Impact of β-Lactamase Inhibition on the Activity of Ceftaroline against Mycobacterium tuberculosis and Mycobacterium abscessus

Abstract

The production of β-lactamases Bla_Mab and BlaC contributes to β-lactam resistance in Mycobacterium abscessus and Mycobacterium tuberculosis, respectively. Ceftaroline was efficiently hydrolyzed by these enzymes. Inhibition of M. tuberculosis BlaC by clavulanate decreased the ceftaroline MIC from ≥256 to 16 to 64 μg/ml, but these values are clinically irrelevant. In contrast, the ceftaroline-avibactam combination should be evaluated against M. abscessus since it inhibited growth at lower and potentially achievable drug concentrations.

TEXT

Mycobacterium tuberculosis is responsible for more than 1 million deaths annually, and the emergence of multidrug-resistant strains constitutes a serious threat for the control of the tuberculosis pandemic (1). In this context, there is an urgent need for new antituberculosis drugs. Mycobacterium abscessus raises radically different issues (2). In recent years, this fast-growing mycobacterium has emerged as an important opportunistic pathogen responsible for lung infections, especially in cystic fibrosis patients (3, 4). It also causes epidemics of skin and soft tissue infections, mainly after iatrogenic or accidental inoculation (5, 6). M. abscessus is naturally highly resistant to most antibiotics (7), including antituberculosis drugs. The recommended treatment of pulmonary infections comprises a combination of clarithromycin, amikacin, and an intravenous β-lactam, either cefoxitin or imipenem, for at least 12 months (2). This treatment achieves microbiological cure (sputum conversion without relapse) in ca. 60% of cases (8, 9). However, the cure rate is only 25 to 40% in case of clarithromycin resistance mediated by an Erm rRNA methylase (9–11), which is produced by 40 to 60% of isolates (11).

Mycobacteria are intrinsically resistant to most β-lactams due to the production of broad-spectrum β-lactamases in combination with low affinity of the drug targets and limited permeability of the cell wall (12–14). The former mechanism can be counteracted by β-lactamase inhibitors, which display various efficiencies for enzymes from different species (15, 16). Clavulanate irreversibly inactivates the Ambler class A β-lactamase BlaC from M. tuberculosis (17), whereas Bla_Mab from M. abscessus hydrolyzes this drug (18). We have recently shown that avibactam, a non-β-lactam β-lactamase inhibitor, efficiently inhibits Bla_Mab and extends the spectrum of β-lactams active against M. abscessus (19).

Ceftaroline is a novel broad-spectrum cephalosporin with activity against bacteria producing low-affinity penicillin-binding proteins (PBPs), including methicillin-resistant Staphylococcus aureus (20) and cephalosporin-resistant Streptococcus pneumoniae (21). Ceftaroline fosamil, a ceftaroline prodrug, has recently been approved for the treatment of skin and soft tissue infection and community-acquired pulmonary infections. In this study, we investigate the in vitro activity of ceftaroline in combination with clavulanate or avibactam against β-lactamase-producing or β-lactamase-deficient strains of M. tuberculosis and M. abscessus.

(This work was presented in part at the 54th Interscience Conference on Antimicrobial Agents and Chemotherapy [22].)

To assess β-lactamase activity, BlaC and Bla_Mab were purified and kinetic parameters for ceftaroline hydrolysis were determined by spectrophotometry (Δε of 6,300 M⁻¹ cm⁻¹ at 306 nm), as previously described (18). Both enzymes hydrolyzed ceftaroline with relatively high catalytic efficiencies (k_cat/K_m = 5.5 × 10⁴ M⁻¹ s⁻¹ and 1.6 × 10⁴ M⁻¹ s⁻¹, respectively) in comparison to those of other cephalosporins (Table 1). This observation suggests that being combined with β-lactamase inhibitors could increase the antibacterial activity of ceftaroline.

TABLE 1.

Kinetic parameters for hydrolysis of β-lactams by BlaC and Bla_Mab^a

β-Lactam	BlaC			BlaMab
	Km (μM)	kcat (s−1)	kcat/Km (s−1 M−1)	Km (μM)	kcat (s−1)	kcat/Km (s−1 M−1)
Amoxicillin	44 ± 6	5.8 ± 0.2	1.3 × 105	890 ± 200	780 ± 50	8.8 × 105
Nitrocefin	71 ± 11	35 ± 2	4.9 × 105	24 ± 7.0	1000 ± 70	4.3 × 107
Cephalothin	260 ± 50	23 ± 2	8.8 × 104	17 ± 1.0	6.7 ± 0.1	4.1 × 105
Ceftriaxone	>200	>0.06	1.7 × 102	>350	>0.3	5.0 × 102
Cefoxitin	140 ± 30	1.1 ± 0.1	7.9 × 103	500 ± 270	0.0003 ± 0.0001	6.7 × 100
Ceftaroline	>300	>4.5	1.6 × 104	>400	>22	5.5 × 104
Imipenem	140 ± 30	0.13 ± 0.02	8.3 × 102	90 ± 40	2.7 ± 0.3	3.0 × 104
Meropenem	1.0 ± 0.2	0.0003 ± 0.00002	3.3 × 102	120 ± 20	1.8 ± 0.2	1.5 × 104

^aData for β-lactams other than ceftaroline are from reference 18. K_m and k_cat data include the plus-or-minus standard error of the best fit.

Ceftaroline antituberculosis activity was assessed using the alamarBlue test (23). Briefly, M. tuberculosis strains H37Rv and CDC1551 were grown to the exponential phase (optical density of 1 at 600 nm) in Middlebrook 7H9 medium (Difco) supplemented with 0.05% (vol/vol) Tween 80, 0.2% (vol/vol) glycerol, 0.085% (wt/vol) NaCl, 0.2% (wt/vol) dextrose, and 0.5% (wt/vol) bovine serum albumin (24). The bacterial suspension was diluted 200-fold in the same medium and used to inoculate sterile round-bottom 96-well plates containing serial 2-fold dilutions of ceftaroline with or without 2.5 μg/ml clavulanate. The total volume of each well was 200 μl. After 6 days of incubation at 37°C, 20 μl of a sterile 0.025% (wt/vol) resazurin solution was added to each well, and the plates were incubated overnight. The MIC was defined as the lowest drug concentration preventing the resazurin color change from blue to pink or violet. Ceftaroline alone did not inhibit the growth of β-lactamase-producing M. tuberculosis at clinically relevant concentrations (MIC, ≥256 μg/ml) (Table 2). The addition of clavulanate or deletion of the blaC gene (12) decreased the ceftaroline MIC to 16 to 64 μg/ml, indicating that the production of BlaC limits the antibacterial activity of ceftaroline.

TABLE 2.

Activity of ceftaroline against Mycobacterium tuberculosis

Day(s) of dosinga	Strain	Ceftaroline MIC (μg/ml)b
		Without Clav	With Clav
0	H37Rv	256	32
	H37Rv ΔblaC	32	16
	CDC1551	≥256	64
0 and 3	H37Rv	256	32
	H37Rv ΔblaC	16	16
	CDC1551	≥256	48

^aCeftaroline was dosed once at the beginning of the experiment (day 0) or additionally at day 3 (days 0 and 3) to compensate for spontaneous drug hydrolysis.

^bData are the median results from at least three independent experiments. MICs were determined in the absence or presence of clavulanic acid (Clav) at 2.5 μg/ml.

Since the MIC determinations required extended incubation of slow-growing M. tuberculosis, spontaneous hydrolysis of ceftaroline was assessed in the culture medium by spectrophotometry, revealing a half-life of 4.8 days at 37°C (data not shown). Compensating for spontaneous drug hydrolysis by the addition of ceftaroline after 4 days of incubation (Table 2) showed that the low activity of ceftaroline was not the consequence of its chemical instability. Together, these results indicate that ceftaroline is of little interest for the treatment of tuberculosis since the maximal drug concentration achieved in plasma (21 μg/ml) for an intravenous injection of 600 mg (25) does not exceed the MICs observed in the absence of a functional β-lactamase (16 to 64 μg/ml).

The in vitro activity of ceftaroline was also assessed against M. abscessus strain CIP104536 and its β-lactamase-deficient (Δbla_Mab) derivative (19) (Table 3). The MICs of ceftaroline and the reference β-lactams cefoxitin and imipenem were determined using the microdilution method in cation-adjusted Mueller-Hinton (CaMH) broth, as recommended by the Clinical and Laboratory Standards Institute (26). As previously described (19), avibactam had no impact on the MICs of cefoxitin and imipenem (Table 3). Ceftaroline alone was not active against the wild-type strain (MIC, >256 μg/ml) but inhibited the growth of the Δbla_Mab derivative at 16 μg/ml. For the wild-type strain, the inhibition of Bla_Mab by avibactam (4 μg/ml) led to a MIC of ceftaroline of 32 μg/ml. Thus, ceftaroline was as active as the reference β-lactam cefoxitin in the absence of functional Bla_Mab.

TABLE 3.

Activities of β-lactams against M. abscessus

Growth mediuma	Strainb	MIC (μg/ml)c
		Cefoxitin	Imipenem	Ceftaroline
CaMH	CIP104536	32	8	>256
CaMH + Avi	CIP104536	32	8	32
CaMH	ΔblaMab strain	32	4	16
7H9sB	CIP104536	16	4	8
7H9sB + Avi	CIP104536	8	2	1
7H9sB	ΔblaMab strain	8	2	2

^aCaMH, cation-adjusted Mueller-Hinton broth; 7H9sB, Middlebrook 7H9 medium supplemented with 0.05% (vol/vol) Tween 80 and 10% (vol/vol) oleic acid, albumin, dextrose, and catalase; + Avi, the medium was supplemented with 4 μg/ml of avibactam.

^bΔbla_Mab strain, β-lactamase-deficient derivative of CIP104536 obtained by deletion of gene bla_Mab.

^cData are the median results from at least three independent experiments. The breakpoints recommended by CSLI for cefoxitin and imipenem MICs determined in CaMH medium are as follows. Cefoxitin: susceptible, ≤16 μg/ml; intermediate, 32 to 64 μg/ml; resistant, ≥128 μg/ml. Imipenem: susceptible, ≤4 μg/ml; intermediate, 8 to 16 μg/ml; resistant, ≥32 μg/ml.

We have previously compared the in vitro activities of β-lactams against M. abscessus in the reference medium, CaMH, and in Middlebrook 7H9 medium supplemented with 0.05% (vol/vol) Tween 80 and 10% (vol/vol) oleic acid, albumin, dextrose, and catalase (OADC; Becton-Dickinson). The latter medium (7H9sB) supports faster growth of M. abscessus and provides lower estimates of the MICs of imipenem and, to a lesser extent, cefoxitin, as determined for a collection of clinical isolates (27). Avibactam is also less active in CaMH than in 7H9sB medium (19). In the current study, the MICs of ceftaroline were >32- and 8-fold higher in CaMH than in 7H9sB medium against M. abscessus CIP104536 and its Δbla_Mab derivative, respectively (Table 3). These differences might result from membrane permeabilization by Tween 80, which is only present in the 7H9sB medium (27). Breakpoints have been defined by CLSI for cefoxitin and imipenem (Table 3, footnote c), but the predictive value of these breakpoints for in vivo efficacy remains poorly evaluated.

In the absence of functional Bla_Mab following deletion of bla_Mab or inhibition of the enzyme by avibactam, the MICs of ceftaroline were intermediary between those of cefoxitin and imipenem in CaMH medium and similar to those of the most active β-lactam (imipenem) in 7H9sB medium (Table 3). The activity of ceftaroline alone or in combination with avibactam was also evaluated in 7H9sB medium against 10 unrelated clinical isolates of M. abscessus subsp. abscessus and M. abscessus subsp. bolletii, mostly from cystic fibrosis patients (27, 28) (Table 4). In the absence of avibactam, the MICs of ceftaroline ranged from 2 to 64 μg/ml. In the presence of 4 μg/ml of avibactam, the MICs of ceftaroline decreased to 0.5 to 2 μg/ml. These observations indicate that the combination of avibactam and ceftaroline should be evaluated for the treatment of pulmonary infections due to clarithromycin-resistant M. abscessus, which are associated with therapeutic failures.

TABLE 4.

MICs of ceftaroline alone or in combination with avibactam determined in 7H9sB^a against clinical isolates of M. abscessus subsp. abscessus and M. abscessus subsp. bolletii

Isolate	Subspecies	Colony morphotypeb	MIC (μg/ml)c
			Ceftaroline	Ceftaroline + avibactam
1	abscessus	S	4	0.5
2	abscessus	R	64	2
3	abscessus	S	4	0.5
4	abscessus	S	4	1
5	abscessus	S	4	0.5
6	abscessus	R	16	1
7	abscessus	R	8	2
8	abscessus	R	16	2
9	bolletii	S	2	1
10	bolletii	R	8	2

^a7H9sB, Middlebrook 7H9 medium supplemented with 0.05% (vol/vol) Tween 80 and 10% (vol/vol) oleic acid, albumin, dextrose, and catalase.

^bS, smooth morphotype; R, rough morphotype.

^cData are the median results from three independent experiments. Avibactam was used at 4 μg/ml.