Late-Stage Functionalization of the Rifamycin Core via Click Chemistry Toward New Antibacterial Derivatives

Abstract

Antimicrobial resistance (AMR) threatens global health, particularly through the rise of multidrug-resistant tuberculosis (MDR-TB) and other critical bacterial infections such as methicillin-resistant Staphylococcus aureus (MRSA). Rifamycins remain frontline antibiotics but are increasingly undermined by resistance. Here, we introduce a click-enabled platform for the synthesis of C8-functionalized rifamycins, which can be converted in a single additional step into efficacious 3′-hydroxy-5′-aminobenzoxazinorifamycins (bxRifs) and enzymatically into 25-deacetylated rifamycins (deAcRifs), providing access to novel antibacterial scaffolds that expand beyond the scope of traditional C8 modifications. Accordingly, we establish a modular strategy for late-stage analog development of the complex natural product rifamycin S, wherein azido and alkyne functionalities are installed via tailored core chemistry and converted into 1,2,3-triazoles through copper(I)-catalyzed click chemistry. Another key feature of this work is the development of systematic HPLC purification methods, enabling the isolation of analytically pure compounds despite structural complexity. The resulting analogs exhibit distinct antibacterial profiles, notably against Gram-positive bacteria including MRSA and Streptococcus mutans, informing structure–activity relationships and offering a foundation for further optimization. This approach supports the rapid diversification of rifamycin scaffolds to combat the escalating threat of AMR, while also establishing a foundation for future discovery through bioorthogonal applications.

1. Introduction

Antimicrobial resistance (AMR) is a growing global health crisis, linked to 4.95 million deaths in 2019 [1]. According to the World Health Organization (WHO), AMR weakens the effectiveness of antibiotics, making common infections harder to treat and increasing the risks of surgery and cancer therapies [2,3]. Its impact extends beyond humans, threatening animals and ecosystems [1,2,3]. Without urgent action, AMR could lower global life expectancy by 1.8 years by 2035, while costing $412 billion annually in healthcare and $443 billion in productivity losses [4,5].

Mycobacterium tuberculosis (MTB), the bacterial agent of tuberculosis (TB), is typically treated by a regimen of four first-line antibiotics: isoniazid, rifampicin, ethambutol, and pyrazinamide [3,6,7]. Despite susceptibility to these compounds, TB remained the world’s second-leading cause of death from a single infectious agent in 2022, following COVID-19, causing an estimated 1.5 million deaths. Approximately 25% of the world’s population has latent TB that can be activated by common viral infections, such as HIV or SARS-CoV-2 [3,6,7,8,9]. Between 2000 and 2022, an estimated 66 million lives were saved following TB diagnosis and treatment, yet multidrug-resistant TB (MDR-TB) is an escalating public health crisis [3,6,10]. In 2022 alone, approximately 410,000 people developed MDR-TB or rifampicin-resistant TB (RR-TB) [7]. The direct costs of treating TB vary significantly depending on drug resistance, ranging from $20,211 for drug-susceptible TB to $567,708 for extensively drug-resistant TB (XDR-TB) [3,6]. Beyond TB, drug-resistant infectious diseases resulted in at least 700,000 deaths globally in 2019, a figure that could escalate to 10 million annually by 2050, including 2.4 million deaths in high-income countries, without sustained intervention [2,11,12,13].

Rifamycins, particularly rifampicin, are crucial components of TB therapy, as they bind to the β subunit of MTB RNA polymerase (RNAP) and inhibit RNA synthesis [14,15,16]. The divergence between bacterial and human RNAP sequences makes bacterial RNAP a promising target for broad-spectrum antibiotic development [15,16,17]. Rifamycins, as the most clinically important and extensively studied class of broad-spectrum antibiotics targeting bacterial RNAP, hold immense value in treating TB and other deadly bacterial infections caused by Gram-positive and Gram-negative pathogens with multidrug-resistant properties, typified by ESKAPE pathogens (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species) [13,18,19,20,21,22,23,24,25]. In this study, clinically relevant methicillin-resistant S. aureus (MRSA) and Escherichia coli were employed as model organisms to evaluate the antibacterial activity of the analogs against Gram-positive and Gram-negative bacteria. Furthermore, multi-drug resistance in Streptococcus mutans, the primary pathogenic bacterium responsible for dental caries, has been reported, and this bacterium has also been implicated in several systemic diseases, including cerebral hemorrhage, inflammatory bowel disease, tumors, and infective endocarditis [26,27]. Therefore, evaluating rifamycins for their activity against S. mutans is crucial for both oral and systemic health.

The efficacy of rifamycins is compromised when mutations in the rifampicin resistance-determining region (RRDR) of MTB RNAP disrupt key interactions with the drug [14,17,28,29,30]. Based on structure–activity relationship (SAR) studies of rifamycins, the hydoxyl (-OH) group at C8 is critical for drug target binding [14,28,30]. However, strategic C8 modifications, specifically substitution of the hydroxyl group with thiomethyl (-SCH₃) or a (2-hydroxyethyl)pyrazole moiety, retain or enhance activity against rifampicin-resistant strains carrying the clinically dominant S450L mutation, which occurs in approximately 41% of RR-TB isolates [14,15,17,28,29]. Previous efforts to modify the C8 position of rifamycins are scarce and often relied on multi-step, non-orthogonal syntheses and yielded limited improvements in potency [15,28]. However, one study revealed that both small polar and bulky nonpolar groups can be tolerated at C8, suggesting untapped chemical flexibility [28].

Additionally, rifamycin analogs with additional rings linking C3 and C4, 3′-hydroxy-5′-aminobenzoxazinorifamycins (bxRifs), which include rifalazil and its derivatives, are reported to show exceptional activities against both drug-susceptible and RR-TB by creating additional contacts with RNAP and also show reduced hepatic Cyp450 induction mediated by Human Pregnane X Receptor (hPXR), a major clinical liability of rifampicin when used in treating patients’ coinfections [15,29,30,31,32,33,34,35]. The synthesis of bxRifs from rifamycin S is a one-step reaction [29,30]; therefore, the development of improved bxRifs from C8-modified rifamycins will be straightforward. It is also known that C25-deacetylated metabolites of representative rifamycins, such as rifampicin, rifabutin, and rifapentine, reduce the induction of drug-metabolizing enzymes in the human liver, thereby mitigating rifamycin-induced hepatotoxicity [36,37]. The enzymatic conversion of rifamycins into 25-deacetylated rifamycins (deAcRifs) appears to be readily achievable using human arylacetamide deacetylase (AADAC), the principal enzyme responsible for rifamycin deacetylation [23,36,37,38].

By leveraging the versatile chemistry of the rifamycin core, particularly through copper(I)-catalyzed azide–alkyne cycloaddition (CuAAC), commonly known as click chemistry, we are enabling the rapid development of C8-functionalized rifamycin derivatives that support systematic exploration of SAR and serve as synthetic precursors to bxRifs and deAcRifs with translational potential [39,40,41,42,43,44,45,46,47]. More importantly, our click-enabled strategy allows late-stage functionalization of the rifamycin scaffold, laying the groundwork for broader biomedical applications via bioorthogonal conjugation strategies, even though pursuing such downstream applications is beyond the current scope of this study [46,48,49,50].

Furthermore, rifamycins are notoriously difficult to purify due to their lipophilicity, susceptibility to degradation, and overlapping impurity profiles. To date, high-performance liquid chromatography (HPLC) protocols for rifamycin S derivatives have not been reported. Existing methods, primarily developed for rifampicin, focus on analytical quantification in complex matrices rather than on the semi-preparative isolation of structurally diverse analogs [51,52,53]. Here, we detail a robust reverse phase semi-preparative HPLC method with optimized solvent gradients, column selection, and diode array detection. This approach enables consistent purification of rifamycin S derivatives in acceptable yields, a critical advancement for medicinal chemistry efforts involving this complex natural product scaffold.

In this paper, we report the first synthesis, purification, isolation, and antibacterial activity of rifamycin derivatives with azido, alkyne, and triazole functionalities at C8. Our approach allows a 2–3 step modular conversion from rifamycin S to triazole analogs to target prevalent RR-TB strains and other clinically relevant pathogens, while also establishing a foundation for future applications through bioorthogonal chemistry.

2. Results and Discussion

Chemistry: Based on the previous literature, activation of the C8 peri-phenolic functionality has enabled modifications at the C8 position of rifamycin S [28,47]. Accordingly, sulfonyl, amino, alkynyl, azido, and triazole moieties were selected in this study to probe the chemistry at C8, introduce modular handles for diversification via click chemistry, and assess SAR potential within a broader platform exploration of C8 chemical space as shown in Figure 1. The reaction of rifamycin S (1) with p-toluenesulfonyl- and (4-fluorobenzene)sulfonyl chlorides yielded compounds 8 and 2, respectively. Among these, compound 8 is the only previously reported C8-sulfonate derivative of rifamycin S that has been evaluated for antibacterial activity against MTB RNAP in vitro and E. coli in vivo [28]. The sulfonates were purified via recrystallization using 5–10% ethyl acetate in hexanes to result in 86–89% yields. However, tosylate (8) was not used for the synthesis of 8-amino congeners of rifamycin S (3 and 7), as sulfonate 2, bearing a more electron-withdrawing fluorinated sulfonyl group, exhibited enhanced reactivity in functionalization screens. This improved reactivity facilitated the synthesis of novel 8-aminorifamycin S analogs 3 and 7, incorporating azido and alkynyl substituents through the enabling reaction at C8 of 2 with the nitrogen-based nucleophiles. We further modified azido-substituted 8-aminorifamycin S compound (3) to generate 1,2,3-triazoles with substituents at C4 of the triazole ring via Cu (I) catalyzed click reaction. These transformations afforded analogs 4–6 in estimated yields of 37–71% following HPLC purification. Reaction conditions were not further optimized, even when yields were suboptimal. Multiple attempts to extend the alkynyl-substituted 8-aminorifamycin S compound (7) to synthesize a series of substituted or unsubstituted 1,2,3-triazoles by reacting with sodium azide and a series of short-chain azidoamines were unsuccessful, likely due to the positioning of the alkynyl group within the complex rifamycin structure. Further investigation will involve modifying the length of the methylene chain in amino-alkynyls.

Figure 1. Synthesis of modified rifamycin S analogs using rifamycin and click chemistry. Reagents and reaction conditions: (i) corresponding benzenesulfonyl chloride, acetonitrile, 4–25 °C, 18 h (2 in 86% and 8 in 89% yield); (ii) 7N ammonia in methanol, acetonitrile, 4–25 °C, 18 h (9 in 26% yield); (iii) 3-azidopropylamine, acetonitrile, 4–25 °C, 18 h (3 in 63% yield); (iv) sodium ascorbate, CuSO₄.5H₂O, tetrahydrofuran, 25 °C, 18 h, corresponding alkyne: 1-pentyne (4 in 69%), 1-cyclopentylacetylene (5 in 71%), 1-phenylacetylene (6 in 37%); (v) propargylamine, acetonitrile, 4–25 °C, 18 h, (7 in 78%). Reported yields are after HPLC purification and the purity is determined to be ≥95% by analytical HPLC with diode array detection, with no detectable rifamycin S or related rifamycin-derived impurities.

Given that previously reported rifamycin derivatives had been purified using flash column chromatography, we initially attempted the same approach [28,30]. However, flash column chromatography failed to yield pure compounds, and under the conditions we employed, we observed significant degradation along with the appearance of multiple stereoisomers on thin-layer chromatography (TLC), indicating compound instability during purification. Consequently, we opted for reverse phase semi-preparative HPLC purification, despite the absence of established methods in the literature. Purification of the modified rifamycin analogs was challenging, but we systematically developed efficient HPLC methods using the U3000 Prep HPLC System with a Diode Array Detector (DAD) and a reverse phase semi-preparative scale HPLC column. As shown in Figure 2, the first HPLC method we developed (method 1), with a total run time of 16 min, improved separation between the second peak (containing the intended product) and the third peak (see chromatogram for HPLC method 1). However, only fraction 4 contained the pure product, while several other fractions (3, 5, and 6) contained products with impurities. To improve efficacy and optimize separation, HPLC method 2 was designed by replacing acetonitrile with THF and adjusting the dual solvent gradients. This modification resulted in the intended product being eluted three minutes earlier and improved the separation between the peaks by 0.247 min compared to method 1 (see chromatogram for HPLC method 2). The modified method not only reduced the run time but also produced several fractions of pure product (fractions 1, 2, and 3), while other fractions contained minimal desired product, if any. The most efficacious HPLC method, method 3, reduced the total run time to 11 min and achieved the same purity in fractions as method 2, which had a total run time of 15 min. Method 3 resulted in several pure product fractions (1, 2, and 3), with fraction 4 being impure, which was then fed into the next purification cycle.

Figure 2. HPLC chromatograms and TLC of HPLC fractions using methods 1–3 for purification of 8-(3-azidopropyl)aminorifamycin S (3).

Overall, both HPLC methods 2 and 3 yielded pure products in sufficient quantities for compound characterization and biological evaluation, typically requiring only 3–5 preparative HPLC injections per derivative, thereby saving time and resources. All compounds used for biological evaluation were confirmed to be ≥95% pure by analytical HPLC with DAD, with no detectable rifamycin S or related rifamycin-derived impurities. Upon HPLC purification, all compounds were characterized by standard Nuclear Magnetic Resonance Spectroscopy (NMR), High-Resolution Electrospray Ionization Mass Spectrometry (HRESIMS), and Fourier Transform Infrared Spectroscopy (FTIR), as described in the experimental section. The digital NMR spectra are provided under the Supporting Information.

Biological Evaluation: Accurately docking modified rifamycins to bacterial RNAPs was challenging due to the size difference between rifamycins and the binding site, along with ansa ring flexibility, making in silico biological activity assessment impossible without synthesis and evaluation. We utilized model organisms MRSA, S. mutans, E. coli, along with E. coli RNAP in vitro assays, to evaluate the new analogs against Gram-positive, Gram-negative, and MTB pathogens. While MRSA was used to assess the response of clinically significant, drug-resistant bacterial strains representing systemic Gram-positive infections, S. mutans served as a model to evaluate the analogs’ potential to target the oral bacterial microbiota.

Most modified rifamycins (10 µM) significantly inhibited MRSA (USA300, LAC strain) growth at 3 or 6 h post-inoculation (hpi), except for compounds 5 and 6 (Figure 3).

Figure 3. Fold change in MRSA growth relative to DMSO treated bacteria in response to rifamycin analogs. Bacterial growth inhibition studies were conducted using rifamycin analogs with a DMSO vehicle control at three and six hours post-inoculation (hpi). Tryptic soy broth (TSB) cultures were treated with either 0.1% DMSO or 10 µM of the indicated compound and inoculated with MRSA, USA300, LAC strain at 37 °C. A Student’s t test was used to determine statistical significance for each compound compared to the DMSO vehicle control, where *** p < 0.0001, ** p < 0.001, and * p < 0.01.

Several compounds, particularly 4 and 9, exhibited improved MRSA growth inhibition compared to the original scaffold, rifamycin S. However, the difference in activity between rifamycin S and any of the effective modified rifamycins was not statistically significant when analyzed with the unpaired parametric Student’s t test with Welch’s correction. Given the excellent biological activity of modified rifamycins against virulent MRSA, determining MIC values using the modified Gompertz function was crucial to fully understand bacterial inhibition and compound potency [54]. MIC, defined as the lowest concentration achieving 90% growth inhibition (Table 1 and Figure 4A), provided key insights. Interestingly, while rifamycin S had the lowest MIC value, some modified rifamycins with higher activity in screening assays displayed higher MIC values. Further studies will be conducted to determine whether this observation is key to distinguishing the modified rifamycins as bactericidal or bacteriostatic agents against MRSA [55,56,57]. Although preliminary given the limited number of each derivative type evaluated in this study, when evaluating the SAR, derivatives bearing 8-sulfonyl and 8-amino groups exhibited high potency against MRSA, whereas alkynyl, azido, and triazole analogs (3–7) were less active. Among the triazoles, the linear alkyl-substituted compound (4) outperformed the bulkier cyclic and aryl analogs (5 and 6), suggesting steric sensitivity at the binding site.

Table 1. Modified rifamycin activity against MRSA, USA300, LAC strain, S. mutans, and E. coli K-12 derivative in vivo, as well as E. coli RNAP in vitro. Analog 1 is rifamycin S, the precursor for analogs 2-9, while analog 10 is rifampicin.

Analog Number	Post-HPLC Yield Percentage	Substituent at C8	MRSA MIC (µM)	S. mutans MIC (µM)	E. coli MIC (µM)	E. coli WT RNAP Inhibition Range (µM)
1	Commercially purchased		0.001710	0.08989	102.4	0.01 *
2	86% purified by recrystallization		0.2966	2.342	>2000	0.0333–0.333
3	63%		19.73	20.58	>2000	0.0333–0.333
4	69%		1.596	18.92	1088	3.33–10
5	71%		>100	4.590	1844	0.333–3.33
6	37%		>100	36.38	1526	0.333–3.33
7	78%		2.186	2.731	>2000	0.333–3.33
8	89% purified by recrystallization		0.1310	0.1472	>2000	0.021 *
9	56%		0.2080	0.09002	>2000	4.4 *
10	Commercially purchased		0.0061	0.1512	41.71	0.01

* Data is reported against MTB RNAP in https://doi.org/10.1016/j.bmcl.2011.08.054.

Figure 4. Minimal inhibitory concentration (MIC) curves of 8-(3-azidopropyl)aminorifamycin S (3) against (A) MRSA, USA300, LAC strain and (B) E. coli K-12 derivative. MICs were defined as the lowest concentration of antimicrobial agent resulting in 90% inhibition of bacterial growth.

The modified rifamycins exhibited modest but consistent activity against S. mutans, with limited fold change across analogs, in contrast to the broader potency range observed against MRSA (Table 1). The amino derivative (9) showed comparable activity to the original rifamycin core (1), while the C8-tosylate analog (8) exhibited similar potency to rifampicin (10), a representative bxRif. This is an encouraging outcome for the further development of new rifamycin analogs aimed at improving oral health.

When E. coli K-12 was used in antibacterial screening assays to evaluate the effects of modified rifamycins on Gram-negative bacteria, no significant inhibition was observed. Determination of MIC values for E. coli with each modified rifamycin revealed a 10- to 20-fold decrease in potency relative to the original scaffold, rifamycin S (Table 1 and Figure 4B). The relatively insignificant antibacterial activity of modified rifamycins against E. coli (BSL1) compared to virulent S. aureus can be attributed to the role of efflux pumps in antibiotic resistance in Gram-negative bacteria [58,59,60,61,62,63]. E. coli K-12 strains possess multiple drug efflux pumps, including AcrAB-TolC, which actively expel larger antibiotics such as rifamycins, reducing effectiveness [17,61,62,63]. We are conducting studies in E. coli to identify and optimize the synergistic effects of known small-molecule efflux pump inhibitors (EPIs) with modified rifamycins, aiming to enhance their efficacy against Gram-negative bacterial infections [64,65,66,67].

Screening compounds for inhibition of WT MTB RNAP (Table 1) and rifamycin-resistant MTB RNAP (S450L) (Supporting Information) revealed a 3- to 100-fold reduction in activity due to modifications at the 8-OH position of parent molecule, rifamycin S. This aligns with previous biochemical and crystallographic studies highlighting the crucial role of the C-8 hydroxyl group in rifamycin binding to E. coli and MTB RNAPs [14,15,17,28,29,30]. We are developing a series of bxRifs of C-8-modified rifamycins to create additional interactions with the σ region 3.2 and other regions of the RNAP complex, with the goal to enhance RNAP inhibitory activity against MTB while blocking hPXR binding to improve pharmacokinetics [15,29,30,31,32,33,34,35]. In parallel, we are generating a selective set of 8-modified deAcRifs via commercially available recombinant AADAC-mediated deacetylation as a proof of concept [36,37].

In this study, we developed and evaluated six C8-aminorifamycins, along with two tosylate derivatives and the parent compounds, rifamycin S and rifampicin, all bearing an oxygen at the C8 position. Overall, the C8-amino derivatives of rifamycin S were found to be less effective as antibacterials, which was not unexpected; however, they occasionally exhibited comparable activity, which was encouraging. However, the primary objective of our work was to explore structural diversification through the incorporation of click chemistry, rather than to specifically aim at surpassing the potency of the parent compound, 1. We are in the process of strategically developing click-based rifamycin analogs that retain an oxygen atom at the C8 position; however, the synthesis has proven difficult, and the yields have been consistently poor.

In Figure 5, a heatmap of log₁₀-transformed MIC ratios (E. coli/MRSA and S. mutans/MRSA) highlights the selectivity profiles of each analog. Higher values indicate greater selectivity for MRSA over the comparator strain. Elevated E. coli/MRSA ratios reflect strong Gram-positive selectivity, while S. mutans/MRSA ratios help differentiate systemic from oral Gram-positive activity. Rifamycin S (1) displayed the highest selectivity for MRSA across both comparisons, consistent with its established potency. Notably, a few negative selectivity ratios in Figure 5 indicate that compounds 5, 6, and 9 were more potent against S. mutans than MRSA, with compound 9 exhibiting superior activity against S. mutans, even surpassing that of rifampin (10). These findings suggest their potential utility as oral niche-specific antimicrobials. The consistently lower selectivity of the analogs for E. coli supports their use as mechanistic probes to disentangle rifamycin transport from RNAP target engagement. Overall, many of the new analogs demonstrated enhanced selectivity and potency toward MRSA, a clinically significant, drug-resistant pathogen, while sparing commensals, underscoring their potential therapeutic relevance.

Figure 5. Bacterial selectivity of rifamycin analogs against MRSA, USA300, LAC strain relative to S. mutans and E. coli K-12 derivative. The heatmap shows the log₁₀ transformed selectivity ratios of MIC values for each compound (µM), calculated as S. mutans/MRSA and E. coli/MRSA. Higher values (green–yellow) indicate greater selectivity toward MRSA over the respective comparator strain. Selectivity values >3 log₁₀ units correspond to ≥1000-fold potency differences.

3. Experimental Section

General Chemical Methods: Rifamycin S was purchased from AAPharmaSyn, Ann Arbor, MI. All the other reagents were purchased from Sigma-Aldrich, St. Louis, MO and used without purification. All supplies and HPLC grade solvents were purchased from Fisher Scientific International, Inc. (Waltham, MA, USA) and Lab Alley (Austin, TX, USA). Glassware was oven-dried before use under anhydrous conditions. Air- and moisture-sensitive reactions were performed in Argon gas. NMR spectra were recorded using Bruker (Billerica, MA, USA) BioSpin, AVANCE, 500MHz NMR with TXI probe, Bruker BioSpin, AVANCE, 300 MHz NMR with BBO probe and Sample Changer, or Bruker BioSpin, AVANCE III, 400 MHz with BBFO probe at the State-wide NMR Facility, and the mass spectra were recorded on a Shimadzu LCMS-IT-TOF Liquid Chromatograph Mass Spectrometer utilizing the electrospray ionization (ESI) mode at the State-wide Mass Facility, Fayetteville, AR. FTIR studies were performed using Thermo Scientific Nicolet 6700 FTIR instrument. HPLC method development and purifications are specified below. Reactions were monitored by TLC using pre-coated silica gel 60 F254 plates. Chromatography was performed with silica gel 60 Å (70–230 mesh).

Synthesis of 8-(4-fluorobenzene)sulfonyloxyrifamycin S (2): Diisopropylethylamine (112 mg, 0.867 mmol, 1.2 eq) was added dropwise into an ice-cold solution of rifamycin S (1) (500 mg, 719 µmol) in acetonitrile (4 mL). Then, p-fluorobenzenesulfonyl chloride (154 mg, 0.791 mmol, 1.1 eq) was added to the reaction mixture portion-wise. This resultant mixture was slowly warmed to room temperature and stirred at room temperature for 18 h before being diluted with dichloromethane (40 mL). This solution was washed with 1 N aqueous NH₄Cl (2 × 30 mL) followed by saturated brine (30 mL), dried over anhydrous sodium sulfate, and concentrated to yield a red-to-orange solid. The crude product was purified by recrystallization with 5–10% ethyl acetate in hexanes (2, 529 mg, 86%), resulting in a yellow solid. Reported purity is ≥95% pure by analytical HPLC with DAD, with no detectable rifamycin S or related rifamycin-derived impurities. HRESIMS [M + H]⁺ observed m/z 854.2831, calculated for C₄₃H₄₉FNO₁₄S⁺ m/z 854.2858; ¹H NMR (700 MHz, CDCl₃) δ: 8.23 (s, 1H), 8.09–8.05 (m, 2H), 7.82 (s, 1H), 7.33–7.28 (m, 2H), 6.35 (dd, J = 10.4, 15.8 Hz, 1H), 6.24 (d, J = 10.3 Hz, 1H), 6.11 (dd, J = 7.1, 15.8 Hz, 1H), 6.00 (d, J = 12.1 Hz, 1H), 5.09 (dd, J = 5.2, 12.3 Hz, 1H), 4.91 (d, J = 10.2 Hz, 1H), 3.82 (d, J = 4.8 Hz, 1H), 3.66 (d, J = 9.7 Hz, 1H), 3.47–3.44 (m, 1H), 3.41 (s, 1H), 3.08 (s, 3H), 3.00–2.96 (m, 1H), 2.31 (q, J = 7.9 Hz, 1H), 2.18 (s, 3H), 2.04 (d, J = 4.4 Hz, 6H), 1.77 (s, 4H), 1.65–1.59 (m, 1H), 0.99 (d, J = 7.0 Hz, 3H), 0.85 (d, J = 7.0 Hz, 3H), 0.67 (d, J = 6.8 Hz, 3H), 0.06 (d, J = 7.1 Hz, 3H) (Figure S1); ¹³C NMR (176 MHz, CDCl₃) δ: 192.43, 181.75, 176.90, 173.92, 172.64, 169.46, 167.13, 165.66, 151.97, 142.59, 141.30, 139.59, 134.46, 133.09, 132.20, 132.18, 131.64, 131.58, 129.78, 125.33, 124.08, 120.00, 118.04, 116.95, 116.82, 116.07, 114.96, 108.13, 78.49, 73.88, 73.79, 60.37, 57.68, 37.98, 37.87, 37.23, 33.08, 21.20, 20.92, 19.92, 17.20, 14.19, 11.13, 9.93, 8.88 (Figure S2).

Synthesis of 8-(3-azidopropyl)aminorifamycin S (3): 3-azidopropylamine (24 mg, 24 µL, 239 µmol) was added dropwise to an ice-cold solution of 8-(4-fluorobenzene)sulfonyloxyrifamycin S (2) (200 mg, 235 µmol) in acetonitrile (4 mL). The resulting mixture was stirred at room temperature for 18 h before being diluted with dichloromethane (30 mL). This solution was washed with 1 N aqueous NH₄Cl (2 × 20 mL) followed by saturated brine (15 mL), dried over anhydrous sodium sulfate, and concentrated to yield a dark purple solid. The crude product was purified by the reverse phase semi-preparative HPLC method reported in this manuscript (3, 116 mg, 63%). Reported purity is ≥95% pure by analytical HPLC with DAD, with no detectable rifamycin S or related rifamycin-derived impurities. HRESIMS [M + H]⁺ observed m/z 778.3591, calculated for C₄₀H₅₂N₅O₁₁⁺ m/z 778.3663; FTIR spectrum showed a strong absorption peak at 2100 cm⁻¹ for N₃ asymmetric vibration; ¹H NMR (300 MHz, CDCl₃) δ: 9.79 (t, J = 4.9 Hz, 1H), 7.66–7.57 (m, 2H), 7.43 (dd, J = 3.3, 5.7 Hz, 1H), 7.16 (s, 1H), 6.35–5.89 (m, 1H), 4.94 (dd, J = 7.1, 12.7 Hz, 1H), 4.68 (d, J = 9.3 Hz, 1H), 4.17–4.06 (m, 3H), 3.73 (dt, J = 5.6, 19.0 Hz, 2H), 3.55 (d, J = 9.5 Hz, 1H), 3.42–3.36 (m, 2H), 3.34–3.29 (m, 1H), 3.01 (s, 2H), 2.97 (s, 1H), 2.38 (s, 2H), 1.95 (s, 3H), 1.92–1.86 (m, 2H), 1.66 (complex pattern), 1.61 (s, 3H), 1.33 (s, 3H), 1.23–1.21 (m, 3H), 1.15 (s, 3H), 0.91 (d, J = 6.9 Hz, 2H), 0.58 (d, J = 6.9 Hz, 2H), 0.15–0.08 (m, 2H) (Figure S3); ¹³C NMR (75 MHz, CDCl₃) δ: 173.17, 167.76, 158.23, 143.73, 142.19, 140.03, 134.06, 130.88, 128.80, 124.20, 115.76, 115.51, 109.17, 107.79, 80.49, 73.96, 73.66, 68.16, 57.10, 48.65, 43.94, 38.73, 38.25, 37.74, 37.29, 32.83, 30.36, 30.32, 30.28, 29.71, 28.93, 23.75, 22.99, 22.18, 21.03, 20.04, 17.09, 14.06, 12.73, 11.23, 10.97, 10.87, 8.87 (Figure S4).

Synthesis of 8-(3-(4-propyl-1,2,3-triazole)propyl)aminorifamycin S (4): The reaction was set up with 8-(3-azidopropyl)aminorifamycin (3) (15 mg, 19 µmol) in tetrahydrofuran (1.5 mL) at room temperature. Sodium ascorbate (19 mg, 5 eq) and CuSO₄.5H₂O (4.8 mg, 1 eq) were dissolved in 1.5 mL of distilled water, and the mixture was quickly added to the reaction vessel. Following that, 1-pentyne (1.3 mg, 1.7 µL, 1 eq) was added using a micropipette to the reaction immediately. The resulting mixture was stirred at room temperature for 18 h before being diluted with dichloromethane (30 mL). This solution was washed with 1 N aqueous NH₄Cl (2 × 20 mL) followed by saturated brine (15 mL), dried over anhydrous sodium sulfate, and concentrated to yield a dark purple-to-maroon solid. The crude product was purified by the reverse phase semi-preparative HPLC method reported in this manuscript (4, 11 mg, 69%). Reported purity is ≥95% pure by analytical HPLC with DAD, with no detectable rifamycin S or related rifamycin-derived impurities. HRESIMS [M+H]⁺ observed m/z 846.4292, calculated for C₄₅H₆₀N₅O₁₁⁺ m/z 846.4289; FTIR spectrum showed the disappearance of a strong absorption peak at 2100 cm⁻¹ of the starting material; ¹H NMR (400 MHz, CDCl₃) δ: 9.94 (t, J = 4.8 Hz, 1H), 8.51 (s, 1H), 7.76 (s, 1H), 7.60–7.46 (m, 1H), 7.35 (s, 1H), 7.28 (s, 1H), 6.46–6.14 (m, 5H), 6.03 (dd, J = 6.8, 15.7 Hz, 1H), 5.31 (s, 2H), 5.07 (dd, J = 6.9, 12.5 Hz, 1H), 4.77 (d, J = 11.3 Hz, 1H), 4.52 (t, J = 6.7 Hz, 2H), 3.95 (d, J = 4.4 Hz, 1H), 3.67 (d, J = 9.7 Hz, 2H), 3.58 (s, 1H), 3.41 (dd, J = 2.8, 7.1 Hz, 1H), 3.12 (s, 3H), 3.06–2.99 (m, 1H), 2.71 (t, J = 7.6 Hz, 2H), 2.40 (s, 3H), 2.36–2.32 (m, 3H), 2.06 (s, 6H), 1.93 (ddd, J = 3.0, 7.0, 10.4 Hz, 1H), 1.71 (s, 3H), 1.00 (d, J = 7.3 Hz, 5H), 0.87 (d, J = 6.8 Hz, 3H), 0.70 (d, J = 6.9 Hz, 3H), 0.23 (d, J = 7.1 Hz, 3H) (Figure S5); ¹³C NMR (101 MHz, CDCl₃) δ: 191.05, 182.94, 181.88, 173.06, 171.63, 169.27, 158.15, 143.98, 142.10, 139.99, 134.25, 133.97, 130.59, 129.10, 124.16, 120.94, 115.70, 115.54, 110.52, 109.40, 108.55, 108.42, 107.85, 80.79, 73.81, 73.56, 73.54, 57.02, 53.43, 47.09, 43.51, 38.35, 37.65, 37.33, 32.80, 31.44, 27.66, 22.70, 22.22, 21.04, 20.05, 17.05, 13.81, 12.63, 11.25, 11.09, 8.88 (Figure S6).

Synthesis of 8-(3-(4-cyclopentyl-1,2,3-triazole)propyl)aminorifamycin S (5): The reaction was set up with 8-(3-azidopropyl)aminorifamycin (3) (20 mg, 26 µmol) in tetrahydrofuran (1.5 mL) at room temperature. Sodium ascorbate (25.4 mg, 5 eq) and CuSO₄.5H₂O (6.4 mg, 1 eq) were dissolved in 1.5 mL of distilled water and the mixture was quickly added to the reaction vessel. Following that, 1-cyclopentylacetylene (2.4 mg, 2.9 µL, 1 eq) was added using a micropipette to the reaction immediately. The resulting mixture was stirred at room temperature for 18 h before being diluted with dichloromethane (30 mL). This solution was washed with 1 N aqueous NH₄Cl (2 × 20 mL) followed by saturated brine (15 mL), dried over anhydrous sodium sulfate, and concentrated to yield a dark purple-to-maroon solid. The crude product was purified by the reverse phase semi-preparative HPLC method reported in this manuscript (5, 15.5 mg, 71%). Reported purity is ≥95% pure by analytical HPLC with DAD, with no detectable rifamycin S or related rifamycin-derived impurities. HRESIMS [M + H]⁺ observed m/z 872.4417, calculated for C₄₇H₆₂N₅O₁₁⁺ m/z 872.4446; FTIR spectrum showed the disappearance of a strong absorption peak at 2100 cm⁻¹ of the starting material; ¹H NMR (400 MHz, CDCl₃) δ: 9.95 (t, J = 5.0 Hz, 1H), 8.52 (s, 1H), 7.77 (s, 1H), 7.60–7.47 (m, 1H), 7.33 (s, 1H), 7.28 (s, 2H), 6.42 (dd, J = 10.2, 15.7 Hz, 1H), 6.27 (d, J = 10.1 Hz, 1H), 6.17 (d, J = 12.0 Hz, 1H), 6.03 (dd, J = 6.8, 15.7 Hz, 1H), 5.07 (dd, J = 7.1, 12.5 Hz, 1H), 4.77 (d, J = 10.5 Hz, 1H), 4.53–4.48 (m, 2H), 3.95 (d, J = 4.7 Hz, 1H), 3.83–3.65 (m, 4H), 3.57 (s, 1H), 3.41 (dd, J = 2.8, 7.5 Hz, 1H), 3.25–3.14 (m, 2H), 3.12 (s, 3H), 3.07–3.00 (m, 1H), 2.40 (s, 3H), 2.34 (t, J = 6.8 Hz, 4H), 2.07 (s, 10H), 1.93 (td, J = 3.6, 7.2 Hz, 2H), 1.82–1.76 (m, 3H), 1.72 (s, 4H), 1.62 (s, 11H), 1.27 (s, 1H), 1.00 (d, J = 7.0 Hz, 3H), 0.87 (d, J = 6.9 Hz, 3H), 0.70 (d, J = 6.9 Hz, 3H), 0.23 (d, J = 7.0 Hz, 3H) (Figure S7); ¹³C NMR (101 MHz, CDCl₃) δ: 191.04, 182.93, 181.88, 173.06, 171.63, 169.26, 158.17, 144.00, 142.11, 139.98, 134.27, 133.98, 130.60, 124.15, 119.93, 115.70, 115.55, 110.51, 109.38, 107.86, 77.33, 77.01, 76.69, 57.02, 47.08, 43.53, 38.35, 37.66, 37.33, 36.77, 33.26, 33.24, 32.80, 31.46, 25.17, 22.23, 21.05, 20.05, 17.06, 12.65, 11.26, 11.10, 8.88 (Figure S8).

Synthesis of 8-(3-(4-phenyl-1,2,3-triazole)propyl)aminorifamycin S (6): The reaction was set up with 8-(3-azidopropyl)aminorifamycin (3) (20 mg, 26 µmol) in tetrahydrofuran (1.5 mL) at room temperature. Sodium ascorbate (25.4 mg, 5 eq) and CuSO₄.5H₂O (6.4 mg, 1 eq) were dissolved in 1.5 mL of distilled water and the mixture was quickly added to the reaction vessel. Following that, 1-phenylacetylene (2.6 mg, 2.8 µL, 1 eq) was added using a micropipette to the reaction immediately. The resulting mixture was stirred at room temperature for 18 h before being diluted with dichloromethane (30 mL). This solution was washed with 1 N aqueous NH₄Cl (2 × 20 mL) followed by saturated brine (15 mL), dried over anhydrous sodium sulfate, and concentrated to yield a dark purple-to-maroon solid. The crude product was purified by the reverse phase semi-preparative HPLC method reported in this manuscript(6, 8.3 mg, 37%). Reported purity is ≥95% pure by analytical HPLC with DAD, with no detectable rifamycin S or related rifamycin-derived impurities. HRESIMS [M + H]⁺ observed m/z 880.4115, calculated for C₄₈H₅₈N₅O₁₁⁺ m/z 880.4133; FTIR spectrum showed the disappearance of strong absorption peak at 2100 cm⁻¹ of the starting material; ¹H NMR (300 MHz, CDCl₃) δ: 9.94 (t, J = 4.4 Hz, 1H), 8.50 (s, 1H), 7.86 (s, 1H), 7.85–7.73 (m, 3H), 7.58 (d, J = 8.3 Hz, 1H), 7.42 (td, J = 7.9, 17.0, 17.7 Hz, 4H), 7.27 (s, 2H), 6.41 (dd, J = 10.2, 15.6 Hz, 1H), 6.26 (d, J = 10.2 Hz, 1H), 6.17 (d, J = 12.6 Hz, 1H), 6.02 (dd, J = 7.0, 15.9 Hz, 1H), 5.05 (dd, J = 7.0, 12.4 Hz, 1H), 4.75 (d, J = 10.4 Hz, 1H), 4.61 (t, J = 6.6 Hz, 2H), 3.94 (d, J = 4.7 Hz, 1H), 3.79 (dq, J = 6.8, 20.0 Hz, 2H), 3.66 (d, J = 9.6 Hz, 1H), 3.56 (s, 1H), 3.42–3.37 (m, 1H), 3.10 (d, J = 1.3 Hz, 3H), 3.06 (s, 2H), 2.42 (s, 3H), 2.05 (d, J = 3.5 Hz, 6H), 1.80 (d, J = 7.3 Hz, 2H), 1.70 (s, 5H), 1.26 (s, 3H), 0.99 (d, J = 7.1 Hz, 3H), 0.85 (d, J = 6.8 Hz, 3H), 0.69 (d, J = 7.2 Hz, 3H), 0.23 (d, J = 6.5 Hz, 3H) (Figure S9); ¹³C NMR (75 MHz, CDCl₃) δ: 191.06, 182.92, 181.89, 173.13, 171.65, 169.25, 158.17, 148.06, 144.11, 142.08, 141.15, 140.37, 139.94, 134.22, 133.96, 130.64, 128.91, 128.34, 125.70, 124.14, 119.92, 115.59, 115.53, 110.52, 109.44, 108.61, 107.89, 80.89, 73.82, 73.50, 57.01, 47.45, 43.55, 38.81, 38.65, 38.37, 37.62, 37.34, 32.78, 31.39, 29.71, 22.24, 21.05, 20.08, 17.04, 12.70, 11.28, 11.17, 8.89 (Figure S10).

Synthesis of 8-(prop-2-ynyl)aminorifamycin S (7): Propargylamine (24 mg, 24 µL, 436 µmol) was added dropwise to an ice-cold solution of 8-(4-fluorobenzene)sulfonyloxyrifamycin S (2) (200 mg, 235 µmol) in acetonitrile (4 mL). The resulting mixture was stirred at room temperature for 18 h before being diluted with dichloromethane (40 mL). This solution was washed with 1 N aqueous NH₄Cl (2 × 20 mL) followed by saturated brine (15 mL), dried over anhydrous sodium sulfate, and concentrated to yield a dark purple solid. The crude product was purified by the reverse phase semi-preparative HPLC method reported in this manuscript to result (7, 134 mg, 78%). Reported purity is ≥95% pure by analytical HPLC with DAD, with no detectable rifamycin S or related rifamycin-derived impurities. HRESIMS [M + H]⁺ observed m/z 773.3336, calculated for C₄₀H₄₉N₂O₁₁⁺ m/z; 773.3336; ¹H NMR (400 MHz, CDCl₃) δ: 12.51 (s, 1H), 8.40 (s, 1H), 7.85 (s, 1H), 7.28 (s, 3H), 6.40–6.20 (m, 3H), 5.96 (dd, J = 7.1, 15.2 Hz, 1H), 5.32 (s, 6H), 5.09 (dd, J = 7.8, 12.4 Hz, 1H), 4.64 (dd, J = 1.8, 10.5 Hz, 1H), 4.28–4.19 (m, 1H), 3.77–3.70 (m, 2H), 3.70–3.63 (m, 2H), 3.63–3.54 (m, 2H), 3.39 (s, 1H), 3.13 (s, 2H), 3.00 (dd, J = 3.6, 10.2 Hz, 1H), 2.35 (s, 3H), 2.10–2.06 (m, 6H), 2.05 (s, 2H), 1.73 (s, 3H), 1.26 (d, J = 11.3 Hz, 3H), 1.03 (d, J = 7.0 Hz, 3H), 0.86 (d, J = 7.1 Hz, 3H), 0.68 (d, J = 7.0 Hz, 3H), 0.23 (d, J = 7.1 Hz, 2H) (Figure S11); ¹³C NMR (101 MHz, CDCl₃) δ: 191.79, 184.81, 182.25, 173.19, 172.51, 169.56, 166.85, 144.82, 142.24, 139.05, 133.82, 130.95, 130.51, 124.08, 117.26, 115.57, 115.50, 110.69, 108.44, 81.52, 73.57, 73.48, 71.27, 70.63, 70.06, 69.15, 63.65, 56.88, 53.42, 38.92, 37.45, 37.29, 32.69, 29.71, 22.27, 21.06, 20.10, 16.89, 13.92, 11.67, 11.36, 8.81, 7.56 (Figure S12).

Synthesis of 8-(4-methylbenzene)sulfonyloxyrifamycin S (8) and Synthesis of 8-aminorifamycin S (9) were completed based on the literature [28]. Reported purity of each is ≥95% pure by analytical HPLC with DAD, with no detectable rifamycin S or related rifamycin-derived impurities. Digital ¹H and ¹³C NMR spectra of each are reported in the Supporting Information (Figures S13–S16).

Development and optimization of HPLC purification method for modified rifamycins (3–7): Method development was performed using the Ultimate U3000 Prep HPLC System with a Diode Array Detector (DAD) and a reverse phase semi-preparative scale HPLC column (150 × 10 mm, 12 µm HYPERSIL GOLD) from ThermoFisher Scientific (Waltham, MA, USA). Ultimate purity of HPLC-purified compounds was confirmed using TLC analysis and HRMS. Method development began with a dual solvent gradient system, 95:5 water–acetonitrile (v/v) and 95:5 acetonitrile–water (v/v), over 20 min at 25 °C. Effective gradients were systematically developed, transitioning from 80–30% 95:5 water–acetonitrile (v/v) to 0% over 5–15 min, as well as from 20–70% 95:5 acetonitrile–water (v/v) to 100% at a flow rate of 10 mL/min. Optimal purification was achieved with a gradient of 40-0% 95:5 water–acetonitrile (v/v) over 15 min in 60-100% 95:5 acetonitrile–water (v/v) at the same flow rate (Figure 2—HPLC method 1). Subsequently, our method was further refined using a different dual solvent gradient system of 95:5 water–acetonitrile (v/v) and 95:5 tetrahydrofuran (THF)–water (v/v), over 15 min at 25 °C (Figure 2—HPLC method 2). While an effective purification was achieved with a gradient of 50-0% 95:5 water–acetonitrile (v/v) over 13 min in 50–100% 95:5 THF–water (v/v) at the same flow rate (Figure 2—HPLC method 3), the optimal purification was achieved with a gradient of 45-7.5% 95:5 water–acetonitrile (v/v) over 11 min in 55–92.5% 95:5 tetrahydrofuran-water (v/v) at a flow rate of 10 mL/min (Figure 2—HPLC method 3). Detection was performed at 254 nm or 190 nm, and all HPLC data were processed using the Chromeleon 7 Chromatography Data System.

Antibacterial Activity Screening MRSA, USA300, LAC strain was cultured overnight in tryptic soy broth (TSB) shaking (225 rpm) at 37 °C. TSB was inoculated from overnight LAC cultures based on OD_600nm and incubated at 37 °C shaking (225 rpm) in the presence of experimental compounds or dimethyl sulfoxide (DMSO, Fisher) vehicle control at a final concentration of 10 µM or 0.1%, respectively. OD_600nm was measured for each culture at 3 and 6 h post-inoculation using a DS-11 FX+ spectrophotometer/fluorometer (Denovix, Wilmington, DE, USA). Experiments were conducted in triplicate, and the resulting data were statistically analyzed using Student’s t test (specifically an unpaired parametric t test with Welch’s correction) using GraphPad Prism Version (10.4.1).

Determination of minimal inhibitory concentration (MIC) of modified rifamycins against Gram-positive S. aureus: S. aureus LAC constitutively expressing green fluorescent protein (GFP) was cultured overnight through shaking (225 rpm) at 37 °C in TSB containing erythromycin (10 µg/mL, Sigma). Serial dilutions in TSB were made for each rifamycin stock solution in dimethyl sulfoxide (DMSO) and inoculated from overnight cultures based on OD_600nm. Untreated TSB was also inoculated as a growth control. Bacteria were cultured for 6 h in a 96-well, black-wall, glass-bottom plate through shaking (230 rpm) at 37 °C. GFP signal was measured (Ex_485nm/Em_528nm) using a Synergy H1 plate reader (BioTek, Winooski, VT, USA). MICs were defined as the lowest concentration of antimicrobial agents resulting in 90% inhibition of bacterial growth. Experiments were conducted in triplicate, and the resulting data were statistically analyzed using GraphPad Prism (version 10.3.0). Statistical methods were selected based on established techniques in the literature using modified Gompertz function [54]. Quality control measures included standard antibiotic controls and verification of bacterial inoculum sterility and accuracy.

Determination of MIC of rifamycins against Gram-negative E. coli: E. coli K-12 expressing GapA-sfGFP from the chromosome was cultured in Luria Broth (LB) at 37 °C overnight through shaking (250 rpm). Purified rifamycin S derivatives were prepared as stock solutions in dimethyl sulfoxide (DMSO) at 100 mM concentration and stored at −20 °C. Based on the broth microdilution method recommended by the Clinical and Laboratory Standards Institute (CLSI), serial dilutions of rifamycins were prepared in LB in a 96-well, black-wall, glass-bottom plate to achieve final concentrations ranging from 1000 µM to 0.98 µM, inoculated from overnight E. coli cultures based on OD_600nm, and cultured for 6 h shaking (250 rpm) at 37 °C in the Jitterbug (Boekel Scientific, Long Branch, NJ, USA). Each microplate included appropriate controls, including DMSO (vehicle control) and untreated LB (growth control). GFP signal was measured (Ex_485nm/Em_528nm) using a Synergy H1 plate reader (BioTek). MICs were defined as the lowest concentration of the antimicrobial agents resulting in 90% inhibition of bacterial growth. Experiments were conducted in triplicate, and the resulting data were statistically analyzed using GraphPad Prism (version 10.3.0). Statistical methods were selected based on established techniques in the literature using modified Gompertz function [54]. Quality control measures included standard antibiotic controls and verification of bacterial inoculum sterility and accuracy.

Determination of MIC of modified rifamycins against Gram-positive S. mutans: MIC was determined by the standard broth microdilution method as recommended by Clinical and Laboratory Standards Institute (CLSI) guidelines with some modifications as reported earlier [68,69,70]. Briefly, bacterial strain S. mutans UA159 was cultured overnight in Brain Heart Infusion (BHI) broth. Compounds were weighed and dissolved in dimethyl sulfoxide (DMSO) to make the required stock concentration. In the first rows of a 96-well plate, compounds dissolved in DMSO were mixed with BHI broth to obtain a starting concentration of 32 μg/mL. Fresh BHI broth was added, and 2-fold dilutions of the compounds were performed down each column, yielding a concentration of 0.25 μg/mL, or 0.03125 μg/mL for compounds 1, 8, 9, and 10, in the last wells of the row. Overnight bacterial culture of S. mutans UA159 was used to prepare bacterial suspensions at the 0.5 McFarland standard in PBS and diluted (1:100) in sterile Cation-adjusted Mueller Hinton broth (CAMHB) to make a final concentration of ~2 × 10⁶ CFU/mL. The plates were then incubated at 37 °C at 5% CO₂ for 20 h. MIC values were then determined as the lowest concentration of compound at which visible growth of bacteria was absent. All the MIC assays were performed in triplicates. Vancomycin was used as positive control and DMSO as a negative control. MIC was confirmed in at least two of three experiments with fresh bacterial cultures conducted on three different days.

Determination of the in vitro transcriptional activity of E. coli RNAPs: Activity of wild-type (WT) and rifamycin-resistant mutant (S450L) E. coli RNAPs was measured via a rolling circle transcription assay using the template in the absence and presence of rifamycins and a single-stranded DNA template developed in our laboratory and a previously reported protocol [17,71].

4. Conclusions

In summary, this work establishes the first integration of tailored rifamycin chemistry and click chemistry to generate a modular series of rifamycin S analogs featuring C8-azido, C8-alkynyl, and C8-triazole substituents. This platform enables systematic SAR studies and downstream bioconjugation for probes, imaging agents, and nanoparticle–antibiotic hybrids. We also report the first preparative-scale HPLC purification of rifamycin derivatives, a key advance for natural product analog isolation. The new analogs showed promising anti-MRSA activity and selective effects on S. mutans, suggesting utility for systemic and oral microbiota targeting. In contrast, reduced activity against E. coli likely reflects efflux-mediated resistance, while decreased RNAP binding is consistent with the loss of C8 hydroxyl group. Ongoing efforts to synthesize click-enabled, C8-modified bxRifs and deAcRifs aim to deliver RNAP inhibitors with improved selectivity and to explore their utility in mechanism-based probe development and Gram-negative sensitization strategies. Our work paves the way for next-generation, modular rifamycin therapeutics with broad applicability.