21 January 2021

Paper by David Griffin

“It is possible that in ten years’ time, penicillin itself will be a back number and will be replaced by something better. It is quite certain though, that to displace penicillin, any newcomer will have to be very, very good.”

Sir Alexander Fleming
In 'Truman Hails Fleming For Penicillin Drug', New York Times (26 Jul 1945)

Download as PDF

Fleming’s prophecy was correct; newer and better antibacterial agents did soon follow penicillin. The purification and characterization of penicillin resulted in identification of next generation penicillins leading to the discovery of different classes of antibiotics, which has had a profound impact and has saved a multitude of lives throughout the last 75 plus years. However, the ability of the newer and better β-lactams to remain “very, very good” became increasingly difficult as bacteria started to produce enzymes, the β-lactamases, that progressively inactivated each of these new generation agents. The discovery of clavulanic acid, the first commercialised β-lactamase inhibitor, addressed this resistance issue and restored the activity of amoxicillin, at least partially. It was not long, however, before the β-lactamases produced by Gram-negative bacteria also became “very, very good”. As a result, newer and better BLIs were sought to overcome this ever-advancing threat. This paper reviews the past, present and future of BLIs, discussing the progress made and the issues that have been or may be encountered. 

The rise of the β-lactamases
The discovery of penicillin in 1929 was a milestone in the history of medicine. Since the introduction of this first β-lactam in the 1940s, this class of antibiotic has remained the cornerstone of treatment for bacterial infections [1] and, consequently, is the most prescribed of all the antibiotic classes [2]. Extensive use, however, led to the emergence and spread of resistance with enzyme-mediated resistance being just one of the mechanisms by which bacteria avoid the killing activity of the β-lactams. These enzymes, the β-lactamases, are produced by both Gram-positive and Gram-negative bacteria and were actually identified in E. coli prior to penicillin’s first use in man [3]. The relevance of this was not recognised at the time because penicillin use was focused mainly on infections caused by Gram-positive pathogens [1].

Whilst the β-lactam class has undergone continuous development over the years leading to significant improvement in potency, spectrum of activity, pharmacokinetic properties and safety profile [1], this has been largely mirrored by the evolution of the β-lactamases. In the 1970s, only a handful of β-lactamases were known but currently, there are over 850 identified β-lactamases, and some 6,970 enzymes have been characterised (www.bldb.eu, accessed 12th January 2021). The importance and threat of this mechanism of resistance is exacerbated as an increasing number of bacterial strains express multiple β-lactamases, and the genes for these enzymes are frequently located on transmissible, multi-resistance plasmids [1, 4, 5].

β-lactamases are generally classified using the Ambler protein sequence classification system [6] which classifies them into four molecular classes: Ambler Classes A to D [6, 7]. Classes  A, C, and D are enzymes that hydrolyse their substrates using an active site serine whereas Class B enzymes are metallo-β-lactamases (MBLs) that have at least one active-site zinc ion.

Class A β-lactamases are the classic enzymes, such as the TEM and SHV families, which inactivate penicillins and narrow-spectrum cephalosporins but also include the CTX-M class that inactivate extended-spectrum β-lactams (e.g. ceftazidime, ceftriaxone) and are referred to as extended-spectrum β-lactamases (ESBLs). Some members of the TEM and SHV families are also ESBLs and other Class A members, such as the KPC, IMI and SME β-lactamases, hydrolyse carbapenems (e.g. meropenem, imipenem) [8].

Class C β-lactamases are the AmpC enzymes (e.g. CMY, ACT and DHA) which confer resistance to penicillins and some cephalosporins [8].

Class D enzymes are the OXA family of β-lactamases that are able to hydrolyse penicillins, cephalosporins, and carbapenems, and the emergence of Pseudomonas aeruginosa and Acinetobacter baumannii that produce these enzymes is particularly worrying [8, 9].

Class B metallo-β-lactamases can hydrolyse all classes of β-lactams except monobactams (e.g. aztreonam). Amongst this family of enzymes, NDM, VIM and IMP enzymes are amongst the most clinically important MBLs which possess carbapenemase activity [8].  

β-lactamase inhibitors: development timelines and spectrum of activity
In the early 1970s, the first β-lactamase inhibitor (BLI), clavulanic acid, was introduced into clinical use and was quickly followed by sulbactam and tazobactam in 1978 and 1980, respectively [9]. These 1st generation BLIs are structural similar to penicillin and are effective against organisms expressing Class A β-lactamases, including CTX-M and the ESBL derivatives of the TEM and SHV family. They are less effective or inactive, however, against Class B, C, and D β-lactamases and the Class A KPC carbapenemases [8, 9].

This limited spectrum of activity, along with the appearance of β-lactamase types that are not susceptible to the 1st generation BLIs and the paucity of new antibiotics in discovery, led to a search for more effective BLIs. Whilst several  molecules were investigated, it was only in 2015 that the first new BLI, avibactam, was approved for clinical use combined with ceftazidime. This BLI belongs to the diazabicyclooctanones (DBO) class and is a potent inhibitor of the Class A (including KPC enzymes) and C β-lactamases but is ineffective against Class B (MBLs) and Class D (OXA-type enzymes) [1, 8]. 

Shortly after in 2017, vaborbactam, a cyclic boronic acid pharmacophore BLI was approved in combination with meropenem [1, 8]. Vaborbactam is a potent inhibitor of classes A and C β-lactamases but not the Class D OXA enzymes or the Class B MBLs [1].

In early 2020, relebactam, also a DBO, was approved for use with imipenem. Relebactam shows a similar β-lactamase inhibitory profile to that of avibactam [1], but its PK/PD profile better mirrors that of imipenem.

Of these 3 β-lactam/BLI combinations, ceftazidime/avibactam has the broadest therapeutic coverage and includes organisms producing KPC, OXA-48 and carbapenem-resistant organisms producing ESBLs/AmpC enzymes. Imipenem/relebactam has potent activity against KPC-producing Enterobacterales and against P. aeruginosa resistant to carbapenems due to impermeability but not against OXA-48 producers. Meropenem/vaborbactam has a narrower spectrum of activity focused more on KPC, with low activity against OXA-48 [10].

Avibactam, relebactam and vaborbactam combinations have dramatically improved treatment options for serious Gram-negative bacterial infections but none are effective against the Class B MBLs which present a major challenge for β-lactamase inhibitor development. Whilst the search continues, an effective inhibitor of MBLs has yet to be developed [1, 10]. 

Pipeline BLIs
There are currently several BLIs in development including enmetazobactam (AAI101), a penam sulphone BLI structurally related to tazobactam, several DBO BLIs including nacubactam, zidebactam, durlobactam, ETX1317, GT-055, and WCK 4234, and taniborbactam, a boronate BLI [8, 10, 11]. 

All DBO BLIs in development inhibit Class A (including KPCs), C and at least some Class D β-lactamases but have no inhibitory activity against MBLs (Class B). Nacubactam,  zidebactam, durlobactam, GT-055 and ETX1317 also have inherent PBP inhibitory properties [8, 11]. GT-055, when combined with the novel siderophore-based cephalosporin, GT-1, also has potent activity against Class B producing Enterobacterales in addition to activity against the other 3 classes [11]. WCK 4234 has potent inhibitory activity against Class D oxacillinases and it’s inhibitory kinetic constants are superior to both avibactam and relebactam against class A and D β-lactamases [11].

The boronate BLI, taniborbactam, is a potent inhibitor of all four β-lactamase classes including the MBLs [11].

Thus, significant progress is being made in the field of BLIs with the promise of agents that will have therapeutic utility against problematic Gram-negative pathogens including ESBL-producing Enterobacterales and carbapenem-resistant P. aeruginosa and A. baumannii. The spectrum of inhibition has now been expanded beyond the Class A β-lactamases that the 1st generation BLIs covered to include Class C, and D β-lactamases and, for some of the newer BLIs, also the Class B MBLs [11]. It is anticipated that many of these newer advanced BLIs will become available for clinical use within the next decade, thus providing some additional therapeutic options for infections caused by MDR pathogens.

Issues in development
The choice of β-lactam (BL) to combine with the BLIs is an important consideration. The current general practice is to partner a BLI with a BL that is a well-established treatment for serious infections with a known efficacy and safety profile and which has a similar elimination half-life, tissue distribution and metabolic pathways to ensure both agents are present at the site of infection. However, this does not guarantee optimal efficacy against all potential infecting organisms. Furthermore, a fixed combination of BL and BLI may not be the best combination if multiple β-lactamases are being produced because each enzyme differ in its affinity for the BLI [12].

Despite improved inhibitory activity, a fixed combination of BL/BLI could become limited in use over time due to development of resistance. For example, shortly after release into clinical use, resistance against ceftazidime/avibactam emerged during treatment of Enterobacterales and P. aeruginosa, driven by a single amino acid substitutions in some β-lactamases, including the KPC enzyme [10, 11]. This has raised the questions as to whether a BLI could be approved on its own without a partner BL. A regulatory path for such a strategy has yet to be defined and current regulatory guidance for approval of antimicrobial agents do not make this such an easy task [12].

A second issue is the determination of the ratio for the BL and BLI. Generally, the BL and BLI are developed in a fixed ratio despite the need in some clinical situations where being able to modify the ratio may offer better BLI exposures, for example, when the infection has a high bacterial load. Whilst we now have considerable experience with both in vitro and in vivo models that can be used to provide information for optimal ratio determination, and various mathematical models that can further help define the ratio, this can still be a challenge to ensure adequate coverage of the major pathogens.

The pharmacokinetic/pharmacodynamic (PK/PD) index, the percentage of time non-protein bound concentrations remain above the MIC (%fT>MIC) is the established index that predicts efficacy for a BL. However, for a BLI, it is likely that the inhibitory effect will be concentration-dependent, i.e., there is a critical concentration at which enzyme inhibition is maximised. Therefore, the specific PK/PD properties of the BLI needs to be determined before the optimal combination of BL and BLI can be defined. For example, for tazobactam,  the percentage of time that tazobactam concentrations remain above a threshold concentration (%Time>CT) is the index that best correlates with tazobactam efficacy [12]. Our understanding of PK/PD for combination therapy is evolving and new approaches based on this knowledge will need to be conducted in order to determine the optimal dose selection for the BL and BLI.

Susceptibility testing of clinical isolates is essential to avoid inappropriate treatment. For a BL alone, this is straightforward. However, for a BL/BLI, susceptibility to some combinations is determined using a fixed inhibitor concentration (e.g., 4 mg/L of tazobactam) with a range of concentrations for the BL. The resulting MIC is used to categorise the bacterial isolate as either susceptible or resistant based on established breakpoints for efficacy. But this assumes that varying concentrations of the BLI has little or no effect on susceptibility which may not be correct and the resulting susceptibility data may not always correlate with in vivo efficacy. It could be argued that testing the BL and BLI in a fixed ratio may better reflect the in vivo concentration ratio achieved following dosing and thus provide better insights of in vivo efficacy [12]. 

The continuous development of BLs has provided significant improvement in potency and spectrum of activity but the continued evolution and dissemination of MDR bacteria and β-lactamases is an inevitable consequence of BL use. Protecting a BL with a BLI has addressed this to some extent but newer and better BLIs will be needed in future to keep up with the continual rise in resistance. Our evolving understanding of PK/PD of combination therapy will provide new approaches to the development of BL/BLI combinations and assist with the selection and rational dosing of these combinations.


  1. Tooke, C.L., et al., beta-Lactamases and beta-Lactamase Inhibitors in the 21st Century. J Mol Biol, 2019. 431(18): p. 3472-3500.
  2. Klein, E.Y., et al., Global increase and geographic convergence in antibiotic consumption between 2000 and 2015. Proc Natl Acad Sci U S A, 2018. 115(15): p. E3463-E3470.
  3. Abraham, E.P. and E. Chain, An enzyme from bacteria able to destroy penicillin. 1940. Rev Infect Dis, 1988. 10(4): p. 677-8.
  4. Rozwandowicz, M., et al., Plasmids carrying antimicrobial resistance genes in Enterobacteriaceae. J Antimicrob Chemother, 2018. 73(5): p. 1121-1137.
  5. Partridge, S.R., et al., Mobile Genetic Elements Associated with Antimicrobial Resistance. Clin Microbiol Rev, 2018. 31(4).
  6. Ambler, R.P., The structure of beta-lactamases. Philos Trans R Soc Lond B Biol Sci, 1980. 289(1036): p. 321-31.
  7. Bush, K. and G.A. Jacoby, Updated functional classification of beta-lactamases. Antimicrob Agents Chemother, 2010. 54(3): p. 969-76.
  8. Tehrani, K. and N.I. Martin, beta-lactam/beta-lactamase inhibitor combinations: an update. Medchemcomm, 2018. 9(9): p. 1439-1456.
  9. Drawz, S.M. and R.A. Bonomo, Three decades of beta-lactamase inhibitors. Clin Microbiol Rev, 2010. 23(1): p. 160-201.
  10. Vazquez-Ucha, J.C., et al., New Carbapenemase Inhibitors: Clearing the Way for the beta-Lactams. Int J Mol Sci, 2020. 21(23).
  11. Papp-Wallace, K.M., The latest advances in beta-lactam/beta-lactamase inhibitor combinations for the treatment of Gram-negative bacterial infections. Expert Opin Pharmacother, 2019. 20(17): p. 2169-2184.
  12. Abodakpi, H., A. Wanger, and V.H. Tam, What the Clinical Microbiologist Should Know About Pharmacokinetics/Pharmacodynamics in the Era of Emerging Multidrug Resistance: Focusing on beta-Lactam/beta-Lactamase Inhibitor Combinations. Clin Lab Med, 2019. 39(3): p. 473-485.