In Vitro Antibacterial Activity and In Vivo Efficacy of Sulbactam-Durlobactam against Pathogenic Burkholderia Species

The Gram-negative bacterial genus Burkholderia includes several hard-to-treat human pathogens: two biothreat species, Burkholderia mallei (causing glanders) and B. pseudomallei (causing melioidosis), and the B. cepacia complex (BCC) and B. gladioli, which cause chronic lung infections in persons with cystic fibrosis. All Burkholderia spp. possess an Ambler class A Pen β-lactamase, which confers resistance to β-lactams.

ABSTRACT

The Gram-negative bacterial genus Burkholderia includes several hard-to-treat human pathogens: two biothreat species, Burkholderia mallei (causing glanders) and B. pseudomallei (causing melioidosis), and the B. cepacia complex (BCC) and B. gladioli, which cause chronic lung infections in persons with cystic fibrosis. All Burkholderia spp. possess an Ambler class A Pen β-lactamase, which confers resistance to β-lactams. The β-lactam–β-lactamase inhibitor combination sulbactam-durlobactam (SUL-DUR) is in clinical development for the treatment of Acinetobacter infections. In this study, we evaluated SUL-DUR for in vitro and in vivo activity against Burkholderia clinical isolates. We measured MICs of SUL-DUR against BCC and B. gladioli (n = 150), B. mallei (n = 30), and B. pseudomallei (n = 28), studied the kinetics of inhibition of the PenA1 β-lactamase from B. multivorans and the PenI β-lactamase from B. pseudomallei by durlobactam, tested for bla_PenA1 induction by SUL-DUR, and evaluated in vivo efficacy in a mouse model of melioidosis. SUL-DUR inhibited growth of 87.3% of the BCC and B. gladioli strains and 100% of the B. mallei and B. pseudomallei strains at 4/4 μg/ml. Durlobactam potently inhibited PenA1 and PenI with second-order rate constant for inactivation (k₂/K) values of 3.9 × 10⁶ M⁻¹ s⁻¹ and 2.6 × 10³ M⁻¹ s⁻¹ and apparent K_i (K_iapp) of 15 nM and 241 nM, respectively, by forming highly stable covalent complexes. Neither sulbactam, durlobactam, nor SUL-DUR increased production of PenA1. SUL-DUR demonstrated activity in vivo in a murine melioidosis model. Taken together, these data suggest that SUL-DUR may be useful as a treatment for Burkholderia infections.

INTRODUCTION

Burkholderia is a genus of Gram-negative bacteria that includes several important human pathogens. Burkholderia cepacia complex (BCC), a group of >20 different species, and Burkholderia gladioli cause difficult-to-treat chronic infections in individuals with cystic fibrosis (CF), some of whom can develop necrotizing pneumonia and bacteremia known as cepacia syndrome (1, 2). Burkholderia pseudomallei is a biothreat pathogen causing melioidosis, a disease with a high fatality rate requiring extended antibiotic treatment (3, 4). Burkholderia mallei, the cause of the zoonotic disease glanders, is a clonal derivative of B. pseudomallei and another potential biothreat agent because of a high mortality rate when human infection occurs by inhalation or when the infection becomes disseminated (5). Burkholderia organisms are intrinsically multidrug resistant, including having resistance to the last-resort antibiotics colistin and polymyxin B (2, 6).

β-Lactam antibiotics, especially meropenem and ceftazidime, are recommended therapies for Burkholderia infections (4, 5, 7). Successful therapy may be compromised, however, by the expression of bacterial β-lactamase enzymes that destroy the drugs (6). β-Lactamases have been divided into 4 classes based on genetic relatedness (8). Classes A, C, and D include the serine β-lactamases, which possess a nucleophilic serine residue in their active sites. Class B includes the metallo-β-lactamases, which rely on zinc ions in the active site. Burkholderia species encode a variety of class A β-lactamases (prefixed Pen) (9), which currently appear to confer the majority of β-lactamase-mediated β-lactam resistance (6), although members of the other classes may be present in Burkholderia genomes (10, 11). Burkholderia class A β-lactamases vary substantially in their amino acid sequences and β-lactam substrate spectra (9, 12, 13). In BCC, the bla gene, which encodes the Pen β-lactamase, is regulated transcriptionally by a LysR-type transcriptional regulator, PenR. This regulation is linked to cell wall remodeling similar to the AmpC/AmpR systems in Enterobacterales (14, 15).

β-Lactamase inhibitors (BLIs) can be coadministered with β-lactams to overcome degradation of the drugs by β-lactamases. Moreover, several BLIs themselves possess some intrinsic β-lactam activity, including sulbactam, clavulanic acid, and tazobactam. The effectiveness of these BLIs can be limited by hydrolysis catalyzed by β-lactamases. Indeed, Papp-Wallace et al. (13) showed that sulbactam, clavulanic acid, and tazobactam are degraded by Burkholderia multivorans PenA. Recently, non-β-lactam BLIs have been introduced, including the diazabicyclooctanes (DBO) avibactam and relebactam and the boronic acid vaborbactam (16). β-Lactam–BLI combinations can be effective in vitro against Burkholderia strains that are resistant to standard-of-care β-lactams. Papp-Wallace and colleagues showed that 36% of 146 BCC isolates from CF patients were ceftazidime resistant, but avibactam could restore ceftazidime susceptibility to 90% of them (18), while Van Dalem et al. (19) and Caverly et al. (20) found that 81% of 91 and 97% of 150 BCC CF isolates, respectively, were susceptible to ceftazidime-avibactam.

The DBO BLI durlobactam is currently in phase 3 clinical testing in combination with sulbactam for the treatment of pneumonia caused by carbapenem-resistant Acinetobacter baumannii-calcoaceticus complex pathogens (21); https://clinicaltrials.gov/ct2/show/NCT03894046). Durlobactam inhibits a broad spectrum of class A, C, and D β-lactamases, including carbapenemases. Sulbactam, which inhibits penicillin-binding proteins 1a, 1b, and 3 (22), is in this case the active β-lactam drug, and durlobactam protects it from degradation by serine β-lactamases. The antibacterial activity of sulbactam is limited to only a few bacterial genera. In addition to Acinetobacter spp., sulbactam has intrinsic activity against Neisseria and Burkholderia spp. (23, 24). We investigated the effectiveness of the sulbactam-durlobactam (SUL-DUR) combination against clinical isolates of pathogenic Burkholderia spp. in vitro. We also measured the durlobactam inhibition kinetics of PenA1 β-lactamase from B. multivorans—a member of the BCC—and PenI β-lactamase from B. pseudomallei. We tested whether bla_PenA1 gene expression was induced by SUL-DUR. Finally, we evaluated SUL-DUR for in vivo efficacy in a murine melioidosis model.

RESULTS

SUL-DUR has potent in vitro activity against the Burkholderia genus.

Clinical isolates of BCC (B. ambifaria, B. arboris, B. cenocepacia, B. cepacia, B. contaminans, B. diffusa, B. dolosa, B. multivorans, B. pseudomultivorans, B. pyrrocinia, B. seminalis, B. stabilis, B. ubonensis, and B. vietnamiensis) and B. gladioli were subjected to antimicrobial susceptibility testing. The range and MIC_50/90 values for sulbactam, durlobactam, SUL-DUR, and comparators (18, 25, 26) are shown in Table 1. A preliminary susceptibility breakpoint of ≤ 4/4 µg/ml was selected for comparison purposes based on population pharmacokinetic (PK) analyses (27). Sulbactam alone was inactive against this collection (MIC₉₀ > 64 µg/ml), while durlobactam alone showed modest activity (MIC₉₀ = 16 µg/ml). In contrast, the SUL-DUR combination was notably active, with MICs of ≤4 µg/ml for 87.3% of strains and MIC_50/90 of 2/8 µg/ml. This degree of activity was nearly as high as that of ceftazidime-avibactam, to which 90% of strains were susceptible.

TABLE 1.

Susceptibility of 150 BCC and B. gladioli clinical isolates to sulbactam, durlobactam, or the combination versus comparator antibiotics

Antibiotic	Range, µg/ml	MIC50, µg/ml	MIC90, µg/ml	% susceptiblea
Sulbactam	1 to >64	>64	>64	11.3
Durlobactam	2 to 64	8	16	23.3
SUL-DURb	0.5 to 16	2	8	87.3
Tobramycin	0.25 to >512	128	512	4.8
Ciprofloxacin	0.25 to 512	4	128	18.6
Minocycline	0.25 to 256	8	128	42.8
Imipenem	0.25 to 256	32	128	10.3
Ceftazidime	1 to >512	8	128	62.7
Aztreonam	0.5 to 512	32	512	24.8
Trimethoprim-sulfamethoxazolec	0.25/4.75 to >16/>304	2/38	16/304	62.8
Ceftazidime-avibactamd	0.5 to >128	4	8	90.0
Aztreonam-avibactamd	0.5 to 512	4	16	78.6
Piperacillin-tazobactame	0.25 to >256	16	256	58.0

^aMIC values for tobramycin, ciprofloxacin, minocycline, imipenem, ceftazidime, aztreonam, trimethoprim-sulfamethoxazole, ceftazidime-avibactam, and aztreonam-avibactam were previously published in reference 18. MIC values for piperacillin-tazobactam were previously published in reference 26. As breakpoints are not available, and for comparison purposes only, possible interpretative breakpoints of ≤4 µg/ml for susceptible and ≥8 µg/ml for resistant were selected for sulbactam, SUL-DUR, and durlobactam. Pseudomonas aeruginosa CLSI breakpoints were used to interpret the following B. cepacia MICs, as breakpoints are not available for B. cepacia: tobramycin, ≤4 µg/ml = susceptible, 8 µg/ml = intermediate, and ≥16 µg/ml = resistant; ciprofloxacin, ≤1 µg/ml = susceptible, 2 µg/ml = intermediate, and ≥4 µg/ml = resistant; imipenem, ≤2 µg/ml = susceptible, 4 µg/ml = intermediate, and ≥8 µg/ml = resistant; aztreonam, ≤8 µg/ml = susceptible, 16 µg/ml = intermediate, and ≥32 µg/ml = resistant; and piperacillin-tazobactam, ≤16/4 µg/ml = susceptible, 32/4 to 64/4 µg/ml = intermediate, and ≥128/4 µg/ml = resistant. For B. cepacia, CLSI breakpoints are as follows: minocycline, ≤4 µg/ml = susceptible, 8 µg/ml = intermediate, and ≥16 µg/ml = resistant; trimethoprim-sulfamethoxazole, ≤2/38 µg/ml = susceptible and ≥4/76 µg/ml = resistant; and ceftazidime, ≤8 µg/ml = susceptible, 16 µg/ml = intermediate, and ≥32 µg/ml = resistant. For ceftazidime-avibactam and aztreonam-avibactam, the former breakpoints for ceftazidime and aztreonam were used for interpretation, respectively.

^bSulbactam (SUL) and durlobactam (DUR) were tested at a 1:1 ratio; the singular ratio MIC value is reported. Bold has been added to highlight the combination.

^cTrimethoprim-sulfamethoxazole were used at ratio of 1:19. Both values are reported.

^dThe β-lactam concentrations were increased by 2-fold dilutions in the presence of β-lactamase inhibitor concentrations fixed at 4 µg/ml.

^ePiperacillin and tazobactam were used at a ratio of 8:1; only the piperacillin MIC value is listed.

A diverse collection of B. mallei (n = 30) and B. pseudomallei (n = 28) clinical isolates were also tested for susceptibility to sulbactam, durlobactam, or the combination, as well as a comparator antibiotic, doxycycline (Table 2). While these isolates showed moderate susceptibility to both sulbactam and durlobactam alone (MIC₉₀s = 16 and 32 µg/ml, respectively), the MIC₉₀ of the combination was 1 µg/ml, with 100% of isolates susceptible to the SUL-DUR combination at ≤2/4 µg/ml. With the exception of sulbactam, B. mallei isolates were more susceptible to the test agents than was B. pseudomallei.

TABLE 2.

Susceptibilities of 58 B. mallei and B. pseudomallei clinical isolates to sulbactam, durlobactam, or the combination versus doxycycline^a

Antibiotic	All strains (n = 58)			B. mallei (n = 30)			B. pseudomallei (n = 28)
	MIC range	MIC50	MIC90	MIC range	MIC50	MIC90	MIC range	MIC50	MIC90
Sulbactam	2 to 32	8	16	2 to 32	8	16	4 to 16	8	8
Durlobactam	0.25 to 32	8	32	0.25 to 8	8	8	8 to 32	32	32
SUL-DURb	<0.03 to 2	0.5	1	<0.03 to 1	0.25	0.5	0.5 to 2	1	2
Doxycycline	0.015 to >8	0.25	8	0.015 to 0.5	0.06	0.12	0.25 to 8	4	8

^aMICs are shown in micrograms per milliliter.

^bThe durlobactam concentration was fixed at 4 µg/ml. The MIC value is the sulbactam concentration.

Durlobactam inactivates the major Pen β-lactamases found in Burkholderia.

Kinetic constants for inhibition of B. multivorans PenA1 and B. pseudomallei PenI β-lactamases by durlobactam are shown in Table 3. The experimental data are in Fig. S1 and S2 in the supplemental material. The corrected apparent K_i (K_iapp) value of durlobactam for PenA was 15 ± 1 nM, much lower than the K_iapp of avibactam (500 ± 50 nM) (18), showing that durlobactam interacts more favorably than avibactam to PenA1 prior to formation of the covalent complex. The second-order rate constant for inactivation (k₂/K) by durlobactam ([3.9 ± 0.2] × 10⁶ M⁻¹ s⁻¹) was about double that of avibactam ([2 ± 1] × 10⁶ M⁻¹ s⁻¹), indicating that durlobactam is twice as potent a PenA1 inhibitor as avibactam. The dissociation rate constant of durlobactam ([5.5 ± 0.4] × 10⁻³ s⁻¹) was somewhat higher than that of avibactam ([2 ± 1] × 10⁻³ s⁻¹), meaning that in the absence of free inhibitor, activity of the PenA1 would be restored by dissociation of durlobactam more quickly than by dissociation of avibactam. These values correspond to half-lives (t_1/2) of 2.1 min and 5.8 min for durlobactam and avibactam, respectively. However, in the restricted space of the periplasm of the bacteria, the released inhibitor is likely to react again with the enzyme if it is released intact following recyclization. This phenomenon of recyclization and dissociation of intact DBO from β-lactamases has been demonstrated for both avibactam (28) and durlobactam (29). Against PenI, durlobactam possessed a K_iapp of 241 ± 25 nM, k₂/K of 3.9 (± 0.3) × 10⁴ M⁻¹ s⁻¹, and a dissociation rate constant (k_off) of (7.5 ± 0.1) × 10⁻⁴ s⁻¹; thus, PenI is inactivated slower than PenA1, but the PenI-durlobactam complex is slower to dissociate, with a t_1/2 of 15.4 min.

TABLE 3.

Kinetic constants for inhibition of B. multivorans PenA1 and B. pseudomallei PenI by durlobactam^a

Parameter	Value for:
	PenA1	PenI
Kiapp	15 ± 1 nM	241 ± 25 nM
k2/K	(3.9 ± 0.2) × 106 M−1 s−1	(3.9 ± 0.3) × 104 M−1 s−1
koff	(5.5 ± 0.4) × 10−3 s−1	(7.5 ± 0.1) × 10−4 s−1
t1/2	2.1 min	15.4 min

^aThe dissociation half-life t_1/2 is ln2/k_off.

Durlobactam forms stable adducts with Pen β-lactamases.

The masses of the durlobactam-PenA1 and durlobactam-PenI covalent complexes, prepared by mixing equimolar amounts of the enzyme and inhibitor, were measured by electrospray ionization mass spectrometry (ESI-MS) to be 29,690 ± 5 Da and 29,627 ± 5 Da, respectively. The complexes showed no change in mass and no appearance of free PenA1 or PenI after 24 h at room temperature (Fig. 1), revealing that PenA1 and PenI are unable to hydrolyze durlobactam. The masses of PenA1 alone and PenI alone were 29,414 ± 5 Da and 29,350 ± 5 Da, respectively. The durlobactam adduct mass was therefore 276 ± 7 Da. The mass of protonated durlobactam is 277 Da. These results demonstrate that durlobactam forms a stable covalent adduct with PenA1 and PenI that consists of the full mass of durlobactam. Since durlobactam was also shown to dissociate from PenA1 and PenI, the data taken together demonstrate that durlobactam recyclizes and dissociates intact from PenA1 and PenI.

FIG 1.

ESI-MS of PenA1, PenI, and the durlobactam-PenA1, durlobactam-PenI covalent complexes.

SUL-DUR does not induce expression of bla_PenA1 in B. multivorans, a species within BCC.

As bla_Pen genes are inducible in the BCC, to determine whether sulbactam, durlobactam, or SUL-DUR induces expression of the bla_PenA1 gene, B. multivorans ATCC 17616 cells were grown in the presence of the compounds for 1 h at sub-MIC (1/4×) concentrations, which were chosen based on their limited impact on bacterial growth. Imipenem was used as a positive control. Western blots were prepared from the treated cells using an anti-PenA1 peptide antibody. An anti-RecA antibody was used as a loading control. A representative blot is shown in Fig. 2. There was no detectable induction of bla_PenA1 by sulbactam, durlobactam, or SUL-DUR.

FIG 2.

Western immunoblot of PenA1 expression upon exposure of B. multivorans ATCC 17616 to imipenem, sulbactam, durlobactam, or SUL-DUR for 1 h during log-phase growth. RecA was used as the loading control. Lanes: 1, 250 ng of PenA1; 2, untreated cells; 3, cells treated with 1 μg/ml of imipenem as a positive control; 4, cells treated with 8 μg/ml of sulbactam; 5, cells treated with 0.5 μg/ml each of sulbactam and durlobactam; and 6, cells treated with 2 μg/ml of durlobactam.

SUL-DUR increases the survival of mice in a melioidosis infection model.

Kaplan-Meier survival plots for mice intranasally infected with B. pseudomallei K96243 and given various drug treatments are shown in Fig. 3. All 10 of the untreated (vehicle control) mice died by the third day of infection. After 45 days, 3 of the ciprofloxacin-treated mice, 4 of the doxycycline-treated mice, and 6 of the SUL-DUR-treated mice (both dosages) were still alive. The survival differences between ciprofloxacin, doxycycline, and both treatment arms of SUL-DUR were not statistically significant by log rank test, but a higher median of survivors was determined for SUL-DUR relative to the other comparators. For animals that survived to 45 days postchallenge, all treatment arms, including doxycycline and ciprofloxacin, demonstrated counts of 10⁷ to 10⁸ CFU/g in the spleen (Table 4). Lung counts, by comparison, were generally sporadic and lower (0 to 10⁵ CFU/g).

FIG 3.

Kaplan-Meier survival plots for B. pseudomallei-infected mice treated with sulbactam- durlobactam or comparator agents.

TABLE 4.

Residual bacterial burdens in surviving animals at day 45 per survival data in Fig. 3 after B. pseudomallei infection

Regimen	Mouse	Bacterial burden (CFU/g) in:
		Spleen	Liver	Lung
Ciprofloxacin, 40 mg/kg	A	1 × 108	1.8 × 108	6 × 104
	B	1 × 108	2.8 × 107	4 × 105
	C	2 × 108	0	0
Doxycycline 40 mg/kg	A	4 × 103	0	0
	B	2 × 105	2.6 × 104	0
	C	1.9 × 107	0	0
SUL-DUR, 100:200 mg/kg	A	2 × 108	3 × 105	2.4 × 107
	B	6 × 107	0	0
	C	3.6 × 107	2.2 × 104	9 × 105
SUL-DUR, 400:200 mg/kg	A	1 × 108	6 × 106	2.6 × 106
	B	2 × 108	1.7 × 105	1.8 × 105
	C	3 × 108	1.4 × 107	1.1 × 105

DISCUSSION

Burkholderia infections can be very difficult to treat with antibiotics, in part due to the presence of β-lactamases (9), as well as other defense mechanisms such as efflux pumps (6). For example, the recommended intensive initial therapy for melioidosis consists of 2 g of intravenous (i.v.) ceftazidime 4 times a day or 1 g of i.v. meropenem 3 times a day for at least 14 days, followed by eradication therapy consisting of oral trimethoprim-sulfamethoxazole twice a day for at least 3 months (4). β-Lactam susceptibility can be increased in many cases by the addition of a BLI. In a study of 91 BCC strains isolated from CF patients in Belgium between 2012 and 2016, only 63% were susceptible to ceftazidime, but 81% were susceptible to ceftazidime-avibactam (19). In a study of 146 BCC isolates by Papp-Wallace et al. (18) (Table 1), 63% of strains were susceptible to ceftazidime, but 90% were susceptible to ceftazidime-avibactam. In addition, an isolate of B. pseudomallei which overproduced bla_PenI due to an increase in gene copy number also had susceptibility to ceftazidime restored upon the addition of avibactam (30). Notably, 8 of the 15 ceftazidime-avibactam-resistant or -intermediate strains (Table S1) were susceptible to SUL-DUR, using an interpretive susceptibility breakpoint of 4/4 μg/ml. SUL-DUR also was highly active against the biothreat pathogens B. mallei and B. pseudomallei (Table 2). Moreover, Zeiser et al. recently showed that 99% of this collection of BCC and B. gladioli CF isolates were susceptible to piperacillin-avibactam or the quadruple combination of piperacillin-tazobactam-ceftazidime-avibactam (26).

An advantage of durlobactam over avibactam is the generally greater inhibitory potency of durlobactam against Ambler class A, C, and D serine β-lactamases (21). Although the k₂/K differential was relatively modest (at 2-fold) for the class A B. multivorans PenA1 (Table 2), the various class A enzymes found in Burkholderia species have substantial sequence and functional variation (9, 13). It is likely that these enzymes will be inhibited to different extents by each BLI. Furthermore, many BCC strains express both class A and C enzymes (26), and class D β-lactamase sequences are present in strains of B. pseudomallei, B. thailandensis, B. cepacia, and B. cenocepacia, according to genome sequence data (https://www.burkholderia.com). Although avibactam is an effective inhibitor of class A and C enzymes, its spectrum of action against class D enzymes is limited compared with that of durlobactam (21).

In a murine model of melioidosis, SUL-DUR demonstrated better activity than comparators ciprofloxacin and doxycycline, with a median survival time (the time at which the survival curve crosses 50% survival) greater than 45 days postchallenge. Although the log rank test failed to statistically differentiate the treatment arms of SUL-DUR and the comparators, 90% of the mice treated with 400:200 mg/kg (of body weight) of SUL-DUR survived to 40 days postchallenge. Pre- and postexposure prophylaxis of experimental B. pseudomallei infection has been studied previously against the same K96243 strain in mice using doxycycline administered orally at a dose of 40 mg/kg every 12 h (q12h) for 10 days, while monitoring survival for 21 days (31). In this study, K96243 was delivered via an aerosol challenge and therapy was initiated at 48 h prior to inoculation and 0, 10, 24, and 48 h postinoculation. Although the design of the present study is different, with therapy initiating 4 h postchallenge and a dosing duration of 6 days, the survival with doxycycline treatment and the tissue bacterial burden levels were similar by 21 days and mimicked the survival curve observed by Sivalingham et al. (31). Since the oral bioavailability of doxycycline is greater than 70%, the dose-response of doxycycline should be independent of the route of administration. It is important to note that these comparators have been used in the development of the murine melioidosis model in a setting of pre- and postexposure prophylaxis (30, 32) and were included in the present study as controls for the published animal models. However, their evaluation in the murine model does not necessarily reflect their use as first-line clinical therapies for the treatment of B. pseudomallei infections. The PK/PD of SUL-DUR has been studied extensively against Acinetobacter baumannii, with exposure targets of fT>MIC (the percentage of a 24-h time period that the unbound drug concentration exceeds the MIC) of 50% for sulbactam and fAUC_0–24/MIC (the area under the unbound drug concentration-time curve divided by the MIC) of 30 for durlobactam for the combination to achieve 1-log kill (33). The sulbactam dosages of 100 mg/kg and 400 mg/kg q4h correspond to fT>MIC of 46% and 62%, respectively, suggesting that the PK/PD exposure target for sulbactam was achieved with the upper dosage of 400 mg/kg q4h. The durlobactam dosage of 200 mg/kg q4h corresponded to an fAUC_0–24 of 350 µg·h/ml, which, when considering the K96243 MIC of 1 µg/ml, is >30× the exposure requirement for durlobactam. The dose level of SUL-DUR of 400:200 mg/kg q4h is predicted to exceed exposure requirements for >1-log kill for A. baumannii and also is efficacious versus B. pseudomallei, as shown here (Fig. 3).

As shown in Table 4, no bacteria were recovered from the lungs of surviving mice receiving doxycycline at 40 mg/kg q12h. A similar finding was obtained with oral administration of co-trimoxazole (31). Although there were bacteria isolated from the lungs of surviving SUL-DUR-treated mice, there was a trend toward greater survival than for both the doxycycline and ciprofloxacin treatment arms. These opposing trends in tissue burden and survival data suggest that further investigation into the pathogenesis of B. pseudomallei, drug distribution, and regimen schedules is warranted. In the absence of a PK/PD understanding of SUL-DUR versus B. pseudomallei, exposure targets and regimens remain to be elucidated to maximize efficacy and reduce the incidence of relapse in pre- and postexposure prophylaxis.

Conclusion.

Historically, the only clinical use of sulbactam for the treatment of Burkholderia infections has been in combination with cefoperazone, as described in reports from Southeast Asia (34, 35). Our results suggest the intrinsic antibacterial activity of sulbactam against pathogenic Burkholderia spp. can be restored by the addition of the novel, broad-spectrum BLI durlobactam.

The results of the present study demonstrate that SUL-DUR has promising in vitro activity against pathogenic Burkholderia spp., suggesting that it could be developed as a treatment for infections caused by these organisms, including those from the BCC and B. gladioli, B. mallei, and B. pseudomallei. Additional studies in vivo will be required to further understand PK/PD and projected dose regimens of SUL-DUR that could be potentially utilized in a clinical setting for efficacy.

MATERIALS AND METHODS

Compounds.

Durlobactam was synthesized at Entasis Therapeutics (21). Sulbactam was from U.S. Pharmacopeia. Nitrocefin was purchased from Thermo Scientific Oxoid. Piperacillin-tazobactam, tobramycin, and imipenem-cilastatin were pharmacy grade. Ceftazidime was from RPI. Aztreonam was from Chem-Impex. Avibactam was from AstraZeneca. Ciprofloxacin, minocycline, trimethoprim, and sulfamethoxazole were from Sigma.

(i) Source of strains.

All studies on BCC and B. gladioli were performed at the VA Northeast Healthcare System in Cleveland, OH. One hundred fifty multidrug-resistant (MDR) clinical strains, including 140 BCC strains of 14 different species and 10 B. gladioli strains, were obtained from the Burkholderia cepacia Research Laboratory and Repository (University of Michigan) (12, 18, 26). These isolates were recovered from respiratory specimens from 150 individuals with CF receiving care in 68 cities in 36 states in the United States. Each isolate was identified to the species level by using species-specific PCR, recA restriction fragment length polymorphism (RFLP), and/or DNA sequencing of the recA gene (36, 37). If the strain could not be confidently placed in 1 of the 22 named BCC species, it was categorized as “BCC indeterminate.”

(ii) Susceptibility testing.

MICs for the bacterial isolates were determined by the cation-adjusted Mueller-Hinton (MH) agar dilution method. The MIC measurements were performed using a Steers replicator that delivered 10 µl containing 10⁴ CFU of overnight culture grown in MH broth at 37°C and diluted in MH broth. MIC results were interpreted using Clinical and Laboratory Standards Institute (CLSI) breakpoints, if available (38). A total of 15 independent experiments were conducted to test 150 strains. The mode of 3 independent measurements was reported. The tested concentrations of the combination of SUL-DUR consisted of 2-fold dilutions of a 1:1 fixed ratio of SUL-DUR. The strain B. multivorans ATCC 17616 was tested on every MIC panel to control for variability between experiments. The control tested within one doubling dilution of the mode MIC. The MIC ranges observed for the control strain were 8 to 32 μg/ml for sulbactam, 4 to 8 μg/ml for durlobactam, and 1 to 2 μg/ml for SUL-DUR.

(iii) MIC interpretations.

For comparison purposes, a preliminary interpretative susceptibility breakpoint of ≤4 µg/ml was used for SUL-DUR based on recent population PK analyses (27). For B. cepacia, CLSI susceptible/intermediate/resistant breakpoints (in micrograms per milliliter) are as follows: minocycline, ≤4/8/≥16, and ceftazidime, ≤8/16/≥32. For trimethoprim-sulfamethoxazole, breakpoints (in micrograms per milliliter) are as follows: ≤2/38 for susceptible and ≥4/76 for resistant. Pseudomonas aeruginosa CLSI susceptible/intermediate/resistant breakpoints were used to interpret the MIC values (in micrograms per milliliter), since those are not available for B. cepacia, and are as follows for the antibiotics in this study: tobramycin, ≤4/8/≥16; ciprofloxacin, ≤1/2/≥4; imipenem, ≤2/4/≥8; aztreonam, ≤8/16/≥32; and piperacillin-tazobactam, ≤16/4/32/4 to 64/8/≥128/4. For ceftazidime-avibactam and aztreonam-avibactam, the former breakpoints for ceftazidime and aztreonam were used for interpretation, respectively.

(iv) PenA1 preparation.

Escherichia coli DE3 Origami 2 cells carrying pGEX-6p2 bla_PenA1 were used for protein expression and purification, as previously described (13, 39). The nomenclature for Pen β-lactamases in Burkholderia species is nebulous. The Pen β-lactamase designations herein are based on references 9 and 12. Briefly, cells were grown in super optimal broth and then isopropyl-β-d-1-thiogalactopyranoside was added to induce expression. Cells were pelleted and frozen at –80°C. Subsequently, the cells were lysed, and the β-lactamase was purified using glutathione S-transferase (GST)-affinity chromatography and verified by electrospray ionization mass spectrometry (ESI-MS), as previously described, with >95% purity and a final molecular weight of 29,414 Da (13, 39).

(v) Enzyme kinetics.

The apparent K_i (K_iapp) value, acylation rate constant (k₂/K), and dissociation rate constant (k_off) for durlobactam and PenA1 were obtained in 10 mM phosphate-buffered saline, pH 7.4 (PBS), with an Agilent 8453 diode array spectrophotometer at room temperature using previously described methods (40, 41).

K_iapp was determined for durlobactam with PenA1 by a direct competition assay under steady-state conditions. The PenA1 concentration was 5 nM and the durlobactam concentration was varied. Nitrocefin (NCF) was used as the reporter substrate at a fixed concentration of 100 µM. PenA1, durlobactam, and nitrocefin were mixed manually, and the initial reaction velocity was monitored. Data were linearized by plotting the inverse initial reaction velocities (1/v₀) versus inhibitor concentrations (I). K_iapp was determined by dividing the value for the y-intercept by the slope of the line and accounting for the use of nitrocefin as a reporter using equation 1.

$$K_{i\mathrm{app}}=K_{i\mathrm{app}}\left(\text{observed}\right)/\left(\text{1\hspace{0.17em}}+\left[\text{S}\right]/K_{m,\text{NCF}}\right)$$ (1)

where K_m,NCF = 142 μM.

To obtain the acylation rate constant, progress curves were obtained by mixing PenA1 at 5 nM with increasing concentrations of durlobactam using nitrocefin at 100 µM as a reporter substrate. Progress curves were fit to equation 2 to obtain k_obs values.

$$y=A_0+V_{f}t+\left(V_0-V_f\right)\left[\text{1\hspace{0.17em}}-\text{\hspace{0.17em}exp}\left(k_{\text{obs}}t\right)\right]/k_{\text{obs}}$$ (2)

where y is the 482 nm absorbance at time t, V_f is final velocity, V₀ is initial velocity, and A₀ is initial absorbance. The data were plotted as k_obs versus [durlobactam]. The k₂/K value was obtained by correcting the value obtained for the slope of the line for the use of nitrocefin as an indicator substrate ([S]) using equation 3.

$$k_2/K=k_2/K\left(\text{observed}\right)\text{x}\left(\text{1\hspace{0.17em}}+\left[\text{S}\right]/K_{m,\text{NCF}}\right)$$ (3)

The dissociation rate constant for the durlobactam-PenA1 complex was determined by incubating 1 µM PenA1 with 3 µM durlobactam for 5 min at room temperature, diluting the mixture 1:100,000, and adding 100 µM nitrocefin. Progress curves measuring nitrocefin hydrolysis were collected for 2 h and the data were fit to equation 2, where k_obs = k_off. PenA1 alone was used as a control.

Individual data points were obtained in triplicate, while each experiment was conducted at least in duplicate.

(vi) ESI-MS.

The masses of intact PenA1 with and without durlobactam were measured by ESI-MS on a Waters Synapt G2-Si quadrupole-time-of-flight mass spectrometer. The Synapt G2-Si was calibrated with a sodium iodide solution using a 50 to 2,000 m/z mass range. A 10 μM concentration of PenA1 was incubated with 10 μM durlobactam at a 1:1 ratio for 24 h in PBS. PenA1 was also incubated alone as a control. The reactions were terminated by the addition of a final concentration of 0.1% formic acid and 1% acetonitrile. The samples were run using a Waters Acquity H class ultraperformance liquid chromatograph (UPLC) and an Acquity UPLC BEH C₁₈ 1.7-µm, 2.1- by 100-mm column. The aqueous phase consisted of 0.1% (vol/vol) formic acid in water (phase A) and the organic phase was acetonitrile with 0.1% formic acid (phase B). The flow rate was 0.5 ml/min. The elution was performed as follows: 1 min at 10% phase B, a linear gradient from 10% phase B to 81% phase B for 3 min, 85% phase B for 30 s, and 90% phase B for 30 s. The tune settings were as follows: capillary voltage at 3.5 kV, sampling cone at 35, source offset at 35, source temperature at 100°C, desolvation temperature at 500°C, cone gas at 100 liters/h, desolvation gas at 800 liters/h, and nebulizer bar at 6.0. The spectra were analyzed using MassLynx v4.1 and deconvoluted using the MaxEnt1 program.

(vii) Immunoblotting.

B. multivorans ATCC 17616 was grown in lysogeny broth to log phase at an optical density at 600 nm (OD₆₀₀) between 0.6 and 0.7. The cells were treated with sub-MICs of imipenem, sulbactam, durlobactam, or SUL-DUR for 1 h. The cells were pelleted and lysed to prepare crude extracts, as previously described (42). These crude extracts, as well as purified full-length PenA protein, were subjected to SDS-PAGE and transferred to polyvinylidene difluoride membranes. The membranes were blocked in 5% (wt/vol) nonfat dry milk in 20 mM Tris-Cl with 150 mM NaCl (pH 7.4) (TBS) for 1 h and probed in 5% nonfat dry milk in TBS with 1 μg/ml of polyclonal anti-PenA-peptide antibody and 1 μg/ml of polyclonal anti-RecA antibody. Both antibodies were raised in rabbits by New England Peptide using a selected PenA 18-amino-acid peptide and the RecA protein as the antigens, respectively (11, 12). Membranes were washed five times for 10 min with TBS plus 0.05% Tween 20 (TBST) and then incubated for 1 h with a 1:5,000 dilution of horseradish peroxidase (HRP)-conjugated anti-rabbit secondary antibody in 5% nonfat dry milk in TBS. Blots were washed five times for 10 min with TBST and developed using the ECL-Plus developing kit (GE Healthcare Life Sciences, Marlborough, MA) or the SuperSignal West Femto chemiluminescent substrate (Thermo-Fisher Scientific, Waltham, MA) according to the manufacturers’ instructions. A Fotodyne Luminary/FX was used to capture images.

(i) Susceptibility testing.

All studies on B. mallei and B. pseudomallei were performed at the U.S. Army Medical Research Institute of Infectious Diseases (USAMRIID) in Frederick, MD. Broth microdilution susceptibility testing was conducted according to CLSI guidelines, including testing control compounds against quality control (QC) organisms (38, 43). The strains were chosen to represent geographical diversity and susceptibility. B. pseudomallei and B. mallei strains were grown on sheep blood agar at 37°C for 24 h and 48 h, respectively. The bacteria from the overnight agar plates were resuspended in PBS. The OD₆₀₀ was measured, and the concentration of the resuspension was calculated based on the CFU/milliliter of each strain at an OD₆₀₀ of 1. The resuspension was then diluted in cation-adjusted Mueller-Hinton broth (CAMHB) to get a final concentration in the assay of 5 × 10⁵ CFU/ml. MICs were determined by the microdilution method in 96-well microplates. For these tests, the SUL-DUR combination was assayed by 2-fold dilution of sulbactam in the presence of a fixed concentration of 4 μg/ml of durlobactam. Bacterial suspension was added to the compound plate to a final volume of 100 µl (5 × 10⁴ CFU). B. pseudomallei and B. mallei were incubated for 18 to 24 h and 40 to 48 h, respectively. The MIC was the lowest concentration of compound with no visible growth. Escherichia coli ATCC 25922 and Staphylococcus aureus ATCC 29213 were used as QC strains for doxycycline, with MIC ranges of 0.5 to 2 μg/ml and 0.12 to 0.5 μg/ml, respectively.

(ii) PenI preparation, enzyme kinetics, and ESI-MS.

E. coli DE3 Origami 2 cells carrying pET24a(+) bla_PenI carrying the S72F amino acid substitution were used for protein expression and purification, as previously described (13, 39). Briefly, cells were grown in super optimal broth and then isopropyl-β-d-1-thiogalactopyranoside was added to induce expression. Cells were pelleted and frozen at –80°C. Subsequently, the cells were lysed, and the β-lactamase was purified using preparative isoelectric focusing and verified by ESI-MS, as previously described, with >95% purity and a final molecular weight of 28,349.8 Da (13). Steady-state inhibition kinetics and ESI-MS with purified PenI and durlobactam were conducted as described above for PenA1, with the following modifications. The K_iapp and k₂/K values were obtained with 200 nM PenI. The K_m,NCF for PenI is 35 μM. The dissociation rate constant for the durlobactam-PenI complex was determined by incubating 1 µM PenI with 1 µM durlobactam for 5 min at room temperature, diluting the mixture 1:200,000, and adding 100 µM nitrocefin.

(iii) Sulbactam and durlobactam PK in mice.

Pharmacokinetic (PK) investigations in Crl:CD-1 mice were carried out with sulbactam and durlobactam administered as a 4:1 ratio spanning a range of doses from 10:2.5 mg/kg up to 200:50 mg/kg. These studies were done in support of in vivo efficacy studies completed in CD-1 mice utilized in neutropenic thigh and lung infection studies. PK in CD-1 mice mimics that obtained in BALB/c mice (data not shown) and was therefore used to provide exposure estimates in the present study for doses used in the melioidosis model. Sulbactam and durlobactam were dissolved in water for injection, and doses were administered subcutaneously at a dose volume of 10 ml/kg. Animals were fasted overnight, with food and water provided ad libitum following dose administration. Blood samples (n = 3/animal) were obtained via two mandibular bleeds, and one terminal sample was taken via cardiac puncture. Sparse PK sampling was completed with 3 animals’ samples obtained per time point. Sampling time points were 0.25, 0.5, 1, 2, 4, 6, 8, and 12 h postdose. Blood samples were collected into 0.5-ml BD microtainers containing 1.0 mg of K₂EDTA, which were then centrifuged at 2,000 × g for 10 min. Plasma was pipetted into 96-well cryotubes and stored at –80°C prior to analysis.

(iv) LC-MS/MS determination of sulbactam and durlobactam.

Plasma samples were thawed on ice prior to bioanalytical processing. Calibration standards and the QC were prepared in blank plasma by serial dilution. An aliquot of 50 µl of each calibration standard, QC (diluted separately), and plasma sample (dilute separately) was transferred to a 96-deep-well plate, followed by protein precipitation with 300 µl of crash solution (100% acetonitrile containing 0.1% formic acid and 250 ng/ml of carbutamide internal standard). The 96-well plate was then vortexed and centrifuged at 3,200 × g for 5 min to pellet the precipitated protein. Approximately 100 µl of supernatant was transferred for liquid chromatography with tandem mass spectrometry (LC-MS/MS) analysis. LC-MS/MS assay conditions are summarized in Table S3. A calibration line was constructed for each analyte using a cocktail of calibration standards. The peak area ratios of sulbactam/carbutamide (internal standard) and durlobactam/carbutamide were calculated, and a linear regression was performed relative to standard sulbactam and durlobactam concentrations. QC and PK sample concentrations of sulbactam and durlobactam were determined by reference to the respective calibration line. The assay had lower and upper limits of quantitation (LLOQ and ULOQ), respectively, of 1.0 ng/ml and 10,000 ng/ml for both durlobactam and sulbactam, with r² of 0.997 for durlobactam and 0.998 for sulbactam.

PK model fitting of the concentration versus time profile data for sulbactam and durlobactam was completed using Phoenix WinNonLin 8.1 (Certara L. P., Princeton, NJ). A 2-compartment model with an extravascular first-order absorption rate input provided the best fit of the subcutaneous PK obtained across the dose range. The mean PK parameter estimates are summarized in Table S2. These values were used to simulate exposures for doses of sulbactam and durlobactam in the melioidosis model.

(v) In vivo efficacy.

SUL-DUR and comparator compounds were evaluated for in vivo efficacy versus B. pseudomallei in an acute melioidosis model in BALB/c mice using the sentinel strain K96423 (SUL-DUR MIC = 1 µg/ml). This strain, which was isolated from a female patient in Thailand who succumbed to the infection (5), is among the most frequently used isolates in preclinical models of melioidosis. Notably, B. pseudomallei K96423 encodes β-lactamases, including PenI and OXA-59 (https://www.burkholderia.com). A lethal challenge with strain K96423 (4 × 10⁴ CFU/mouse or 174× the 50% lethal dose [LD₅₀]) (44) was administered intranasally (n = 10/group). Therapy for 6 consecutive days was initiated 4 h postchallenge via subcutaneous injection of SUL-DUR (q4h) or intraperitoneal administration of positive control compounds doxycycline (q12h) or ciprofloxacin (q12h). Animals receiving only a vehicle generally succumb to the infection within the first 3 days. Survivors were monitored for 45 days after infection for relapse. On the final day of the study, survivors were euthanized and the lungs, spleens, and livers were removed aseptically. The tissues were homogenized in 1 ml of sterile phosphate-buffered saline. One hundred microliters each of homogenate and 10-fold serial dilutions were plated in duplicate on LB agar and incubated for 48 h at 37°C for CFU enumeration. Data were reported as CFU/gram of tissue. Animal research at the USAMRIID was conducted under an animal use protocol approved by the USAMRIID Institutional Animal Care and Use Committee (IACUC) in compliance with the Animal Welfare Act, PHS policy, and other federal statutes and regulations relating to animals and experiments involving animals. The facility where this research was conducted is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALACi) and adheres to the principles set down in the Guide for the Care and Use of Laboratory Animals (45).

PK/PD understanding for effective treatment in a melioidosis model is not available. The doses of SUL-DUR were selected based upon exceeding PK/PD endpoints established for SUL-DUR activity against A. baumannii as determined from in vitro dynamic modeling studies (33) and doses associated with in vivo efficacy in neutropenic thigh and lung models (21). All in vivo procedures were completed in compliance with the Animal Welfare Act Regulations (9 CFR 3) under USAMRID IACUC protocols and under the supervision of a site attending veterinarian.