GC-072, a Novel Therapeutic Candidate for Oral Treatment of Melioidosis and Infections Caused by Select Biothreat Pathogens

Burkholderia pseudomallei, the etiological agent of melioidosis, is a Gram-negative bacterium with additional concern as a biothreat pathogen. The mortality rate from B. pseudomallei varies depending on the type of infection and extent of available health care; in the case of septicemia, left untreated, it can range from 50% to 90%. Current therapy for melioidosis is biphasic, consisting of parenteral acute-phase treatment for 2 weeks or longer, followed by oral eradication-phase treatment lasting several months.

studies (17,51). Bacterial resistance to quinolone compounds is predominantly due to target protein mutations, although efflux-and plasmid-mediated resistance mechanisms are also known (52,53). The amino acid changes in DNA gyrase and topoisomerase IV (Topo IV) leading to fluoroquinolone resistance occur mainly in the quinolone resistance-determining region (QRDR) of respective subunits GyrA and ParC (54,55).
Modifying the structure of existing antibiotics to increase potency and overcome mechanisms associated with resistance provides a more practical approach than finding new antibacterial agents with novel mechanisms of action (56). GC-072 ( Fig. 1) is a 4-oxoquinolizine, a class for which there were a number of publications and patents in the 1990s, mostly from Abbott Laboratories; however, relatively little about this class has been recently reported. There are no available literature reports indicating clinical development of 4-oxoquinolizines. The therapeutic targets of oxoquinolizines, such as GC-072, are the bacterial DNA gyrase and Topo IV enzymes, collectively known as type IIA topoisomerases. Oxoquinolizines, which include GC-072, possess potent antimicrobial activity against a broad spectrum of organisms, including Gram-positive, Gramnegative, and resistant bacteria, including those with mutations in the QRDR that confer quinolone resistance (57)(58)(59)(60). The position of nitrogen in the heteroaromatic structure of GC-072 confers different physicochemical and biological properties than those of the fluoroquinolone class of antibiotics. The C-8 group has been demonstrated to have importance in terms of functionality for oxoquinolizines (also known as 2-pyridones) (57,60).
The results presented here from in vivo studies indicate GC-072 is a promising candidate for the oral treatment of acute B. pseudomallei-associated pneumonic infection. Current first-line standard-of-care treatment for melioidosis is i.v. ceftazidime (17), which does not have an oral option. The replacement of acute-phase i.v. treatment with an effective oral alternate would greatly enhance health care options for melioidosis treatment, particularly in resource-constrained settings or in the event of an intentional release (biowarfare).

RESULTS
In vitro selectivity of GC-072 for bacterial and human topoisomerases. GC-072 was tested in comparison to ciprofloxacin using gel-based topoisomerase assays ( Table  1). Results of these assays are reported as the half-maximal inhibitory concentration (IC 50 ) for inhibiting activity of bacterial topoisomerases II (DNA gyrase) and IV and  human topoisomerases I and II (Topo I and Topo II, respectively). GC-072 inhibited both Escherichia coli and S. aureus gyrase and Topo IV at concentrations comparable to, or lower than, those for ciprofloxacin. GC-072 more actively inhibited DNA gyrase isolated from quinolone-resistant E. coli than ciprofloxacin, implying GC-072 has a different binding mode than fluoroquinolones such as ciprofloxacin. Furthermore, GC-072 demonstrated no detectable inhibition of human Topo I and II, comparable to that of ciprofloxacin, indicating that it is selective for bacterial topoisomerases and likely to have an acceptable genotoxicity profile.
In vitro susceptibility of B. pseudomallei to GC-072. Susceptibility testing was performed to determine the MICs of GC-072 for 100 strains of B. pseudomallei. These strains represent the geographic distribution across Southeast Asia. The majority are clinical isolates and include strains resistant against "front-line" antibiotic therapies, including ceftazidime, carbapenem, and tetracycline. GC-072 demonstrated good activity against B. pseudomallei, with an MIC 90 of 0.25 g/ml and a range of Յ0.008 to 1 g/ml (Table 2). This compares to MIC 90 s of 8, 8, 1, 64, and 4 g/ml for ciprofloxacin, finafloxacin, meropenem, ceftazidime, and doxycycline, respectively. The quality control strain MIC results were all within CLSI ranges for the comparator antibiotics.
In vitro susceptibility of additional bacterial biothreat pathogens to GC-072. Susceptibility testing was performed to determine the activity of GC-072 and antibiotic comparators against four additional biothreat agents. GC-072 was tested against 30 geographically biodiverse strains each of category A biothreat agents Bacillus anthracis, Yersinia pestis, and Francisella tularensis and category B biothreat agent Burkholderia mallei. Overall, GC-072 exhibited strong in vitro antimicrobial potency against all four biothreat pathogens. The results are summarized in Table 3. The MIC 90 s of GC-072  Mutation frequency. The mutation frequencies to 3ϫ the MICs of GC-072 and ceftazidime for B. pseudomallei 1026b were determined in a total of 11 experiments (6 with GC-072 and 5 with ceftazidime in parallel). The mutation frequency for GC-072 ranged from Ϫ7.1 log to Ϫ8.4 log CFU. The MICs of a subset of colonies that grew on GC-072-supplemented agar ranged from 0.5 to 2 mg/liter compared with an MIC of 0.125 to 0.25 mg/liter for the parent strain. The mutation frequency values for ceftazidime were Ϫ6.6 log to Ϫ8.0 log CFU. The colonies that grew on the ceftazidimesupplemented agar plates had MICs of 4 to 32 mg/liter, compared to an MIC of 1 to 2 mg/liter for the parent strain.
Time-kill study. An in vitro time-kill assay was performed to measure the rate and activity of GC-072 in killing B. pseudomallei at different concentrations. Bacterial suspensions containing B. pseudomallei strain 1026b were left untreated or inoculated with GC-072 at concentrations of 1ϫ, 2ϫ, 4ϫ, 8ϫ, or 16ϫ MIC and incubated over a 24-h period. Samples were collected from each suspension at five time points and enumerated. Untreated bacteria in the control group grew to titers Ͼ1 ϫ 10 9 CFU/ml, with maximal growth noted between 8 and 24 h (Fig. 2). GC-072 provided a dose-response effect on the reduction of viable B. pseudomallei 1026b. GC-072 demonstrated rapid bactericidal activity at all concentrations. As shown in Fig. 2, more than 75% of the full effect was observed within 5 h. Regrowth was not seen in any of the treatment arms, indicating that the killing effect of GC-072 was sustained. While complete killing was not observed at any of the GC-072 concentrations tested, at 24 h, the 8ϫ and 16ϫ MIC bacterial counts were at, or near, the lower limit of quantitation (100 and 15 CFU/ml, respectively). The limits of solubility of GC-072 in aqueous media may have resulted in reduced incremental killing between the 8ϫ and 16ϫ concentrations compared with the increments in killing between the lower concentrations. Alternatively, slow precipitation of the drug out of solution over time may have contributed to the lack of additional killing past the 8-h time point and, hence, the lack of sterilization of the cultures. Bacterial samples of GC-072-supplemented agar were not plated after treatment; therefore, it is unknown if resistant subpopulations may have emerged as a result of drug exposure in this particular study.
GC-072 is active against drug-resistant B. pseudomallei. Two studies were conducted to investigate the activity of GC-072 against B. pseudomallei strains with characterized antibiotic resistance mechanisms. An isogenic panel of B. pseudomallei 1026b-based efflux-proficient and -compromised strains that either expressed or lacked defined efflux pumps were used to assess the efflux propensity of GC-072. Doxycycline and ciprofloxacin were used as control antibiotics, as they are prone to efflux, particularly by the BpeEF-OprC pump in B. pseudomallei (41,48). The observed MIC for GC-072 against B. pseudomallei strain 1026b was within the previously observed MIC range for diverse B. pseudomallei strains, including other studies with 1026b ( Table 4). As demonstrated by an MIC comparable to the parental 1026b strain, GC-072 does not appear to be effluxed by strains solely expressing AmrAB-OprA (e.g., Bp227), and efflux by BpeAB-OprB is barely discernible (e.g., Bp58). GC-072 appears to be a substrate of BpeEF-OprC, as indicated by the increased MIC in strain Bp282 compared to that of strains either not expressing (e.g., 1026b) or lacking (e.g., Bp320) the BpeEF-OprC efflux system. However, the degree of efflux of GC-072 by BpEF-OprC was significantly lower than that observed with comparators ciprofloxacin and doxycycline, and the observed susceptibility (MIC, 0.5 to 1 g/ml) is within the range reported above for 100 clinical isolates.
In a separate study, a panel of select agent-excluded B. pseudomallei strain Bp82 derivatives was employed to assess single drug class resistance due to mutations causing ␤-lactam resistance (e.g., ceftazidime resistance due to PenA ␤-lactamase point mutations or overexpression or trimethoprim resistance due to folA mutation). The results show that GC-072 was fully active against the panel of drug-resistant B. pseudomallei strains tested (i.e., ceftazidime-, clavulanate-and trimethoprim-resistant strains) and its activity was superior to that of ciprofloxacin in this panel (Table 5).
GC-072 is active in a model of B. pseudomallei intracellular survival. The activity of GC-072 against intracellular B. pseudomallei 1026b was assessed via intracellular survival assays. The MIC for B. pseudomallei strain 1026b measured for this experiment was 0.25 g/ml and was used to set the multiples of the MICs for the intracellular survival assay. Cells were either left untreated or treated with 0.25 g/ml or 2.5 g/ml of GC-072. The results from this study indicate GC-072 is effective at inhibiting growth of intracellular B. pseudomallei in murine macrophage cells in vitro. As shown in Fig. 3, at dose levels close to the MIC (0.25 g/ml) or approximately 10ϫ the MIC (2.5 g/ml), GC-072 effectively inhibited the growth of intracellular B. pseudomallei in a rapid timeand dose-dependent manner. After 1 h, the numbers of bacteria were significantly reduced in cultures treated with the higher GC-072 concentration compared to those in untreated cultures (P Ͻ 0.0023) (see Tables S1 and S2 in the supplemental material). At 5 h, there was a significant reduction in B. pseudomallei at both GC-072 concentra- a All strains used in this study are derived from B. pseudomallei 1026b as previously described, with the exception of Bp282 and Bp320. The latter two strains were derived as follows: Bp282 is a ciprofloxacin-resistant derivative of Bp207 obtained using passive selection. It contains a bpeT point mutation causing an S280P BpeT amino acid substitution. The passive selection experiment was conducted prior to 4 December 2012, and its performance and mutant possession did not require US Federal Select Agent Program approval. Bp320 is a Δ(bpeEF-oprC) derivative of Δ(amrAB-oprA) Δ(bpeAB-oprB) strain Bp282 and was constructed using the previously described methods. This strain does not express any known efflux pumps (48). b The MIC against each strain was determined using broth microdilution assays. The data shown represent three biological replicates, each in technical duplicates, reporting modes from 6 data points. c The control strains E. coli ATCC 25922 and P. aeruginosa ATCC 27853 were verified in parallel with all MIC testing to ensure that MICs for the control antibiotics were within the expected range, according to CLSI guidelines. tions compared to untreated B. pseudomallei, both having a P value of Ͻ0.0001 for cultures treated with 0.25 and 2.5 g/ml. Twenty-four hours after initiation of treatment, there were no viable bacteria detectable at either the lower (0.25 g/ml) or higher (2.5 g/ml) concentrations of GC-072, compared to a mean of 3.5 ϫ 10 5 bacteria per well in the untreated culture (Tables S1 and S2). Ceftazidime is traditionally not employed as a comparator drug in intracellular survival models due to poor macrophage permeability; ceftazidime concentrations as high as 10 g/ml have no effect on intracellular survival of B. pseudomallei in macrophages (61). Thus, ceftazidime was not used as a positive control in this study.
GC-072 is efficacious against B. pseudomallei aerosol infection in a mouse model following exposure to a 24-LD 50 bacterial challenge. An in vivo murine melioidosis model of pneumonic illness was employed to determine the efficacy of GC-072 against an aerosol challenge with B. pseudomallei strain 1026b. BALB/c mice  (47) was used to assess single drug class resistance due to target mutations causing ␤-lactam resistance (e.g., ceftazidime resistance due to PenA ␤-lactamase point mutations or overexpression) or trimethoprim resistance (due to folA point mutations). b The MIC against each strain was determined using broth microdilution assays. The data shown represent three biological replicates, each in duplicates, reporting modes from 6 data points. c Values for ceftazidime and ciprofloxacin obtained with E. coli and P. aeruginosa type strains are within the expected CLSI range. (n ϭ 10 mice/group) were treated starting at 8 or 24 h postchallenge with either GC-072 (1, 3, 10, or 30 mg/kg) administered three times a day intragastrically (i.g.), 150 mg/kg of ceftazidime delivered intraperitoneally (i.p.) four times daily, or vehicle administered i.g. thrice daily. For the challenge, mice were divided into two exposure groups balanced such that each cohort received five animals from each of the two aerosol runs. The actual bacterial exposures were 6.37 ϫ 10 4 and 1.08 ϫ 10 5 CFU/mouse for each of the two groups. These represented 17.7 and 30 LD 50 s, respectively, with a mean of 23.8 LD 50 s.
When treatment was initiated 8 h postchallenge, 0%, 0%, 70%, and 90% survival was observed in mice administered 1, 3, 10, and 30 mg/kg of GC-072, respectively, and 100% survival was observed in the positive-control group which received ceftazidime (Fig. 4A). When treatment with GC-072 was initiated 16 h postchallenge (Fig. 4B), 50% at the beginning of the study from causes deemed unrelated to exposure or drug; therefore, this group had a total of 9 mice. *, P Յ 0.05. and 100% survival was observed in the 10 and 30 mg/kg groups, respectively. None of the placebo-treated animals survived the challenge.
The median survival times were 3.125 days for vehicle control and 3.25 and 4 days for the 1-and 3-mg/kg GC-072 groups, respectively, under the 8-h treatment initiation condition. The median survival time for the 10-mg/kg GC-072 16-h postchallenge treatment initiation group was 43 days. All of the animals in the ceftazidime group survived, as did the majority of the animals in the 8-h 10-and 30-mg/kg GC-072 and 16-h 30-mg/kg GC-072 treatment groups; thus, median survival times could not be calculated.
Under the 8-h treatment initiation condition, a statistically significant (P Ͻ 0.0001) dose-response trend was observed for all doses of GC-072 (see Table S3). The 3-, 10-, and 30-mg/kg GC-072 groups showed significant efficacy over that of the control (P ϭ 0.0002, P Ͻ 0.0001, and P Ͻ 0.0001, respectively). Intraperitoneal treatment with ceftazidime showed a significant advantage over the groups administered 1 and 3 mg/kg GC-072 i.g. (P Ͻ 0.0001) but was not significantly different from the 10-and 30-mg/kg groups. Survival of mice treated with either of the GC-072 dose levels (10 and 30 mg/kg) in the 16-h treatment initiation group was superior to that of the vehicle-treated control group (P Ͻ 0.0001). For the 16-h initiation, administration of 30 mg/kg GC-072 showed a significant advantage over administration of 10 mg/kg GC-072 (P ϭ 0.0164).
GC-072 is efficacious against B. pseudomallei aerosol infection in a mouse model following exposure to a 339-LD 50 bacterial challenge. The next study further investigated the efficacy of GC-072 against B. pseudomallei strain 1026b using a challenge level approximately 10-fold higher than the 24-LD 50 challenge. BALB/c mice (n ϭ 10/group) were treated i.g. with GC-072 (37.5, 75, or 150 mg/kg) thrice daily, 150 mg/kg i.p. of ceftazidime four times a day, or vehicle thrice daily beginning 8 or 24 h following inhalational challenge (Fig. 5A). Mice were again divided into two challenge runs with actual bacterial exposure being 1.24 ϫ 10 6 and 1.20 ϫ 10 6 CFU/ mouse, representing 345 and 333 LD 50 s, respectively, with a mean of 339 LD 50 s.
When initiated 8 h postexposure, the groups treated i.g. with GC-072 had a significant survival benefit compared to both the positive comparator ceftazidime (P Ͻ 0.03) and vehicle control (P Յ 0.0006) groups (see Table S4). Mice in the 37.5-, 75-, and 150-mg/kg GC-072 treatment groups demonstrated 90%, 90%, and 80% survival, respectively, compared to 30% in the ceftazidime and 0% in the vehicle control groups. Median survival time was 3.75 days for vehicle and 9.25 days for ceftazidime; median survival times could not be calculated for the three groups administered GC-072, as most animals survived to the end of the experiment. No dose dependence was observed in this portion of the study, as there did not appear to be a significant difference in survival for any of the GC-072 doses tested.
In the 24-h treatment initiation groups, overall survival was low; only four animals from the 150-mg/kg group survived to the end of the experiment (Fig. 5B). However, each dose of GC-072 provided a significant survival advantage compared to vehicle control (P Յ 0.0006). There did not appear to be a significant difference between the 37.5-and 75-mg/kg GC-072 groups, but treatment with 150 mg/kg GC-072 provided a significant advantage over both lower-dose groups. Median survival times were 4.5 days for both the groups treated with ceftazidime and 75 mg/kg GC-072, 4.75 days for the group treated with 37.5 mg/kg GC-072, and 5.63 days for the group treated with 150 mg/kg GC-072.

DISCUSSION
GC-072 is a novel 4-oxoquinolizine antibiotic candidate with potent activity against Gram-positive and Gram-negative bacteria, including quinolone-resistant strains. The antimicrobial data presented here and structure-based modeling with GC-072 and target topoisomerases (data not shown) indicate that the position of nitrogen in the heteroaromatic structure confers different physicochemical and biological properties than those for fluoroquinolone antibiotics. Additionally, the substituent in the C-8 position of GC-072 also takes a different conformation from that of the piperidine ring in ciprofloxacin, potentially contributing to the differential activities seen on fluoroquinolone-resistant bacterial strains. Consistent with modeling studies, results from in vitro topoisomerase assays indicate GC-072 more actively inhibits S. aureus topoisomerases than ciprofloxacin, which aligns with higher MICs for ciprofloxacin on Gram-positive bacteria that were observed in susceptibility studies (data not shown). Additionally, results from topoisomerase studies indicate that GC-072 and ciprofloxacin similarly inhibited quinolone-susceptible E. coli DNA gyrase and Topo IV; however, GC-072 was more active against quinolone-resistant E. coli gyrase than ciprofloxacin, consistent with a different binding mode for these molecules as well as a potentially different binding site in the QRDR. These data are consistent with GC-072 having improved activity and a different binding mode compared to that of fluoroquinolones.
In vitro studies show GC-072 is highly effective against B. pseudomallei and other biothreat pathogens that are resistant to current standard-of-care antibiotics. GC-072 outperformed the comparator antibiotics when tested against all five sets of biothreat agents in category A (B. anthracis, F. tularensis, and Y. pestis) and category B (B. pseudomallei and B. mallei), which included strains with known resistance. B. pseudomallei is considered naturally refractory to treatment with ciprofloxacin and other fluoroquinolones. Of the strains evaluated in this study, 20% had MICs of 8 to 16 g/ml for ciprofloxacin. Based on the activity observed for GC-072, only 7% of strains had MICs higher than 0.5 g/ml for GC-072, and only a single strain, resistant to ciprofloxacin, had an MIC of 1 g/ml. The data also suggest that GC-072 is less affected by other resistance mechanisms, as a number of ceftazidime-, carbapenem-, and tetracycline-resistant strains were also included in this study.
In mutational frequency studies, when B. pseudomallei strain 1026b was exposed to 3ϫ the MIC of GC-072, mutations arose once every 7.1 log to 8.4 log CFU compared to once every 6.6 log to 8.0 log CFU when exposed to 3ϫ the MIC of ceftazidime. The mutation frequencies for GC-072 were consistently lower than for ceftazidime in all 5 experiments run in parallel, with ceftazidime having a 4.4-fold higher frequency of mutations than GC-072. The wild-type MICs for ceftazidime were on average 7.2-fold higher than the MICs for GC-072, and the MICs of isolates with reduced susceptibilities were on average 7.5-fold higher for ceftazidime than for GC-072. For both antimicrobials, the frequency of mutation studies demonstrated emergence of subpopulations with reduced susceptibilities to GC-072 and ceftazidime.
B. pseudomallei is known to have a number of efflux and other resistance mechanisms, which render the organism naturally resistant to a number of classes of antibiotics. However, the low MIC range observed for GC-072 in our studies indicates this candidate may be unaffected by these specific mechanisms. This observation was further investigated in a study of GC-072's efflux propensity and resistance to strains with defined mutations. The results confirm that GC-072 has in vitro activity against wild-type, multidrug-resistant, and single-drug-resistant B. pseudomallei strains and that this activity is superior to that of ciprofloxacin. Previous studies showed that BpeEF-OprC-mediated efflux is the main fluoroquinolone resistance mechanism in B. pseudomallei (48). Using a panel of isogenic efflux-proficient and -compromised B. pseudomallei strains, we observed that GC-072 was less prone to efflux than the fluoroquinolone ciprofloxacin and doxycycline. Although GC-072 is a substrate of the BpeEF-OprC pump (which confers resistance to ciprofloxacin and doxycycline), efflux of GC-072 does not appear to confer significant resistance (MIC, 0.25 to 1 g/ml) and is likely well within the therapeutic range, even for an overexpressing strain.
Intragastric treatment with a research and development formulation of GC-072 demonstrated equivalent or improved efficacy over ceftazidime in an in vivo murine inhalational melioidosis model, indicating this compound may be a promising candidate for the oral treatment of acute melioidosis and other biodefense indications. In the 24-LD 50 study, the 30-mg/kg dose of GC-072 given i.g., initiated at both 8 and 16 h postexposure, showed equivalence to 8-h initiation of treatment with the positive control ceftazidime given i.p. In the context of exposure to a high-concentration bacterial challenge (ϳ339 LD 50 s) with B. pseudomallei, GC-072 demonstrated superiority over the positive-control ceftazidime in the 8-h treatment initiation model, even at the lowest dose of 37.5 mg/kg (P Ͻ 0.004). When treatment was initiated at 8 h postchallenge, only 40% of the ceftazidime group survived, while groups treated with GC-072 had an overall survival rate of 87%, indicating treatment with GC-072 provided a significant advantage. Although very few animals survived in the 24-h treatment initiation groups, there was a clear extension in survival time in the groups treated with GC-072, particularly in the 150-mg/kg group. In other studies using this animal model, a positive bacterial burden has been observed with lower bacterial counts, even when treatment was continued for an additional 7 days (62). The results from both efficacy studies indicate that treatment with GC-072 requires further investigation.
However, the oxoquinolizine GC-072 shows promising results as an oral treatment for melioidosis, which would be advantageous in resource-constrained environments and necessary in the context of a widespread, intentional biothreat event or accidental exposure to B. pseudomallei. GC-072 was shown to have improved in vitro potency against fluoroquinolone-resistant B. pseudomallei strains. Isogenic B. pseudomallei strains expressing defined efflux pumps, a common mechanism of fluoroquinolone resistance, were also more susceptible to GC-072 than fluoroquinolones (Table 5) (48).
These factors greatly improve the potential use of GC-072 as a first-line antibiotic treatment for melioidosis.
Future studies with this compound should aim to define the therapeutic window, further optimize the treatment regimen, define pharmacokinetic/pharmacodynamic relationships (unbound therapeutic exposure levels) for human treatment, and examine extended lengths of treatment in an effort to clear residual bacteria that may cause recurrence of disease through regrowth of the pathogen. Overall, even with an efficacy equivalent to that of current standard of care, oral treatment with GC-072 could provide an advantage, as there are currently no approved oral treatment options for acutephase melioidosis.

MATERIALS AND METHODS
Antimicrobial compounds. GC-072 potassium salt was manufactured by AMRI (Albany, NY) and supplied by Emergent BioSolutions Inc. (Gaithersburg, MD). Material was supplied as powder and stored at 4°C until use. Comparator antibiotics, including ciprofloxacin, ceftazidime, meropenem, and doxycycline were purchased from U.S. Pharmacopoeia (Rockville, MD). Finafloxacin was purchased from MedChem Express (Monmouth Junction, NJ). All stocks were stored at Ϫ70°C until use.
Bacterial strains. A geographically and genetically diverse set of 100 strains of B. pseudomallei and 30 strains each of B. anthracis, Y. pestis, F. tularensis, and B. mallei were used for MIC determinations. Use of these sets establishes reasonable MIC 90 and MIC 50 values for each compound against the holistic diversity set. A compound's performance against naturally occurring or unknown strains of each pathogen is expected to fall within these MIC 90 and MIC 50 values, because the diversity set provides a good representation of isolates that may be encountered anywhere in the world and reflects many common resistance mechanisms against "front-line" antibiotic therapies.
All procedures involving select agents were performed in select agent approved biosafety level 3 (BSL3) facilities at the University of Florida or USAMRIID using approved select agent compliant procedures and protocols. B. pseudomallei 1026b, obtained through the NIH Biodefense and Emerging Infections Research Resources Repository (NIAID, NIH), was used for the mutation frequency, time-kill, efflux propensity, activity against drug-resistant strains, invasion, and in vivo efficacy studies (45,63). For assessing single drug class resistance, B. pseudomallei strain Bp82 (64) and its penA (37,65) and folA (47) mutant derivatives were used. Strain Bp82 is an attenuated derivative of strain 1026b and excluded from select agent regulations (15). All experiments with strain Bp82 and its derivatives were conducted at BSL2 facilities with institutional biosafety committee approval at the University of Florida. Growth media used for Bp82 and its derivatives were supplemented with 40 g/ml (broth) or 80 g/ml (agar) adenine.
Topoisomerase assays. Gel-based topoisomerase assays were developed based on the TopoGEN Inc. and Inspiralis assay systems, where purified bacterial Topo IV and DNA gyrase from E. coli and S. aureus and human Topo I and II were exposed to different concentrations of the compounds in the presence of DNA. These assays were performed using methods previously described by Nitiss et al. (66).
In vitro susceptibility testing. Bacterial inocula were prepared by suspending in cation-adjusted Mueller-Hinton broth (CAMHB) colonies from 18-to 24-h B. anthracis, B. pseudomallei, and B. mallei plates or 42-to 48-h F. tularensis and Y. pestis plates that were incubated at 35°C. Sheep blood agar plates were used for Y. pestis and B. anthracis, and chocolate agar plates were used for F. tularensis, B. pseudomallei, and B. mallei. Suspended cultures were diluted with CAMHB to a bacterial cell density of 10 5 CFU/ml adjusted based on comparison to a 0.5 McFarland standard. MICs were determined by the microdilution method in 96-well plates according to CLSI (67). Antibiotics were serially diluted 2-fold in 50 l of CAMHB. For all steps with F. tularensis, CAMHB was supplemented with 2% IsoVitaleX (Becton, Dickinson).
B. pseudomallei, B. anthracis, and B. mallei plates were incubated at 35°C for 18 to 24 h and F. tularensis, and Y. pestis were incubated for 42 to 48 h. Antibiotic concentrations tested for the diverse set of B. pseudomallei ranged from 128 to 0.004 g/ml for GC-072, ciprofloxacin, ceftazidime, and doxycycline, 16 to 0.008 g/ml for finafloxacin, and 63 to 0.03 g/ml for meropenem. Tested concentration ranges for the diversity sets of B. anthracis, F. tularensis, and Y. pestis were 1 to 0.0005 g/ml and were 8 to 0.004 g/ml for the diversity sets of B. mallei. Quality control of antibiotic stocks was established using the recommended CLSI reference strains E. coli ATCC 25922, S. aureus ATCC 29213, and Pseudomonas aeruginosa ATCC 27853.
Mutation frequency. Microdilution broth and agar dilution MICs and mutation frequencies were determined for GC-072 and ceftazidime for B. pseudomallei 1026b. For this study, a 1-g/liter solution of ceftazidime was made by dissolving the drug powder in sterile water and passed through a 0.22-m syringe. Similarly, GC-072 was solubilized in dimethyl sulfoxide (DMSO) and diluted in sterile water before 0.22-m filtration.
B. pseudomallei 1026b was grown overnight from a frozen stock on a blood agar plate (BD Diagnostics, Sparks, MD) that was incubated at 35°C in ambient air. The following morning, colonies were taken from the culture and inoculated into BBL cation-adjusted Mueller-Hinton II broth (BD Diagnostics, Sparks, MD). The suspension was gently vortexed, and the density of the bacterial suspension was visually adjusted to a 0.5 McFarland standard and then diluted with medium to a concentration of approximately 10 6 CFU/ml. Suspensions of P. aeruginosa ATCC 27853, used as an internal control for susceptibility studies with ceftazidime, were prepared using the same protocol. A 1-ml volume of the suspension was inoculated on each of five Difco Mueller-Hinton agar plates (BD Diagnostics, Sparks, MD) supplemented with GC-072-K or ceftazidime at a concentration of 3ϫ the agar MIC value observed with B. pseudomallei for the respective antibiotic agent. After the agar plates were incubated at 35°C for 48 h, the colonies on the drug-free and drug-supplemented agar plates were counted.
MICs were determined for a subset of colonies that grew on the GC-072-and ceftazidimesupplemented agar plates. The mutation frequency was calculated by dividing the number of colonies per milliliter on the antibiotic-supplemented agar by the number of colonies per milliliter on the drug-free agar plates.
Time-kill study. Thirty milliliters of 10 6 CFU/ml of a B. pseudomallei 1026b suspension was added to each of 6 flasks. GC-072-K was added to the flasks to achieve final concentrations of 0, 1, 2, 4, 8, and 16 times the MIC (0.125 g/ml) for B. pseudomallei 1026b. The flasks were placed on a shaker (200 rpm) in an ambient air incubator set at 35°C. At 0, 2, 5, 7.5, and 24 h of incubation, 1 ml of the suspension was collected from each flask, washed, quantitatively cultured on blood agar plates (BD Diagnostics, Sparks, MD), and enumerated after 48 h of incubation. Bacterial concentrations were log transformed and graphed.
Efflux propensity and activity against drug-resistant strains. For these studies, working stocks of GC-072 solubilized in DMSO (Fisher BioReagent, Pittsburgh, PA) were made by further diluting in Mueller-Hinton broth (MHB), and doxycycline and ciprofloxacin obtained from Gold Biotechnology (St. Louis, MO) were used as control antibiotics. The antimicrobial susceptibilities of strains to GC-072 were assessed by determining the MIC according to CLSI guidelines (67,68). A colony of the tester bacterium grown at 37°C on Lennox LB (MO BIO Laboratories, Carlsbad, CA) agar plates was selected and placed in 1 ml MHB. A sufficient volume of this suspension was added to 3 ml sterile saline (0.85% NaCl) to match a 0.5 McFarland turbidity standard. Media were supplemented with 80 g/ml adenine for Bp82-derived strains. Expression from the tac promoter in Bp82-P tac -penA ϩ was induced with 1 mM isopropyl-␤-Dthiogalactopyranoside (Gold Biotechnology, St. Louis, MO). MIC data were determined in technical duplicates and biological triplicates on three separate days, and plates were read after 20 h of incubation at 37°C. Intracellular activity. Intracellular growth assays were performed as previously described (69). Briefly, RAW 264.7 cells (obtained from ATCC via David Pascual, University of Florida [UFL]) were infected with B. pseudomallei 1026b at a multiplicity of infection (MOI) of 10 using an aminoglycoside protection assay. After coincubation of bacteria and cells for 1 h at 37°C, the Dulbecco's modified Eagle medium (DMEM) was then removed and the cells were washed three times with phosphate-buffered saline (PBS). Extracellular bacteria, suspended in DMEM, were killed by the addition of 250 g/ml amikacin (Gold Biotechnology) and 250 g/ml kanamycin (Gold Biotechnology), and plates were incubated for 1 h at 37°C. After adding fresh medium containing amikacin, kanamycin, and 0, 0.25, or 2.5 g/ml of GC-072, the cells were incubated at 37°C. Samples were removed at 0, 1, 5, and 24 h after initiation of treatment, washed three times with PBS, and then lysed by addition of 0.2% Triton X-100 (Amresco, Solon, OH) diluted in PBS. The lysates were 10-fold serially diluted in PBS, and three 10-l samples of each dilution were spotted onto Lennox LB agar plates. Bacterial cells growing on the plates were enumerated after overnight incubation at 37°C.
Animal care and use. Female BALB/c mice 6 to 8 weeks old (average weight, 19.75 g) were obtained from the National Cancer Institute/Charles River Laboratories (Frederick, MD). Animals were acclimated for 1 week prior to the challenge and had free access to food and water (ad libitum) throughout the study.
Guidelines in the Guide for the Care and Use of Laboratory Animals were adhered to for all experimental procedures. Research was conducted in the BSL3 laboratory at UFL in compliance with the Animal Welfare Act and other federal statutes and regulations. The facility is fully accredited by the American Association for the Accreditation of Laboratory Animal Care.
Aerosol infection. B. pseudomallei 1026b was grown overnight in brain heart infusion broth. For the aerosol challenge, the overnight culture was adjusted on the basis of the optical density at 600 nm (OD 600 ) to give a target-challenge aerosol dose, approximately 24 or 339 LD 50 s. To verify final starting bacterial concentrations, the adjusted bacterial cultures were serially diluted and plated on tryptic soy agar (TSA) plates. Colonies were enumerated following overnight incubation of the plates at 35°C.
The target inhaled dose of B. pseudomallei was administered to mice by whole-body aerosol, which was generated using a three-jet Collison nebulizer (70). In this model, one LD 50 was equivalent to approximately 3 ϫ 10 3 CFU/mouse. All aerosol procedures were controlled and monitored using the automated bioaerosol exposure system (71) operated with a whole-body rodent exposure chamber. Integrated air samples were obtained from the chamber using an all-glass impinger during each exposure. The AGI collections were serially diluted and plated on TSA, as described above. The inhaled dose (CFU/mouse) of B. pseudomallei was estimated using mouse respiratory rates per Guyton's formula (72) and calculated according to Roy and Pitt (73).
24-LD 50 bacterial challenge efficacy study. To create a liquid solution for administration via oral gavage, the potassium salt of GC-072, which has modest aqueous solubility and good permeability, was homogenized in water containing 0.5% methylcellulose to enhance wetting of the surface of the particles. Eight groups of 10 mice each were included in this study. Beginning at 8 or 16 h after challenge, GC-072 was administered as doses of 1, 3, 10, or 30 mg/kg every 8 h (q8h) i.g. via oral gavage for 14 days. The positive comparator, ceftazidime (150 mg/kg), was administered i.p. every 6 h (q6h) for 14 days. A vehicle control group received 0.5% methylcellulose in sterile water for injection (SWFI) by oral gavage q8h. Mortality was assessed and recorded every 6 h during antibiotic administration for 14 days and at least twice a day thereafter for 45 days postchallenge. The experiment was terminated at day 45.
339-LD 50 bacterial challenge efficacy study. This experiment included 9 cohorts of 10 mice each. Beginning 8 or 24 h postchallenge, GC-072 was administered as doses of 37.5, 75, or 150 mg/kg q8h by oral gavage for 14 days. The positive comparator, ceftazidime (150 mg/kg), was administered i.p. q6h for 14 days. A vehicle control group received SWFI by oral gavage q8h. Mortality was assessed and recorded every 6 h for 14 days during antibiotic administration and at least twice daily thereafter for 37 days after the challenge.
Statistical analysis. All graphs were generated using GraphPad Prism version 7.00 for Windows (GraphPad Software, La Jolla, CA, USA).
MIC 90 and MIC 50 values from in vitro susceptibility testing of 100 strains of B. pseudomallei were calculated by determining the 50th percentile (median) and 90th percentile, respectively, using Prism version 7.00. MIC 90 and MIC 50 values from in vitro susceptibility testing of additional biodefense pathogens (B. anthracis, F. tularensis, Y. pestis, and B. mallei) were determined by rank-ordering MIC values from highest to lowest in a Microsoft (Redmond, WA) Excel spreadsheet and identifying the MIC 90 (corresponding to the 27 th strain in the ranking of 30-strain sets) and the MIC 50 (corresponding to the 15th strain in each rank-ordered set), respectively.
For analysis and graphing of efflux propensity, activity against drug-resistant strains, and intracellular activity studies, GraphPad Prism version 6.00 and Excel 2016 for Windows were used. The errors of the means from three biological replicates for efflux propensity and activity against drug-resistant strains were calculated using Microsoft Excel. Intracellular activity data were analyzed in Prism version 7.00 using a 2-way analysis of variance and Tukey's test for individual comparisons.
Analysis of survival data from in vivo efficacy studies was performed by employing a stratified Kaplan-Meyer analysis with a log rank (Mantel-Cox) test as implemented in Prism version 7.00.

SUPPLEMENTAL MATERIAL
Supplemental material for this article may be found at https://doi.org/10.1128/AAC .00834-19.