4.3.1. β-Lactams and Clavulanic Acid

*Discovery.* The first class of synthetic drugs that successfully treated bacterial infections were the sulfa drugs in the early 1930s [163]; however, by the early 1940s, penicillin had quickly eclipsed the sulfa drugs. Penicillin, a β-lactam compound, was active against Gram-positive and Gram-negative, yet penicillin, like the sulfa drugs, had no activity against *M. tuberculosis*. For penicillin, the lack of activity was due to the presence in *M. tuberculosis* of a potent β-lactamase BlaC. The innate resistance of *M. tuberculosis* to β-lactams decreased when blaC was deleted from *M. tuberculosis* or when the β-lactamase irreversible inhibitor clavulanic acid [248] was used in combination with β-lactams [249]. Hugonnet and colleagues used this knowledge to validate the combination of meropenem, a carbapenem with poor affinity for BlaC, and clavulanic acid as an efficacious *M. tuberculosis* inhibitor [250].

*Activity.* The combination meropenem plus clavulanate is part of the group D3 (add-on agents) of second-line TB drugs. Meropenem is a potent inhibitor of drug-susceptible and XDR *M. tuberculosis* strains (MIC range from 0.6 μm to 3.3 μm) when combined with clavulanate [250]. Interestingly, this combination also inhibits non-replicating *M. tuberculosis* [250]. In a chronic model of *M. tuberculosis* infection in mice, meropenem had a modest activity in reducing lung burdens [251]. Surprisingly, the addition of clavulanate to the meropenem treatment did not increase meropenem activity in mice. In mice intravenously infected with *M. tuberculosis*, the lung burden had increased by almost a 1 log10 after 4 weeks of treatment with meropenem and clavulanate showing no inhibitory activity of this combination in vivo [252]. However, studies in mice are problematic since the mice dehydropeptidase (DHP) rapidly cleaves the β-lactam ring.

Meropenem is administered via intravenous injection and new regimens are steering away from injectables. Therefore, Dhar and colleagues tested faropenem, an *oral* β-lactam, and demonstrated activity against *M. tuberculosis* (MIC 4.6 μm) without clavulanate addition (meropenem MIC is 8-fold higher without clavulanate) [253]. Faropenem is also more soluble in water than meropenem (solubility 14.7 g/L vs. 8 g/L for meropenem). Using the dehydropeptidase inhibitor, probenecid, Dhar and colleagues demonstrated that *M. tuberculosis*-infected mice had a small but significant reduction in lung burden after 9 days of treatment with a combination faropenem/clavulanate/probenecid. In a study using DHP-I knockout mice [254], mice were infected intratracheally with *M. tuberculosis* H37Rv and treated for 8 consecutive days, 10 to 12 days post infection. Meropenem was given subcutaneously three times a day in combination with amoxicillin and clavulanate. Another group of mice received faropenem orally with amoxicillin and clavulanate. A control group was treated with moxifloxacin. Treatment also consisted of a dose of probenecid prior to drug administration. While the lung burden was 7.7 log10 CFU/mouse at the beginning of the treatment, the lung burdens at the end of the treatment were 7.8 and 7.5 log10 CFU/mouse for the meropenem and faropenem groups, respectively. Only the mice receiving moxifloxacin had a 2-log10 reduction in lung burden [254].

Experiments performed in other animal models (rabbits and monkeys) revealed that intravenous injections of meropenem with or without clavulanate and dehydropeptidase inhibitor led to serious adverse effects (diarrhea, weight loss and death) [251].

*Mode of action.* Meropenem binds to and inhibits the L,D-transpeptidase LdtMt1 (Rv0116c), which catalyzes the 3→3 cross-linkage in the peptidoglycan [255,256]. Faropenem is 14-fold more potent in inactivating LdtMt1 than meropenem, but this did not correlate with increasing killing against intracellular *M. tuberculosis* [253]. Dhar and colleagues demonstrated that the addition of faropenem to *M. tuberculosis* quickly arrested growth, which did not translate into an immediate lysis of the cells. This is opposite to the current understanding that β-lactams induce cell lysis by inhibiting peptidoglycan cross-linking while cells are still actively dividing. Dhar and colleagues concluded that the mechanism of *M. tuberculosis* killing by β-lactam should be revisited [253].

*Clinical trial.* A clinical study on MDR and XDR TB patients receiving meropenem/clavulanate (intravenous injections three times a day for an average of 85 days) in combination with an MDR/XDR drug regimen showed that the addition of meropenem/clavulanate did not improve: (1) sputum smear or culture conversion; (2) treatment outcome; or (3) treatment success [257]. A phase 2 clinical trial evaluated the early bactericidal activity, safety and tolerability of intravenously administered meropenem with amoxicillin and clavulanate versus orally-administered faropenem with amoxicillin and clavulanate (NCT02349841). Results have not been posted yet.
