Tunicamycin was isolated from liquid cultures of Streptomyces chartreusis NRRL B-12338, sourced from the USDA Agricultural Research Service Culture Collection located in Peoria, IL. The compounds TunR1 and TunR2 are not commercially available, and are synthetic compounds developed by the USDA that were synthesized from the native Tun through selective catalytic hydrogenations, as detailed in prior studies (Price et al., 2023; Hering et al., 2020; Price et al., 2017b). Compounds were purified according to Price et al. (2016), and the purity of the compounds was checked by high-performance liquid chromatography (HPLC), matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF/MS), and nuclear magnetic resonance spectroscopy (NMR). Sodium deoxycholate (DOC, 5β-cholan-24-oic acid-3α, 12α-diol sodium salt) was purchased from Sigma-Aldrich, St. Louis, MO, United States. The stock solution of Tun, TunR1, and TunR2 was prepared at a concentration of 10 mg/mL, in DMSO or combined with 5 mg/mL of DOC in water and dissolved at 50°C for 10 min. Amikacin, gentamicin, carbenicillin, cefquinome, ceftazidime, thiazolidine (TZ), and N-acetyl-cysteine were obtained from Sigma-Aldrich (St Louis, MO, United States). Cysteamine, glycolaldehyde TZ, and glyceraldehyde TZ were prepared synthetically as described previously (Ramazani et al., 2017). The stock solutions of the antibiotics and small thiols were prepared in distilled water, with the exception of ceftazidime, which was prepared in DMSO. The stock solutions were stored at −20°C until use.

Several Mycobacterium avium subspecies were used in this study (Table 1). A total of 10 strains including seven strains of Map, two strains of M. avium subsp. hominissuis (Mah), and M. avium subsp. avium (Maa) strain 35,713. All Map strains were cultured in Middlebrook 7H9 broth supplemented with 10% oleic acid-albumin-dextrose-catalase (OADC) (Becton Dickinson) and 2 μg/mL mycobactin J (Allied Monitor, Fayette, MO, United States). These were incubated at 39°C for 6 to 8 weeks. Mah and Maa strains were grown in Middlebrook 7H9 broth containing 0.5% glycerol and 10% OADC, at 37°C for 2 and 4 weeks, respectively. Furthermore, M. smegmatis strain mc²155 and M. marinum strain M were cultured in Tryptone-Glucose-Yeast Extract (TGY) broth at 37°C for 3 days and 30°C for 1–2 weeks, respectively, and were included as controls in selected assays.

Tun, TunR1 and TunR2 in DOC were evaluated by HPLC and MALDI-TOF/MS as previously described by Orwa et al. (2002) and Price et al. (2016). A 1D and 2D-NMR spectroscopy was also used. ¹H NMR measurements were carried out with a Varian Unity Inova 500 spectrometer at 500 MHz (Varian, Germany). All NMR spectra were recorded at 298 K using D₂O. The FT-IR spectra of samples were re-corded on a PerkinElmer FT-IR spectrometer in the range of 400–4,000 cm⁻¹ at 298 K by averaging 120 scans with a resolution of 4 cm⁻¹. Each of the dried samples was used to prepare the KBr pellets for FT-IR analysis.

Also, the anti-mycobacterial activity of tunicamycin compounds in DMSO and DOC were evaluated in Map strain K-10 and M. smegmatis as described below, in order to comparate the potency of the compounds in the different drug vehicles.

The minimal inhibitory concentration (MIC) was determined using the previously described resazurin microtiter assay (REMA) in 96-well microplates as per EUCAST recommendations (EUCAST, 2019; Viale et al., 2021; Schön et al., 2020). Map, M. avium subsp. avium (Maa), and Mycobacterium avium subsp. hominissuis (Mah) were grown in Middlebrook 7H9 media supplemented with OADC, along with 2-fold serial dilutions of Tun, TunR1, TunR2 or drug vehicle alone (DMSO or DOC) as a growth control, ranging from 0.25 to 32 μg/mL. For Map, mycobactin J (2 mg/L) was also added to the culture medium. Amikacin and gentamicin were used as a control of the antimicrobial susceptibility test (AST) method. Mah cultures were incubated for 7 days, and Map and M. avium cultures were incubated for 21 days. For Mah and Map, a modified colorimetric microtiter plate-based assay containing resazurin was used to measure metabolic activity during incubation. The resazurin indicator changes color from purple to pink when metabolic activity is detected indicating viability. The MIC₉₀ was defined as the lowest drug concentration that showed no visible growth or ≥90% of inhibition, and MIC₅₀ as the lowest drug concentration that showed 50% of growth inhibition The percentage of inhibition was calculated, measuring the optical density (OD) at 570 nm, as:

In addition, the minimum bactericidal concentration (MBC) was determined by plating the contents of wells showing no metabolic activity onto Middlebrook 7H10 agar medium supplemented with OADC and mycobactin, without tunicamycin or its derivatives. The MBC was defined as the lowest drug concentration that showed a 3-log reduction from the original inoculum (Santos et al., 2020).

For the MIC determination of M. marinum and M. smegmatis, the inocula were prepared at a density equivalent to McFarland turbidity standard 0.5 in Tryptone-Glucose-Yeast Extract (TGY) broth, and incubated with 2-fold serial dilutions of the antibiotic for 3 and 6 days, respectively. The MIC was defined as the lowest drug concentration that showed no visible growth.

The Middlebrook 7H9 culture medium used for determining the MIC of Map, Maa, and Mah contains 5 mg/mL of bovine serum albumin (BSA). Since Tun and its derivatives are lipophilic drugs that bind extensively to proteins such as BSA, and only the unbound (free) drug can exhibit activity against the bacteria (Celestin and Musteata, 2021), we evaluated the effect of varying concentrations of BSA (5–25 mg/mL) on the MIC of M. smegmatis.

The MIC of β-lactams (carbenicillin, cefquinome, and ceftazidine), and small thiols (thiazolidine (TZ), N-acetyl-cysteine, cysteamine, glycolaldehyde TZ, and glyceraldehyde TZ) for Map and M. smegmatis were also tested to evaluate their potential to be used together with TunR2 to create synergistic killing effect.

Preparation of mycobacterial samples for transmission microscopy has been described previously (Bannantine et al., 2003). All mycobacteria were cultured in Middlebrook 7H9 media at an OD₆₀₀ = 0.06 for ~10 days with or without TunR2 (16–256 μg/mL) prior to fixation. The fixation and staining procedures were conducted at room temperature. Cells were fixed overnight in 2.5% glutaraldehyde –0.1 M cacodylate buffer, pH 7.4. Fixed cells were washed in the same buffer three times and were postfixed in 1% OsO₄ in 0.1 M cacodylate buffer, pH 7.4, for 1 h. After washing in the same buffer, cells were incubated with 30% ethanol for 10 min. The cells were further dehydrated with a graded series of ethanol and embedded in epoxy resin (Embed 812). Ultrathin sections for electron microscopy were obtained and stained with uranyl acetate and Reynolds lead citrate and the bacterial cell morphology was observed under a Tecnai G2 Spirit BioTWIN electron microscope.

An MTT assay, which measures the enzymatic reduction of the MTT reagent (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide), was performed to measure cytotoxicity on the MDBK cell line, bovine peripheral blood mononuclear cells (PBMCs), and bovine monocyte-derived macrophages (bMDMs). PBMCs and bMDMs were obtained from healthy Holstein cows housed at NADC, as previously described by Colombatti Olivieri et al. (2023) and Colombatti Olivieri et al. (2021). Cells were seeded in 96-well plates (Corning Costar) at 5 × 10⁴ cells/well for MDBK and bMDMs, and 2.7 × 10⁵ cells/well for PBMCs. Two-fold serial dilutions of Tun, TunR1, TunR2, in DOC aquous solution, were added to each plate and incubated for 16 h (MDBK and PBMC) or 24 and 48 hs (bMDM) at 37–39°C in 5% CO₂. The concentration of each compound ranged from 0.0039 μg/mL to 128 μg/mL. As a negative control, cells were left untreated or treated with DOC alone. Cell fixed with ethanol were used as a positive control (100% inhibition). The MTT reagent was added to plates and incubated for an additional 4 h before measuring on a microplate reader (SpectraMax 340 PC, Molecular Devices) at OD_590nm. To calculate the half-maximal (50%) inhibitory concentration (IC₅₀) the following equation was used:

Bovine erythrocytes (RBCs), obtained from Holstein cows, were used in the hemolysis assay at a dilution of 1:20 in Dulbecco’s Phosphate-Buffered Saline (D-PBS, Thermo Fisher Scientific Inc.). Two-fold serial dilutions of Tun, TunR1, and TunR2 in sodium DOC were added to each plate. The negative control consisted of RBCs in D-PBS, representing 0% hemolysis and RBCs in distilled water, representing 100% hemolysis. All wells were incubated at room temperature for 2 h and then centrifuged at 700 × g for 3 min. The supernatants were measured at OD_420nm. The following equation was used to calculate the half-maximal (50%) hemolytic concentration (HC₅₀):

Bovine monocyte-derived macrophages were infected with a multiplicity of infection (MOI) of 10, using the Map K-10 strain. After 4 h, the bMDM were washed with warm D-PBS to remove the extracellular bacteria, and different concentrations of Tun, TunR1 and TunR2 were added to the culture media. Drug concentrations with less than 20% cytotoxicity were selected to treat the infected bMDM. After 24 and 48 h post-infection, cells were lysed with 100 μL of 0.1% Triton X-100. The dilutions of lysed cells were plated in 7H10 agar to assess bacterial colony forming units (CFUs).

One-way ANOVA and Dunnett’s multiple comparison test were used to assess the analysis of the transmission electron microscopy data. Two-way ANOVA and Tukey multiple comparison test were use for the hemolytic and MTT assays. Finally, one-way ANOVA and Šídák multiple comparison test were used for the bioassay. A p-value below 0.05 was considered significant. Data and graphical representation were based on GraphPad Prism version 10.3.1 (GraphPad Software Inc., San Diego, CA, United States).

The present work undertaken to develop a water-soluble Tun, TurR1, and TunR2 formulation solubilized by sodium deoxycholate as an alternative to DMSO, to reduce the solvent toxicity. All three tunicamycin-based compounds are completely solubilized in water when mediated by emulsification with DOC. The TunR1 and TunR2 solutions were also stable after cooling, and remained in solution even after standing for 18 h at room temperature (Supplementary Figures S1, S2).

Several strains of Map, Mah, and Maa were tested to determine susceptibility to the Tuns. M. marinum was used as a control. All M. avium subspecies showed equal sensitivity to TunR1 and TunR2, but were more susceptible to Tun. Among the seven Map strains, K-10 exhibited the highest susceptibility to both native and synthetic compounds, with MICs <0.25 μg/mL, 1 μg/mL, and 4 μg/mL for Tun, TunR1, and TunR2, respectively (Table 1 and Figure 1). However, the range of resistance among Map strains was broad, spanning from 4 μg/mL (K-10) to 32–64 μg/mL (Kay and 285 strains) for TunR2. The Mah strains were generally more resistant to the Tuns compared to Map strains (Table 1). The MBC was equal to or 2× higher than the MIC (Table 1 and Supplementary Figure S3). On the other hand, M. marinum had the same sensitivity (0.025 μg/mL) to Tun and its synthetic derivatives (TunR1 and TunR2) (Table 1). Furthermore, the efficacy of tunicamycin compounds in DMSO and DOC was compared in Map K-10 and M. smegmatis strains, revealing the same MIC value and confirming that the compounds maintained the potency in the DOC aqueous solution (data not shown). The impact of serum albumin on the minimum inhibitory concentration (MIC) of Tun, TunR1, and TunR2 was also assessed, given that the Middlebrook 7H9 culture medium contains 5 mg/mL of bovine serum albumin (BSA), which can bind free Tun. As the concentration of BSA increases, the MIC value for M. smegmatis also increases (Supplementary Table S1).

We evaluated the MIC of Map for β-lactams, and small thiols, as Tun and its derivatives have shown synergistic activity with other antibiotics that target cell wall synthesis (Price et al., 2017a). Small thiols have also shown promise in treating tuberculosis in humans (Vilchèze and Jacobs, 2023). Our results show that Map has a low sensitivity to the tested compounds, in contrast to M. smegmatis which only had low sensitivity to carbenicillin and N-acetyl-cysteine (Table 2). Therefore, the combination of these compounds with TunR2 was not evaluated.

Map K-10 was exposed to varied concentrations of TunR2 and observed by transmission electron microscopy. No discernable effect in cell morphology was observed between 4–16 μg/mL TunR2. Bacteria with cytoplasm detached from the cell wall, clearly swollen forms (like protoplasts), ghost cells, degraded forms, and even a significant reduction in the number of bacilli per field were detected as morphological changes using high concentrations of TunR2 compared to the control (Figure 2B). The reduction of number of bacilli per field and the percentage of Map with morphological changes was only statistically significant with 256 μg/mL of TunR2 compared to bacilli cultured at 0 μg/mL TunR2 (Figures 3A,B). Those bacilli did not have sharp margins at the periphery of the cells compared to the culture with no TunR2 (Figure 2A, left column). In addition, we evaluated the percentage of growth inhibition by REMA assay on the same samples, and from ≥32 μg/mL, the % of growth inhibition is significant and coincides with the morphological changes observed by EM (Figure 3C). Collectively, these data demonstrate a toxic effect of high concentration of TunR2 on Map.

The viability of bovine cells was quantified by an MTT assay, which is among the most used assays to measure cell viability. The inhibitory concentration 50 (IC₅₀) from this assay was lower for Tun and TunR1 than TunR2, with the PBMCs as the most susceptible cells (Figures 4A, 5A,B and Table 3). The percent of bovine red blood cell hemolysis increased as the Tun concentrations increased (Figure 4B). TunR2 invoked 100% hemolysis at higher concentrations than Tun or TunR1. An average hemolysis of 38% was observed at a concentration of 15.6 μg/mL of TunR2, while complete hemolysis occurred at a concentration of 62.5 μg/mL. The hemolytic concentration 50 (HC₅₀) for Tun and TunR2 ranged between 15.6 and 31.7 μg/mL, while for TunR1, it ranged between 7.8 and 15.62 μg/mL (Figure 4B). The drug vehicle, DOC, was not cytotoxic.

Administering microdoses of Tun, TunR1, or TunR2 to infected macrophages resulted in a notable decrease in the count of Map CFU. However, this reduction was temporary since the decrease was only observed 24 h post-infection, and not at 48 h post-infection (Figure 5C and Supplementary Table S2). Collectively, these findings suggest that these treatments might not have a significant impact on controlling the infection.