Mycobacterium tuberculosis DprE1 Inhibitor OPC-167832 Is Active against Mycobacterium abscessus In Vitro

ABSTRACT

The antituberculosis candidate OPC-167832, an inhibitor of DprE1, was active against Mycobacterium abscessus. Resistance mapped to M. abscessus dprE1, suggesting target retention. OPC-167832 was bactericidal and did not antagonize activity of clinical anti-M. abscessus antibiotics. Due to its moderate potency compared to that against Mycobacterium tuberculosis, the compound lacked efficacy in a mouse model and is thus not a repurposing candidate. These results identify OPC-167832–DprE1 as a lead-target couple for a M. abscessus-specific optimization program.

INTRODUCTION

The nontuberculous mycobacterium (NTM) and opportunistic pathogen Mycobacterium abscessus can establish extremely difficult to treat lung infections (1–3). Complex antibiotic regimens, typically consisting a macrolide (clarithromycin or azithromycin), the aminoglycoside amikacin, and a β-lactam (cefoxitin or imipenem), are administered often for years and show low cure rates (4–8). In brief, there is no reliable cure for M. abscessus lung disease. The M. abscessus drug pipeline is thinly populated (9), and new repurposing candidates and lead-target couples are sorely needed (10).

Decaprenylphosphoryl-β-d-ribose oxidase (DprE1) has emerged as an attractive target for antituberculosis (anti-TB) drug development (11–13). The enzyme, catalyzing the formation of decaprenyl-phospho-arabinose (DPA), is essential for growth and viability of Mycobacterium tuberculosis (14–16). DPA serves as a precursor for the synthesis of arabinogalactan, a critical component of the mycobacterial cell wall (14). Inhibitors of M. tuberculosis DprE1 have been identified from various structural scaffolds and show potent activity in vitro and in mouse models (13, 17, 18). BTZ-043, PBTZ-169, OPC-167832, and TBA-7371 have progressed to phase I or II clinical trials for TB (19).

Transposon mutagenesis studies have shown that M. abscessus dprE1 (mab_0192c) is genetically essential (20). Whether M. abscessus DprE1 is a vulnerable target whose inhibition would translate into whole-cell antimicrobial activity has not been established. BTZ-043 and its analog PBTZ-169 have been tested for activity against M. abscessus, and both were found to be inactive (21, 22). This is likely due to an amino acid polymorphism in M. abscessus DprE1. BTZ-043 and PBTZ-169 form covalent adducts with cysteine 387 in M. tuberculosis DprE1 as their on-target mechanism of action (22–24). M. abscesuss DprE1 has alanine at the corresponding amino acid residue position, thus preventing covalent adduct formation and enzyme inhibition by the covalent inhibitors (22–24).

Here, we tested the growth inhibitory activity of the DprE1 inhibitors OPC-167832 and TBA-7371, which do not form covalent adducts with their target (25, 26). The MIC of the compounds against the type strain M. abscessus subsp. abscessus ATCC 19977 (American Type Culture Collection) was determined in Middlebrook 7H9 broth (BD) using the broth microdilution method with optical density at 600 nm (OD₆₀₀) as the readout as described previously (27). The MIC was defined as 90% growth inhibition compared to the drug-free culture. While TBA-7371 (MedChem Express) was inactive (MIC > 100 μM), the dihydrocarbostyril OPC-167832 (MedChem Express) was found to be active (MIC = 5.2 μM). To determine whether the activity of OPC-167832 against the type strain was retained against the broader M. abscessus complex (28), MICs were measured against the reference strains of the two other subspecies, M. abscessus subsp. bolletii CCUG 50184^T and M. abscessus subsp. massiliense CCUG 48898^T (Culture Collection University of Goteborg), and against a panel of clinical isolates which include erm41-harboring macrolide-resistant strains (29, 30). Potency was largely consistent across the members of the M. abscessus complex, with MICs ranging from 5.2 to 15 μM (Table 1).

TABLE 1.

MIC of OPC-167832 against M. abscessus complex

M. abscessus strain	erm41 sequevara	Clarithromycin susceptibility	MIC (μM) ofb:
			OPC-167832	Clarithromycin
Reference strains
M. abscessus subsp. abscessus ATCC 19977	T28	Resistant	5.2	1.2
M. abscessus subsp. bolletii CCUG 50184-T	T28	Resistant	5.4	3.2
M. abscessus subsp. massiliense CCUG 48898-T	Deletion	Sensitive	6	0.6
Clinical isolatesc
M. abscessus subsp. abscessus Bamboo	C28	Sensitive	10	0.8
M. abscessus subsp. abscessus K21	C28	Sensitive	15	2
M. abscessus subsp. abscessus M9	T28	Resistant	10	5.4
M. abscessus subsp. abscessus M199	T28	Resistant	10	4.8
M. abscessus subsp. abscessus M337	T28	Resistant	7	2.9
M. abscessus subsp. abscessus M404	C28	Sensitive	8.5	0.8
M. abscessus subsp. abscessus M421	T28	Resistant	9.3	1.3
M. abscessus subsp. bolletii M232	T28	Resistant	8.8	5.4
M. abscessus subsp. bolletii M506	C28	Sensitive	6	0.6
M. abscessus subsp. massiliense M111	Deletion	Sensitive	11	0.7

^aerm41 is the methylase gene responsible for inducible clarithromycin resistance. The C28 and deletion sequevars are inactive erm41 alleles and result in susceptibility to clarithromycin, while the T28 sequevar is functional and confers inducible resistance against clarithromycin (29, 30).

^bMIC determination was carried out three times independently, and the results are presented as mean values. Clarithromycin (Sigma-Aldrich) was used as the assay control (34).

^cM. abscessus Bamboo, M. abscessus K21, and the other clinical isolates were characterized and reported previously (32, 39, 40).

The micromolar concentration activity against the NTM is in stark contrast to the nanomolar concentration activities of OPC-167832 reported for M. tuberculosis (25). The dramatic in vitro potency difference of the TB drug candidate against M. abscessus suggests that OPC-167832 is likely not a repurposing candidate for the treatment of this lung disease. This was confirmed by in vivo pharmacokinetic-pharmacodynamic analyses. All experiments involving live animals were approved by the Institutional Animal Care and Use Committee of the Center for Discovery and Innovation, Hackensack Meridian Health (no. 269.030 and no. 265.015).

The plasma concentration-time profile upon oral administration of OPC-167832 in uninfected CD-1 mice (Charles River Laboratories) was determined by measuring the plasma concentrations of the compound via high-pressure liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) as described previously (31). Dosing at 20 or 100 mg/kg of body weight resulted in a plasma concentration versus time curve above the MIC for M. tuberculosis (25) for most of a 24-h interval; however, the MIC for M. abscessus was not reached (Fig. 1A). As increasing the dose from 100 to 200 mg/kg did not result in a significant increase of exposure (Fig. 1A), 100 mg/kg was chosen as the highest dose for an efficacy study in a M. abscessus mouse model. NOD.CB17-Prkdc^scid/NCrCrl mice (NOD SCID; Charles River Laboratories) were infected with M. abscessus K21 as described previously (32) and treated once daily for 10 days with orally administered OPC-167832 (50 or 100 mg/kg), clarithromycin (250 mg/kg) as the positive control, or drug-free vehicle. As expected, OPC-167832 treatment did not result in a statistically significant reduction of the lung bacterial burden (Fig. 1B). Plasma concentrations of OPC-167832 were measured 3 h and 24 h after the last dose, confirming similar concentrations in infected and naive mice (Fig. 1A and C). Together, these in vivo analyses suggest that OPC-167832 is not a repurposing candidate for M. abscessus lung disease due to its moderate micromolar concentration in vitro potency compared to its nanomolar concentration activity against M. tuberculosis. It is interesting to note that OPC-167832 at 100 mg/kg showed a weak effect on the bacterial burden in the spleen, similar to the positive control, clarithromycin, at 250 mg/kg (Fig. 1B). The reason for this apparent organ-specific effect remains to be determined and may involve differential drug penetration and/or differences in the pathophysiology of the bacteria.

FIG 1.

Pharmacokinetic profile and activity of OPC-167832 in mice. (A) Plasma concentration versus time profile of OPC-167832. Female CD-1 mice received a single dose of 20, 100, or 200 mg/kg of OPC-167832 formulated in 5% (wt/vol) gum arabic solution (Sigma-Aldrich) by oral gavage. Blood samples were collected from the tail vein at 0.5, 1, 3, 5, 7, and 24 h after drug administration, and the plasma concentration of OPC-167832 was measured by high-pressure liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS). The MIC of OPC-167832 against M. abscessus K21 (15 μM [6,852.6 ng/mL]) (Table 1) is indicated by the red dotted line. The reported MIC of OPC-167832 against M. tuberculosis H37Rv (1.1 nM [0.5 ng/mL]) is indicated by the green dotted line (25). (B) In vivo efficacy of OPC-167832 against M. abscessus in a NOD SCID mouse model. NOD SCID mice were infected intranasally with M. abscessus K21. Starting 1 day postinfection (D1), OPC-167832 (50 or 100 mg/kg, formulated in 5% gum arabic solution), the positive control, clarithromycin (250 mg/kg, formulated in 0.5% carboxymethyl cellulose–0.5% Tween 80–sterile water), or drug-free OPC-167832 vehicle was orally administered to infected mice for 10 consecutive days via oral gavage. Twenty-four hours after the last dose (11 days postinfection), all mice were euthanized, and their lungs and spleen were aseptically removed prior to homogenization. Serial dilutions of organ homogenates were plated onto Middlebrook 7H11 agar (BD) to quantify lung and spleen bacterial load on day 1 postinfection and after administration of drug-free vehicle (D11 Vehicle), OPC-167832 (OPC), and clarithromycin (CLR). The mean and standard deviation are shown for each treatment group (n = 6). Statistical significance was determined using one-way analysis of variance (ANOVA) for multiple comparisons and Dunnett’s posttest (*, P < 0.05; **, P < 0.01). The experiment was carried out twice, showing similar results, and one representative data set is shown. (C) Plasma concentrations of OPC-167832 in infected NOD SCID mice 3 and 24 h after the last dose in the efficacy experiment shown in panel B. The graphs were generated using GraphPad Prism 9 software.

To determine whether OPC-167832 retains DprE1 as its target in M. abscessus and inform future lead optimization efforts, spontaneously resistant mutants were isolated using the type strain M. abscessus ATCC 19977 on Middlebrook 7H10 agar as described previously (31). A total of 20 × 10⁹ CFU were plated on agar medium containing 16× MIC (MIC = 5.2 μM) (Table 1), the lowest OPC-167832 concentration that suppressed growth of wild-type colonies, resulting in a frequency of resistance of 10⁻⁹/CFU. The experiment was repeated with another independently grown culture, yielding a similar frequency of resistance. Twelve randomly selected OPC-167832-resistant strains (OPC_RM1 to OPC_RM12) from the two selection experiments achieved pronounced resistance with an ~30- to 100-fold-higher MIC than the parent strain (Table 2). Whole-genome sequencing (Novogene Corporation, Inc.), followed by Sanger sequencing (Genewiz, Inc.), revealed that the 12 OPC-167832-resistant strains comprised two genotypic classes. Six strains (OPC_RM1 to OPC_RM6) harbored four different missense mutations in the M. abscessus homolog of dprE1, while the other six resistant strains (OPC_RM7 to OPC_RM12) harbored five different missense mutations in the homolog of sigA (mab_3009), which encodes the essential sigma factor A that assists the RNA polymerase in recognizing promoters of target genes (Table 2) (33).

TABLE 2.

Characterization of spontaneous and engineered OPC-167832-resistant M. abscessus ATCC 19977 strains

Strain	MIC (μM) ofa:		Candidate resistance gene	Polymorphisms in candidate resistance geneb	Polymorphisms in other genesc
	OPC-167832	Clarithromycin
Wild type (WT)	5.2	1.2	NAd	NA	NA
Spontaneous mutants (culture batch)e
OPC_RM1 (1)	>500	1.5	dprE1	G745A/Gly249Arg	None
OPC_RM2 (1)	230	1.6	dprE1	G103A/Ala35Thr	mab_0020 A104C/Gly249Arg
OPC_RM3 (1)	220	1.8	dprE1	T872G/Val291Gly	None
OPC_RM4 (2)	>500	1.6	dprE1	G745A/Gly249Arg	None
OPC_RM5 (2)	180	1.2	dprE1	C484T/Pro162Ser	None
OPC_RM6 (2)	380	1.2	dprE1	G103A/Ala35Thr	mab_0341 C38T/Ala13Val
OPC_RM7 (1)	>500	2	sigA	G787A/Gly263Ser	mab_1186c Δ29G
OPC_RM8 (1)	>500	2.3	sigA	C822A/Phe274Leu	mab_2152 Δ−14G
OPC_RM9 (2)	>500	1.8	sigA	A509G/Tyr170Cys	mab_1612 G901A/Glu301Lys
OPC_RM10 (2)	>500	1.6	sigA	T589G/Tyr197Asp	None
OPC_RM11 (2)	>500	1.8	sigA	G787A/Gly263Ser	mab_4225c Δ167CT
OPC_RM12 (2)	>500	2	sigA	G895A/Ala299Thr	mab_0938c Δ752T; mab_0389c A223C/Lys75Glu
Engineered strainsf
WT/pMV262 empty	6	1.8	NA	NA	NDg
WT/pMV262 dprE1*	>500	1.8	NA	NA	ND
WT/pMV262 dprE1	27	1.5	NA	NA	ND
OPC_RM7/pMV262 empty	>500	1.8	sigA	G787A/Gly263Ser	mab_1186c: Δ29G
OPC_RM7/pMV262 sigA	80	1.4	sigA	G787A/Gly263Ser	mab_1186c: Δ29G

^aMIC determination was carried out three times independently and the results are presented as mean values. Clarithromycin was used as the assay control.

^bThe spontaneously resistant strains OPC_RM1 to OPC_RM12 were subjected to whole-genome sequencing, followed by confirmation by Sanger sequencing. The primers used for Sanger sequencing are described in Table S1. The identified mutations in the resistant strains are detailed by the changes in the DNA/amino acid sequence of the affected genes and proteins, respectively.

^cNon-dprE1 and non-sigA polymorphisms detected by whole-genome sequencing. Consistent polymorphism in other genes were not observed. The identified mutations in the resistant strains are detailed by the changes in the DNA/amino acid sequence of the affected genes and proteins, respectively.

^dNA, not applicable.

^eTwelve spontaneous OPC-167832-resistant strains (OPC_RM1 to OPC_RM12), isolated from two independent culture batches, were randomly selected for characterization.

^fWT/pMV262 empty, wild-type M. abscessus ATCC 19977 strain harboring the pMV262-hsp60 expression system without any inserted genes; WT/pMV262 dprE1*, wild-type strain expressing mutant dprE1 (from the spontaneous mutant strain OPC_RM1) carried by pMV262 under the control of hsp60 promoter; WT/pMV262 dprE1, wild-type strain expressing wild-type dprE1 carried by pMV262 under the control of hsp60 promoter; OPC_RM7/pMV262 empty, OPC-167832 resistant strain OPC_RM7 (possessing mutant sigA) harboring the pMV262-hsp60 expression system without any inserted genes; OPC_RM7/pMV262 sigA, resistant strain OPC_RM7 expressing wild-type sigA carried by pMV262 under the control of hsp60 promoter.

^gND, not determined.

To confirm that the observed polymorphisms detected in dprE1 and sigA are indeed responsible for resistance to OPC-167832, merodiploid strains were engineered using custom-synthesized pMV262-hsp60-based expression systems (Genewiz, Inc.) as described previously (34). To confirm involvement of dprE1 missense mutations, a copy of the mutant dprE1 allele from a representative resistant strain (OPC_RM1) (Table 2) was constitutively expressed under the control of the hsp60 promoter in wild-type M. abscessus ATCC 19977. As expected, the strain expressing the mutant dprE1 allele displayed high-level resistance to OPC-167832 (Table 2). To exclude the possibility that the observed resistance was caused by mere overexpression of the dprE1 gene, as opposed to the missense mutation harbored by the mutant dprE1 allele, the wild-type allele of dprE1 was expressed under the control of the hsp60 promoter in wild-type M. abscessus ATCC 19977. This resulted in low-level resistance to OPC-167832 (Table 2), indicating that the missense mutations are the major contributors to the resistance phenotype. Together, these genetic analyses suggest that DprE1 is a target of OPC-167832 in M. abscessus. To confirm that the observed polymorphisms in sigA cause resistance to OPC-167832, one representative sigA mutant strain (OPC_RM7) (Table 2) was complemented with a wild-type copy of the sigA gene that was constitutively expressed under the control of the hsp60 promoter. Expression of wild-type sigA in the mutant background partially restored sensitivity to OPC-167832 (Table 2), suggesting that the missense mutations observed in sigA are the cause of resistance to OPC-167832. It is interesting to note that sigA mutations have not been reported to be involved in resistance of M. tuberculosis to DprE1 inhibitors. However, a few reports describe mutations in sigA causing drug resistance in other bacteria, apparently by reprogramming the transcriptome (35, 36). How mutations in M. abscessus sigA cause resistance against OPC-167832 remains to be determined.

To further evaluate the attractiveness of OPC-167832–DprE1 as a lead-target couple, in vitro bactericidal activity and in vitro drug-drug potency interactions with anti-M. abscessus antibiotics were determined using M. abscessus ATCC 19977 as described previously (27). OPC-167832 was bactericidal, with a 3-log kill at 4× MIC (Fig. 2). The absence of antagonism with clarithromycin, amikacin (Sigma-Aldrich), cefoxitin (MedChem Express), or imipenem (Cayman Chemical) (Table 3), together with the clean drug-drug interaction profile of OPC-167832 as required under multidrug TB therapy (25, 37, 38), suggests that dihydrocarbostyril analogs are compatible with the current standard of care for M. abscessus lung disease.

FIG 2.

In vitro time-kill activity of OPC-167832 against M. abscessus ATCC 19977. M. abscessus cultures were grown in Middlebrook 7H9 and treated with 0.25×, 1×, 4×, and 16× MIC of OPC-167832 (MIC = 5.2 μM) (Table 1) over a period of 3 days, and CFU per milliliter were measured by plating samples on Middlebrook 7H10 agar. Clarithromycin was used as a negative control at 64× MIC (MIC = 1.2 μM) (Table 1). The experiment was carried out three times independently, and the results are presented as mean values with standard deviations displayed as error bars. The graph was generated using GraphPad Prism 9 software.

TABLE 3.

In vitro drug-drug potency interaction between OPC-167832 and selected clinically used drugs against M. abscessus ATCC 19977

Druga	Class	Target	MIC (μM)b		FICIc	Outcomec
			Alone	In combination
OPC-167832	3,4-Dihydrocarbostyril	DprE1	5.2	1.9	0.6	Additivity
Clarithromycin	Macrolide	50S ribosomal subunit	1.2	0.3
OPC-167832	3,4-Dihydrocarbostyril	DprE1	5.2	1.6	0.9	Additivity
Amikacin	Aminoglycoside	30S ribosomal subunit	30	17
OPC-167832	3,4-Dihydrocarbostyril	DprE1	5.2	1.7	0.6	Additivity
Cefoxitin	β-Lactam	Peptidoglycan biosynthesis transpeptidases	26	7.8
OPC-167832	3,4-Dihydrocarbostyril	DprE1	5.2	1.5	0.4	Synergy
Imipenem	β-Lactam	Peptidoglycan biosynthesis transpeptidases	25	3.6

^aTo determine possible antagonisms between OPC-167832 and clinically used drugs, checkerboard analyses were carried using a 96-well plate format (41, 42). The effect of serially diluted OPC-167832 ranging from 0.39 μM to 25 μM was tested against the partner drugs ranging from 0.49 μM to 250 μM.

^bMIC determination was carried out three times independently, and the results are presented as mean values.

^cThe fractional inhibitory concentration index (FICI) was calculated as [(MIC of partner drug in combination/MIC of partner drug alone) + (MIC of OPC-167832 in combination/MIC of OPC-167832 alone)]. An FICI value of ≤0.5 indicates synergy, an FICI value of 0.5 to 4 indicates additivity (no interaction), and an FICI value of >4 indicates antagonism (43).

In conclusion, we identified OPC-167832 as the first whole-cell active inhibitor of M. abscessus DprE1, thus validating DprE1 as a vulnerable target in the opportunistic pathogen. The 1,000-fold-weaker activity of OPC-167832 against M. abscessus compared to M. tuberculosis results in unfavorable pharmacokinetic-pharmacodynamic parameters and lack of efficacy in a mouse model of M. abscessus infection. Thus, the TB drug candidate is unlikely to present a repurposing candidate for the treatment of M. abscessus lung disease. The reason for the pronounced potency difference against the two mycobacterial species remains to be determined and may involve target binding, uptake/excretion, or intrabacterial metabolism (10). If the basis for the differential potency can be elucidated, OPC-167832 may present an attractive chemical starting point for a rational, pathogen-specific lead optimization program.