Microbiological Profile of Sarecycline, a Novel Targeted Spectrum Tetracycline for the Treatment of Acne Vulgaris

Sarecycline is the first narrow-spectrum tetracycline-class antibiotic being developed for acne treatment. In addition to exhibiting activity against important skin/soft tissue pathogens, sarecycline exhibits targeted antibacterial activity against clinical isolates of Cutibacterium acnes.

ethoprim) are associated with significant side effects and a risk for resistance development (6).
The antimicrobial action of tetracyclines against C. acnes occurs via inhibition of protein synthesis (7). In addition, C. acnes also produces proteins/enzymes that play a role in inflammation (e.g., lipase), which would also be downregulated as a consequence of inhibition of protein synthesis and which may account for the antiinflammatory properties observed with sarecycline and other tetracyclines (7,8).
While several tetracycline agents are available for acne treatment, the superior efficacy of one agent over another has never been determined. Therefore, side effect profiles may serve as a primary consideration in the choice of therapy. Notably, in recent years, the role of the human microbiome in maintaining health (9)(10)(11) and the impact of broad-spectrum antibiotics on dysbiosis (12) have garnered increasing attention. As minocycline and doxycycline exhibit potent broad-spectrum antimicrobial activity beyond their targeted pathogens (13), their widespread use is associated with off-target antibacterial effects on the human microbiome (i.e., intestinal flora), which may manifest clinically as diarrhea, fungal overgrowth (in the intestine and vagina), and vaginal candidiasis, especially in patients undergoing acne treatment, which involves prolonged oral administration (the typical duration of oral doxycycline or minocycline therapy is ϳ12 weeks) (14)(15)(16). A potential association with inflammatory bowel disease has also been attributed to the widespread use of doxycycline and minocycline (17). Furthermore, the increased and prolonged use of tetracyclines is associated with the development of antibiotic resistance (8), though the risk may be lower than that from erythromycin, which is also utilized in acne treatment (18).
Sarecycline is a novel oral aminomethylcycline with a unique and stable modification at position C-7-7-{[methoxy(methyl)amino]methyl} ( Fig. 1) and has recently completed evaluation in two phase 3 clinical trials, in which it was found to meet the 12-week primary efficacy endpoint for the treatment of moderate to severe acne (19,20). The aim of the current study was to determine the spectrum of in vitro activity of sarecycline and comparator tetracyclines against clinical isolates of a broad panel of both aerobic and anaerobic bacteria, including C. acnes, and to assess its in vivo efficacy, mode of action, and potential for resistance development.

RESULTS
Activity against clinical isolates of C. acnes. To assess the antibacterial activity of sarecycline and the comparators versus C. acnes, 55 clinical isolates of C. acnes were tested (the results are summarized in Table 1). The MIC values for sarecycline ranged from 0.5 to 16 g/ml; the MIC 50 was 0.5 g/ml, and the MIC 90 was 4 g/ml. The comparator agents, which included tetracycline, doxycycline, and minocycline, exhibited similar antibacterial activity. A subset of the C. acnes isolates included organisms with high-level resistance to the macrolide erythromycin (MICs Ն 128 g/ml). All the tetracycline agents, including sarecycline, exhibited activity against the macrolideresistant organisms ( Table 2). Only one strain (strain 16099) exhibited an elevated sarecycline MIC of 16 g/ml, and elevated MICs of the comparator tetracyclines were observed as well. amino]methyl}. The positions marked in orange at C-7 and C-9 have been modified to create tetracycline derivatives to potentially overcome tetracycline resistance mechanisms and to change bacterial ribosome binding. The figure is modified from that at http://www.chemspider.com/Chemical-Structure.28540486.html.

Spectrum of activity.
Tetracyclines are known to have a broad spectrum of antibacterial activity when used for the treatment of acne, which includes activity against the normal microbiome (an unwanted off-target effect). Therefore, it was of interest to test the activity of sarecycline and the comparators against a broad collection of organisms encompassing the normal flora across the human body-including Gram-positive cocci, Enterococcus spp., Enterobacteriaceae, and Gram-positive and -negative anaerobes-to assess sarecycline's spectrum of activity.
The activity of sarecycline and comparators against a broad collection of clinically important Gram-positive cocci is summarized in Table 3. Among the isolates of Staphylococcus aureus, sarecycline, like the other tetracyclines, maintained equivalent activity against both methicillin-susceptible and -resistant isolates, with an MIC 90 value of 0.5 g/ml. Sarecycline was also active against clinical isolates of Staphylococcus epidermidis, with MIC 90 values of 2 g/ml against both methicillin-susceptible and -resistant phenotypes. Sarecycline was more active than tetracycline and doxycycline against Staphylococcus haemolyticus, with an MIC 90 value of 2 g/ml, whereas the MIC 90 s of tetracycline and doxycycline were Ͼ32 and 16 g/ml, respectively. Among the clinical isolates of Streptococcus pyogenes and Streptococcus agalactiae tested, sarecycline also exhibited activity equivalent to that of minocycline, with MIC 90 s of 8 g/ml and 16 g/ml, respectively.
The susceptibility results for sarecycline and the comparator agents against the Enterococcus spp. are also summarized in Table 3. Sarecycline had limited antibacterial activity against the vancomycin-susceptible isolates of Enterococcus faecalis, with an MIC 50 value of 32 g/ml, whereas the MIC 50 s of doxycycline and minocycline were 8 g/ml. Although sarecycline was also less active against both vancomycin-susceptible and -resistant isolates of Enterococcus faecium than against the other species tested, it had activity that was either equivalent to or slightly less than that of the comparator tetracyclines.
The activity of sarecycline and the comparator agents against common members of the Enterobacteriaceae is summarized in Table 4. Among the isolates of Enterobacter cloacae tested, sarecycline was the least active of the agents tested, with an MIC 50 of  A separate study was also conducted to assess the activity of sarecycline and the comparators against 389 contemporary clinical isolates consisting of 10 members of the Enterobacteriaceae and the normal flora in the human gastrointestinal tract. The organisms were collected in 2015 and 2016 from patients at the common acne age of 11 to 40 years. The list of organisms and the activity of sarecycline are summarized in Table 5. Among the isolates of the Citrobacter freundii species complex, sarecycline was the least active agent tested, with an MIC 50 and an MIC 90 of 16 g/ml and 128 g/ml, respectively, whereas the MIC 50 was 0.5 or 1 g/ml for the other tetracyclines and the MIC 90 was 8 and 16 g/ml for minocycline and doxycycline, respectively. Tetracycline's MIC 90 value of 128 g/ml was as low as that of sarecycline. Similar results were observed for E. coli, K. pneumoniae, and Klebsiella oxytoca, for which the MIC 50 value for sarecycline was 16 g/ml, whereas it ranged from 0.5 to 2 g/ml for tetracycline, doxycycline, and minocycline. The MIC 90 value of sarecycline against these species was 128 to 256 g/ml, whereas it was 8 to 32 g/ml for minocycline and doxycycline, but the lower activity of sarecycline was similar to that of tetracycline, which showed an MIC 90 of 128 to 256 g/ml. Against the isolates of Enterobacter aerogenes, sarecycline was also the least active agent, with an MIC 50 and an MIC 90 of 8 g/ml and 16 g/ml, respectively, whereas the MIC 50 was 1 to 2 g/ml and the MIC 90 was 4 to 8 g/ml for the other tetracyclines. Similar results were observed for the Enterobacter cloacae species complex, against which the MIC 50 and MIC 90 values of sarecycline were 32 g/ml and 64 g/ml, respectively, whereas the MIC 50 and MIC 90 of the other tetracyclines were 2 g/ml and 8 g/ml, respectively. For the isolates of Serratia marcescens, the MIC 50 value of sarecycline was 32 g/ml, whereas those of tetracycline, doxycycline, and minocycline were 2 to 16 g/ml. The MIC 90 value of sarecycline was 64 g/ml, whereas the MIC 90 values of minocycline and doxycycline were 4 and 8 g/ml, respectively, but the lower activity of sarecycline was similar to that of tetracycline. Sarecycline showed no activity against the Morganella morganii and P. mirabilis clinical isolates tested, with MIC 50 and MIC 90 values of Ͼ256 g/ml, whereas the MIC 50 and MIC 90 values of the other tetracyclines ranged from 2 to 128 g/ml. Sarecycline, in common with the other tetracyclines, was largely inactive against the isolates of Providencia stuartii that were tested. The comparative MIC distributions for sarecycline and the comparators against 389 contemporary clinical isolates from 10 members of the Enterobacteriaceae and the normal flora found in the human intestinal track are shown as a Finlandogram in Fig. 3. Similar to the trend shown in Fig. 2A Fig. 2B. The MIC distributions for sarecycline are shifted to the right compared to those for minocycline and doxycycline, reflecting a 4-to 8-fold reduced potency against the representative Gram-positive anaerobes.
The MIC distributions for sarecycline and the comparators against 18 isolates of Gram-negative anaerobes were tested as well. These anaerobes included Bacteroides   Fig. 2C. Sarecycline was the least active tetracycline against the representative Gram-negative anaerobes.
Effect on macromolecular biosynthesis. Protein synthesis is a vital precursor to macromolecular biosynthesis, which is an essential component of microbial growth and homeostasis. Tetracyclines elicit their antimicrobial activity by targeting protein synthesis. To assess sarecycline's mechanism of action in comparison to that of the other tetracyclines, the inhibitory effects of each molecule on major biosynthetic endpoints,   Table 6). Sarecycline, doxycycline, and minocycline were effective in treating the systemic infection caused by S. aureus, with 50% protective dose (PD 50 ) values of 0.25, 0.3, and

was determined by measurement of the incorporation of [ 3 H]thymidine, [ 3 H]uridine, [ 3 H]leucine, [ 3 H]Nacetylglucosamine, and [ 3 H]glycerol, respectively. Control agents included ciprofloxacin (a DNA synthesis inhibitor), linezolid (a protein synthesis inhibitor), cerulenin (a lipid synthesis inhibitor), vancomycin (a cell wall biosynthesis inhibitor)
, and rifampin (a RNA synthesis inhibitor). Data represent the median with 95% confidence intervals (n ϭ 3). 0.03 mg/kg of body weight, respectively. In contrast, sarecycline did not demonstrate in vivo efficacy against the systemic infection caused by E. coli PBS1478, even at the highest dose of 40 mg/kg tested, whereas both minocycline and doxycycline exhibited protective effects at doses that were Յ7 mg/kg. To assess the comparative efficacies of sarecycline and doxycycline against S. aureus RN450-1, a murine neutropenic thigh infection model was utilized to represent a tissue-based infection (the results are summarized in Table 7). Sarecycline achieved a 2-log 10 reduction in the bacterial burden in the thigh at a dose comparable to that of doxycycline, with 50% effective dose (ED 50 ) values of 8.23 and 8.32 mg/kg, respectively.
Spontaneous mutation frequency and antimicrobial resistance. Single-step resistance development studies were performed, using three clinical isolates and one American Type Culture Collection (ATCC) isolate of C. acnes and a total of 4 clinical isolates of S. epidermidis (methicillin susceptible) and S. aureus (methicillin susceptible), to assess sarecycline's propensity for resistance development. Sarecycline had low spontaneous mutation frequencies ranging from 10 Ϫ9 to 10 Ϫ11 for C. acnes at 4-and 8-fold the MIC, similar to those for the comparator agents minocycline and vancomycin (Table 8). Sarecycline also showed low spontaneous mutation frequencies ranging from 10 Ϫ9 for S. aureus and 10 Ϫ8 for S. epidermidis at 4-and 8-fold the MIC, similar to those for vancomycin (Table 9).
Active efflux and ribosomal protection are two common mechanisms of tetracycline resistance that have been identified (21). Acquisition of the gene tet(K), tet(L), or tet (38) confers the ability for active efflux, whereas the acquisition of the gene tet(M), tet(O), tet(S), or tet(W) confers ribosomal protection (8,22). The most common genes that  confer tetracycline resistance among S. aureus strains are tet(M) and tet(K) in combination (23)(24)(25)(26). In order to assess susceptibility to sarecycline in the presence of known tetracycline resistance mechanisms, a defined collection of tetracycline-resistant S. aureus strains in which resistance is conferred by tet(K), tet(M), and tet (38) was tested against comparator agents (the results are summarized in Table 10). Strains of S. aureus were utilized, as strains of C. acnes with tetracycline resistance mechanisms were not available at the time that the study was conducted. Sarecycline was more active than tetracycline against strains of the tet(K) genotype, with the MICs of sarecycline ranging from 0.12 to 0.5 g/ml and those of tetracycline ranging from 16 to 65 g/ml. However, in common with the other tetracyclines, sarecycline displayed elevated MICs against strains containing a combination of both tet(M) and tet (38).

DISCUSSION
Tetracyclines such as doxycycline and minocycline have been widely used for acne treatment due to their improved bioavailability and lipophilicity over those of tetracycline, providing better tissue penetration, including uptake into the pilosebaceous unit, than tetracycline (27). Doxycycline and minocycline also exhibit more potent broadspectrum antibacterial activity than the other tetracycline derivatives, along with better absorption from the gastrointestinal tract (4,28,29). Due to their potent broadspectrum antibacterial activity, both doxycycline and minocycline are approved for multiple indications, in addition to acne treatment (13). However, recent advances in the characterization of the human intestinal microbiome have shed greater light on the impacts of the widespread use of oral broad-spectrum antibiotics, including the tetracyclines (30)(31)(32). Treatment with tetracyclines may predispose acne patients to the overgrowth of tetracycline-resistant (and, potentially, multidrug-resistant) organisms, including the overgrowth of Candida albicans, ultimately resulting in intestinal and vaginal dysbiosis (33)(34)(35). There are also reports that the widespread use of minocycline  and doxycycline presents an increased risk for inflammatory bowel disease (17). Although sarecycline retains antibacterial activity against C. acnes and important skin pathogens, such as staphylococci, the results of the present study demonstrated reduced activity-compared to that of doxycycline and minocycline-against aerobic enteric Gram-negative bacteria, as well as representative anaerobes that comprise the normal intestinal flora. Sarecyline's unique narrow spectrum of antibacterial activity may result in reduced dysbiosis of the intestinal flora (and, potentially, vaginal flora), manifesting as reduced overgrowth of resistant bacteria and C. albicans yeast infections as well as reduced gastrointestinal adverse effects, such as diarrhea. Two pivotal identically designed, placebo-controlled, phase 3 trials including approximately 2,000 patients evaluated the efficacy and safety of once-daily sarecycline treatment at 1.5 mg/kg for 12 weeks in patients aged 9 to 45 years with moderate to severe facial acne vulgaris (19,20). The rates of adverse events associated with the gastrointestinal tract, such as nausea, vomiting, and abdominal pain, were generally low, while the rate of diarrhea in patients treated with sarecycline was comparable to that in patients treated with placebo (19,20). Additionally, vulvovaginal candidiasis and vulvovaginal mycotic infection were rare (19,20). In contrast, other tetracyclineclass antibiotics may be associated with gastrointestinal tract side effects, phototoxicity (typically seen with doxycycline), candidiasis, or vestibular side effects (observed with minocycline) (5), whereas sarecycline produced such side effects at low rates (19,20). Furthermore, the administration of doxycycline, minocycline, and tetracycline has been associated with disruption of the gut microbiome (36)(37)(38). Additional studies of sarecycline testing its putative reduced effects on the intestinal microbiome by comparing its effects with those of doxycycline and minocycline using in vitro intestinal/gut models (39) or studies in acne patients or healthy volunteers would be beneficial for a more comprehensive assessment of the potential benefits of minimizing intestinal dysbiosis.
Acne remains one of the most prevalent skin conditions among adolescents (40). Although several treatment options are available (28), there are several challenges that remain, including limiting resistance among C. acnes isolates, simplifying treatment regimens, and developing new agents with more favorable safety profiles. Sarecycline is the first narrow-spectrum tetracycline-derived antibiotic that may reduce the potential for gastrointestinal dysbiosis, adverse effects, and concerns regarding resistance development during therapy.

MATERIALS AND METHODS
Bacterial isolates. A total of 55 individual clinical isolates of C. acnes were obtained from a culture collection that was tested centrally at the R. M. Alden Research Laboratory (Culver City, CA). C. acnes isolates were collected from 2003 to 2010 from various clinical specimens, such as acne facial skin, diabetic foot infection skin, blood, and specimens from infection sites, such as abdominal abscesses and perirectal abscesses. The isolate collection also included a subset of isolates demonstrating high-level resistance to erythromycin (MICs Ն 128 g/ml). Among the other representative anaerobes tested in the profiling study were 19  Antibiotics and in vitro susceptibility testing. Sarecycline powder was provided by Allergan plc (Irvine, CA). Tetracycline, doxycycline, minocycline, clindamycin, and erythromycin were purchased from Sigma-Aldrich (St. Louis, MO) and were tested as comparator agents to benchmark the activity of sarecycline. All anaerobic bacteria were tested using the reference agar dilution method in accordance with CLSI guidelines (41). All aerobic bacteria were tested using the reference broth microdilution (BMD) method in accordance with CLSI guidelines (42).
Mode of action. Sarecycline and the comparator agents were evaluated for their effects on macromolecular biosynthesis (DNA, RNA, cell wall, protein, and lipid synthesis) in S. aureus ATCC 29213. For DNA, RNA, and protein synthesis, the effects of the test agents on the incorporation of [ 3 H]thymidine (DNA), [ 3 H]uridine, or [ 3 H]leucine were studied. Sarecycline was tested at multiples of the MIC value in triplicate in 96-well microtiter plates. The S. aureus ATCC 29213 culture was used after the growth reached early exponential phase (optical density at 600 nm ϭ 0.2 to 0.3) in either Mueller-Hinton broth (DNA synthesis) or M9 minimal medium (protein synthesis). Following a 5-min incubation of the bacterial culture containing sarecycline, either [ 3 H]thymidine (DNA synthesis), [ 3 H]uridine (RNA synthesis), or [ 3 H]leucine (protein synthesis) was added at 0.5 to 1.0 Ci per reaction mixture. The reactions were allowed to proceed at room temperature for 15 to 30 min and then stopped by adding 12 l of cold 5% trichloroacetic acid (DNA and RNA synthesis) or 5% trichloroacetic acid-2% Casamino Acids (protein synthesis). The reaction mixtures were incubated on ice for 30 min, and the samples were collected and counted using a Beckman LS 3801 liquid scintillation counter.
For cell wall biosynthesis, the test agents were evaluated for their effects on the incorporation of [ 3 H]N-acetylglucosamine, while [ 3 H]glycerol was used for lipid synthesis. In a process similar to that used for DNA synthesis, following a 5-min incubation of the bacterial culture containing sarecycline at room temperature, [ 3 H]N-acetylglucosamine (0.5 Ci/reaction mixture) was added. The mixture was allowed to incubate for 30 min at room temperature, and samples were collected and counted using a Beckman LS 3801 liquid scintillation counter.
The control agents evaluated in the macromolecular synthesis assays included ciprofloxacin (a DNA synthesis inhibitor), linezolid (a protein synthesis inhibitor), cerulenin (a lipid synthesis inhibitor), vancomycin (a cell wall biosynthesis inhibitor), and rifampin (an RNA synthesis inhibitor).
In vivo efficacy. The murine systemic intraperitoneal infection model with E. coli strain 1478 and S. aureus strain RN450-1 was used to evaluate the in vivo efficacy of sarecycline and the comparator agents. E. coli strain 1478 was obtained from the Paratek Pharmaceuticals culture collection and was originally derived from Bristol-Myers Squibb strain SC8294. The S. aureus RN450-1 strain was produced by taking S. aureus RN450 from the Paratek Pharmaceuticals culture collection (original strain, NCTC8325) and passaging it in vivo to increase its virulence, and then the S. aureus RN450-1 strain was cultured from blood collected from mice infected with S. aureus RN450. Six-week-old specific-pathogen-free male CD-1 mice weighing 18 to 30 g (Charles River, Hartford, CT) were used for all experiments.
For the S. aureus RN450-1 infection, the bacterial culture grew overnight in Mueller-Hinton broth to approximately 1 ϫ 10 9 CFU/ml. Serial dilutions of the bacterial suspension were performed in phosphatebuffered saline (PBS; Fisher Scientific, Boston, MA) to obtain the infectious dose. Septicemia was induced by infecting mice intraperitoneally with 3.5 ϫ 10 6 to 7.4 ϫ 10 8 CFU of bacteria in PBS with a 5% bacteriological mucin (VWE Scientific, Pittsburg, PA) suspension. The inoculum represented approximately 100-fold the 50% lethal dose (LD 50 ). At 1 h postinfection, the mice were treated with a single dose of sarecycline, doxycycline, and/or minocycline administered subcutaneously at doses ranging from 0.01 to 0.5 mg/kg in a vehicle of sterile water. All drug doses were adjusted to account for the percentage of the active moiety.
For the E. coli infection, the bacterial culture grew overnight in Mueller-Hinton broth to approximately 2 ϫ 10 9 CFU/ml. Serial dilutions of the bacterial suspension were performed in PBS to obtain the infectious dose. An inoculum of 6.5 ϫ 10 5 to 1.6 ϫ 10 7 CFU was inoculated intraperitoneally and represented approximately 100 times the 50% lethal dose (LD 50 ). At 1 h postinfection, the animals were treated with sarecycline, doxycycline, or minocycline subcutaneously at doses ranging from 0.5 to 40 mg/kg in a vehicle of sterile water. All drugs doses were adjusted to account for the percentage of the active moiety. Sarecycline was tested at a high dose of 40 mg/kg to evaluate efficacy against an enteric Gram-negative organism, such as E. coli. For both organisms, the 50% protective dose (PD 50 ) was assessed at 48 h postinfection. The PD 50 was defined as the dose required to achieve 50% survival.
A murine neutropenic thigh wound infection model was also utilized. Female SD-1 mice were rendered neutropenic by injecting cyclophosphamide (Sigma-Aldrich, St. Louis, MO) at 150 and 100 mg/kg on days Ϫ4 and Ϫ1 before infection, respectively. Severe neutropenia (Ͻ100 neutrophils/ mm 3 ) developed by day 0, when the infection studies were initiated. Before infecting the mice, S. aureus RN450-1 from a frozen stock was cultured overnight. By following the same inoculum preparation procedures described above, 1 ϫ 10 5 CFU/mouse of S. aureus RN450-1 was injected intramuscularly into the left thigh. At 2 and 6 h postinfection, sarecycline or doxycycline therapy was administered intravenously to each mouse at a dose of 0.33, 1, 3, or 9 mg/kg in a vehicle of sterile water. At 24 h, the thighs were removed and S. aureus RN450-1 was cultured in Trypticase soy agar (TSA) II plates with 5% sheep blood (Northeast Labs, Waterville, ME) from thigh tissue homogenate. The bacterial burden was assessed to determine the 50% effective dose (ED 50 ), defined as the dose required to achieve a 2-log 10 reduction in the bacterial burden compared with that for the nontreated control. The genotype information for the S. aureus RN450-1 and E. coli 1478 strains tested in these studies was not available.
Resistance development. Single-step resistance development studies were conducted in vitro. C. acnes was grown on brucella agar for 48 h under anaerobic conditions and suspended in brucella broth. For C. acnes, 4 different isolates, including three clinical isolates and one American Type Culture