Identification of Novel Benzoxa-[2,1,3]-diazole Substituted Amino Acid Hydrazides as Potential Anti-Tubercular Agents

Discovery and development of new therapeutic options for the treatment of Mycobacterium tuberculosis (Mtb) infection are desperately needed to tackle the continuing global burden of this disease and the efficacy and cost limitations associated with current medicines. Herein, we report the synthesis of a series of novel benzoxa-[2,1,3]-diazole substituted amino acid hydrazides in a two-step synthesis and evaluate their inhibitory activity against Mtb and selected bacterial strains of clinical importance utilising an end point-determined REMA assay. Alongside this, their potential for undesired cytotoxicity against mammalian cells was assessed employing standard MTT assay methodologies. It has been demonstrated using modification at three sites (the hydrazine, amino acid, and the benzodiazole) it is possible to change both the antibacterial activity and cytotoxicity of these molecules whilst not affecting their microbial selectivity, making them attractive architectures for further exploitation as novel antibacterial agents.


Introduction
Tuberculosis (TB) remains the predominant bacterial infectious disease globally, with approximately 10.4 million new cases and 1.7 million deaths recorded in 2016 [1]. Additionally, estimates suggest that over 20% of the population is latently infected with Mycobacterium tuberculosis (Mtb), the causative agent of TB [2]. Current treatment regimens of TB require patients to administer drug combinations for an extended period of time, at least six months, which is costly, prone to differential response rates, and exhibits significant problems with patient compliance and adherence [3]. The current treatment regimen involves combinations of rifampicin (RIF), isoniazid (INH), ethambutol (EMB), and pyrazinamide. However, resistance to these drug therapies has exacerbated the management issues of TB. In 2016 alone, almost 500,000 new cases of multidrug-resistant TB (resistant to both RIF and INH) and over 100,000 cases of RIF-resistant TB were identified [1]. Consequently, the identification of Consequently, the identification of novel therapeutic options is an absolute necessity for the management of TB in the future [4]. In this context, in recent years several new drugs and/or regimens have been reported for TB, including bedaquiline [5], delamanid [6], clofazimine [7], bedazuilineprotomanid-linezolid [8]. Worryingly, resistance to several of these approaches has already been reported, primarily because of similar drug resistance mechanisms and target pathways [9,10]. Consequently, the development of new drug therapies for TB requires the discovery of new drug targets and novel structures which circumvent resistance mechanisms, whilst also enabling shorter treatment regimens.

Results and Discussion
In the context of a study to identify novel antibacterial agents designed to overcome antimicrobial resistance, a small library of diverse bioactive compounds had previously been synthesised within our team. Using the Resazurin Microtiter Assay (REMA) [14][15][16], these compounds were screened for antibacterial activity at a fixed concentration (128 μg/mL) against a range of drug-susceptible bacteria including Gram-positive, Gram-negative and mycolata bacteria (Supporting Information, Table S1) which revealed that many possessed little utility, even at these high concentrations. However, benzo- [2,1,3]-diazole architectures 1-12 were shown to possess antibacterial activity, including activity against mycolata bacteria and Mtb ( Figure 2).

Results and Discussion
In the context of a study to identify novel antibacterial agents designed to overcome antimicrobial resistance, a small library of diverse bioactive compounds had previously been synthesised within our team. Using the Resazurin Microtiter Assay (REMA) [14][15][16], these compounds were screened for antibacterial activity at a fixed concentration (128 µg/mL) against a range of drug-susceptible bacteria including Gram-positive, Gram-negative and mycolata bacteria (Supporting Information, Table S1) which revealed that many possessed little utility, even at these high concentrations. However, benzo- [2,1,3]-diazole architectures 1-12 were shown to possess antibacterial activity, including activity against mycolata bacteria and Mtb ( Figure 2). To gain an improved understanding of the antibacterial potency and scope of these compounds, a dose-range REMA assay was performed (128-0.125 μg/mL, converted to μM if active) ( Table 1).   To gain an improved understanding of the antibacterial potency and scope of these compounds, a dose-range REMA assay was performed (128-0.125 µg/mL, converted to µM if active) ( Table 1).  The results of the endpoint REMA assay revealed a mix of activity against the organisms tested, with simple substituted benzodiazole compounds 1-4 providing broad spectrum activity. Whilst the nitrobenzoxa- [2,1,3]-diazole 3 showed the highest levels of activity against Mtb, the lack of selectivity was a cause for concern. Replacing the nitro group with a sulphonamido amino acid moiety greatly improved specificity for mycolata bacteria, with substituted benzoxa- [2,1,3]-diazole 6 showing much higher activity than 5, suggesting poor cell wall penetration of 5 is due to the carboxylic acid moiety. Notwithstanding this, replacement of the benzoxa- [2,1,3]-diazole with benzothia-[2,1,3]-diazoles 7 and 8 led to a complete loss of activity suggesting that the benzoxa- [2,1,3]-diazole plays a crucial role in these compounds antibacterial activity.
Further analysis of the results revealed that conversion of the ester to an aryl hydrazide 9-12 provided compounds more consistent activity across a range of structures. Consequently, substituted benzoxa- [2,1,3]-diazoles were chosen as the partner to amino acid hydrazides for further investigation, via a SAR study to further understand the importance of the amino acid (AA) and the hydrazine (R 1 ) on anti-mycobacterial activity.

Chemical Synthesis of Benzoxa-[2,1,3]-diazole Amino Acid Hydrazides
To undertake this investigation, a two-step synthesis was engaged starting from N-Boc amino acids (Scheme 1). DCC coupling with a monosubstituted hydrazine produced the intermediate protected amino acid hydrazine which, following deprotection and condensation with benzoxa- [2,1,3]-diazole sulphonyl chloride afforded the desired products in moderate to good overall yields following flash chromatography purification (Table 2). Interestingly, several of the compounds exhibited restricted rotation as demonstrated by VT-1 H-NMR spectroscopy (Supporting Information, Figure S2). The results of the endpoint REMA assay revealed a mix of activity against the organisms tested, with simple substituted benzodiazole compounds 1-4 providing broad spectrum activity. Whilst the nitrobenzoxa- [2,1,3]-diazole 3 showed the highest levels of activity against Mtb, the lack of selectivity was a cause for concern. Replacing the nitro group with a sulphonamido amino acid moiety greatly improved specificity for mycolata bacteria, with substituted benzoxa- [2,1,3]-diazole 6 showing much higher activity than 5, suggesting poor cell wall penetration of 5 is due to the carboxylic acid moiety. Notwithstanding this, replacement of the benzoxa- [2,1,3]-diazole with benzothia-[2,1,3]-diazoles 7 and 8 led to a complete loss of activity suggesting that the benzoxa- [2,1,3]-diazole plays a crucial role in these compounds antibacterial activity.
Further analysis of the results revealed that conversion of the ester to an aryl hydrazide 9-12 provided compounds more consistent activity across a range of structures. Consequently, substituted benzoxa- [2,1,3]-diazoles were chosen as the partner to amino acid hydrazides for further investigation, via a SAR study to further understand the importance of the amino acid (AA) and the hydrazine (R1) on anti-mycobacterial activity.

Structure-Activity Relationships
Following the synthetic approach, Boc amino acid hydrazides 13a-22a and benzoxa-[2,1,3]diazole amino acid hydrazide 9, 10, 14b-22b were screened in the same way against the same range of drug-susceptible bacteria as described above. Importantly, in line with the preliminary screening data these compounds showed high selectivity for mycolata bacteria (Supporting Information, Table  S3). Focusing on the Mtb response, initially exploring the role of the amino acid, fixing the hydrazide and increasing the bulk of the amino acid substituent 13a-17a resulted in diminished antibacterial activity of this component (Table 2). Subsequently, fixing the amino acid to glycine, we then evaluated the role of the hydrazine component (18a-22a, Error! Reference source not found.2). Introduction of an unsubstituted aromatic hydrazine 18a alongside halogenated hydrazines 19a-22a did not provide any significant enhancement in activity although a marked increase in cytotoxicity was observed. For both series, enhanced antibacterial activity was restored on coupling to the benzoxa-[2,1,3]-diazole 9, 10, 14b-22b albeit at the cost of increased cytotoxicity, as noted for this subunit [17]. Scheme 1. Synthesis of amino acid hydrazides and the desired benzoxa- [2,1,3]-diazole amino acid hydrazides.

Structure-Activity Relationships
Following the synthetic approach, Boc amino acid hydrazides 13a-22a and benzoxa-[2,1,3]-diazole amino acid hydrazide 9, 10, 14b-22b were screened in the same way against the same range of drug-susceptible bacteria as described above. Importantly, in line with the preliminary screening data these compounds showed high selectivity for mycolata bacteria (Supporting Information, Table S3). Focusing on the Mtb response, initially exploring the role of the amino acid, fixing the hydrazide and increasing the bulk of the amino acid substituent 13a-17a resulted in diminished antibacterial activity of this component (Table 2). Subsequently, fixing the amino acid to glycine, we then evaluated the role of the hydrazine component (18a-22a). Introduction of an unsubstituted aromatic hydrazine 18a alongside halogenated hydrazines 19a-22a did not provide any significant enhancement in activity although a marked increase in cytotoxicity was observed. For both series, enhanced antibacterial activity was restored on coupling to the benzoxa-[2,1,3]-diazole 9, 10, 14b-22b albeit at the cost of increased cytotoxicity, as noted for this subunit [17].

Discussion
Worryingly, as drug-resistant bacterial infections are on the rise and with the recent removal of antibiotic drug discovery programmes, there will be a significant demand for new chemical entities to address this condition. This study has identified that benzoxa- [2,1,3]-diazole substituted amino acid hydrazides have considerable potential as selective and potent agents against Mtb.
Throughout this study, the benzoxa- [2,1,3]-diazole core appears to be essential for activity. Whilst, as observed in some examples, the use of this unit is commonly associated with cytotoxicity this can be effectively modulated through the addition of the amino acid hydrazine. For example, we observed that conjugation with the amino acid hydrazides 19a or 21a provides a reduction in cytotoxicity (19b, 21b, Table 2).
Excitingly, this modulation led to the use of a simple unsubstituted hydrazide 18a which, although in isolation showed significant cytotoxicity, when conjugated with the benzoxadiazole 10 provides a compound with good level of activity against Mtb and no observable cytotoxicity.

Discussion
Worryingly, as drug-resistant bacterial infections are on the rise and with the recent removal of antibiotic drug discovery programmes, there will be a significant demand for new chemical entities to address this condition. This study has identified that benzoxa- [2,1,3]-diazole substituted amino acid hydrazides have considerable potential as selective and potent agents against Mtb.
Throughout this study, the benzoxa- [2,1,3]-diazole core appears to be essential for activity. Whilst, as observed in some examples, the use of this unit is commonly associated with cytotoxicity this can be effectively modulated through the addition of the amino acid hydrazine. For example, we observed that conjugation with the amino acid hydrazides 19a or 21a provides a reduction in cytotoxicity (19b, 21b, Table 2).
Excitingly, this modulation led to the use of a simple unsubstituted hydrazide 18a which, although in isolation showed significant cytotoxicity, when conjugated with the benzoxadiazole 10 provides a compound with good level of activity against Mtb and no observable cytotoxicity.

Discussion
Worryingly, as drug-resistant bacterial infections are on the rise and with the recent removal of antibiotic drug discovery programmes, there will be a significant demand for new chemical entities to address this condition. This study has identified that benzoxa- [2,1,3]-diazole substituted amino acid hydrazides have considerable potential as selective and potent agents against Mtb.
Throughout this study, the benzoxa- [2,1,3]-diazole core appears to be essential for activity. Whilst, as observed in some examples, the use of this unit is commonly associated with cytotoxicity this can be effectively modulated through the addition of the amino acid hydrazine. For example, we observed that conjugation with the amino acid hydrazides 19a or 21a provides a reduction in cytotoxicity (19b, 21b, Table 2).
Excitingly, this modulation led to the use of a simple unsubstituted hydrazide 18a which, although in isolation showed significant cytotoxicity, when conjugated with the benzoxadiazole 10 provides a compound with good level of activity against Mtb and no observable cytotoxicity.

Synthesis of Hydrazides-General Procedure
A solution of N-Boc amino acid (0.25 g, 1.43 mmol, 1 equiv.), HOBt (2 equiv.) and DCC (1.2 equiv.) was dissolved in THF (7.5 mL), cooled to 0 • C and stirred for 15 min. The solution was treated with N-Aryl/Alkyl hydrazine * (1.2 equiv.) before warming to room temperature and stirring for a further 1.5 h. The mixture was then poured into sat. aq. NH 4 Cl (20 mL) before separating and extracting the aqueous layer with EtOAc (40 mL). The organic layer was further washed with sat. aq. NaHCO 3 (20 mL) and then brine (20 mL

Biological Assessment
Bacterial strains and growth media used in this study (Supporting Information, Table S4).

Bacterial Growth Inhibition Assays
The minimum inhibitory concentration of the compounds against all strains using stand REMA assay protocols [14]. Briefly, 100 µL of relevant growth media was added to all wells of a sterile 96-well plate (Corning Incorporated, Corning, NY, USA). The wells in rows A to H in columns 1 received 94.88 µL of growth medium (7H9 media was supplemented with 0.2% casamino acids, 24 µg/mL pantothenate and 10% OADC, Beckton Dickinson, Sparks, MD, USA). Compounds were added to rows A1-H1 (quadruplet per compound) followed by 1:2 serial dilutions across the plate to column 11 were 100 µL of excess medium was discarded from the wells in column 11. The bacterial cultures at 0.5 McFarland standard diluted 1:25 was added to the wells in rows A to H in columns 1 to 11 (100 µL), where the wells in column 12 served as drug-free controls (positive and negative). The plates were sealed with parafilm TM and incubated at 37 • C, unless 30 • C was stated as the optimum for the organism. Freshly prepared filter sterilised resazurin (0.2% w/v, Sigma Aldrich, Dorset, UK) was filter sterilised and 10 µL added to all wells and re-incubated at 37 • C or 30 • C for 24 h or until the positive and negative controls showed a clear result.

Mammalian Cytotoxicity Determination Using the MTT Assay
In vitro chemosensitivity of Human NCI-H460 lung carcinoma cells to the agents were determined using the MTT assay, described elsewhere [18]. Cells were exposed to the amino acid hydrazides or benzoxa- [2,1,3]-diazole amino acid hydrazides (10 µM), or solvent (dimethyl sulphoxide; DMSO) in quadruplicate. Solvent concentrations did not exceed 0.1% and were not cytotoxic. Chemosensitivity and cell survival was assessed following 96 h compound exposure, with cytotoxicity relative to vehicle control subsequently determined.

Conclusions
As shown by the present study, this interplay between cytotoxicity and antibacterial activity can be readily manipulated through the substitution patterns on each component, aromatic hydrazides, the size of the amino acid side chain and the benzoxa-[2,1,3]-diazole ( Figure 3). The ease of manipulation makes this an attractive template and a full examination of all these parameters is the subject of ongoing efforts which will be reported in due course. The minimum inhibitory concentration of the compounds against all strains using stand REMA assay protocols [14]. Briefly, 100 μL of relevant growth media was added to all wells of a sterile 96well plate (Corning Incorporated, Corning, NY, USA). The wells in rows A to H in columns 1 received 94.88 μL of growth medium (7H9 media was supplemented with 0.2 % casamino acids, 24 μg/mL pantothenate and 10% OADC, Beckton Dickinson, Sparks, MD, USA). Compounds were added to rows A1-H1 (quadruplet per compound) followed by 1:2 serial dilutions across the plate to column 11 were 100 μL of excess medium was discarded from the wells in column 11. The bacterial cultures at 0.5 McFarland standard diluted 1:25 was added to the wells in rows A to H in columns 1 to 11 (100 μL), where the wells in column 12 served as drug-free controls (positive and negative). The plates were sealed with parafilm TM and incubated at 37 °C, unless 30 °C was stated as the optimum for the organism. Freshly prepared filter sterilised resazurin (0.2% w/v, Sigma Aldrich, Dorset, UK) was filter sterilised and 10 μL added to all wells and re-incubated at 37 °C or 30 °C for 24 h or until the positive and negative controls showed a clear result.

Mammalian Cytotoxicity Determination Using the MTT Assay
In vitro chemosensitivity of Human NCI-H460 lung carcinoma cells to the agents were determined using the MTT assay, described elsewhere [18]. Cells were exposed to the amino acid hydrazides or benzoxa- [2,1,3]-diazole amino acid hydrazides (10 μM), or solvent (dimethyl sulphoxide; DMSO) in quadruplicate. Solvent concentrations did not exceed 0.1% and were not cytotoxic. Chemosensitivity and cell survival was assessed following 96 h compound exposure, with cytotoxicity relative to vehicle control subsequently determined.

Conclusions
As shown by the present study, this interplay between cytotoxicity and antibacterial activity can be readily manipulated through the substitution patterns on each component, aromatic hydrazides, the size of the amino acid side chain and the benzoxa-[2,1,3]-diazole ( Figure 3). The ease of manipulation makes this an attractive template and a full examination of all these parameters is the subject of ongoing efforts which will be reported in due course.  Table S1: Initial screen of 99 compounds against a range of Gram-positive, Gram-negative and mycolata bacterial in a 96-well plate REMA assay at 128 μg/ml. Blue cells indicate no bacterial growth. Empty cells indicate bacterial growth, Figure S1: Several of the amino acid hydrazides presented as a mixture of rotamers by NMR, Table S2: Antibacterial activity against Grampositive, -negative and mycolata bacteria and Mammalian Cell Toxicity of amino acid hydrazides and benzoxa-[2,1,3]-diazole amino acid hydrazides, expressed as MIC (μM) or percentage relative to control (100 %). -= No  Table S1: Initial screen of 99 compounds against a range of Gram-positive, Gram-negative and mycolata bacterial in a 96-well plate REMA assay at 128 µg/mL. Blue cells indicate no bacterial growth. Empty cells indicate bacterial growth, Figure S1: Several of the amino acid hydrazides presented as a mixture of rotamers by NMR, Table S2: Antibacterial activity against Gram-positive, -negative and mycolata bacteria and Mammalian Cell Toxicity of amino acid hydrazides and benzoxa- [2,1,3]-diazole amino acid hydrazides, expressed as MIC (µM) or percentage relative to control (100%). -= No activity from the REMA assay at 128 µg/mL, Table S3: Bacterial strains and growth media used in the REMA assays, 1 H NMR, 13 C NMR for all compounds.