Why Some Antibiotics Just Don’t Stand a Chance: Understanding β-Lactamases

I was sharing one of my posts with a friend who’s an internist, and he said something that stuck with me: “I really like your posts, but it would be even better if you made one explaining how to interpret an antibiogram—there’s a lot of nuance there.” I couldn’t agree more. In fact, just recently during our ID conference, we were discussing what seemed like a “straightforward” case—but even then, we weren’t sure about the mechanism of resistance or whether ceftriaxone would be a reliable choice for the isolate.

So let’s start with that last point. In Enterobacterales, I truly see the antibiogram in as a reflection of the mechanism of resistance. The possible mechanisms are generally due to:

• Enzymatic inhibition: β-lactamases

• Decreased permeability: porins

• Promotion of antibiotic efflux: pumps

• Altered target sites

• Protection of target sites

In this post, we start with the most common mechanism: enzymatic inhibition - β-lactamases.

With a few basic principles — I think we can discuss antimicrobial resistance and interpret antimicrobial susceptibility tests (AST) more easily. (Of note, I will be focusing on Enterobacterales; Gram-positive cocci, especially Enterococci, deserve their own post in the future!).

β-lactamases, are basically an enzyme that deactivates a beta lactic by splitting the beta lactam ring. So back in 1970 Ambler classified beta lactamases depending on their structure (amino acid sequence) and classified in A, B, C, and D. This will make sense to you as each class has it’s “own” characteristic.

Class A β-lactamases

Class A beta-lactamases are prone to inhibition by beta-lactamase inhibitors, so in the antimicrobial susceptibility test (AST), they would appear susceptible to clavulanate or tazobactam. This includes:

Broad-spectrum (not to confuse with extended-spectrum) TEM-1 and SHV

• TEM-1 (named after Temoneira, the patient from Greece whom the Escherichia coli strain carrying this enzyme was isolated) is the most common β-lactamase in Gram-negative bacteria. It can hydrolyze penicillins and narrow-spectrum cephalosporins (basically first-generation).

• SHV-type β-lactamases are found primarily in Klebsiella pneumoniae strains. Usually, third-generation cephalosporins (e.g., ceftriaxone) are stable to the activity of TEM-1 and SHV. In the antibiogram, this would appear as resistant to ampicillin and cefazolin, but susceptible to ceftriaxone (also typically susceptible to piperacillin-tazobactam and amoxicillin-clavulanate).

If AST shows resistance to third-generation cephalosporins, the Enterobacteriaceae most likely expresses an ESBL (see below), which can be mediated by CTX-M or by subtle point mutations on TEM-1 and SHV β-lactamases (referred to as TEM-1 and SHV-derived ESBLs).

Extended-spectrum β-lactamases (ESBLs):

• By far, Cefotaxime-M β-lactamases (CTX-M) are the most prevalent ESBLs worldwide. They hydrolyze ceftriaxone better than ceftazidime and can also act on aztreonam. They are inhibited more by tazobactam than clavulanic acid, although point mutations can increase activity against ceftazidime.

• In the antibiogram, you would typically see ceftriaxone resistance (sometimes ceftazidime), while remaining piperacillin-tazobactam susceptible.

• In syndromes that involve these isolates and are not uncomplicated UTIs, carbapenems are preferred. But why not piperacillin-tazobactam? The MERINO trial demonstrated a worse mortality rate in patients receiving piperacillin-tazobactam vs meropenem for bloodstream infections. Whole-genome sequencing also revealed that some Enterobacterales harbor OXA (Class D), which is not inhibited by tazobactam (see below).

• One last clue to identify an ESBL using AST is cefoxitin. Cefoxitin is a cephamycin and is resistant to ESBL hydrolysis. So if your isolate shows ceftriaxone resistance, cefoxitin susceptibility, and amoxicillin-clavulanate or piperacillin-tazobactam susceptibility, you have clear evidence of what most likley is CTX-M or Class A ESBL.

KPC (Klebsiella pneumoniae carbapenemase)

• Is a Class A carbapenemase. Phenotypic AST alone does not reliably identify carbapenemase production, so laboratories are advised to perform nucleic acid or antigen testing to confirm. AST will typically show resistance to meropenem and imipenem. If only ertapenem is resistant and meropenem/imipenem remain susceptible, the mechanism is less likely a carbapenemase and more likely porin disruption. In some cases, KPC isolates may appear clavulanate-susceptible in vitro — but this does not correlate clinically (no one would treat an carbapenem resistant with clavulanic acid). Preferred treatments include ceftazidime-avibactam, meropenem-vaborbactam, and imipenem-cilastatin-relebactam.

Class B β-lactamases (Metallo-β-lactamases, MBLs)

Class B β-lactamases all use a Zn²⁺ cation for hydrolysis therefore they are called metallo-beta-lactamases (MBL). They are not inhibited by clavulanic acid or tazobactam, and they hydrolyze nearly all β-lactams except monobactams (aztreonam).

Isolates with MBLs are resistant to all beta-lactam antibiotics, including cefazolin, cefoxitin, ceftriaxone, cefepime, meropenem, and imipenem, and also to piperacillin-tazobactam and amoxicillin-clavulanate.

Clinically important MBLs include:

• NDM-1 (New Delhi metallo-β-lactamase–1) — Enterobacterales

• IMP (Imipenemase)

• VIM (Verona integron-encoded metallo-β-lactamase)

• SPM (Sao Paulo metallo-β-lactamase)

• GIM (German imipenemase)

• SIM (Seoul imipenemase)

Treatment options: aztreonam alone can sometimes work, but MBL-producing organisms often co-harbor ESBLs. The recommended approach is avibactam plus aztreonam, or ceftazidime-avibactam plus aztreonam, allowing aztreonam to remain active while the other agents are sacrificed. Another option is cefiderocol, the “Trojan horse” antibiotic, which is stable against MBLs and uses bacterial iron transport systems to enter cells, bypassing porin mutations and efflux pumps.

Class C β-lactamases (AmpC)

Class C are mainly cephalosporinases, so they will hydrolyze all cephalosporins (yes, at least from first to fourth generation, but also penicillins). They are also known as AmpC… AmpC β-lactamases are primarily chromosomal enzymes and are not susceptible to β-lactamase inhibitors such as clavulanic acid or tazobactam. So an isolate with AmpC sometimes will basically show resistance to cefazolin (1st generation) and cefoxitin (2nd generation), while remaining susceptible to ceftriaxone (3rd generation) and cefepime (4th generation), and maybe even piperacillin-tazobactam.

But wait — I mentioned before that these enzymes hydrolyze most cephalosporins and penicillins, so what happened? AmpC production in Gram-negative bacilli is normally repressed. However, a transient increase in production (10- to 100-fold) can occur in the presence of β-lactam antibiotics in the following species that possess inducible AmpC enzymes: Enterobacter (usually cloacae), Klebsiella aerogenes, Citrobacter freundii, but also Serratia, Morganella morganii, Providencia, Hafnia alvei, Yersinia enterocolitica, and Pseudomonas aeruginosa.

From these, Hafnia alvei, Enterobacter cloacae, Klebsiella aerogenes, Citrobacter freundii, and Yersinia enterocolitica are more likely to be clinical AmpC inducers — hence the acronym HECK Yeah. In organisms with inducible AmpC, cefoxitin is a particularly strong inducer, so in vitro you may see resistance. You also have to be careful because isolates that initially test susceptible to ceftriaxone may become non-susceptible after treatment, as AmpC induction can be observed after even a few doses of ceftriaxone, cefotaxime, or ceftazidime — this also applies to piperacillin-tazobactam.

So, the treatment will be cefepime. But wait — I also said AmpC hydrolyzes 4th-generation cephalosporins… fortunately, cefepime is a poor substrate for AmpC and is usually effective if MIC ≤4. Some observational studies have demonstrated that isolates with cefepime MIC >4 might harbor an ESBL or acquired mutations making them more likely to hydrolyze cefepime. If MIC >4, a carbapenem is the safest choice, as you cannot exclude the presence of an ESBL or an acquired mutation that increases cefepime hydrolysis.

Class D β-lactamases (OXA, Oxacillinases)

Alright, last but not least, Class D. These are Oxacillinases (OXA). They were first described in the 1960s–70s as enzymes that hydrolyzed oxacillin and penicillins, which were different from TEM and SHV (remember Class A?) because they were poorly inhibited by clavulanic acid — so you already expect that organisms with OXA would be resistant to amoxicillin/clavulanate and piperacillin/tazobactam.

OXAs are very similar to Class A in that if an isolate has a broad-spectrum OXA, it would be resistant to first-generation cephalosporins and penicillins, including amoxicillin/clavulanate. If the isolate had an “OXA ESBL,” it would be resistant to first- to fourth-generation cephalosporins and penicillins, but in this case also to piperacillin-tazobactam — unlike Class A ESBLs. As explained above, this is not necessarily clinically relevant, as a carbapenem is most likely the best choice in this scenario.

Most recently, OXA carbapenemases — initially of limited clinical concern in Enterobacterales — became more significant with the discovery of OXA-48 in Klebsiella pneumoniae in Turkey in 2001. OXA-48 spread globally due to its ability to hydrolyze carbapenems while sparing extended-spectrum cephalosporins. Therefore, treatment for isolates demonstrating OXA-48 consists of ceftazidime-avibactam and cefiderocol. (Meropenem-vaborbactam and imipenem-cilastatin-relebactam have limited activity against OXA-48 and should not be used in this scenario.)

In summary, think of Class D as Class A with the exception that on AST they will most likely be resistant to piperacillin-tazobactam and amoxicillin/clavulanate, although again this is not very relevant in clinical practice. For OXA-48, a carbapenemase, ceftazidime-avibactam and cefiderocol are the agents of choice.

Thanks for staying to the end! Most likely this will need to be read multiple times, as it can be slightly confusing. But remember — these are just enzymatic mechanisms of antibiotic resistance, and there are at least four more mechanisms I will discuss in future posts!

MUSICAL CODA 🎶

Could this be Temoneira?

Do you want to practice?

Isolate: Escherichia coli
Key AST Pattern: Ampicillin R, Cefazolin R, Ceftriaxone S, Piperacillin/Tazobactam S
Likely Mechanism: Class A broad-spectrum β-lactamase (TEM-1)
Rationale: TEM-1 hydrolyzes penicillins and 1st-gen cephalosporins; 3rd-gen cephalosporins are stable. Can safely treat with Ceftriaxone.

Isolate: Klebsiella pneumoniae
Key AST Pattern: Ampicillin R, Cefazolin R, Ceftriaxone R, Piperacillin/Tazobactam S
Likely Mechanism: Class A ESBL (CTX-M type)
Comment / Rationale: 3rd-gen cephalosporin resistance suggests ESBL; inhibition by β-lactamase inhibitors explains piperacillin/tazobactam susceptibility. Carbapenem for invasive disease is suggested.

Isolate: Enterobacter cloacae
Key AST Pattern: Cefoxitin R, Ceftriaxone S, Cefepime S, Piperacillin/Tazobactam S
Likely Mechanism: Class C - AmpC β-lactamase
Comment / Rationale: Cefoxitin is a strong inducer; Cefepime is a poor AmpC substrate, so is preferred for treatment

Isolate: Klebsiella pneumoniae
Key AST Pattern: Meropenem R, Imipenem R, Piperacillin/Tazobactam R
Likely Mechanism: Class A carbapenemase (KPC)
Comment / Rationale: Carbapenem-resistant; labs should confirm with molecular/antigen test; treatment with Ceftazidime/Avibactam, can also use meropenem-vaborbactam, and imipenem-cilastatin-relebactam.

Isolate: Escherichia coli
Key AST Pattern: Cefazolin R, Cefoxitin R, Ceftriaxone R, Meropenem R, Piperacillin/Tazobactam R, Aztreonam S
Likely Mechanism: Class B metallo-β-lactamase (NDM-1, VIM, IMP)
Comment / Rationale: Resistant to all β-lactams except monobactams; combination therapy (aztreonam + avibactam) or cefiderocol recommended.

Isolate: Klebsiella pneumoniae
Key AST Pattern: Ampicillin R, Cefazolin R, Ceftriaxone R, Piperacillin/Tazobactam R, Amoxicillin/Clavulanate R, Meropenem R
Likely Mechanism: Class D β-lactamase (OXA-48 carbapenemase)
Comment / Rationale: OXA-48 carbapenemase treated with ceftazidime-avibactam or cefiderocol.