|Home | About | Journals | Submit | Contact Us | Français|
Cefiderocol (CFDC; S-649266), a novel parenteral siderophore cephalosporin conjugated with a catechol moiety, has a characteristic antibacterial spectrum with a potent activity against a broad range of aerobic Gram-negative bacterial species, including carbapenem-resistant strains of Enterobacteriaceae and nonfermenting bacteria such as Pseudomonas aeruginosa and Acinetobacter baumannii. Cefiderocol has affinity mainly for penicillin-binding protein 3 (PBP3) of Enterobacteriaceae and nonfermenting bacteria similar to that of ceftazidime. A deficiency of the iron transporter PiuA in P. aeruginosa or both CirA and Fiu in Escherichia coli caused 16-fold increases in cefiderocol MICs, suggesting that these iron transporters contribute to the permeation of cefiderocol across the outer membrane. The deficiency of OmpK35/36 in Klebsiella pneumoniae and the overproduction of efflux pump MexA-MexB-OprM in P. aeruginosa showed no significant impact on the activity of cefiderocol.
Nosocomial infections caused by Gram-negative bacteria are increasingly difficult to treat due to the global spread of multidrug-resistant (MDR) strains which are resistant to several antibiotics, such as carbapenems, cephalosporins, aminoglycosides, and quinolones (1). The WHO has listed carbapenem-resistant Enterobacteriaceae (CRE), carbapenem-resistant Pseudomonas aeruginosa, and carbapenem-resistant Acinetobacter baumannii as the pathogens against which urgent development of new antibiotics are needed, since the emergence of these resistant pathogens poses serious public health issues due to the limited number of treatment options (2, 3).
Since the 1980s, numerous attempts to conjugate iron-binding functional groups onto β-lactams have been made to hijack the iron uptake systems of Gram-negative bacteria and circumvent the outer membrane barriers (4,–6). However, none of the molecules have been approved for clinical use for various reasons, such as a lack of correlation between in vitro and in vivo efficacies (7,–9). Cefiderocol (CFDC; S-649266), a novel catechol-substituted siderophore cephalosporin, is structurally different from other recently developed hydroxypyridone-substituted siderophore monobactam antibiotics such as BAL30072, MB-1, and MC-1 and has been reported to have potent antibacterial activity against MDR Gram-negative pathogens, including carbapenem-resistant strains of Enterobacteriaceae, P. aeruginosa, and A. baumannii, as well as potent in vivo efficacy against multiple clinical strains of Gram-negative bacteria in mouse lung infection models (10,–14, 34; I. Ghazi, M. L. Monogue, M. Tsuji, and D. P. Nicolau, submitted for publication). This is the first report evaluating the in vitro features of cefiderocol, including its antibacterial spectrum against Gram-negative and Gram-positive bacteria and its mode of action, such as penicillin-binding protein (PBP) binding affinity and morphological changes, as well as the impacts of various β-lactamases, efflux pump overexpression, and deficiency of porin or iron transporter on the in vitro activity.
The MICs of cefiderocol were ≤2 μg/ml against a broad range of Gram-negative bacterial strains, including Enterobacteriaceae such as Enterobacter spp., Escherichia coli, Klebsiella spp., Proteus spp., Providencia spp., Salmonella spp., and Yersinia spp., as well as Vibrio species (Table 1). Cefiderocol showed in vitro activity against nonfermenting bacteria such as Acinetobacter spp., Pseudomonas spp., and Burkholderia spp.; cefiderocol also showed in vitro activity against the intrinsically MDR bacteria of Stenotrophomonas maltophilia and Elizabethkingia meningoseptica, as well as the causative pathogens for respiratory tract infections, such as Haemophilus spp., Moraxella catarrhalis, and Bordetella parapertussis. On the other hand, the MICs of cefiderocol against Campylobacter jejuni ATCC 33560 and 794009 as well as ceftriaxone-resistant Neisseria gonorrhoeae 868339 were relatively high (MIC, >4 μg/ml) for the Gram-negative bacteria tested, although the MICs of cefiderocol against two other N. gonorrhoeae strains, including an azithromycin-resistant strain, were ≤0.5 μg/ml. The MICs of cefiderocol against aerobic Gram-positive bacteria growing in an aerobic or microaerophilic atmosphere were ≥4 μg/ml, except for those against Streptococcus pneumoniae ATCC 49619, Streptococcus pyogenes ATCC 10389, and Micrococcus luteus ATCC 9341, which were 2, 1, and 4 μg/ml, respectively. The activities against these strains of Gram-positive bacteria were weaker than those of other tested β-lactam compounds. The MICs of cefiderocol against anaerobic Gram-negative and Gram-positive bacteria showed variation within genera and were higher than those of cefepime or meropenem, with cefiderocol MICs of 0.5 to >32 μg/ml, except for that against Fusobacterium necrophorum, which was ≤0.031 μg/ml (Table 2). Although cefiderocol showed activity against some ATCC strains of Bacteroides spp., Prevotella spp., and Clostridium spp., with MICs of 1 to 2 μg/ml, cefiderocol did not show potent activity against multiple clinical isolates of Bacteroides spp., Prevotella spp., or Clostridium difficile, of which the MIC50s were 32 μg/ml or higher (Table 3).
Cefiderocol exhibited potent in vitro activity against 33 strains of Gram-negative bacteria harboring various kinds of β-lactamases, including extended-spectrum β-lactamases (ESBLs) and carbapenemases (Table 4). The MICs of cefiderocol were 8 μg/ml or lower against all the test strains, including carbapenemase producers such as Klebsiella pneumoniae harboring NDM-1, KPC, or GES-4, P. aeruginosa harboring VIM or IMP-1, and A. baumannii harboring OXA-23 and/or OXA-51-like or OXA-58. Other tested antibiotics, including various classes of β-lactams, amikacin, and ciprofloxacin, showed less activity against most of these carbapenemase producers, with MICs of 16 μg/ml or more, while colistin showed MICs of 1 μg/ml or lower.
The affinities (50% inhibitory concentrations [IC50s]) of cefiderocol for PBPs of E. coli NIHJ JC-2, K. pneumoniae SR22291, P. aeruginosa ATCC 27853, and A. baumannii ATCC 17978 were determined (Table 5). The IC50s of cefiderocol against PBP3 of E. coli NIHJ JC-2, K. pneumoniae SR22291, P. aeruginosa ATCC 27853, and A. baumannii ATCC 17978 were 0.04, 0.062, 0.06, and 0.67 μg/ml, respectively, which were lower than those of ceftazidime, indicating a higher affinity of cefiderocol for PBP3 than that of ceftazidime. Other than its affinity for PBP3, cefiderocol had an affinity for PBP2 of K. pneumoniae SR22291 as ceftazidime did (IC50s of cefiderocol and ceftazidime were 0.063 and 0.41 μg/ml, respectively), and cefiderocol also had an affinity for PBP1a of P. aeruginosa ATCC 27853 as ceftazidime did (IC50s of cefiderocol and ceftazidime were 0.85 and 3.62 μg/ml, respectively). Morphological changes of these four bacteria were examined by phase-contrast microscopy after exposure to cefiderocol (see Fig. S1 and S2 in the supplemental material). Filamentous cells were observed in all the test strains after exposure to cefiderocol, similar to what was observed after exposure to ceftazidime. In the time-kill study with the four strains E. coli NIHJ JC-2, K. pneumoniae SR22291, P. aeruginosa ATCC 27853, and A. baumannii ATCC 17978, cefiderocol reduced the bacterial counts in a manner similar to that of ceftazidime after exposure at 1, 4, or 16 times the MIC, and the killing rates were similar between 4 times and 16 times the MIC (Fig. 1 and and22).
The effects of the deficiency of iron transporters were examined using P. aeruginosa and E. coli (Table 6 and and7).7). The MICs of cefiderocol against all the test P. aeruginosa PAO1 derivative mutant strains which have a transposon (Tn) insertion in the genes of iron transporters, including major siderophore receptors for pyoverdine (fpvA and fpvB), pyochelin (fptA), and enterobactin (pirA), ranged from 0.063 to 0.125 μg/ml, equivalent to the MIC against the parent strain PAO1, with the exception of the MICs against the strains having a Tn insertion in the probable iron transport receptor gene piuA (Table 6). The MIC of cefiderocol increased to 2 μg/ml (PW8599) after Tn insertion into piuA, which was complemented by the introduction of wild-type PiuA (SR-L00252). MICs of ceftazidime ranged from 1 to 2 μg/ml against all the tested strains with a Tn insertion in iron transporter-related genes as well as the parent strain PAO1. The MIC of cefiderocol against E. coli BW25113 with deletion of iron transporter gene cirA or fiu was 0.063 or 0.125 μg/ml, within 2-fold of that against the parent strain, whereas the MIC of cefiderocol increased 16-fold by the double knockout of cirA and fiu (Table 7).
The MICs of cefiderocol against PAO1 derivative mutant strains which have a Tn insertion in the genes of multidrug efflux pump MexAB-OprM, its transcriptional regulator, and porin OprD, which are involved in β-lactam resistance, were examined (Table 6). MICs of aztreonam against the strains with a Tn insertion in either mexB (PW1781) or oprM (PW1783), which lost the function of the MexAB-OprM efflux pump, were 16-fold lower than that against PAO1, while the decreases in cefiderocol MICs due to a Tn insertion were 2- or 4-fold. The MIC of cefiderocol was also determined against PW1781 and PW1783 in cation-adjusted Mueller-Hinton broth (CAMHB), which contains ferric iron, and the cefiderocol MICs were 0.125 and 0.063 μg/ml, respectively, which were 2- or 4-fold lower, respectively, than that against PAO1, which was 0.25 in CAMHB. The MICs of ceftazidime, aztreonam, and ciprofloxacin against the strains with a Tn insertion in either mexR (PW1776) or nalD (PW7066), which leads to overexpression of the MexAB-OprM efflux pump, were 4-fold higher than that against PAO1, while the increases in cefiderocol MICs due to a Tn insertion were within 2-fold. The MIC of cefiderocol was also determined against PW1776 and PW7066 in CAMHB, and the increases in cefiderocol MICs were also 2-fold compared to that against PAO1. Against the strain with a Tn insertion in oprD, imipenem showed an 8-fold higher MIC than against the parent strain, while the increases in MIC of cefiderocol and other test antibiotics were within 2-fold.
The MIC of cefiderocol was determined against K. pneumoniae NVT2001S and its derivative mutant strains, which were deficient in porin ompK35 and/or ompK36, which are involved in resistance to various classes of β-lactam antibiotics, including carbapenems (Table 7). The MIC of meropenem against the double deletion mutant strain was 8-fold higher than that against the parental strain NVT2001S, whereas the increases in MICs of cefiderocol and ceftazidime against strains with deletion of ompK35 and/or ompK36 were 2- to 4-fold that against the parental strain.
In comparison to carbapenem antibiotics, cefiderocol has more potent in vitro activity against a broad range of Gram-negative bacteria, including carbapenem-resistant strains of Enterobacteriaceae and nonfermenting bacteria that produce carbapenemases as well as ESBLs and class C β-lactamases, but weaker activity against Gram-positive bacteria and anaerobic bacteria. Cefiderocol has an antibacterial spectrum that is significantly different from that of carbapenem antibiotics, which have been mainly used for the treatment of Gram-negative bacterial infections. The reason for this unique antibacterial spectrum is the active ferric iron uptake by the siderophore, which is observed only in aerobic Gram-negative bacteria. The detailed mode of action on the activity against some of the strains or species among Gram-positives and anaerobic bacteria has not been sufficiently observed. The global surveillance studies of cefiderocol, SIDERO-WT-2014, using recent clinical isolates (n = 9,205) against Enterobacteriaceae and nonfermenting bacteria such as P. aeruginosa, A. baumannii, and S. maltophilia, including MDR strains, and other studies to evaluate the antibacterial activity of cefiderocol against well-characterized carbapenem-resistant Gram-negative pathogens have shown that cefiderocol also has potent activity against these problematic Gram-negative pathogens (10,–12).
This study revealed that the antibacterial action of cefiderocol is to inhibit mainly PBP3 of Enterobacteriaceae and nonfermenting bacteria, resulting in morphological changes of filamentous cells, similar to the action of ceftazidime. Although the impact of the differences in PBP3 affinity on the in vitro activities of cefiderocol and ceftazidime is not clear, the antibacterial activity of cefiderocol determined by the time-kill experiment was similar to that of ceftazidime. The key features of cefiderocol are its active uptake mechanisms by Gram-negative bacteria under iron-depleted conditions and its improved stability to various types of β-lactamases reported previously (15). In this study, we showed that PiuA is one of the iron transporters of P. aeruginosa that is responsible for the active transport of cefiderocol into bacterial cells resulting in the activity of cefiderocol, as is the case for MC-1 (16). However, the MIC results also showed that PirA, which was reported to be one of the iron transporters responsible for the activity of MC-1 (16), did not contribute to the in vitro activity of cefiderocol. Moreover, this study demonstrated the contribution of both cirA and fiu of E. coli, which has been reported to be involved in the monomeric catechols and whose production is regulated by the availability of ferric iron, to the in vitro activity of cefiderocol (17).
The MIC results with mutant strains showed that the effect of the deficiency of the porin OmpK35/36 of K. pneumoniae, which is reported to be one of the resistance determinants coordinated with various β-lactamases against carbapenems (18, 19), on the activity of cefiderocol was not significant, which may also contribute to the potent activity of cefiderocol against such carbapenem-resistant K. pneumoniae strains. In terms of the efflux pump MexAB-OprM of P. aeruginosa, the decrease in the MIC of cefiderocol against the strains with a deficient efflux pump indicates that cefiderocol could be a substrate for the efflux pump MexAB-OprM. However, the overproduction of the efflux pump increased the cefiderocol MIC only slightly, even under the condition with ferric iron in the medium, in which the function of active transport for cefiderocol is weak; this indicates that the effect of the overproduction on the activity of cefiderocol should be limited and that cefiderocol is not taken into bacterial cells faster than it is extruded from the bacterial cells by the efflux pump. On the other hand, it has been reported that BAL30072 is a substrate for the efflux pumps MexAB-OprM and MexEF-OprN (20) and that MC-1 is a substrate for the efflux pump MexAB-OprM (16). Those reports suggest that cefiderocol has different profiles of transport into P. aeruginosa with other siderophore-conjugated β-lactams. These studies are limited in clarifying and understanding the differences in the mechanisms of action of cefiderocol and other siderophore-conjugated β-lactams, and further detailed studies are required.
In summary, this study showed that cefiderocol has potent in vitro activity against a broad range of aerobic Gram-negative bacteria, including MDR strains, and that the antibacterial activity of cefiderocol is based mainly on the inhibition of PBP3. This study also revealed that iron transporters such as PiuA of P. aeruginosa and CirA and Fiu of E. coli are involved in the permeation of cefiderocol into bacterial cells. The characteristics of cefiderocol indicate that cefiderocol may be a promising option for the treatment of infections caused by a broad range of Gram-negative pathogens, including MDR Enterobacteriaceae and MDR nonfermenting bacteria.
A number of type strains were obtained from the American Type Culture Collection (Manassa, VA), the National Collection of Type Cultures (Salisbury, United Kingdom), and the National Institute of Technology and Evaluation Biological Resource Center (Tokyo, Japan). A number of clinical strains were kindly provided by the Bicêtre Hospital (Le Kremlin-Bicêtre, France), Singapore General Hospital (Singapore), and GlaxoSmithKline plc (Middlesex, United Kingdom). Other test strains were obtained from various hospitals, mainly in Japan. Species-appropriate quality control (QC) strains, which were obtained from the ATCC, were used as described in Clinical and Laboratory Standards Institute (CLSI) guidelines (21,–24). Transposon (Tn) insertion mutant strains of P. aeruginosa PAO1 were kindly provided by the University of Washington (25). A PAO1 derivative of SR-L00016 was constructed from PW8599 by deletion of the pirA gene according to the method described by Alexeyev et al. and Schweizer et al. (26, 27). The PiuA expression plasmid was constructed by using an In-Fusion HD cloning kit (TaKaRa Bio, Inc., Shiga, Japan) with pMMB67HE-Gm, and PW8599 was transformed with the plasmid to obtain SR-L00252. The cirA and/or fiu deletion mutants of E. coli BW25113 were constructed according to the methods described by Datsenko et al. (28). Detailed procedures for constructing the plasmid and recombinant strains are described in the supplemental material (see Method S1). K. pneumoniae NVT2001S and its ompK35 and/or ompK36 deletion mutants were kindly provided by the National Health Research Institutes in Taiwan.
Cefiderocol, ceftolozane, and avibactam were synthesized at the research laboratories of Shionogi & Co., Ltd. (Osaka, Japan). Commercial-grade antibiotics were obtained as follows: ceftazidime, tazobactam, amikacin, and aztreonam from Chem-Impex International, Inc. (Wood Dale, IL); cefepime and metronidazole from U.S. Pharmacopeia (Rockville, MD); meropenem, colistin, and gentamicin from Wako Pure Chemical Industries, Ltd. (Osaka, Japan); and ciprofloxacin and piperacillin from LKT Laboratories, Inc. (St. Paul, MN).
MICs were determined by using broth microdilution or agar dilution according to the CLSI (21,–24), except for the MIC for Bordetella parapertussis, which was determined by the method described by Mortensen and Rodgers (29). For the determination of cefiderocol MIC, iron-depleted cation-adjusted Mueller-Hinton broth (ID-CAMHB) was prepared as previously described and used according to the recommendations of the CLSI (30), except for the cases that are required to determine MICs under specific conditions (Method S2). The quality control MIC ranges of cefiderocol approved by the CLSI were 0.06 to 0.5 μg/ml for both E. coli ATCC 25922 and P. aeruginosa ATCC 27853 (31). For anaerobic bacteria, brucella agar (Becton, Dickinson and Company, NJ) supplemented with hemin, vitamin K1, and laked sheep blood was used. For recombinant strains, test medium was supplemented with 10 μg/ml of gentamicin and/or 0.1 mM IPTG (isopropyl-β-d-thiogalactopyranoside) (Wako Pure Chemical Industries) when required.
The affinities of cefiderocol and ceftazidime for PBPs of E. coli NHIJ JC-2, K. pneumoniae SR22291, and P. aeruginosa ATCC 27853 were determined by using benzylpenicillin [benzyl-14C]potassium (American Radiolabeled Chemicals, Inc., St. Louis, MO) as described by Spratt (32). For A. baumannii ATCC 17978, the method using Bocillin FL penicillin sodium salt (Life Technologies, Inc., Carlsbad, CA) reported by Vashist et al. (33) was used. Detailed procedures are described in the supplemental material (Method S3).
A bacterial suspension of the log-phase bacteria was smeared on compound-containing thin-layer Mueller-Hinton agar (Becton, Dickinson and Company, NJ) coated on a glass slide. After incubation at 35°C for 4 to 6 h, morphological changes of bacterial cells were observed with Leica DM2500 microscopy (Leica, Germany). Detailed procedures are described in the supplemental material (Method S4).
An overnight culture of the test strain was diluted into fresh medium to yield an inoculum of approximately 106 CFU/ml. The ID-CAMHB and CAMHB media were used for cefiderocol and ceftazidime, respectively. Concentrations of antibiotics were 0 (control), 0.25, 1, 4, or 16 times the MIC. Incubation was performed at 35°C, and the sampling times were 2, 4, and 6 h after initiation of incubation. MICs against E. coli NIHJ JC-2, K. pneumoniae SR22291, P. aeruginosa ATCC 27853, and A. baumannii ATCC 17978 were 0.25, 0.008, 0.063 and 0.016 μg/ml, respectively, for cefiderocol (ID-CAMHB) and 0.25, 0.063, 1 and 4 μg/ml, respectively, for ceftazidime (CAMHB).
We thank Colin Manoil at the University of Washington and Koh Tse Hsien at Singapore General Hospital for providing strains. We thank A. Naito and T. Yamaguchi at Shionogi & Co., Ltd., and Yutaka Jinushi, Keiichiro Hirooka, Hayato Matsuda, Ryuichiro Nakai, Toshihiko Hori, Saya Nishimori, and Makoto Iwasaki at Shionogi TechnoAdvance Research Co., Ltd., for their experimental advice and support.
A. Ito, T. Sato, M. Ota, M. Takemura, T. Nishikawa, S. Toba, N. Kohira, S. Miyagawa, N. Ishibashi, S. Matsumoto, R, Nakamura, M. Tsuji, and Y. Yamano are employees of Shionogi and have no conflicts to declare.
Supplemental material for this article may be found at https://doi.org/10.1128/AAC.01454-17.