In Vitro and Intracellular Activities of Omadacycline Against Legionella pneumophila.

Omadacycline is an aminomethylcycline antibiotic with in vitro activity against pathogens causing community-acquired bacterial pneumonia (CABP). This study investigated the activity of omadacycline against Legionella pneumophila strains isolated between 1995 and 2014 from nosocomial or community-acquired respiratory infections. Omadacycline exhibited extracellular activity similar to comparator antibiotics; intracellular penetrance was found by day 3 of omadacycline exposure. These results support the utility of omadacycline as an effective antibiotic for the treatment of CABP caused by L. pneumophila.

MIC 90 values observed, followed by telithromycin, omadacycline, azithromycin, doxycycline, and erythromycin. Intracellular activity of omadacycline was determined against five strains of L. pneumophila serogroup 1. The mononuclear cell method (22) was performed using 48-well flat cell culture microplates. RPMI 1640 (with 10% heat-inactivated fetal bovine serum), mononuclear cells (U-937; 1 ϫ 10 6 to 2 ϫ 10 6 cells/ml), and Legionella inoculum (10 4 to 10 5 CFU/ml) were used. After a 1-h exposure in a shaking incubator, 150 l of infected cultures was maintained without shaking for 7 days at 37°C in 5% CO 2 and 95% air. After 24 h (day 1), infected cultures were washed three times (300 l). Antibiotics (150 l of diluted antibiotic at 1ϫ MIC) were added for a final volume of 300 l, and cultures were incubated for 2 days. After 72 h (day 3), cultures were washed three times and split into two groups-one with the same antibiotic and one without antibiotic (to observe potential intracellular postantibiotic effect)-for 4 days of incubation. Monocytes in a 20 l sample taken at time zero and every 24 h until day 7 were diluted by 10-fold dilutions and lysed with distilled water. CFU/ml counts were determined in duplicate using BCYE agar at each time point.
A reduction of 3 log 10 CFU/ml or 99.9% of L. pneumophila serogroup 1 grown in macrophages was reached only with omadacycline and moxifloxacin after 3 days of antibiotic exposure (Fig. 1). Compared with erythromycin, azithromycin, and levofloxacin, delayed regrowth of intracellular L. pneumophila was observed with omadacycline, moxifloxacin, and doxycycline after drug washout, day 3. A similar reduction and delayed regrowth of intracellular L. pneumophila was obtained at 2ϫ MIC, 8ϫ MIC, and 16ϫ MIC with omadacycline and moxifloxacin (data not shown).
The MIECs of omadacycline and comparators (doxycycline, azithromycin, and moxifloxacin) inhibiting intracellular human monocyte growth (22,23) were determined for the five strains of L. pneumophila serogroup 1. At days 1 and 3 of exposure, each strain was exposed to antibiotic concentrations of 1, 1/2, 1/4, 1/8, or 1/16 times the MIC required to determine the precise MIEC. Cultures were incubated with antibiotic for 4 days. CFU/ml counts were performed daily in duplicate using BCYE agar. MIEC was defined as the lowest MICs that produced intracellular reductions of Ն1 log 10 (CFU/ml) of L. pneumophila and was calculated at days 3 and 5 of exposure.
Mean reduction of intracellular activity (Ն92%) of L. pneumophila growth in macrophages was detected at day 5 of omadacycline exposure, with an MIEC/MIC ratio of 0.24 (1/4ϫ MIC) and MIEC of 0.06 mg/liter (Table 2). At day 3 of omadacycline exposure, an MIEC/MIC ratio of 0.5 (1/2ϫ MIC) and MIEC of 0.12 mg/liter were observed against all tested strains of L. pneumophila (Fig. 2).  and ϳ2,520 for tested strains of L. pneumophila, respectively. These important intracellular findings suggest an achievable level of omadacycline at the infection site and support the potency and clinical efficacy of omadacycline for the treatment of CABP caused by susceptible strains of L. pneumophila. Even when MIC results for doxycycline, moxifloxacin, and azithromycin were lower or higher than those for omadacycline, the MIEC/MIC ratio of omadacycline at day 5 (0.24 or 1/4ϫ MIC) was consistently lower than the MIEC/MIC ratio of moxifloxacin (0.5 or 1/2ϫ MIC), doxycycline (1 or 1ϫ MIC), and azithromycin (Ͼ1 or Ͼ1ϫ MIC).
Omadacycline demonstrated potent in vitro activity against L. pneumophila serogroups 1 to 6. Based on the MIC 90 values, omadacycline was 4-fold more potent by weight than doxycycline and erythromycin; omadacycline MIC 90 values were 2-fold lower by weight than that of azithromycin. Omadacycline was 10-fold less potent by weight than telithromycin and fluoroquinolones tested. Noteworthy was the activity of omadacycline against L. pneumophila serogroup 1, the most common serotype isolated from nosocomial or community-acquired respiratory tract infections. Although L. pneumophila strains were isolated from patients across broad time frames, no change in MIC values was seen for omadacycline or comparators, indicating stable susceptibility across 20 years.
L. pneumophila is isolated as the cause of CAP in ϳ2% to 5% of cases, but this incidence increases as much as 2-fold in hospitalized patients and the elderly (7). L. pneumophila is an intracellular pathogen, and understanding the intracellular activity, extracellular activity, and cellular penetration of an antibiotic is necessary to evaluate its potential utility. The current study results indicate that omadacycline demonstrates relative intracellular penetrance against L. pneumophila serogroup 1, comparable to other antibiotics used for CABP treatment. Findings also support those from a phase 3 study of CABP in which omadacycline was comparable to moxifloxacin, with a 87% early clinical success rate among 37 patients for whom L. pneumophila was identified as the causative pathogen (25). Thus, omadacycline may be a potential option for empirical therapy for CABP, particularly when atypical bacteria, especially L. pneumophila, are suspected.

ACKNOWLEDGMENTS
Editorial support in the form of development of the first draft of the manuscript was provided by Richard Perry, PharmD. Editorial support of the revised manuscript was provided by Theresa E. Singleton, PhD, of Innovative Strategic Communications. This work was supported by Paratek Pharmaceuticals, Inc., King of Prussia, PA. The author Jacques Dubois is also president of M360 Inc., an organization that contracted this work with Paratek Pharmaceuticals Inc., and declares conflicts of interest relevant to this study. The authors Maïtée Dubois and Jean-François Martel declare no conflicts of interest relevant to this study. Jacques Dubois, Maïtée Dubois, and Jean-François Martel contributed equally to this study and to the review and revision of the manuscript.