Synthesis and Evaluation of the 2-Aminothiazoles as Anti-Tubercular Agents

The 2-aminothiazole series has anti-bacterial activity against the important global pathogen Mycobacterium tuberculosis. We explored the nature of the activity by designing and synthesizing a large number of analogs and testing these for activity against M. tuberculosis, as well as eukaryotic cells. We determined that the C-2 position of the thiazole can accommodate a range of lipophilic substitutions, while both the C-4 position and the thiazole core are sensitive to change. The series has good activity against M. tuberculosis growth with sub-micromolar minimum inhibitory concentrations being achieved. A representative analog was selective for mycobacterial species over other bacteria and was rapidly bactericidal against replicating M. tuberculosis. The mode of action does not appear to involve iron chelation. We conclude that this series has potential for further development as novel anti-tubercular agents.


Introduction
Tuberculosis (TB) remains a deadly global threat and continues to be one of the leading causes of death worldwide [1]. Although significant strides have been made toward combating this disease, coordinated global health efforts have fallen short, in part due to patient incompliance to current treatment regimens [1]. The current minimum treatment regimen for TB requires a combination therapy of isoniazid (INH), rifampicin (RIF), pyrazinamide (PZA), and ethambutol (EMB) for two months, followed by INH and RIF for an additional four months [2]. This intensive therapy is becoming increasingly ineffective with the emergence of both multidrugresistant TB (MDR-TB) and extensively drug-resistant TB (XDR-TB). MDR-TB is resistant to isoniazid and rifampicin, whereas XDR-TB is resistant to isoniazid, rifampicin, one of the fluoroquinolones, as well as one injectable drug. Both MDR-TB and XDR-TB have the potential to render our current treatment regimen ineffective. Therefore, the discovery and development of novel TB drugs is urgent.
Researchers have relied on a variety of screening techniques to identify promising compound series for TB drug development. A large number of compound series have been give 2. The amine intermediate (5) prepared in this manner was subsequently acylated to give 30-40 or treated with cyclopentyl isocyanate or phenyl isocyanate to give urea 41 or 42, respectively. Analogs 43-61, which examine different substituents at the C-2 position of the thiazole core, were prepared in a similar manner.
The C-4 ketone (62) and carboxamide (63) were prepared from carboxylic acid 6, which is accessible from commercially available ethyl 2-aminothiazole-4-carboxylate (Fig 3). Treatment of compound 6 with 2-chloropyridine in THF under reflux conditions followed by conversion to Weinreb amide and Grignard in the presence of methyl magnesium bromide gave 62 in 49% yield. Similarly, acylation of 6 gave intermediate 7, which after a similar Weinreb amide conversion afforded 63.
We also prepared analogs which are comprised of modifications to the thiazole core. Thiadiazole 64 was prepared from commercial 2-aminothiadiazole (7) via standard amidation chemistry. Thiadiazole 65 and oxadiazole 66 were prepared in two steps (Fig 4). Treating hydrazide 8 with isothiocyanate and /or isocyanate 9 gave intermediate 10, which after acidcatalyzed ring-closure afforded thiadiazole 65. Treatment of 10 with EDC gave oxadiazole 66.
The pyrimidine-based analog 67 was prepared according to The synthesis of pyrazole analogs 69 and 70 is shown in  was reacted with methoxyacrylonitrile (14) in the presence of base to give pyridylpyrazole intermediate 15, which was readily converted to 69 via standard acylation protocols. Treatment of 15 with isocyanate afforded urea 70.

Structure-Activity Relationship (SAR) studies
The goal of our research efforts was to explore the AT compound class, and establish SAR with respect to activity against M. tuberculosis and selectivity over mammalian cells. Compounds were designed to explore various segments of the hit structure 1 including the thiazole core. The 2-pyridyl was previously reported to be a preferred residue at the C-4 position [5]. Therefore, beginning with a 2-pyridyl substituent fixed at C-4 position of the thiazole, we explored the C-2 position for optimal substitution pattern (Fig 7).
We noted increased anti-tubercular activity in compounds with amide or urea substituents at C-2 position, as compared to amines. For example, the cyclohexylamide 35 had 14-fold greater activity as compared to the analogous amine 27, and the amide 33 had 3-fold greater activity over the analogous amine 26. Both 33 and 35 had good separation from cytotoxicity, with respective selectivity index (SI = MIC/TC 50 ) of 37 and 28. The potency and selectivity of these compounds highlighted the benefit of incorporating an amide linkage to the core, thus providing a good starting point for compound optimization.
Urea analogs were also more active relative to their aniline counterparts; for example the cyclopentyl urea analog 41 demonstrated good activity against M. tuberculosis, as well as selectivity (SI = 26). A phenyl urea analog 42 had 3-fold higher activity as compared to the aniline 21. Furthermore, lipophilic amides retained activity, while attempts to introduce small alkyl (31, and 32) or small polar (34, 36 and 37) amides generally resulted in loss of activity.
Consistent with prior reports, we found that the placement of a 2-pyridyl unit at C-4 position of the thiazole core conferred good activity [10,11]. Given the limited scope of these reports on C-4 SAR, we decided to further explore requirements for activity at this position. We designed, synthesized and evaluated a diversity of substituents ranging in structural and functional scope to include alkyls, aromatics, heteroaromatics and specific 2-pyridyl surrogates (Fig 8).  These analogs were designed as combinations with selected C-2 residues known to confer good activity. The 3-pyridyl (51) isomer was devoid of any activity, an indication of a specific preference for the 2-pyridyl residue at C-4 position. In general, C-4 alkyls, such as 48, or plain aromatics (43, 45, 49 and 55) were inactive. 2-Pyridyl replacements with 2-methoxyphenyl (46) or 2-hydroxyphenyl (47) yielded analogs devoid of any activity. Several other 2-pyridyl isosteric replacements tested showed no activity. The C-4 carboxamide or ketone (62 and 63) analogs also tested inactive. 58),61) and 2-quinolyl (60) analogs showed moderate activity. However, these compounds also showed moderate cytotoxicity. Accommodation of these residues at C-4 however provides an opportunity for modulating physicochemical properties of the compound series without much penalty in activity.
We investigated isosteric replacement of the thiazole scaffold in amine 19 (MIC = 4.5 μM) compared to thiadiazole 65, oxadiazole 66, and pyrimidine 67. In addition, we investigated thiazole replacement of 33 (MIC = 0.70 μM) by comparing the activity against M. tuberculosis of related thiadiazole 64 and pyrimidine 68. We also looked into replacing the thiazole with a pyrazole (69 and 70). However, all modifications resulted in total loss of activity. These findings demonstrate that isosteric replacement of the thiazole core by other five-or six-membered heterocyclics do not necessarily produce compounds with retained biological activity i.e. they are not bioisosteres. Therefore, we conclude that the thiazole core must play a significant role in the activity against M. tuberculosis, in addition to the structural display of key residues.

Aminothiazoles have limited broad spectrum activity
Two compounds were selected for testing against other bacterial species-we used the nonpathogenic mycobacterial species Mycobacterium smegmatis, as well as Staphylococcus aureus (Gram positive) and Escherichia coli (Gram negative) ( Table 1). Both compounds were potent against M. tuberculosis, but also had activity against M. smegmatis, suggesting that the series targets mycobacteria broadly. One compound also had activity against S. aureus, suggesting that there may be limited broad spectrum activity, but neither compound was active against E. coli.

Aminothiazoles have bactericidal activity against M. tuberculosis
We determined the effectiveness of a representative compound in killing aerobically-growing M. tuberculosis. Compound 20 had rapid killing activity, resulting in complete sterilization of cultures (>4 log kill) within 7 days (Fig 10), even at 0.625 μM. The minimum bactericidal concentration (MBC), defined as 3 log kill in 21 days, was less than the MIC i.e. < 0.5 μM, confirming that this compound is bactericidal (defined as MBC/MIC ratio <4).

Lack of iron chelating activity
Due to the proximity of the heteroatomic pyridine and an electron-rich aminothiazole ring system, these aminothiazoles have the potential to chelate iron via the coordination of iron to the nitrogen atoms of both the pyridine and aminothiazole rings. This could be the mechanism of activity against M. tuberculosis, since iron is an essential nutrient for bacterial growth. To determine whether this is the case, we measured the capability of the representative compound 20 to inhibit growth in the presence of excess iron ( Fig 11); we monitored growth over 7 days in six different concentrations bracketing the MIC in larger scale culture (growth tubes). As expected compound 20 completely prevented bacterial growth at the highest concentrations (2-5 μM); growth was also inhibited at 1 μM. When additional iron was provided there was no growth at concentrations of 1-5 μM, thus demonstrating that iron was not available to overcome compound-mediated inhibition and in fact if anything, the bacteria were slightly more susceptible.
We also tested iron supplementation in the form of hemin on solid medium with compound 20 and three other compounds, but no difference in MIC was seen (data not shown). This  suggests that chelation of iron from the medium is not the mechanism of action for this class of compounds.

Conclusions
Chemical library screening has successfully identified a number of compounds with excellent activity against M. tuberculosis including the 2-aminothiazoles [5]. We conducted an SAR assessment of different substitutions at the C-2 and C-4 positions, as well as possible replacements for the thiazole core. These modifications are in agreement with other aminothiazole analogs reported previously [10,11]. Concurrent with previous studies, a 2-pyridyl moiety at the C-4 position is essential for bacterial activity, as replacement of the pyridine ring resulted in a loss of activity. The efforts around the C-2 position indicate flexibility to various modifications with amine and amide all showing activity. 4-(Pyridin-2-yl)-N-(pyridin-3-yl)thiazol-2-amine (20) was the most potent analog prepared. Unfortunately, as with other active compounds presented in this manuscript and those reported previously [10,11], the activity and cytotoxicity of 20 are strongly correlated. However, a few analogs demonstrated improved selectivity. Amide-linked phenyl substituents at the C-2 position provided potent analogs with good separation of cytotoxicity, similar to what has been previously reported [10]. In this study, compound 33, consisting of a 4-tert-butylbenzamide at the C-2 position, gave an MIC of 0.30 μM and a selectivity index of~28 suggesting that cytotoxicity can be dialed out of the series. Urea-linkages at the C-2 position were not previously reported, but we demonstrated that they also have good potency and some selectivity. We also demonstrated that some ketones and carboxamide residues can confer moderate activity. This expands the opportunities for structure-property relationship (SPR) studies and property modulation for in vivo studies.
Representative compounds demonstrated selectivity towards mycobacteria, suggesting a target that is restricted to this genus. In kill kinetic studies, the representative compound 20 was bactericidal with rapid killing of the bacteria and complete culture sterilization in 7 days. These findings suggest targeting of a particularly vulnerable and essential enzyme or pathway in the bacterium. Although the mechanism of action for the compound series remains unknown, we confirmed that it is not likely to result from iron chelation in the medium. Thus, further development of the AT series is promising.

Determination of minimum inhibitory concentration (MIC)
MICs were determined against M. tuberculosis H37Rv (London Pride) [15] grown in Middlebrook 7H9 medium containing 10% OADC (oleic acid, albumin, dextrose, catalase) supplement (Becton Dickinson) and 0.05% w/v Tween 80 (7H9-Tw-OADC) under aerobic conditions. Bacterial growth was measured by OD after 5 days of incubation at 37°C. The MIC was defined as the minimum concentration at which growth was completely inhibited and was calculated from the inflection point of the fitted curve to the lower asymptote (zero growth).

Cytotoxicity assay
The Vero cell line (African green monkey kidney epithelial cells: ATCC CRL-1587) was grown in DMEM, High Glucose, GlutaMAX™ (Invitrogen), 10% fetal bovine serum (FBS), and 1X penicillin-streptomycin solution (100 U/mL). Compounds were solubilized in DMSO and assayed as a 10-point three-fold serial dilution. Compounds were incubated with cells for 2 days at 37°C, 5% CO 2 . CellTiter-Glo1 Reagent (Promega) was added and relative luminescent units (RLU) measured. Inhibition curves were fitted using the Levenberg-Marquardt algorithm. TC 50 is the compound concentration that gave 50% inhibition of growth.

Activity spectrum
The serial proportion method was used to determine MIC 99 on solid medium [16]. LB was used for E. coli and S. aureus. Middlebrook 7H10 supplemented with 10% v/v OADC was used for mycobacteria. E. coli and S. aureus were incubated at 37°C overnight. M. smegmatis was incubated at 37°C for 3 days. The MIC 99 was the lowest concentration of compound, which yielded less than 1% growth.

Kill kinetics
A late log phase (OD 590 0.6-1.0) culture of M. tuberculosis (H37Rv) was adjusted to OD 590 0.1 in 7H9-Tw-OADC; 50 μL was used to inoculate 5 mL 7H9-Tw-OADC (10 5 CFU/mL). Compound was added to each tube to a final DMSO concentration of 2%. Cultures were incubated at 37°C and serial dilutions plated on 7H10 agar plates to determine CFU/mL. Plates were incubated for 4 weeks before colonies were counted.

Growth curves
Growth curves were conducted in 5 mL of 7H9-Tw-OADC in 16 x 125 mm glass tubes with stir bars. M. tuberculosis was inoculated to a theoretical starting OD 590 of 0.04. OD 590 was measured daily.
General synthesis (procedure I) for compounds 17-29 and 43-61. Compounds 17-29 and 43-61 were prepared following this general protocol unless otherwise noted. To substituted 2-bromoethanone in ethanol was added substituted thiourea (1.02 eq). The mixture was stirred at 70°C. The reaction was monitored via LC/MS. After 2 h, the reaction mixture was cooled to room temperature and precipitate was formed. The precipitate was collected by vacuum filtration and washed with acetone. The solid was dissolved in 2 M NaOH (25 mL) and extracted with EtOAc (3 x 50 mL). The combined organic layers were dried over Na 2 SO 4 and concentrated in vacuo desired product.
General synthesis for compounds 30-40. Compounds 20-30 were prepared following this general procedure unless otherwise noted. To 5 dissolved in THF was added triethylamine (1.0 eq). The mixture was cooled in an ice-bath, and to the mixture was added substituted acid chloride (1.0 eq). The mixture was stirred at room temperature and the reaction was monitored via LC/MS. After 1 h, the reaction solvent was evaporated. The resulting crude was purified by recrystallization from ethanol or purified via flash column chromatography to give desired product.  2, 14.9, 111.9, 120.3, 122.7, 137.0, 149.1, 149.7, 152.1, 159.6, 172.2; HRMS MS ESI m/z calcd for C 12 H 11 N 3 OS (M+H) + 246.0696, found 246.0675 (Δ 2.1 ppm).
(iv) To a solution of tert-butyl 3-(methoxy(methyl)carbamoyl)-1H-pyrazole-1-carboxylate (2.5 g, 9.8 mmol) in dry THF (15 mL) at -78°C under N 2 atmosphere, methyl magnesium bromide (50 mL, 3M solution) was added drop wise and stirred at that temperature for 2 h. Then allowed the reaction to reach room temperature and continued stirring for 16 h. After TLC showed completion, the reaction was quenched with saturated ammonium chloride solution (25 mL) and extracted with EtOAc (3 x 100 mL). The combined organic layer was dried over anhydrous Na 2 SO 4 and concentrated under reduced pressure to obtain 1-(1H-pyrazol- 3-yl)