1. Introduction

Figure 1. Structures of BTZ043 and PBTZ169.

Figure 1. Structures of BTZ043 and PBTZ169.

2. Materials and Methods

3. Results

3.1. Chemistry

Chemical Synthesis

Compounds 1−30 were prepared as shown in Scheme 1. The intermediate 43 and 45 was prepared according to published methods[11]. Click chemistry between 45 and corresponding azide using CuSO₄ and sodium ascorbate in EtOH/H₂O gave final compounds 1−30.

Scheme 1. Synthesis of compound 1−30.

Scheme 1. Synthesis of compound 1−30.

Compounds 31−36 were prepared as shown in Scheme 2. The reaction intermediates 46a−c and 47a−c were prepared according to the published procedure[11]. The final cyclization reaction between 44 and the above intermediates afforded compounds 31−36 in medium yield.

Scheme 2. Synthesis of compounds 31−36.

Scheme 2. Synthesis of compounds 31−36.

Compounds 37−43 were prepared as shown in Scheme 3. The reaction intermediates 48a−c and 49a−c were prepared according to the published procedure[11]. The final cyclization reaction of 44 with the above intermediates provided the target compounds 37−42 in 30%−50% yields.

Scheme 3. Synthesis of compound 31−36.

Scheme 3. Synthesis of compound 31−36.

3.2. MIC Determination and Structure Activity Relationship

The compound antimycobacterial activity was performed as previously described[11]. The triazole moiety has been widely used in drug molecules and regarded as a privileged motif[12]. More importantly, the facile click chemistry to construct triazole is readily achieved through the reaction between a terminal alkyne and an azide, the 1,2,3-triazole linker compounds 1−11 displayed MIC 4−237 ng/mL, The MIC values are shown in Table 1. Specifically, the N-cyclohexyltriazole 1 exhibited MIC 16 ng/mL against Mtb H37Rv. the N-phenyltriazole 2 was 4−fold more potent, with MIC 4 ng/mL. Reduced activities were recorded for N-benzyltriazole 3 and compound 4. Notably, all compounds showed ~5−fold activity improvement as compared to our previously reported 6-methanesulfonyl counterparts[11].

Table 1. Antitubercular activity of compounds 1−11 against Mtb H37Rv.

Table 1. Antitubercular activity of compounds 1−11 against Mtb H37Rv.

entry	R	MIC (ng/mL)	entry	R	MIC (ng/mL)
1		15	7		237
2		4	8		29
3		15	9		111
4		16	10		25
5		22	11		153
6		79	PBTZ169		0.2

Having profiled the tail position, we then defined the phenyl group and intended to further investigate the substitute effect at the phenyl ring (Table 2), aiming to obtain compounds with higher potency and improved physiochemical properties than compound 2. Replacement of one or more hydrogens with a fluorine atom is a commonly employed tactic to elevate the compounds ADME properties[13]. Thus, the analogs 12, 13 and 14 with one fluorine atom at different position were prepared. The meta-F substituted compound 13, MIC 4.0 ng/mL, had the highest potency among these three compounds, compounds 12 and 14 had MIC 6 and13 ng/mL, respectively.

We then explored the meta- position by introducing an electron withdrawing or donating group. Within this frame, compounds 15−21 were generated. Unfortunately, none of them exhibited activity improvement, higher MIC values of 7−101 ng/mL were recorded. Specifically, gradual activity loss was observed from chloro- to bromo- and iodo-. Other groups like methyl, methoxyl, and trifluoromethyl substitution also diminished the Mtb growth inhibition activity. The meta-cyano substituted compound 21 showed the weakest activity. Thus, our optimization choice was turned to phenyl di-substitution, generating compounds 22−29. The di-substitution strategy also failed to provide compounds with further potency increase. Among these compounds, the three difluoro-substituted compounds 22, 23 and 24 showed MICs lower than 10 ng/mL. In contrast, the 2,5-difluoro substituted compound 24 displayed MIC 129 ng/mL.

Table 2. Antitubercular activity of compounds 12−30 against Mtb H37Rv.

Table 2. Antitubercular activity of compounds 12−30 against Mtb H37Rv.

entry	R	MIC (ng/mL)	entry	R	MIC (ng/mL)
12		6	22		7
13		4	23		8
14		13	24		129
15		7	25		7
16		30	26		19
17		78	27		31
18		59	28		15
19		29	29		12
20		21	30		25
21		101	PBTZ169		0.2

Having established that the triazole linker and the phenyl attachment could drop down MICs to less than 10 ng/mL, we next replaced this linker with its bioisosters. Other five-member heterocycles like oxazole, thiazole, oxadiazole or thiodiazole are considered as suitable surrogates of triazole[14]. We postulated that replacement of triazole with the heterocycles likely generates analogs with different physiochemical properties. Compounds 31−42 were designed and synthesized using various linkers (Table 3). It was interesting that the oxazole linker compounds 31−33 and the three 1,3,4-oxadiazole compounds 37−39 displayed antitubercular activity higher than or comparable to compound 2. The MIC of compound 34 with a thiazole linker was 7 ng/mL, while compounds 35 and 36 with substituted phenyl attachment exhibited decreased activity (MIC 16 and 17 ng/mL, respectively). Similarly, the 1,3,4-thiodiazole-linked compounds 40−42 displayed higher MIC values.

Table 3. Antitubercular activity of compounds 31−42 against Mtb H37Rv.

Table 3. Antitubercular activity of compounds 31−42 against Mtb H37Rv.

entry	R	MIC (ng/mL)	entry	R	MIC (ng/mL)
31		3	34		7
32		3	35		16
33		10	36		17
37		3	40		16
38		3	41		12
39		3	42		13

3.3. DprE1 Inhibition Activity

The above compounds with various linkers were selected based on their MIC values to test potential on-target inhibitory activity of the Mtb DprE1 enzyme. Following a previously reported assay method[9], all of the compounds inhibited the DprE1 activity, with the IC₅₀ values ranging from 0.02 to 7.2 μM (Figure 2, and Supplementary Table S1). Compounds with low MICs accordingly exhibited lower IC₅₀ values as observed in the DprE1 inhibitory assays. For example, compounds 2, 31, 34, and 37 had the IC₅₀ values of 0.02−0.75 μM, and the least potent compound 24 displayed an IC₅₀ of 7.2 μM. As a comparison, the positive control PBTZ169 displayed IC₅₀ 0.009 μM, consistent with the value reported in literature[6].

Figure 2. IC₅₀ determination of compounds 2 (●), 24 (▲), 31 (■), 34 (◆), 37 (●) and PBTZ169 (■), against the DprE1 activity.

Figure 2. IC₅₀ determination of compounds 2 (●), 24 (▲), 31 (■), 34 (◆), 37 (●) and PBTZ169 (■), against the DprE1 activity.

3.4. Compound−mediated Cell Wall Inhibition of Mtb

The ability of the compounds 2 and 37 to inhibit the DprE1 enzyme in Mtb H37Rv was examined by [¹⁴C]-acetate metabolic labeling, as described before (Figure 3) [10]. Analysis of the extractable lipids from radiolabeled bacteria by TLC points out to similar profiles for PBTZ169 and the two tested inhibitors, suggesting the same mechanism of action. The cultures treated with the inhibitors accumulated trehalose monomycolates (TMM) and trehalose dimycolates (TDM) compared to the control bacteria, which indicates interference with the build-up of the mycobacterial cell wall core. DprE1 inhibition causes deficiency in the synthesis of arabinan chains serving as attachment sites for mycolic acids, which then incorporate to soluble lipids, TMM and TDM.

Figure 3. TLC analysis of the [¹⁴C]-acetate labelled lipids, extracted from Mtb H37Rv. Control: TDM, trehalose dimycolates; TMM, trehalose monomycolates; PE, phosphatidylethanolamine; CL, cardiolipin.

Figure 3. TLC analysis of the [¹⁴C]-acetate labelled lipids, extracted from Mtb H37Rv. Control: TDM, trehalose dimycolates; TMM, trehalose monomycolates; PE, phosphatidylethanolamine; CL, cardiolipin.

3.5. Metabolic Stability, Cytotoxicity and Solubility of Selected Compounds

Since compounds 2, 31, 34, 37 and 40 showed higher inhibition potentials against DrpE1, their metabolic stability in human microsomes was then examined (Table 4). As expected, the metabolic stability was affected by the linker portion. Among them, compound 2 containing an 1,2,3-triazole linker had the highest stability (T_1/2=100.4 min) and the lowest intrinsic clearance Cl_int (17.31 mL/min/kg), followed by compound 36 with an 1,2,4-oxadiazole linker (T_1/2=68.2 min, and Cl_int=25.48 mL/min/kg). The thiazole linker compound 34 was the least stable.

Table 4. Metabolic stability of compounds 2, 31, 34, 37 and 40 in human microsomes.

Table 4. Metabolic stability of compounds 2, 31, 34, 37 and 40 in human microsomes.

	2	31	34	37	40
T1/2 (min)	100.4	33.6	24.3	68.2	34.3
Clint(mL/min/kg)	17.31	51.71	71.60	25.48	50.68

Compounds 2 and 37 were further examined for cytotoxicity against HepG2 cell using an MTT assay[15]. No inhibition of cell viability was observed at concentration lower than 1.0 μg/mL (Supporting Information, Figure S1). At 5.0 μg/mL, cell viability was reduced to 72−75%.

The compound solubility is closely related to its bioavailability and in vivo activity, thus the kinetic solubility of compounds 2 and 37 was determined. In PBS buffer, both compounds exhibited higher solubility than PBTZ169 (Supporting Information, Table S2), which also correlated with their reduced clogP values.

3.6. Pharmacokinetics and Efficacy of Compounds 2 and 37

To further characterize compounds 2 and 37, we evaluated their pharmacokinetic profiles in mice. BALB/C mice were given compounds 2 or 37 intravenously (i.v., 2 mg/kg) or orally (p.o., 10 mg/kg). As shown in Table 5, both compounds had acceptable PK profiles with good oral bioavailability of 35.7% and 70.6%, respectively. The T_1/2 and C_max values were shorter than PBTZ169[16], which was dosed orally at 25 mg/kg.

Table 5. Pharmacokinetic parameters for compounds 2 and 37.

Table 5. Pharmacokinetic parameters for compounds 2 and 37.

PK parameters	Compound 2		Compound 37
	PO	IV	PO	IV
Cmax(ng·mL-1)	1798.5	1250.0	1891.3	1860.8
t1/2 (h)	1.3	1.1	0.6	0.3
MRT (h)	2.0	1.5	0.9	0.4
AUC (h·ng/mL)	2185.6	1210.6	1865.2	528.1
Clint (mL/h/kg)		1670.2		3837.4
F (%)	35.7		70.6

C_max, maximum concentration of drug in blood plasma; t_1/2, the elimination half-life of drug; MRT, mean residence time; AUC, area under the curve; Cl_int, hepatic clearance; F, oral bioavailability. Dose: i.v., 2 mg/kg; p.o., 10 mg/kg.

The anti-TB potential of the compounds was evaluated using a mouse model acutely infected with Mtb H37Rv. Both compounds were administered, oral dose 50 mg/kg for 4 weeks, once per day except weekends. PBTZ169 killed Mtb by 3 logs in the lungs compared to the control group (Figure 4). By comparison, compound 37 reduced the bacteria load by 1.2 logs, compound 2 barely displayed efficacy.

Figure 4. Efficacy of compound 2 and 37 in BALB/C mice (n = 6) infected with M. tuberculosis H37Rv.

Figure 4. Efficacy of compound 2 and 37 in BALB/C mice (n = 6) infected with M. tuberculosis H37Rv.

4. Discussion

5. Conclusions

References

1. Global Tuberculosis Report 2022; 2022, World Health Organization.
2. Gauri, S. Gauri S. Shetye, Scott G. Franzblau, and Sanghyun Cho, New tuberculosis drug targets, their inhibitors, and potential therapeutic impact. Transl. Res. 2020, 220, 68–97. [Google Scholar]
3. Makarov, V.; Manina, G.; Mikušová, K.; Mollmann, U.; Ryabova, O.; Saint-Joanis, B.; Dhar, N.; Pasca, M. R.; Buroni, S.; Lucarelli, A. P.; Milano, A.; De Rossi, E.; Beláňová, M.; Bobovská, A.; Dianišková, P.; Korduláková, J.; Sala, C.; Fullam, E.; Schneider, P.; McKinney, J. D.; Brodin, P.; Christophe, T.; Waddell, S.; Butcher, P.; Albrethsen, J.; Rosenkrands, I.; Brosch, R.; Nandi, V.; Bharath, S.; Gaonkar, S.; Shandil, R. K.; Balasubramanian, V.; Balganesh, T.; Tyagi, S.; Grosset, J.; Riccardi, G.; and Cole, S. T. Benzothiazinones kill Mycobacterium tuberculosis by blocking arabinan synthesis. Science 2009, 324, 801–804.
4. Trefzer, C.; Škovierová, H.; Buroni, S.; Bobovská, A.; Nenci,S.; Molteni, E.; Pojer, F.; Pasca, M. R.; Makarov, V.; Cole, S. T.; Riccardi, G.; Mikušová, K.; and Johnsson, K., Benzothiazinones are suicide inhibitors of mycobacterial decaprenylphosphoryl-beta-D-ribofuranose 2'-oxidase DprE1. J. Am. Chem. Soc. 2012, 134, 912–915.
5. Neres, J.; F. Pojer, E. M.; Chiarelli, L. R.; Dhar, N.; Boy-Rottger, S.; Buroni, S.; Fullam, E.; Degiacomi, G.; Lucarelli, A. P.; Read, R. J.; Zanoni, G.; Edmondson, D. E.; De Rossi, E.; Pasca, M.; McKinney, R. J. D.; Dyson, P. J.; Riccardi, G.; Mattevi, A.; Cole, S. T.; Binda, C. Structural basis for benzothiazinone-mediated killing of Mycobacterium tuberculosis. Sci. Transl. Med. 2012, 4, 150ra121.
6. Makarov, V.; Lechartier, B.; Zhang, M.; Neres, J.; van der Sar, A. M.; Raadsen, S. A.; Hartkoorn, R. C.; Ryabova, O. B.; Vocat, A.; Decosterd, L. A.; Widmer, N.; Buclin, T.; Bitter, W.; Andries, K.; Pojer, F.; Dyson, P. J. ; Cole. S. T. Towards a new combination therapy for tuberculosis with next generation benzothiazinones, EMBO Mol. Med., 2014, 6, 372–383. [Google Scholar]
7. Zhang, G.; Howe, M.; and Aldrich, C. C.; Spirocyclic and bicyclic 8-nitrobenzothiazinones for Tuberculosis with improved physicochemical and pharmacokinetic properties. ACS Med. Chem. Lett. 2019, 10, 348–351. [CrossRef] [PubMed]
8. Rui, S.; Bin, W.; Giovanni, S.; Xiaomei, W.; Yuanyuan, S.; Yue, W.; Xin, W.; Laurent R., C.; Yu, L.; and Chunhua, Q. Development of 6-methanesulfonyl-8-nitrobenzothiazinone based antitubercular agents, ACS Med. Chem. Lett. 2022, 13, 593–598. [Google Scholar]
9. Liu, L.; Kong, C.; Fumagalli, M.; Savkova, K.; Xu, Y.; Huszar, S.; Sammartino, J. C.; Fan, D.; Chiarelli, L. R.; Mikusova, K.; Sun, Z.; Qiao, C. Design, synthesis and evaluation of covalent inhibitors of DprE1 as antitubercular agents. Eur. J. Med. Chem. 2020, 208, 112773. [Google Scholar] [CrossRef] [PubMed]
10. Neres, J.; Hartkoorn, R. C.; Chiarelli, L. R.; Gadupudi, R.; Pasca, M. R.; Mori, G.; Venturelli, A.; Savina, S.; Makarov, V.; Kolly, G. S.; Molteni, E.; Binda, C.; Dhar, N.; Ferrari, S.; Brodin, P.; Delorme, V.; Landry, V.; de Jesus Lopes Ribeiro, A. L.; Farina, D.; Saxena, P.; Pojer, F.; Carta, A.; Luciani, R.; Porta, A.; Zanoni, G.; De Rossi, E.; Costi, M. P.; Riccardi, G.; Cole, S. T. 2-Carboxyquinoxalines kill Mycobacterium tuberculosis through noncovalent inhibition of DprE1. ACS Chem. Biol. 2015, 10, 705–714. [Google Scholar] [CrossRef] [PubMed]
11. Dongguang F.; Bin W.; Stelitano, G.; Savková, K., Shi, R. Huszár, S., Han, Q., Mikusová, K., Chiarelli, L. R., Lu, Y., Chunhua, Q.; Structure and activity relationships of 6-sulfonyl-8-nitrobenzothiazinones as antitubercular agents. J. Med. Chem. 2021, 64, 14526–14539. [CrossRef] [PubMed]
12. Kumari, S.; Carmona, A. V.; Tiwari, A. K.; Trippier, P. C. Amide bond bioisosteres: strategies, synthesis, and successes, J. Med. Chem. 2020, 63, 12290–12358. [Google Scholar] [CrossRef] [PubMed]
13. Gillis, E. P.; Eastman, K. J.; Hill, M. D.; Donnelly, D. J.; Meanwell, N. A. Applications of fluorine in medicinal chemistry, J. Med. Chem. 2015, 58, 8315–8359. [Google Scholar] [CrossRef] [PubMed]
14. Karabanovich, G., Dusek, J.; Savkova,K.; Pavlis,O.; Pavkova,I.; Korabecny, J.; Kucera,T.; Vlckova, H. K.; Huszar, S.; Konyarikova, Z.; Konecna, K.; Jandourek,O.; Stolarikova, J.; Kordulakova, J.; Vavrova, K.; Pavek, P.; Klimesova,V.; Hrabalek, A.; Mikusova K.; Roh J. Development of 3,5-dinitrophenyl-containing 1,2,4-triazoles and their trifluoromethyl analogues as highly efficient antitubercular agents inhibiting decaprenylphosphoryl-β-D-ribofuranose 2'-oxidase. J. Med. Chem., 2019, 62, 8115–8139.
15. Yang, J.; Liu, Y.; Xue, C.; Yu, W.; Zhang, J.; Qiao, C. Synthesis and biological evaluation of glaucocalyxin A derivatives as potential anticancer agents, Eur. J. Med. Chem. 2014, 86, 235–241. [Google Scholar] [CrossRef] [PubMed]
16. Makarov. V.; Cole. S. T. 2-Piperazine-1-yl-4H-1,3-benzothiazin-4-one derivatives and their use for the treatment of mammalian infections, WO 2012/066518A1.
17. https://www.newtbdrugs.org/pipeline/compound/btz-043.
18. https://www.newtbdrugs.org/pipeline/compound/macozinone-mcz-pbtz-169.
19. Tiwari, R.; Mollmann, U.; Cho, S.; Franzblau, S. G.; Miller, P. A.; Miller, M. J. Design and syntheses of anti-Tuberculosis agents inspired by BTZ043 using a scaffold simplification strategy, ACS Med. Chem. Lett., 2014, 5, 587–591. [Google Scholar]
20. Piton, J.; Vocat, A.; Lupien, A.; Foo, C. S.; Riabova, O.; Makarov, V.; Cole. S. T., Structure-based drug design and characterization of sulfonyl piperazine benzothiazinone inhibitors of DprE1 from Mycobacterium tuberculosis. Antimicrob. Agents & Chemother. 2018, 62, e00681-18.
21. Makarov, V.; Mikušová, K. , Development of Macozinone for TB treatment: an update. Appl. Sci. 2020, 10, 2269. [Google Scholar] [CrossRef]