Orally Administered Recombinant Metallo-β-Lactamase Preserves Colonization Resistance of Piperacillin-Tazobactam-Treated Mice

Antibiotics that are excreted into the intestinal tract may disrupt colonization resistance (i.e., the ability of the indigenous microflora to provide resistance against colonization by potentially pathogenic microorganisms) (1, 5). Because antibiotic activity within the intestinal lumenis is not needed for the treatment of most infections, we hypothesized that the enzymatic inactivation of the portion of antibiotic that is excreted into the intestinal tract would result in the preservation of colonization resistance during parenteral antibiotic therapy. We previously demonstrated that an orally administered recombinant class A β-lactamase preserved the colonization resistance of piperacillin-treated mice against nosocomial pathogens (4). Here, we show that a recombinant class B metallo-β-lactamase that is resistant to tazobactam inactivation preserves colonization resistance during treatment with the more widely used broad-spectrum antibiotic piperacillin-tazobactam.

Targeted recombinant β-lactamase 2 (TRBL-2) containing amino acid residues 31 to 257 of the metalloenzyme of a clinical Bacillus cereus isolate (98ME 1552 from Helsinki University Hospital) was overproduced in Bacillus subtilis using a bacillar secretion vector (4). The hydrolysis rate of TRBL-2 for piperacillin with or without tazobactam was 275 μg/minute/μg of enzyme. Individually housed female CF-1 mice weighing 25 to 30 g (Harlan Sprague-Dawley, Indianapolis, Indiana) were used to examine the efficacy of TRBL-2 in the preservation of colonization resistance against vancomycin-resistant enterococci (VRE). At 24 h and again at 12 h prior to the orogastric inoculation of 10⁴ CFU of VRE strain C68 (4), mice received either subcutaneous (0.2 ml) and orogastric (0.5 ml) phosphate-buffered saline (PBS), subcutaneous piperacillin-tazobactam (4 mg) and orogastric PBS, subcutaneous piperacillin-tazobactam (4 mg) and orogastric TRBL-2 (60 mg/kg of body weight), or subcutaneous piperacillin-tazobactam (4 mg) and orogastric TRBL-2 (60 mg/kg) that had been inactivated by boiling for 10 min. Stool VRE density was monitored as previously described (2). Denaturing gradient gel electrophoresis (DGGE) of PCR-amplified bacterial rRNA genes was performed on stool samples collected 3 days after the inoculation of VRE in order to monitor changes in the indigenous microflora (3, 4). One-way analysis of variance was performed to compare VRE densities among treatment groups, with P values adjusted for multiple comparisons using the Scheffé correction. DGGE similarity indices were compared using Student's t test. Computations were performed using Stata software (version 5.0; Stata, College Station, Texas).

Mice treated with piperacillin-tazobactam developed high-density VRE stool colonization; saline controls and mice treated with piperacillin-tazobactam in conjunction with TRBL-2 did not (P < 0.001) (Fig. 1). The protective effect of TRBL-2 was eliminated by heat inactivation. DGGE analysis showed that piperacillin-tazobactam caused a significant disruption of the indigenous microflora but that piperacillin-tazobactam in conjunction with TRBL-2 caused a relatively minor alteration of the microflora (mean similarity indices of mice treated with piperacillin-tazobactam and mice treated with piperacillin-tazobactam and TRBL-2 in comparison to those of saline controls were 33% and 86%, respectively; P < 0.001) (Fig. 2).

FIG. 1.

Efficacy of oral β-lactamase in preventing piperacillin-tazobactam-induced overgrowth of VRE. The densities (log₁₀ CFU/g) of stool VRE are shown after the orogastric inoculation of 10⁴ CFU of VRE on day 0. Prior to inoculation, none of the mice had detectable levels of VRE (level of detection, ∼2 log₁₀ CFU/g stool). Treatment with subcutaneous piperacillin-tazobactam (▪) resulted in high-density VRE colonization, whereas treatment with normal saline (○) or subcutaneous piperacillin-tazobactam in conjunction with orogastric β-lactamase (▴) did not. Inactivation of the β-lactamase by boiling (□) resulted in a loss of efficacy. The horizontal bars represent standard errors.

FIG. 2.

DGGE analysis of stool microflora of individual mice. Lane 1, controls containing rRNA genes amplified from strains of Bacteroides thetaiotaomicron, Bacteroides uniformis, and Escherichia coli (top to bottom); lanes 2 to 4, saline control mice; lanes 5 to 7, piperacillin-tazobactam-treated mice; lanes 8 to 10, piperacillin-tazobactam- plus β-lactamase-treated mice.

These findings provide further evidence that oral β-lactamase treatment may be an effective means to preserve colonization resistance during therapy with broad-spectrum, parenteral β-lactam antibiotics. Additional studies are needed to determine the efficacy of this strategy as a means to limit the dissemination of nosocomial pathogens in humans.