Bacteria glycerol stocks were stored frozen at –80°C until use. Laboratory strains used for small compound fragment screening included: Staphylococcus aureus, Pseudomonas aeruginosa, Corynebacterium glutamicum, Enterococcus faecium, Proteus vulgaris, Mycobacterium marinum, and Mtb (Tables S3–6).

Antimicrobial testing: The ability of the synthesized small compounds and fragments to inhibit the growth of the microbes, i.e., minimum inhibitory concentration (MIC), was determined. C. glutamicum (control for acid-fast susceptibility), as well as the Gram-positive and Gram-negative bacteria, were grown overnight [Muller Hinton broth (MHB); 37°C]. These overnight cultures were then diluted in fresh MHB and regrown to early logarithmic phase [optical density of ~0.2 at 600 nm (OD₆₀₀)]. The specific OD₆₀₀ of the strains used for MIC testing is available in Table S3. MIC was determined using standard microdilution antibiotic testing methodology. Briefly, the test compounds were added to MHB (100 μL; 96-well clear, round-bottom plates) and then serially diluted (2-fold dilutions). This resulted in test compound concentrations that ranged from 100 μg/mL to 0.1 μg/mL, with the last column being drug-free. Logarithmic bacteria (50 μL) were then added to each well. Rifampicin was used as the positive control. All compounds were tested in biological duplicates, and MICs were noted on Days 1 and 3 as the lowest concentration that caused complete inhibition of growth as observed visually with the use of an inverted enlarging mirror.

All Mtb microbiology assays were conducted as previously described [25]. Mtb cultures were incubated (37°C) in the biosafety level 3 laboratory, while M. marinum was grown as described for Mtb, with the exception that M. marinum cultures were incubated at 30°C in the biosafety level 2 laboratory. M. marinum and M. tuberculosis H37Rv cultures were grown in Middlebrook 7H9 broth medium (Difco) supplemented with 0.2% (v/v) glycerol, 0.05% (v/v) Tween 80, and ADC supplement (BSA 5 g/L; dextrose 2 g/L; and NaCl 0.81 g/L). As needed, these organisms were also cultured in glycerol-alanine-salts-Tween-iron (GAST-Fe) medium consisting of 0.3 g/L bacto-casitone, 4 g/L dibasic potassium phosphate, 2 g/L citric acid, 1 g/L L-alanine, 1.2 g/L magnesium chloride hexahydrate, 0.6 g/L potassium sulfate, 2 g/L ammonium chloride, 0.05 g/L ferric ammonium citrate, 1% glycerol (v/v) and 0.05% Tween 80 (v/v), (pH 6.6).

For MIC testing, Mtb or M. marinum was grown to an optical density at 650 nm (OD₆₅₀) of 0.2–0.3 in 7H9-based or GAST-Fe medium (Tables S4–6). Cells were diluted to a final OD₆₅₀ of 0.0002 (1:1,000 of parent culture) in the desired medium. Two-fold serial dilutions of the test compounds were made in homologous medium (50 μL/well; clear round-bottom 96-well plate; from 50–0.012 μM). An equal volume of cells was then added to all the wells of the 96-well plate. The assay was performed in duplicate for each growth condition. The plates were incubated at 37°C (Mtb) or 30°C (M. marinum) in sealed ziplock bags, and MIC was recorded on Days 7 and 14 (Mtb) or Days 3 and 7 (M. marinum). MIC was noted as the lowest concentration of the test compound that inhibited visible growth at the given time point. INH and rifampicin were used as positive controls for Mtb, and rifampicin was used as the positive control for M. marinum MICs.

From the compounds derived from urea and amide chemotypes (both acyclic and cyclic), only one compound, the 1,2,4-triazole-containing urea 6c (Table 1), demonstrated activity against Mtb in all three media tested, i.e., glucose, cholesterol, and BSA-free dipalmitoyl phosphatidylcholine (DPPC)-containing media. This compound showed the best anti-Mtb activity (4.7 μg/mL) in BSA-free medium. The measured differences in anti-Mtb activity in the presence vs. absence of BSA suggest an interaction between these compounds and BSA (Tables S3–5). Incorporating the azole moiety in these two chemotypes was based on the activity of the azoles as anti-Mtb agents [37, 38]. Of the three triazole-containing ureas, 6c (Table 1) and 6a, 6b (Table S2), the two that demonstrated activity, 6a (in all media tested), 6b (25 μg/mL only in BSA-free media), have similar lipophilicity. However, only the azole-containing urea 6c (Table 1) has promising anti-Mtb activity. It has been shown that azole-containing compounds inactivate the azole-efflux transporters [12, 43]. The presence of the azole in 6c, if active, would allow for its intracellular accumulation. The preparation of the cyclic amide chemotype represented by monocyclic β-lactams 8a–d (Table S2) was intended to evaluate the effect of the lactam ring (as compared to acyclic counterparts) on the antimicrobial activity. Oxidation of the sulfur atom was expected to increase the electrophilicity of the carbonyl carbon of the amides (8a–d, Table S2). However, none of the amides tested demonstrated activity.

Nontuberculous mycobacteria are also causative agents of various opportunistic human infections, which have been on the rise over the last decade. Unfortunately, current treatment has limited efficacy [44, 45]. In the literature, the term “thiocarbamate” is used to describe both thiocarbamate and carbamothioate functionalities. We focused on evaluating compounds with carbamothioate in their structure. Compounds 9a–d (Table 1) and 9e–g (Table S2) are representatives of the carbamothioate chemotype. Compounds with this functionality that are based on the salicylanilide scaffold have been identified as anti-Mtb and anti-nontuberculous mycobacteria (M. avium and M. kansasii) agents [41].

All carbamothioates in our library are diaryl-substituted, featuring either electron-withdrawing groups (EWG) or electron-donating groups (EDG) on the aryl substituents. Of the compounds prepared by us, those tested against M. marinum demonstrated modest anti-mycobacterial activity as compared to their benzyl-substituted carbamothioates. Of the di-phenyl substituted carbamothioates, those with the best MIC for M. marinum are compounds having EWG groups as the phenyl substituents. These are carbamothioate 9a (Table 1, with activity in all media tested, with the best MIC 6.25 μg/mL in GAST-Fe media); compound 9b and 9d (Table 1, activity shown only in GAST-Fe media, MIC 12.5 μg/mL); compound 9c (Table 1, with activity in all media tested, having the best MIC 6.25 μg/mL in GAST-Fe media). Both carbamothioates 9a and 9c (Table 1) feature a powerful EWG at the ortho position to the carbonyl carbon and at least one Cl atom on the thiophenyl substituent. This structural feature indicates the necessity for the carbonyl carbon to be electrophilic enough to facilitate a nucleophilic attack and also to render the aromatic thiol a suitable leaving group. Their mechanism of action most likely involves the release of aromatic thiolate upon nucleophilic attack on the carbonyl carbon. In contrast, none of the di-phenyl carbamothioates with EDG demonstrated anti-mycobacterial activity.

The α,β-unsaturated systems, acting as reversible sulfhydryl binders 10a–k (Table S2) [34, 35], and the irreversible binder 11 (Table 1, Figure 8) [35], all of which act as MA, were prepared to evaluate their anti-Mtb activity. None of the reversible binders have anti-Mtb activity. However, the 1,3,4-thiadiazole-containing compound 11 (Table 1, Figure 8) demonstrated promising anti-Mtb activity in all three media, with the best (0.6 μg/mL) in casitone-containing media. Compound 11 (Table 1, Figure 8) differs from the known activator of human LCAT (lecitin: cholesterol acyltransferase) compound A by one methyl group (in the S-alkyl substituent of the thiadiazole; compound A has an S-ethyl group) [34]. LCAT catalyzes plasma cholesteryl ester formation and is defective in familial LCAT deficiency (FLD), an autosomal recessive disorder characterized by low high-density lipoprotein, anemia, and renal disease [35]. The activity of the MA acceptor 11 (Table 1, Figure 8) could be due to the presence of a thiadiazole substituent, similarly to the triazole-containing compound 13 (Table 1), the only active compound of the urea class. Thiadiazole-containing compounds have been demonstrated to act both as anti-Mtb/antibacterials [46–48] and antivirals [49].

It is possible that compound 11 (Table 1, Figure 8), regardless of being an MA, behaves similarly to the aforementioned trifluoromethyl cinnamic acids, i.e., not acting as an MA [39]. A potential mechanism of action of 11 (Table 1, Figure 8) may reside in its similarity to sulfone-substituted pyrazine AC2P36 (12; Table S2, Figure 8). The latter has been demonstrated to bind to thiols via nucleophilic attack on the pyrimidine, with the sulfone as the leaving group [50]. Compound AC2P36 (12; Table S2, Figure 8) is reported to directly deplete Mtb thiol pools, with enhanced depletion of free thiols at acidic pH [50]. The similarity in this activity for both compounds 11 and 12 is depicted in Figure 8.

In summary, of the compounds from our library, all 42 compounds tested exhibited good selective toxicity since they had no effect on the Gram-positive and Gram-negative organisms or a non-mycobacterial acid-fast microbe screened in this study. In contrast, two compounds exhibited activity against Mtb. The two that demonstrated the best activity against Mtb (6c and 11, Table 1) possess an azole group; thus, our focus will be on their further evaluation as potential anti-mycobacterial agents.