplatin. We note that the prodrug precursor oxPt(IV)-alkyne 5 did not induce filamentation in any strains evaluated in this study (Figure S5), which was presumably due to the slow cellular entry by passive diffusion.⁵⁸

Figure 4.

Representative bright-field and fluorescence micrographs of E. coli (A) K12, (B) K12 ΔfepA, (C) CFT073, and (D) CFT073 ΔfepC treated with oxaliplatin and L/D-EOP (scale bar = 10 μm). All assays were performed in modified M9 medium (11 h, 30 °C with shaking). Bacterial cells after treatment were stained with SYTO 9 (viable, green) and propidium iodide (dead, red).

We then investigated the effects of L/D-EOP on bacterial morphology. We found that L-EOP was more potent than oxaliplatin in inducing filamentation in both K12 and CFT073. Long filaments were observed at all tested concentrations (7.5–60 μM, Figure S6), which were longer than those induced by oxaliplatin (Figures 4A, 4C, and S6). D-EOP also induced long filamentation in both K12 and CFT073 at all tested concentrations (Figures 4A, 4C, and S7). Consistent with the enhanced antibacterial activity of D-EOP, the filaments induced by D-EOP were relatively short compared to those induced by the same concentration of L-EOP, and most filamentous cells were dead as confirmed by LIVE/DEAD staining (Figure S7). We also examined the effect of D-Ent on E. coli morphology and found that the cells were predominantly normal sized following treatment of K12 or CFT073 cultures with D-Ent (Figure S8). These observations indicate that the oxaliplatin warhead in L/D-EOP was delivered in the cytoplasm and exerted activity in E. coli.

We also examined bacterial mutants defective in the Ent transport machinery. Elongation or filamentation was induced by oxaliplatin in K12 ΔfepA, CFT073 ΔfepA ΔiroN, and CFT073 ΔfepC (Figure S4). By contrast, L/D-EOP did not induce filamentation in the K12 ΔfepA or CFT073 ΔfepC mutants (Figures S6 and S7). Most of the K12 ΔfepA and CFT073 ΔfepC cells were normal-sized and viable after treatment with L/D-EOP (Figure 4B and 4D), consistent with the abolished antibacterial activity of D-EOP against these mutants (Figure 3). We note that CFT073 ΔfepA ΔiroN also exhibited filamentous morphology after L/D-EOP treatment (Figures S6 and S7), in agreement with the results of our growth experiment (Figure 3B). Taken together, we conclude that the OM receptor FepA and the IM transporter FepCDG are responsible for the uptake of L/D-EOP into E. coli cytoplasm, indicating that L/D-EOP target Gram-negative bacteria expressing the Ent uptake machinery.

Ent conjugation enhances Pt uptake by bacterial cells while reducing that by human cells.

It is generally believed that Pt(II) anticancer agents like cisplatin and oxaliplatin primarily enter mammalian cells through passive diffusion, with active transport (e.g., the organic cation transporter and the copper transporter 1) as a secondary pathway.^59–60 Ent conjugation is proposed to decrease Pt uptake by human cells, as L/D-EOP are too large for passive diffusion, and human cells lack machinery for Ent active transport, and facilitate Pt uptake by E. coli cells through active transport.²⁴ In support of this notion, studies of Pt uptake by E. coli CFT073 revealed a significant enhancement in Pt uptake for L/D-EOP compared to oxaliplatin and the prodrug precursor oxPt(IV)-alkyne 5. Specifically, Pt uptake was 37-fold and 28-fold greater for E. coli CFT073 when treated with L-EOP (15.7%) or D-EOP (11.6%), respectively, compared to oxaliplatin (0.4 %) (Figure 5A). L-EOP facilitates Pt uptake to a slightly greater extent than D-EOP, but the overall levels are similar, which is consistent with our prior study of L/D-EP.²⁴ By contrast, HEK293T cells showed negligible Pt uptake for L/D-EOP and oxPt(IV)-alkyne (0.02–0.03%), which is ~7-fold lower compared to oxaliplatin treatment (0.14%) (Figure 5B). Collectively, L/D-EOP facilitated targeted Pt uptake by E. coli but lead to negligible Pt uptake by human cells. These results are in agreement with our model and studies of L/D-EP.

Figure 5.

Pt uptake in (A) E. coli CFT073 and (B) HEK293T cells treated with 1 μM oxaliplatin, L/D-EOP, and oxaliplatin-alkyne 5. E. coli CFT073 cells were treated for 30 min in modified M9 medium at 30 °C (n = 4), and HEK293T cells were treated for 6 h in DMEM+1% penicillin/streptomycin at 37 °C, 5% CO₂ (n = 3).

L/D-EOP induce lysis in lysogenic bacteria.

So far, our data have shown that L/D-EOP are less growth inhibitory against E. coli compared to L/D-EP,²⁴ which correlates with the relative antibacterial activities of the oxaliplatin and cisplatin warheads. Moreover, we found that the D-Ent moiety also contributes to the activity of D-EOP at relatively high conjugate concentrations. To further investigate the differing antibacterial activities of Ent–Pt(IV) conjugates, we decided to study their cellular fates in E. coli by probing DNA damage as this activity is suggested by the observed cell filamentation (Figure 4). We started by examining the abilities of oxaliplatin and L/D-EOP to induce lysis in lysogenic bacteria, which is an experimental approach utilized in early and recent studies of Pt compounds.^{24, 27, 61–62} Briefly, lysogenic bacteria are bacteria harboring a prophage, the genetic information of a bacterial virus (bacteriophage λ).⁶¹ The expression of viral genes is repressed under normal conditions but is induced in response to DNA damage, leading to the production of viral particles and ultimately cell lysis. We treated the lysogenic bacterial strain E. coli W3104 with the three Pt compounds and spotted the resulting cell suspensions on a lawn of nonlysogenic E. coli CFT073. Pt-induced DNA damage triggers the production of viral particles, which are released from lysed E. coli W3104, preventing the growth of E. coli CFT073 and resulting in plaque formation on the lawn.²⁷

We observed plaques on a lawn of CFT073 when spotted with a 100-fold diluted suspension of E. coli W3104 that was treated with ≥15 μM oxaliplatin (Figure S9). When we treated E. coli W3104 with 15 μM L/D-EOP, 15 μM L-Ent, or 1% DMSO (control), we observed plaque formation on CFT073 only when W3104 was treated with 15 μM L/D-EOP at 100-fold dilution (Table 1 and Figure S9). No plaques formed in the areas where the compound solutions were spotted, indicating that the plaque formation was due to the release of viral particles from lysed W3104 cells. Overall, the ability of L/D-EOP to induce bacterial filamentation (Figure 4) and initiate lysis in lysogenic bacteria suggests that oxaliplatin and L/D-EOP cause DNA damage in E. coli.

Table 1.

Effect of L/D-EOP on lysogenic E. coli^a

	oxaliplatin	L-EOP	D-EOP	L-Ent	1% DMSO	Untreated
Ec W3104 suspension	+	+	+	−	−	−
Compound solution	−	−	−	−	−	−

^aThe development of plaques in a lawn of non-lysogenic E. coli CFT073 following spotting 100-fold diluted suspensions of E. coli W3104 treated with 15 μM of each compound or solutions containing only the corresponding compound. +, plaque formation; –, no plaque observed. Representative images of agar plates are shown in Figure S8. L-Ent and L/D-EOP were preloaded with 0.9 equivalents of Fe(III).

A comparison of conditions for plaque formation resulting from L-EP²⁴ and L/D-EOP treatment of W3104 revealed two noteworthy differences. First, plaque formation occurred with a higher-fold dilution of W3104 treated with 15 μM L-EP compared to 15 μM L/D-EOP (1000-fold and 100-fold, respectively). Second, plaque formation occurred after W3104 was treated with apo L-EP, whereas L/D-EOP treatment afforded plaque formation only when the conjugate was preloaded with Fe(III). The Ent OM receptors recognize and transport ferric Ent and consequently ferric Ent–Pt(IV), and we reason that treatment with 15 μM apo conjugate results in <1 μM conjugate available for efficient transport into W3014 cells in the Fe-deficient M9 medium (<1 μM Fe). This small quantity of ferric L-EP led to the production of sufficient viral particles for plaque formation; however, the same quantity of L/D-EOP could not yield enough viral particles. Overall, the results from lysogenic bacteria assays indicate that Ent–Pt(IV) conjugates harboring both cisplatin and oxaliplatin induce DNA damage in E. coli. However, the data also suggest that L/D-EP induce more DNA damage to bacterial cells compared to L/D-EOP.

L/D-EP cause more DNA damage in E. coli compared to L/D-EOP.

To further investigate the ability of L/D-EOP and L/D-EP to cause DNA damage in E. coli, we constructed a reporter strain by transforming E. coli JW0334, a ΔlacY mutant from the Keio collection,⁵¹ with a reporter plasmid pSulAp_lacZ. This reporter plasmid places control of the expression of lacZ under the SOS-inducible promoter for sulA. In the SOS regulon, sulA encodes the cell division inhibitor SulA, which is closely associated with the inhibition of cell division and the consequent filamentation when DNA is damaged.^63–64 The sulA promoter has been employed in reporter systems for the SOS response and has been shown to be induced efficiently upon DNA damage.^65-66 The reporter strain E. coli JW0334 pSulAp_lacZ allowed us to compare the extent of DNA damage based on levels of the SOS response induced by each Pt compound, which was determined by measuring the reporter activity using a β-galactosidase activity assay.

We first validated that E. coli JW0334 pSulAp_lacZ was a selective reporter for the SOS response by treating the strain with a sub-MIC concentration of ciprofloxacin, an SOS-inducing agent, or tetracycline, a non-SOS-inducing agent. Following ciprofloxacin treatment, the reporter activity increased in a time- and concentration-dependent manner (Figure S10A). By contrast, tetracycline did not affect the reporter activity (Figure S10A). We also found that the low-Fe condition of the modified M9 medium did not affect the reporter activity compared to growing E. coli JW0334 pSulAp_lacZ in modified M9 with the addition of 5 μM Fe (Figure S10A). Moreover, treatment of the reporter strain with D-Ent resulted in negligible induction of the reporter activity (Figure S10B). Therefore, this reporter assay allowed us to further evaluate the contribution of the imported Pt warheads to the antibacterial activity of our conjugates.

We then measured the levels of SOS response in the reporter strain induced by different Pt compounds. Among them, L/D-EP exhibited the greatest reporter responses in a time- and concentration-dependent manner (Figures 6A and 6B). Cisplatin treatment also led to a time- and concentration-dependent increase, but the overall reported activity was much lower compared to L/D-EP (Figures S10C and Figure 6E). Negligible induction of reporter activity occurred following oxaliplatin treatment (Figure S10D) and D-Ent treatment (Figure S10E). Treatment with L/D-EOP had negligible impact on reporter activity for the first 60 min and a slight increase in reporter activity occurred after 90 min treatment (Figures 6C and 6D). L/D-EOP induced a slightly higher reporter response than oxaliplatin (Figure 6E), presumably due to enhanced Pt uptake resulting from the L/D-Ent-mediated active transport.

Figure 6.

β-Galactosidase activity of E. coli pSulAp_lacZ after treatment with (A) L-EP, (B) D-EP, (C) L-EOP, and (D) D-EOP. (E) β-Galactosidase activity of E. coli pSulAp_lacZ after treatment with 7.5 μM compounds for 90 min (mean ± standard deviation, n = 3). (F) Pt uptake in E. coli pSulAp_lacZ after treatment with 1 μM Pt compounds for 30 min (n = 3). Statistical differences were calculated using two-tailed Student’s t test assuming unequal variances; **P < 0.01, ***P < 0.001, ****P < 0.0001, *****P < 0.00001. All assays were performed in modified M9 medium (30 °C with shaking). See Figure S9 for time- and concentration-dependence data for cisplatin, oxaliplatin, and D-Ent.

To determine whether the observed β-galactosidase activities from the reporter strain are influenced by cellular Pt concentration, we conducted a Pt uptake assay using the reporter strain and the six Pt compounds. The results are consistent with what was observed for Pt uptake with E. coli CFT073 (Figure 5A and ref²⁴) where treatment with the L-enantiomers results in somewhat greater Pt uptake than treatment with the D-enantiomers. For each enantiomer pair (e.g., L-EP and L-EOP, D-EP and D-EOP), Pt uptake indicated somewhat greater uptake of the oxaliplatin conjugates than the cisplatin conjugates (Figure 6E). Overall, our results from the reporter assay and Pt uptake assay show that while L/D-EP treatment leads to slightly lower Pt uptake in E. coli, L/D-EP induce markedly higher levels of SOS response compared to L/D-EOP (Figure 6F). This finding supports the notion that the higher potency of L/D-EP results from the cisplatin cargo inducing more DNA damage in E. coli than the oxaliplatin cargo in L/D-EOP.

Recombination-deficient E. coli exhibits enhanced susceptibility to L/D-EP and L/D-EOP.

As a complementary approach to probing DNA damage, we evaluated the susceptibility of a bacterial DNA repair mutant to the Ent–Pt(IV) conjugates. A prior report showed that major recombination pathways are critical to protect E. coli against cisplatin toxicity, demonstrated by the increased susceptibility of E. coli recombination-deficient mutants to cisplatin relative to the parent strain E. coli AB1157 (a K12 derivative).⁶⁷ We employed E. coli AB1157 and its mutant defective in RecG and RuvC, two enzymes involved in the late steps of the recombination pathway. We chose the ΔrecG ΔruvC mutant due to its hypersensitivity to cisplatin and its ability to grow in the modified M9 medium.

We first evaluated the susceptibility of E. coli AB1157 and ΔrecG ΔruvC to cisplatin. Consistent with the prior report,⁶⁷ growth of the parent strain was partially inhibited by cisplatin, and the hypersensitive ΔrecG ΔruvC mutant was fully killed when treated with ≥7.5 μM cisplatin (Figures 7A and 7C). Oxaliplatin showed negligible growth inhibition against AB1157 (Figure 7B), and concentration-dependent growth inhibition of the ΔrecG ΔruvC mutant with the turbidity almost reduced to the baseline value at 60 μM oxaliplatin (Figure 7D). These data align with the higher potency of cisplatin than oxaliplatin against E. coli. Moreover, the enhanced susceptibility of the ΔrecG ΔruvC mutant to oxaliplatin compared to the parent strain suggest that the recombination pathway is also important for bacterial survival after oxaliplatin treatment, presumably through repairing the DNA damage caused by oxaliplatin.

Figure 7.

Susceptibility of E. coli AB1157 to (A) cisplatin and L/D-EP and (B) oxaliplatin and L/D-EOP; E. coli AB1157 ΔrecG ΔruvC to (C) cisplatin and L/D-EP and (D) oxaliplatin and L/D-EOP. All assays were performed in modified M9 medium (11 h, 30 °C with shaking; mean ± standard deviation, n ≥ 4).

The results for L/D-EP and L/D-EOP followed similar trends. For the cisplatin conjugates, E. coil AB1157 was not inhibited by L-EP and was partially inhibited by D-EP (Figure 7A). Conversely, the ΔrecG ΔruvC mutant was fully killed after treatment with ≥7.5 μM L/D-EP (Figure 7C). The oxaliplatin conjugates L/D-EOP showed negligible inhibition against E. coli AB1157; however, the ΔrecG ΔruvC mutant was inhibited slightly by L-EOP and to a much greater extent by D-EOP (turbidity reduction by 8% or >90% after treatment with 15 μM L- or D-EOP, respectively) (Figures 7B and 7D). These results show that the antibacterial activities of L/D-EP and L/D-EOP can be countered by recombination. Since recombination is one of the DNA repair pathways responsible for managing DNA damage, our data indicate that all four conjugates cause DNA damage. The increased susceptibility of E. coil AB1157 and its recombinant-deficient mutant to L/D-EP compared to L/D-EOP suggest that L/D-EP-induced DNA damage is more severe compared to L/D-EOP-induced DNA damage, which is consistent with our analyses above using lysogenic E. coli and a plasmid-borne reporter.

Discussion

In this study, we report two novel Ent–Pt(IV) conjugates harboring an oxaliplatin-based Pt(IV) prodrug, further expanding the scope of SAC-based drug repurposing and Pt cargos that can be selectively delivered into the E. coli cytoplasm by the Ent uptake machinery. Notably, D-EOP showed enhanced antibacterial activity compared to oxaliplatin and L-EOP, which we attribute to enhanced uptake due to the siderophore combined with the blocked release of Fe by cytoplasmic Ent esterases; though the intrinsic activity of the D-Ent moiety may also contribute, which merits additional examination. Compared to the cisplatin conjugates L/D-EP, L/D-EOP exhibit lower antibacterial activity, which we largely attribute to the lower potency of oxaliplatin compared to cisplatin.

Motivated to gain further insight into the antibacterial activity and cellular fates of Pt agents in E. coli, we investigated the consequences of cisplatin, oxaliplatin, L/D-EOP, and L/D-EP treatment on indicators of DNA damage. We analyzed the abilities of these compounds to induce the lytic cycle in a lysogenic bacterial strain, determined DNA damage levels by measuring the SOS response using an in-cell reporter system and evaluated the susceptibility of an E. coli DNA repair mutant to these Pt complexes. Our data present compelling evidence that cisplatin and L/D-EP cause more DNA damage compared to oxaliplatin and L/D-EOP, which correlates with the higher antibacterial activities of cisplatin and L/D-EP.

The lesser degree of DNA damage in E. coli induced by oxaliplatin and L/D-EOP in our studies is consistent with analyses by others focused on oxaliplatin as an anticancer agent in eukaryotic cells. Prior work showed that oxaliplatin exhibits slower DNA binding kinetics and less Pt–DNA adduct formation on a model oligonucleotide or in human A549 cells compared to cisplatin,⁶⁸ results that could be rationalized based on the slow dissociation of the oxalate ligand in oxaliplatin, which presumably delays aquation of the Pt center and therefore biomolecular cross-linking.^69–70 It was also reported that eukaryotic DNA polymerases bypass oxaliplatin-GG adducts more efficiently than cisplatin-GG adducts;⁷¹ however, these studies were not able to explain the distinct spectrum of antitumor activity of oxaliplatin. Recent investigations provided evidence for an alternative mode of cytotoxicity against human cells, involving the disruption of ribosome biogenesis and induction of the nucleolar stress response.^{42, 68, 70} Moving forward, the antibacterial activities of Ent–Pt(IV) conjugates harboring various Pt-based warheads against other bacterial pathogens and the physiological consequences of Pt on bacterial cells warrant further investigation. We expect that future studies of Ent–Pt(IV) conjugates will provide design principles for newly repurposed antibiotic warheads.

Conclusion

The successful targeting of Ent–Pt(IV) conjugates harboring cisplatin and oxaliplatin to E. coli illustrates the potential of siderophore-based drug repurposing and deepens our understanding of the antibacterial activity of these conjugates. Specifically, this work demonstrates a link between the antibacterial activity of Ent–Pt(IV) and the intrinsic potency of the Pt warhead. Moreover, by probing DNA damage, this study indicates that at least some observations from studies in human cells are recapitulated in bacteria cells and reveals a correlation between the antibacterial activity of Ent–Pt(IV) conjugates and their ability to cause DNA damage in E. coli. Collectively, these findings provide motivation for the future design and investigation of other siderophore-based Pt antibacterials. Such investigations are essential for exploring drug repurposing for combating bacterial infections.

Methods

Instrumentation

High-performance liquid chromatography (HPLC).

Semi-preparative and analytical HPLC were performed by using an Agilent 1200 series HPLC system outfitted with an Agilent Zorbax C18 column (5 μm, 9.4 × 250 mm) at a flow rate of 4 mL/min and a Clipeus C18 column (5 μm, 4.6 × 250 mm; Higgins Analytical, Inc.) at a flow rate of 1 mL/min, respectively. Preparative HPLC was performed by using an Agilent PrepStar system outfitted with a Phenomenex Luna C18 column (10 μm, 21.2 × 250 mm) at a flow rate of 10 mL/min. The multiwavelength detectors were set to read the absorbance at 220, 280, and 316 (catecholate absorption) nm.

Solvent A was Milli-Q water (18.2 MΩ·cm, (18.2 MΩ·cm, 0.22-μm filter) with trifluoroacetic acid (TFA, purchased from Millipore Sigma) that was filtered through a 0.2-μm bottle-top filter before use. Solvent B was HPLC grade acetonitrile (MeCN, purchased from Millipore Sigma) with TFA. The amount of TFA in each eluent is indicated in the synthetic procedures. Each HPLC method began with a four-minute equilibration at 0% B followed by a gradient of increasing %B. For analytical HPLC performed to evaluate conjugate purity, the entire portion of each HPLC-purified compound was dissolved in a mixture of 1:1 H₂O/MeCN, an aliquot was taken for HPLC analysis, and the remaining solution was subsequently frozen and lyophilized to dryness.

Liquid chromatography/mass spectrometry (LC/MS).

LC/MS was performed using a nominal mass Agilent 6125B mass spectrometer with an electrospray (ESI) source attached to an Agilent 1260 Infinity LC. High-resolution mass spectrometry was performed using a high-resolution Agilent 6545 mass spectrometer with a Jet Stream ESI source coupled to an Agilent Infinity 1260 LC system. For all LC/MS analyses, solvent A was 0.1% formic acid/H₂O and solvent B was 0.1% formic acid/MeCN (LC/MS grade MeCN, Millipore Sigma). The samples were analyzed using a solvent gradient of 5–95% B over 6 min with a flow rate of 0.4 mL/min. All LC/MS instruments are housed in the MIT Department of Chemistry Instrumentation Facility (DCIF).

Inductively coupled plasma-mass spectrometry (ICP-MS).

Metal analysis was conducted using an Agilent 7900 ICP-MS system in helium mode outfitted with an integrated autosampler housed in the Center for Environmental Health Sciences (CEHS) Bioanalytical Core Facility at MIT. To quantify Fe concentration, the instrument was calibrated using standards prepared by serial dilution of an environmental calibration standard solution (1000 ppm each of Ca, Fe, K, Mg, Na; 10 ppm each of Ag, Al, As, Ba, Be, Cd, Co, Cr, Cu, Mn, Mo, Ni, Pb, Sb, Se, Th, Tl, U, V, Zn; Agilent, part number 5183–4688). To quantify Pt concentration, the instrument was calibrated using standards prepared by serial dilution of a Pt standard solution (1 ppm Pt; Millipore Sigma). Terbium (1 ppb Tb; Agilent) was used as an internal standard. All samples were prepared as 2 mL solutions in 5% HNO₃ (Honeywell, TraceSELECT; 69.0%) in 5 mL centrifuge tubes, transferred to ICP-MS polypropylene vials (Agilent) and analyzed.

For whole-cell metal analyses of Escherichia coli cells or HEK293T cells, cell pellets diluted into solutions of 3% HNO₃ were liquefied using a Milestone UltraWAVE digestion system housed in the CEHS Core Facility at MIT. A standard microwave protocol (15 min ramp to 200 °C at 1,500 W power; 10 min ramp to 220 °C at 1,500 W power) was used for the acid digestion.

NMR spectroscopy.

¹H NMR spectra were collected on a two-channel Bruker Avance-III HD Nanobay 400 MHz spectrometer, a three-channel Bruker Avance Neo 500 MHz spectrometer, or a four-channel Bruker Avance Neo 600 MHz spectrometer (both 400 MHz and 500 MHz spectrometers are equipped with a 5 mm liquid-nitrogen cooled Prodigy broad band observe cryoprobe; the 600 MHz spectrometer is equipped with a 5 mm helium-cooled QCI-F cryoprobe). ¹⁹⁵Pt NMR and 2-dimentional (2D) NMR spectra were collected on a three-channel Bruker Avance Neo 500 MHz spectrometer (equipped with a 5 mm BBFO SmartProbe) at ambient probe temperature (293 K). All NMR spectrometers are housed in the MIT DCIF.

Optical absorption spectroscopy.

Optical absorption spectra were recorded on a Beckman Coulter DU800 spectrophotometer (1 cm quartz cuvettes, Starna) and used to determine concentrations of Ent stock solutions (vide infra).

Microscopy.

Bright-field and fluorescence microscopy imaging were carried out using a Zeiss Axioplan2 upright microscope equipped with a 100× oil-immersion objective lens. Bright-field images were acquired using the Trans_DIC channel. Differential interference contrast (DIC) imaging was performed on a DeltaVision widefield deconvolution imaging system equipped with a 100x oil immersion objective lens. All microscopy equipment is housed in the W. M. Keck Microscopy Facility at Whitehead Institute. For LIVE/DEAD cell viability assays, the Texas Red (λ_ex = 532–587 nm; λ_em = 608–683 nm) and GFP (λ_ex = 457–487 nm; λ_em = 502–538 nm) channels were used to acquire images of the DEAD (red) and LIVE (green) cells, respectively.

Synthesis

General synthetic methods.

Anhydrous N,N-dimethylformamide (DMF), dichloromethane (DCM), and dimethyl sulfoxide (DMSO) were purchased from Millipore Sigma and used as received. All other chemicals were purchased from Millipore Sigma, VWR, TCI chemicals, or Alfa Aesar in the highest available purity and used as received.

Supelco TLC silica gel 60 matrix plates with fluorescent indicator were used for analytical thin layer chromatography. Supelco PLC silica gel 60 matrix plates with fluorescent indicator of 2-mm thickness were used for preparative TLC. Sigma-Aldrich silica gel (70–230 mesh, 60 Å) was used for flash column chromatography.

Ent and L/D-Ent-PEG₃-N₃ 6,7, and L/D-EP 1,2 were synthesized following published procedures.^{19–20, 24, 72} t-[Pt(DACH)(ox)(OAc)(OH)] (DACH = trans-(1R,2R)-1,2,diaminocyclohexane, ox = oxalato) was synthesized following based on the reported procedure for c, c, t-[Pt(NH₃)₂Cl₂(OAc)(OH)].²⁴

Synthesis of t-[Pt(DACH)(ox)(OAc)(OOCCH₂CH₂C≡CH)] (denoted hereafter as oxPt(IV)-alkyne 5).

The synthesis of oxPt(IV)-alkyne 5 was performed based on reported procedures.²⁴ t-[Pt(DACH)(ox)(OAc)(OH)] (41 mg, 0.087 mmol) was suspended in anhydrous DMF (4 mL). 4-Pentynoic acid (34 mg, 0.35 mmol) was dissolved in anhydrous DMF (800 μL) and combined with N,N’-dicyclohexylcarbodiimide (DCC, 75 mg, 0.36 mmol). The mixture was placed in an ultrasonic bath for 15 min, during which time a white precipitate formed. The mixture was centrifuged (4,500 rpm, 15 min, room temperature), and the supernatant was added slowly to the above suspension. The reaction was purged with N2 and stirred in the dark at room temperature for 16 h, resulting in a clear yellow solution. The reaction mixture was concentrated by air stream to yield a yellow oil, which was diluted with 1:1 H₂O/MeCN, filtered through a 0.45 mm PTFE filter (purchased from VWR), and purified by preparative HPLC (0–100% B over 30 min, 10 mL/min, 0.1% TFA in solvents A and B). The eluate from 15.6–16.6 min was collected and lyophilized to give the pure product as a white powder (29 mg, 61% yield). MS (ESI+): [M+H]⁺ calcd. 554.1097, exptl. 554.1096. [M+Na]⁺ calcd. 576.0917, exptl. 576.0915. ¹H NMR (500 MHz, D₂O): δ 2.82 (m, 2H), 2.50 (m, 2H), 2.36 (m, 2H), 2.21–2.23 (m, 3H), 1.99 (s, 3H), 1.46–1.58 (d, J = 9.7 Hz, 4H), 1.18 (m, 2H). ¹³C NMR (126 MHz, D₂O): δ 181.7, 181.3, 166.4, 83.9, 69.9, 62.0, 61.3, 34.6, 30.8, 23.4, 22.1, 14.6. ¹⁹⁵Pt NMR (108 MHz, D₂O): δ 1577.15.

Synthesis of L-Ent-oxPt(IV) (L-EOP 3).

L-EOP was first synthesized based on the report procedure for L-EP,²⁴ affording a yield of ~5% following HPLC purification. To optimize the conjugation reaction, another two CuAAC methods were tested but did not yield the conjugates as expected, including CuBr(PPh₃)₃ and in situ generation of Cu(I) by incubating Cu(OAc)₂ in tert-BuOH.^73–74 Multiple reaction conditions were tested, and two most effective methods are described below. The second method using CuSO₄+NaAsc afforded the highest yield at 41%.

Method 1. Cu(MeCN)₄PF₆.

L-Ent-PEG₃-N₃ 6 (32 μL of a 17 mM solution in DMF, 0.55 μmol) and oxPt(IV)-alkyne 5 (121 μL of a 9.0 mM solution in DMF, 1.1 μmol) were combined, to which 63.5 μL of DMF was added. tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine (TBTA) (11 μL of a 50 mM solution in DMF, 0.55 μmol) was added to the azide/alkyne mixture, followed by the addition of Cu(MeCN)₄PF₆ (11 μL of a 50 mM solution in DMF, 0.55 μmol). The resulting solution was briefly mixed by a vortex mixer and incubated on a benchtop rotator at room temperature in the dark. After 6 h, the reaction mixture was diluted with 1:1 MeCN:water and purified by preparative HPLC (0–100% B over 30 min, 10 mL/min, 0.005% TFA was used in solvents A and B; this low percent TFA was used to prevent decomposition). The eluant at 21.3 min was collected, flash frozen in liquid N₂, and lyophilized, which afforded L-EOP as a white powder (~4.9 mg, 16%).

Method 2. CuSO₄+NaAsc.^{20, 22–23}

L-Ent-PEG₃-N₃ 6 (73 μL of an 11.3 mM solution in DMF, 0.83 μmol) and oxPt(IV)-alkyne 5 (50 μL of a 50 mM solution in DMF, 2.5 μmol) were combined, to which 100 μL of DMF was added. An aliquot of CuSO₄ (50 μL of a 90 mM solution in water, 4.5 μmol) and TBTA (100 μL of a 50 mM solution in DMF, 5.0 μmol) were combined to give a blue solution, to which sodium ascorbate (NaAsc, 100 μL of a 90 mM solution in water, 9.0 μmol) was added. The color of the solution immediately changed from blue to pale yellow indicating reduction of Cu(II) to Cu(I) and the mixture was added to the alkyne/azide mixture. The reaction was incubated on a benchtop rotator for 20 min in the dark at room temperature. Then, the reaction mixture was diluted with 1:1 MeCN:water and purified by preparative HPLC (0–100% B over 40 min, 10 mL/min; 0.005% TFA was used in solvents A and B to prevent decomposition). The eluate at 24.9 min was collected, flash frozen in liquid N2, and lyophilized, affording L-EOP as a white powder (0.5 mg, 41%). Analytical HPLC traces of purified L-EOP are shown in Figure S1.

HRMS (ESI+): [M+H]⁺ calcd, 1468.3729; found, 1468.3726. ¹H NMR (500 MHz, DMSO_d₆): δ 11.91 (s, 1H), 11.62 (s, 2H), 9.73 (s, 1H), 9.42 (s, 2H), 9.29 (d, 1H, J = 6.9 Hz), 9.10 (d, 2H, J = 7.1 Hz), 8.18–8.43 (m, 5H), 7.93 (d, 1H, J = 2.2 Hz), 7.76 (s, 1H), 7.45 (d, 1H, J = 2.2 Hz), 7.35 (d, 2H, J = 8.1 Hz), 6.97 (d, 2H, J = 9.0 Hz), 6.75 (t, 2H, J = 8.1 Hz), 4.92 (m, 3H), 4.66 (m, 3H), 4.44 (t, 2H, J = 5.4 Hz), 4.41 (m, 3H), 3.78 (t, 2H, J = z, 3H), 1.49 (d, 2H, J = 10.0 Hz), 1.37–1.43 (m, 2H), 1.13–1.19 (m, 2H). ¹⁹⁵Pt signal was detected by ¹H-¹⁹⁵Pt HMQC NMR (500 MHz, DMSO_d₆): δ 1111.71 (s).

D-Ent-oxPt(IV) (D-EOP, 4).

D-EOP 4 was synthesized as described for L-EOP 3 (Method 2. CuSO₄+NaAsc), except that D-Ent-PEG₃-N₃ 7 was employed instead of L-Ent-PEG₃-N₃ 6, affording a yield of 31%. Analytical HPLC traces of purified D-EOP are shown in Figure S1.

HRMS (ESI+): [M+H]⁺ calcd, 1468.3729; found, 1468.3727. ¹H NMR (500 MHz, DMSO_d₆): δ 11.91 (s, 1H), 11.62 (s, 2H), 9.73 (s, 1H), 9.42 (s, 2H), 9.29 (d, 1H, J = 6.9 Hz), 9.10 (d, 2H, J = 7.1 Hz), 8.18–8.43 (m, 5H), 7.93 (d, 1H, J = 2.2 Hz), 7.76 (s, 1H), 7.45 (d, 1H, J = 2.2 Hz), 7.35 (d, 2H, J = 8.1 Hz), 6.97 (d, 2H, J = 9.0 Hz), 6.75 (t, 2H, J = 8.1 Hz), 4.92 (m, 3H), 4.66 (m, 3H), 4.44 (t, 2H, J = 5.4 Hz), 4.41 (m, 3H), 3.78 (t, 2H, J = z, 3H), 1.49 (d, 2H, J = 10.0 Hz), 1.37–1.43 (m, 2H), 1.13–1.19 (m, 2H). ¹⁹⁵Pt signal was detected by ¹H-¹⁹⁵Pt HMQC NMR (500 MHz, DMSO_d₆): δ 1111.71 (s).

Storage and handling of Ent and Ent-oxPt(IV) conjugates

All synthetic precursors, Ent and L/D-EOP were stored as either powders or DMSO stock solutions at −80 °C. Concentrations of Ent stock solutions (~10 mM) were determined using Beer’s law and the reported extinction coefficient for apo Ent in MeOH (ε₃₁₆ = 9500 M⁻¹ cm⁻¹).⁷⁵ An aliquot of the DMSO stock solution was diluted into MeOH for this analysis. Concentrations of L/D-EOP stock solutions were determined by quantifying Pt concentration using ICP-MS. The stock solution concentrations were 5–12 mM. To minimize multiple freeze–thaw cycles, all stock solutions were divided into 10 μL aliquots and stored at −80 °C. Aliquots were routinely analyzed by analytical HPLC to confirm the integrity of the samples. Multiple synthetic batches of L/D-EOP were employed throughout this work.

Working solutions of L/D-EOP were prepared via dilutions in 10% DMSO/H₂O. Working solutions of oxaliplatin were freshly prepared by dissolving oxaliplatin in modified M9 medium before treatment. The Fe(III)-bound L/D-EOP complexes were prepared before all microbiology and imaging assays because the modified M9 medium contains insufficient Fe to fully complex the siderophore following addition to the culture. Fe(acac)₃ (4.1 mM stock in DMSO, concentration determined by ICP-MS) was used to prepare the Fe(III) complexes, which were formed by incubating 10 μL of L/D-EOP working solution with 0.9 μL of a 10× Fe(acac)₃ solution in 10% DMSO/H₂O for 5 min prior to addition to the culture.

Stability evaluation of D-EOP

The DMSO stock solution of D-EOP was diluted in 260 μL of modified M9 medium (Na₂HPO₄ 6.8 g/L, KH₂PO₄ 3 g/L, NaCl 0.5 g/L, NH₄Cl 1 g/L, 0.4% glucose, 0.2% casein amino acids, 2 mM MgSO₄, 0.1 mM CaCl₂, 0.6 g/mL thiamine) to afford a 30 μM D-EOP solution. The resulting mixture was divided into five 50 μL aliquots. The aliquots were incubated at 30 °C with shaking at 150 rpm in the dark for 0, 2, 5, 10, and 20 h, respectively. At each time point, an aliquot was flash frozen in liquid N₂ and then stored at −80°C. Analytical HPLC samples were prepared by thawing each sample and centrifuging the sample (13,000 rpm and 4 °C for 10 min). The resulting supernatants were analyzed by analytical HPLC (0–100 B% in 30 min, 1 mL/min, 0.005%TFA in solvents A and B). The percent of remaining D-EOP at each time point was determined by integrating the peak area of D-EOP in each sample. Decomposition products were analyzed by LC/MS. The results from this study are described below in the Supporting Discussion.

Supporting Discussion

We selected D-EOP to perform the stability evaluation, and we expect similar results for the stability of L-EOP. Analytical HPLC revealed that apo D-EOP decomposed in the modified M9 medium with a t_1/2 ~ 5 h at 30 °C, which is in the same range as L-EP.²⁴ LC/MS analysis afforded m/z values of 1485.1 and 1262.1 for the major decomposition products after 5 h incubation, which correspond to the hydrolysis products of D-EOP where the Ent trilactone moiety was linearized by one or two or ester hydrolysis events, yielding linear D-EOP and (DHBS)₂oxPt(IV) (m/z calcd. 1485.4 and 1262.3, respectively; DHBS, 2,3-dihydroxybenzoyl serine). Therefore, we moved forward using modified M9 medium with 0.6 μg/mL of thiamine and determining bacterial growth and morphology after 11 h treatment for all microbiology assays presented in this work. These are the same conditions that were used in prior studies of L/D-EP.²⁴

We note that bacterial uptake of L/D-EOP hydrolysis products that contain the oxPt(IV) cargo and a hydrolyzed Ent moiety may also occur during the microbiology assays.

Molecular biology methods

Construction of E. coli pSulAp_lacZ.

To make the plasmid pSulAp_lacZ, a plasmid backbone derived from pBAD-HisA (purchased from Invitrogen) was first prepared. pBAD-HisA (50 ng/μL) was double digested by incubation with NdeI (1 U/μL) and HindIII-HF (1 U/μL) in 20 μL 1X Cutsmart buffer for 2.5 h before the enzymes were heat inactivated by incubation at 80 °C for 1 h on a heating block. The linearized plasmid was then amplified with overhangs (containing part of the sulA promoter) with primers pBAD_HindIII-f and pBAD_sulAp-r using Q5 polymerase following the manufacturer’s instructions (Table S2). The amplicon was purified using the PCR Cleanup Kit from Qiagen following the manufacturer’s instructions. To prepare the lacZ fragment, lacZ with overhangs was amplified from E. coli MG1655 gDNA (purified using the Wizard Genomic DNA Purification Kit from Invitrogen) with the primers sulAp-lacZ-f and pBAD_lacZ-r (Table S2) using Q5 polymerase. The amplicon was then purified using the Qiagen PCR Cleanup Kit. The two purified fragments were then ligated together in 1:1 molar ratio (50 ng of the pBAD fragment) using the NEB HiFi DNA Assembly Mix in a 10 μL reaction volume. The isothermal reaction proceeded at 50 °C for 2.5 h in a thermocycler. All 10 μL of the reaction mixture was then used to transform a 200 μL aliquot of chemically competent E. coli TOP10 cells. The isothermal reaction mixture was incubated with the cells on ice for 30 minutes before the cells were heat shocked at 42 °C for 30 s. The cells were then incubated on ice for 5 minutes before 700 μL LB was added, and the cells were rescued for 40 minutes at 37 °C with shaking on an orbital shaker (250 rpm). Cells were then pelleted via centrifugation, resuspended in ~150 μL LB, and plated onto LB-agar supplemented with 100 μg/mL ampicillin. The plates were incubated at 37 °C for ~15 h. Colonies were picked and screened for ligated plasmid by colony PCR using the sulAp-lacZ-f/pBAD_lacZ-r primer pair and analysis via 1% agarose gel electrophoresis. Hits were cultivated in LB supplemented with 100 μg/mL ampicillin for 12 h at 37 °C on an orbital shaker (250 rpm), the plasmid was purified using the Qiagen Miniprep Kit, and the fidelity of the promoter region was verified by Sanger sequencing (Quintara Biosciences). E. coli pSulAp_lacZ was constructed by transforming chemically competent E. coli JW0334 (the ΔlacY mutant from the Keio collection)⁵¹ with pSulAp_lacZ via heat shock at 42 °C for 30 s.

Microbiology, microscopy, and cell culture methods

General materials.

Lysogeny broth (LB; tryptone 10 g/L, yeast extract 5 g/L, NaCl 10 g/L), M9 minimal salts 5×, casamino acids, and agar were purchased from Becton Dickinson (BD). LB medium and Milli-Q water (18.2 MΩ·cm, 0.22-μm filter) used for bacterial cultures or for preparing working solutions of the tested compounds were sterilized in an autoclave. Modified M9 medium was sterilized by passage through a sterile 0.22-μm filter. Sterile polypropylene culture tubes and adhesive PCR film seals were purchased from VWR. Sterile polystyrene 96-well plates and 6-well plates used for culturing were purchased from Corning Incorporated. LIVE/DEAD BacLight Bacterial Viability Kits were purchased from Thermo Fisher (Invitrogen Molecular Probes). Agarose (PCR grade) for microscopy was purchased from Bio-Rad. Microscope slides and microscope cover glasses were purchased from VWR. Cisplatin (≥99.9% trace metals basis) and Fe(acac)₃ (≥99.9% trace metals basis) were purchased from Sigma-Aldrich. Oxaliplatin (cis-[(1R,2R)-1,2-cyclohexanediamine-N,N′]-oxalatoplatinum (II), 98%) was purchased from AmBeed. Synthetic oligonucleotides were purchased from IDT. Enzymes and buffers used for molecular cloning were purchased from NEB.

Materials for tissue culture were kindly provided by the Shoulders lab at MIT. HEK293T cells were purchased from ATCC. DMEM was purchased from Corning Incorporated. Penicillin and streptomycin were purchased from Corning Incorporated.

Bacterial strains.

Bacterial strains employed in this study are summarized in Table S1. Freezer stocks were prepared from single colonies in 25% glycerol/LB medium.

General procedures for bacterial growth assays and microscopy.

Growth of E. coli under Fe-deficient conditions was performed in a modified M9 medium. The Fe content of the modified M9 medium was determined by ICP-MS to be 0.6–0.7 μM Fe. These Fe-deficient conditions cause E. coli to express siderophore biosynthesis and transport machinery, including genes for enterobactin biosynthesis and transport encoded by the enterobactin gene cluster as well as the iroA cluster for salmochelin biosynthesis and transport.²⁴

Working solutions of L/D-EOP were prepared via dilutions in 10% DMSO/H₂O. Working solutions of oxaliplatin were freshly prepared by dissolving oxaliplatin in modified M9 medium before treatment. The Fe(III)-bound L/D-EOP complexes were prepared before all microbiology and imaging assays because the modified M9 medium contains insufficient Fe to fully complex the siderophore following addition to the culture. Fe(acac)₃ (13 mM stock in DMSO, concentration determined by ICP-MS) was used to prepare the Fe(III) complexes, which were formed by incubating 10 μL of L/D-EOP working solution with 0.9 μL of a 10× Fe(acac)₃ solution in 10% DMSO/H₂O for 5 min prior to addition to the culture. In the discussion of the biological assays, L/D-EOP refer to the corresponding ferric complexes. For all microbiology assays, the final cultures contained 1% v/v DMSO, except for cultures with oxaliplatin treatment which do not contain DMSO.

Overnight cultures of E. coli were prepared in 15 mL polypropylene tubes by inoculating 5 mL of medium with the appropriate freezer stock. The overnight cultures were incubated at 37 °C for 16–18 h in a tabletop incubator set at 150 rpm. Each overnight culture was diluted 1:100 into 5 mL of fresh medium at 37 °C with shaking at 250 rpm until OD₆₀₀ reached 0.6 ± 0.1 (measured on a Beckmann Coulter DU800 spectrophotometer). Each culture was subsequently diluted in fresh medium to achieve an OD₆₀₀ value of 0.005 (~5 × 10⁶ CFU/mL). A 90-μL aliquot of the diluted culture was combined with a 10-μL aliquot of a 10× working solution in a 96-well plate. The plate was sealed with an adhesive film (purchased from VWR) and incubated at 30 °C with shaking at 500 rpm for 20 h in the BioTek LogPhase 600 (LP600) microbiology reader (96-well plate format). Growth curves were recorded as OD₆₀₀ values collected every hour in the LP600 microbiology reader. Growth assays of L/D-EOP were performed using a two-fold dilution series spanning 7.5–60 μM or a ten-fold dilution series spanning 0.001–10 μM. Each well condition was prepared in duplicate and at least four independent replicates using two synthetic batches of each conjugate and were performed on different days. The resulting mean OD₆₀₀ values are reported, and the error bars are the standard deviation from the independent replicates. Statistical differences compared to untreated controls were calculated using two-tailed student t test assuming unequal variances.

Samples for microscopy were prepared by taking aliquots of E. coli culture at t = 11 h (mid-log phase). For samples that require only bright-field imaging, a 5 μL aliquot of each culture was pipetted on an agarose pad (1% w/w agarose/Milli-Q water) which was placed on a microscope slide. The sample was then covered with a glass coverslip. Representative micrographs for each condition are shown in the figures. For LIVE/DEAD viability assays, a 90 μL aliquot of each bacterial culture was centrifuged at 3,000 rpm at 4 °C for 15 min. The resulting cell pellet was resuspended in 0.85% NaCl and the OD₆₀₀ was adjusted to 0.2 using 0.85% NaCl. A 25 μL aliquot of each bacterial suspension was incubated with 25 μL of LIVE/DEAD dye mixture (48 μM SYTO9 and 240 μM propidium iodide) at 30 °C for 15 min in the dark, and 5 μL of the suspension was pipetted on an agarose pad which was placed on a microscope slide. The sample was then covered with a glass coverslip. For each type of microscopy experiment, each condition was repeated in at least three biological replicates using two different synthetic batches. Representative micrographs for each condition are shown in figures.

We selected 11 h as the time point to determine inhibitory effects and morphological changes because of the following considerations described previously:²⁴ (i) the 11 h time point is in the mid-log phase of bacterial growth as determined by growth curves; (ii) there is sufficient cell density to determine growth inhibitory effects and have sufficient cells to image; (iii) initial microbiology studies indicated that the stability of L/D-EOP under these conditions was sufficient for our studies.

Image analysis.

The microscopy images were processed using the FIJI software (8-bit image type). For bright-field images, contrast enhancement was performed by setting saturated pixels as 0.1%. For fluorescence images, fluorescence background subtraction was performed using a rolling ball method with a radius of 150 pixels.

Induction of lysogenic bacteria.

The induction of lysogenic bacteria was performed based on a published procedure.²⁷ E. coli W3104 was inoculated in 5 mL of modified M9 medium and grown for 18 h at 37 °C with shaking. A 50 μL aliquot of the overnight culture was diluted in 5 mL of modified M9 medium, which was then incubated at 37 °C until the OD₆₀₀ reached 0.6 (~7 h). Then, 0.5 μL aliquots of bacterial culture were added to 100 μL portions of modified M9 medium containing 0 (untreated), 7.5, 15 and 30 μM oxaliplatin in a 96-well plate. After incubation at 30 °C for 10 h, each culture was diluted 1:10, 1:100, and 1:1000, and 10 μL of each diluted culture was spotted onto an LB agar plate on which a lawn of E. coli CFT073 had been freshly plated. After spotting, the plate was incubated at 37 °C for 12 h to allow plaque formation. To test the effect of L/D-EOP, 0.5 μL aliquot of exponentially growing E. coli W3104 culture was then incubated in 100 μL of the modified M9 medium which contained 15 μM L/D-EP, 15 μM L-Ent, or 1% DMSO, respectively.

The E. coli CFT073 lawn was prepared by inoculating 4 mL of molten top agar (0.5% LB agar) with 100 μL of an overnight culture of E. coli CFT073 grown in LB medium. Following gentle mixing, 3 mL of the inoculated molten top agar was layered atop a preheated (37 °C) LB agar plate and allowed to solidify.

Pt uptake by E. coli.

E. coli CFT073 or E. coli pSulAp_lacZ was inoculated in 5 mL of modified M9 medium and grown for 18 h at 37 °C. A 50 μL aliquot of the overnight culture was diluted in 5 mL of modified M9 medium, which was then incubated at 37 °C until the OD₆₀₀ reached 0.6. A 4 mL portion of the culture was centrifuged at 3,500 rpm and 4°C for 10 min, and the resulting cell pellet was resuspended in 1 mL of fresh modified M9 medium. Then, to a 90 μL aliquot of the diluted culture, a 10 μL aliquot working solution of 10 μM L-EOP, D-EOP, or oxPt(IV)-alkyne 5 (in 10% DMSO/H₂O) was added. After incubation at 30 °C with shaking at 150 rpm for 30 min, cell pellets were harvested by centrifuging each culture at 3,500 rpm and 4°C for 10 min. To measure the Pt content in the supernatant, 90 μL of each supernatant was diluted into 1.91 mL of 5% HNO₃ for ICP-MS analysis. To measure the cell-associated Pt content, the cell pellets were first washed with fresh modified M9 medium, and then washed with fresh modified M9 containing 2% w/w aqueous EDTA. The resulting cell pellets were suspended in 2 mL of 5% HNO₃ and digested for ICP-MS analysis. Mass of cell-associated Pt and mass of Pt in the supernatant of each sample were calculated using Pt contents determined by ICP-MS. Cell-associated Pt% was determined according to equation 1.

$$\text{\hspace{0.5em}Cell-associated\hspace{0.5em}Pt\%}=\frac{\text{\hspace{0.5em}mass\hspace{0.5em}of\hspace{0.5em}cell-associated\hspace{0.5em}Pt}(\text{μg})}{\text{\hspace{0.5em}mass\hspace{0.5em}of\hspace{0.5em}cell-associated\hspace{0.5em}Pt}(\text{μg})+\text{\hspace{0.5em}mass\hspace{0.5em}of\hspace{0.5em}Pt\hspace{0.5em}in\hspace{0.5em}the\hspace{0.5em}supernatant\hspace{0.5em}}(\text{μg})}$$ (eq 1)

Pt uptake by HEK293T cells.

Materials for tissue culture were kindly provided by the Shoulders lab at MIT. HEK293T cells were purchased from ATCC. DMEM was purchased from Corning Incorporated. Penicillin and streptomycin were purchased from Corning Incorporated.

HEK293T cells (passages 10–20) were plated at a density of 750,000 cells per well in a 6-well plate (2.5 mL/well) in DMEM supplemented with 1% penicillin/streptomycin and incubated at 37 °C and 5% CO₂ for 24 h. Working solutions of L-EOP and D-EOP (6 μM) were prepared in 10% DMSO/PBS (PBS purchased from Millipore Sigma). Working solutions of oxaliplatin were freshly prepared by dissolving oxaliplatin in PBS (6 μM) before treatment. Each working solution (0.5 mL) was added to each well of cell culture to give the final treatment concentration as 1 μM. Cells were treated for 6 h at 37°C and 5% CO_2, transferred to a 15 mL conical tube and centrifuged at 500 rcf and 4°C for 5 min. To measure the Pt content in the supernatant, 200 μL of each supernatant was diluted into 1.8 mL of 5% HNO₃ for ICP-MS analysis. To measure the cell-associated Pt content, cell pellets were washed with 2 mL of PBS three times, and the resulting cell pellets were suspended in 2 mL of 5%HNO₃ and digested for ICP-MS analysis. Mass of cell-associated Pt and mass of Pt in the supernatant of each sample were calculated using Pt contents determined by ICP-MS. Cell-associated Pt% were determined by ICP-MS as described above (eq 1).

E. coli JW0334 pSulAp_lacZ reporter assay.^76–78

E. coli JW0334 pSulAp_lacZ was grown to mid-log phase (OD₆₀₀ ≈ 0.6) in modified M9 medium at 37 °C. Culture aliquots (90 μL) were inoculated into the wells of sterile 96-well plates, and 10 μL of appropriately diluted compound stock was added to give the indicated final concentration. Plates were incubated at 30 °C with shaking for 30, 60, and 90 min, respectively. During growth, aliquots of 80 μL of permeabilization solution (25mM Na₂HPO₄, 50 mM KCl, 2 mM MgSO₄, 0.8 mg/mL hexadecyltrimethylammonium bromide (CTAB), 0.4 mg/mL sodium deoxycholate, 50 mM dithiothreitol) were added into another 96-well plates. Aliquots of 85.7 μL of substrate solution (60 mM Na₂HPO₄, 40 mM NaH₂PO₄, 1 mg/mL o-nitrophenyl-β-D-galactoside (ONPG, purchased from VWR), 0.8 mg/mL CTAB, 50 mM dithiothreitol) were added into another 96-well plates. The OD₆₀₀ was measured using a BioTek Synergy HT plate reader, and a 20 μL of the culture was rapidly added to the 80 μL of permeabilization solution to give the sample solution. After the last sample solution was prepared, aliquots of 14.3 μL of sample solution were added to wells containing substrate solution and incubated at 30 °C with shaking. After 30-min incubation, aliquots of 100 μL of stop solution (1 M Na₂CO₃) were added. The plate was agitated and the A₄₂₀ and A₅₅₀ were recorded. β-Galactosidase activity was calculated using equation 2 as Miller units.

$$\text{β-Galactosidase\hspace{0.5em}activity\hspace{0.5em}}=\frac{1000\times \left(A_{420}-1.75\times A_{550}\right)}{\text{O}{\text{D}}_{600}\times \text{\hspace{0.5em}time\hspace{0.5em}}(\text{in}\min )\times \text{\hspace{0.5em}volume\hspace{0.5em}}(\text{in\hspace{0.17em}mL})}$$ (eq 2)

Supplementary Material

The Supporting Information is available free of charge at http://

Tables S1-S2, Figures S1-S15, NMR spectroscopic data, and Supporting References.

Scheme 1.

Synthesis of L/D-EOP via click chemistry.