The beta-lactam antibiotics share a core four-membered lactam ring that is essential for antimicrobial activity. Understanding the structural basis of beta-lactam action, and how that structure is exploited by bacterial resistance mechanisms, provides the conceptual framework for rational antibiotic selection across an enormous and clinically indispensable drug class.
The Beta-Lactam Ring. The beta-lactam ring is a four-membered cyclic amide in which the nitrogen and carbonyl carbon are under ring strain due to the small ring size. This strain makes the carbonyl carbon highly electrophilic and reactive toward nucleophilic attack. In penicillins, the beta-lactam ring is fused to a five-membered thiazolidine ring; in cephalosporins, it is fused to a six-membered dihydrothiazine ring. These bicyclic structures collectively define the penam and cephem scaffolds, respectively, from which the clinical agents are derived. The specific side chains attached to the core scaffold at the 6-position (penicillins) or 7-position (cephalosporins) determine spectrum, pharmacokinetics (PK), and stability to enzymatic degradation. The minimum inhibitory concentration (MIC) of a beta-lactam against a given organism reflects the combined influence of PBP (penicillin-binding protein) affinity, outer membrane penetration, and susceptibility to enzymatic inactivation.1
Bacterial Cell Wall Biosynthesis. Gram-positive and gram-negative bacteria maintain structural integrity through a peptidoglycan (murein) layer composed of linear glycan strands of alternating N-acetylglucosamine (NAG) and N-acetylmuramic acid (NAM) cross-linked by short peptide bridges. The final cross-linking step is catalyzed by transpeptidase enzymes that form peptide bonds between adjacent stem peptides, converting the glycan strands from a loose meshwork into a rigid, mechanically stable sacculus. This cross-linking reaction is the pharmacological target of all beta-lactam antibiotics. Gram-positive bacteria have a thick, multilayered peptidoglycan wall (20-80 nm) exposed directly to the extracellular environment. Gram-negative bacteria have a thinner peptidoglycan layer (2-7 nm) sandwiched between an inner cytoplasmic membrane and an outer membrane containing lipopolysaccharide (LPS), which limits outer membrane permeability to hydrophilic molecules and requires beta-lactams to traverse via porin channels.1
Penicillin-Binding Proteins. The transpeptidases and related enzymes responsible for peptidoglycan synthesis are collectively designated penicillin-binding proteins (PBPs). Each species of bacteria expresses a characteristic set of PBPs, numbered by decreasing molecular weight. In Staphylococcus aureus, the numbered PBP variants PBP1 (transpeptidase), PBP2 (transpeptidase-transglycosylase), PBP3 (transpeptidase for septation), and PBP4 (carboxypeptidase) are expressed under normal conditions; methicillin-resistant Staphylococcus aureus (MRSA) expresses an additional PBP2a (also designated PBP2′) encoded by the mecA gene, which has low affinity for virtually all beta-lactam antibiotics. In Streptococcus pneumoniae, resistance to penicillin is conferred by altered PBP2x and PBP2b proteins with reduced beta-lactam affinity. In Escherichia coli, the essential PBP variants PBP1a (transpeptidase-transglycosylase), PBP1b (transpeptidase-transglycosylase), PBP2 (cell shape), and PBP3 (cell division transpeptidase) each serve distinct roles; inhibition of PBP1 and PBP2 produces cell lysis while PBP3 inhibition causes filamentation without immediate lysis.2
Mechanism of Action: Acylation of PBPs. Beta-lactam antibiotics are structural analogs of the D-alanyl-D-alanine terminus of the peptide stem that is the natural substrate of transpeptidase enzymes. The beta-lactam carbonyl carbon forms a covalent acyl-enzyme intermediate with the active-site serine residue of the PBP, which is chemically stable and extremely slow to hydrolyze under physiological conditions. This effectively irreversible acylation permanently inactivates the transpeptidase. As bacteria continue to grow and divide in the presence of a beta-lactam, peptidoglycan cross-linking is inhibited while peptidoglycan-degrading autolysins continue to operate, resulting in weakened cell wall structural integrity, osmotic stress, and ultimately cell lysis. This bactericidal mechanism is time-dependent: the pharmacodynamic (PD) parameter that best predicts efficacy is the percentage of the dosing interval during which the free drug concentration exceeds the minimum inhibitory concentration (MIC), denoted as fT>MIC. For most beta-lactams, sustained fT>MIC above 40-50% of the dosing interval is required for bactericidal effect; continuous infusion or extended infusion strategies exploit this property to maximize efficacy against organisms with elevated MICs.8
Beta-lactam efficacy is governed by fT>MIC (percentage of the dosing interval that free drug exceeds the minimum inhibitory concentration), not by peak concentration. For beta-lactams treating infections caused by organisms with elevated MICs (e.g., Pseudomonas aeruginosa with MIC near the susceptibility breakpoint), prolonged or continuous infusion of piperacillin-tazobactam or meropenem dramatically increases target attainment. Meropenem 2 g infused over 3 hours achieves substantially higher fT>MIC against Pseudomonas than a standard 30-minute infusion of the same dose at equivalent daily doses.8
The penicillin family is most usefully organized by spectrum of activity and beta-lactamase stability rather than by chronology. Each subclass occupies a distinct clinical niche defined by its activity against gram-positive and gram-negative organisms, its oral bioavailability, and its susceptibility to enzymatic inactivation.
Natural Penicillins: Penicillin G and Penicillin V. Penicillin G (benzylpenicillin) and penicillin V (phenoxymethylpenicillin) retain the original narrow-spectrum profile with excellent activity against streptococci (Streptococcus pyogenes, viridans streptococci, non-resistant Streptococcus pneumoniae), Treponema pallidum, Neisseria meningitidis, most oral anaerobes including Fusobacterium species, Clostridium species, Pasteurella multocida, and Listeria monocytogenes. Penicillin G is available only for parenteral administration; its acid lability prevents reliable oral absorption. Penicillin V is acid-stable and orally bioavailable (approximately 60-73%), making it suitable for step-down therapy in streptococcal pharyngitis and skin infections. Neither agent has meaningful activity against gram-negative enteric organisms (enterobacteriaceae), staphylococci (due to near-universal beta-lactamase production), or Pseudomonas aeruginosa. Penicillin G is the drug of choice for primary, secondary, and tertiary syphilis, as Treponema pallidum retains universal susceptibility; no penicillin resistance in this organism has been documented.4
Aminopenicillins: Ampicillin and Amoxicillin. Introduction of an amino group at the alpha position of the acyl side chain (creating the aminopenicillin class) extends gram-negative coverage to include Haemophilus influenzae (non-beta-lactamase-producing strains), Escherichia coli (community-acquired, non-resistant strains), Proteus mirabilis, Salmonella species, and Enterococcus faecalis, while maintaining the gram-positive activity of natural penicillins. Amoxicillin has substantially superior oral bioavailability compared to ampicillin (approximately 80-90% versus 30-55%), is less affected by food, and produces more predictable and higher plasma concentrations at equivalent doses. Ampicillin is therefore preferred for parenteral therapy (such as in Listeria meningitis or enterococcal endocarditis), while amoxicillin is the oral agent of choice for outpatient indications including acute otitis media and community-acquired pneumonia in appropriate patients. Both agents are susceptible to beta-lactamase inactivation and have no useful activity against Klebsiella pneumoniae, Pseudomonas aeruginosa, or MRSA (methicillin-resistant Staphylococcus aureus).4
Antistaphylococcal Penicillins: Nafcillin, Oxacillin, and Dicloxacillin. The antistaphylococcal (isoxazolyl) penicillins were developed specifically to resist hydrolysis by the staphylococcal penicillinase (a class A beta-lactamase) through the introduction of a bulky acyl side chain that provides steric protection of the beta-lactam ring. Nafcillin and oxacillin are parenteral agents used for methicillin-susceptible Staphylococcus aureus (MSSA) infections; nafcillin is predominantly hepatically eliminated (making it a preferred choice in patients with significant renal impairment), while oxacillin is renally eliminated. Dicloxacillin is the oral isoxazolyl penicillin, with bioavailability of approximately 50-76%; it should be taken on an empty stomach as food substantially reduces absorption. Antistaphylococcal penicillins remain the drugs of choice for MSSA bacteremia, endocarditis, and deep-seated MSSA infections, outperforming vancomycin in clinical outcome studies for MSSA.4
Extended-Spectrum and Ureidopenicillins. Carbenicillin was the first penicillin with antipseudomonal activity; it has been superseded by ticarcillin and subsequently by piperacillin, the ureidopenicillin that represents the current standard of the class. Piperacillin extends gram-negative coverage to include Pseudomonas aeruginosa, many Enterobacteriaceae, and anaerobes, while retaining streptococcal and enterococcal activity. Piperacillin is always used in combination with the beta-lactamase inhibitor tazobactam (piperacillin-tazobactam, designated pip-tazo or TZP) to restore activity against beta-lactamase-producing organisms. The combination of piperacillin with tazobactam extends the spectrum to include many extended-spectrum beta-lactamase (ESBL)-producing organisms in vitro, though the clinical efficacy against ESBL bacteremia is an area of ongoing investigation and controversy, with the MERINO (multicenter randomized trial comparing meropenem versus piperacillin-tazobactam) trial for ceftriaxone-resistant enterobacterial bacteremia for definitive treatment of ESBL bacteremia) demonstrating inferior 30-day mortality with piperacillin-tazobactam compared to meropenem for this indication.3
The MERINO trial (2018) randomized patients with bloodstream infections due to ceftriaxone-resistant E. coli or Klebsiella pneumoniae (a phenotypic marker for ESBL or AmpC production) to piperacillin-tazobactam 4.5 g every 6 hours versus meropenem 1 g every 8 hours. Thirty-day mortality was 12.3% with piperacillin-tazobactam versus 3.7% with meropenem, a statistically significant and clinically large difference. This trial established that piperacillin-tazobactam should not be used as definitive therapy for ESBL-producing gram-negative bacteremia, even when in vitro susceptibility testing reports the isolate as susceptible to pip-tazo.
The spectrum of penicillin activity is determined by three intersecting properties: intrinsic affinity for the target organism’s PBPs (penicillin-binding proteins), ability to penetrate to the site of those PBPs (determined by outer membrane architecture and porin expression in gram-negative organisms), and susceptibility to enzymatic inactivation at that site. These properties must be considered alongside pharmacokinetic factors that determine whether adequate drug concentrations are delivered and sustained at the site of infection.
Gram-Positive Spectrum. All penicillins retain strong intrinsic affinity for the PBPs of most gram-positive organisms. Streptococcus pyogenes (group A streptococcus) remains universally susceptible to penicillin G and V; no acquired penicillin resistance has emerged in this species after decades of use, making narrow-spectrum penicillin the drug of choice for streptococcal pharyngitis and skin infections. Among Streptococcus pneumoniae, susceptibility has declined substantially: approximately 25-35% of isolates in the United States now demonstrate reduced susceptibility to penicillin, mediated by altered penicillin-binding protein (PBP) variants, specifically PBP2x and PBP2b, with reduced beta-lactam affinity. Pneumococcal meningitis requires high-dose intravenous penicillin (or a third-generation cephalosporin in areas of resistance) to achieve adequate cerebrospinal fluid (CSF) concentrations. Enterococcus faecalis is susceptible to ampicillin and penicillin G (though bacteriostatic action, not bactericidal); Enterococcus faecium is frequently ampicillin-resistant. The intrinsic tolerance of enterococci to beta-lactam bactericidal killing makes aminoglycoside synergy essential for endocarditis treatment.4
Gram-Negative Spectrum and Outer Membrane Penetration. Gram-negative bacteria present an additional barrier to beta-lactam penetration: the outer membrane with its lipopolysaccharide (LPS) component is largely impermeable to hydrophobic molecules, and beta-lactam antibiotics must traverse this membrane through water-filled porin channels (principally OmpF and OmpC in Enterobacteriaceae). The size and charge of the beta-lactam molecule determines the rate of porin traversal: smaller, more hydrophilic molecules traverse more efficiently, explaining why ampicillin achieves better gram-negative penetration than the bulkier antistaphylococcal penicillins. Loss or downregulation of specific porin channels is a mechanism of resistance in Pseudomonas aeruginosa (OprD porin loss) and Klebsiella pneumoniae. Once inside the periplasmic space, beta-lactams reach the PBPs on the outer surface of the inner membrane; any beta-lactamases secreted into the periplasmic space intercept and inactivate the antibiotic before it reaches the PBP (penicillin-binding protein) target.1
Pharmacokinetics: Distribution and CNS (Central Nervous System) Penetration. Penicillins are moderately protein-bound (penicillin G approximately 55%, ampicillin approximately 20%, nafcillin approximately 87%) and distribute into most tissues and body fluids. Only free (unbound) drug is pharmacologically active, making protein binding an important determinant of effective drug concentration. CNS penetration of penicillins is low under normal conditions (CSF-to-plasma ratio approximately 1-2%) due to efflux by the blood-brain barrier P-glycoprotein and organic anion transporters, but increases substantially (to approximately 5-10%) when the meninges are inflamed, as occurs in bacterial meningitis. This inflammation-dependent penetration supports the use of high-dose intravenous penicillin or ampicillin for meningitis caused by susceptible organisms, but underscores the importance of using agents with adequate CSF penetration when meningitis is present.4
Pharmacokinetics: Renal Elimination and Dose Adjustment. The majority of penicillins are eliminated predominantly by renal excretion through a combination of glomerular filtration and active tubular secretion by the organic anion transporter OAT1 (organic anion transporter 1) in the proximal tubule. This active tubular secretion accounts for the short half-lives of most penicillins (penicillin G 0.5 hours, ampicillin 1-2 hours, piperacillin-tazobactam approximately 1 hour); probenecid, a competitive inhibitor of OAT1, extends penicillin half-life and was historically used to reduce dosing frequency. In patients with renal impairment (estimated glomerular filtration rate (eGFR) below 30-40 mL/min/1.73 m²), dose reduction or interval extension is required for most penicillins to avoid accumulation; in severe renal failure, high-dose penicillin G can accumulate to cause neurotoxicity (myoclonus, seizures). Nafcillin is the notable exception: it is predominantly hepatically eliminated (70-80% biliary excretion) and does not require dose adjustment in renal failure, making it preferred over oxacillin for MSSA (methicillin-susceptible Staphylococcus aureus) infections in patients with renal impairment.4
| Agent | Route | Key Spectrum | Elimination | Renal Adjustment |
|---|---|---|---|---|
| Penicillin G | IV/IM | Streptococci, T. pallidum, oral anaerobes | Renal (60-90%) | Yes (<30 mL/min) |
| Penicillin V | PO | Same as penicillin G; bioavailability 60-73% | Renal | Yes (severe failure) |
| Ampicillin | IV/PO | + H. influenzae, E. coli, Enterococcus faecalis, Listeria | Renal (>75%) | Yes (<30 mL/min) |
| Amoxicillin | PO | Same as ampicillin; BA 80-90% | Renal | Yes (<30 mL/min) |
| Nafcillin | IV | MSSA (antistaphylococcal) | Hepatic (70-80%) | No |
| Oxacillin | IV | MSSA (antistaphylococcal) | Renal/hepatic | Mild (severe failure) |
| Dicloxacillin | PO | MSSA (skin/soft tissue); take fasted | Hepatic/renal | Mild |
| Piperacillin-tazobactam | IV | Extended gram-neg, Pseudomonas, anaerobes; avoid ESBL bacteremia | Renal (>68%) | Yes (<40 mL/min) |
Bacterial resistance to penicillins and all beta-lactam antibiotics arises through four principal mechanisms that are mechanistically distinct and clinically important: enzymatic inactivation by beta-lactamases, production of altered PBPs (penicillin-binding proteins) with reduced drug affinity, active efflux of drug from the bacterial cell, and reduced outer membrane permeability in gram-negative organisms. These mechanisms frequently coexist within a single clinical isolate, and their simultaneous expression constitutes the pharmacological basis of multidrug resistance.
Beta-Lactamases: Ambler Classification. Beta-lactamases are enzymes that hydrolyze the beta-lactam ring, inactivating the antibiotic before it can reach its PBP (penicillin-binding protein) target. The Ambler classification system organizes beta-lactamases by molecular mechanism into four classes. Class A (serine beta-lactamases) includes TEM-1 (TEM: named after patient Temoniera, source of the original isolate) and SHV-1 (SHV: sulfhydryl variable enzyme), which are the most common penicillinases in gram-negative organisms, along with the extended-spectrum beta-lactamases (ESBLs) such as CTX-M (cefotaxime-Munich) enzymes that additionally hydrolyze third-generation cephalosporins. Class A also includes the KPC (Klebsiella pneumoniae carbapenemase) enzymes that hydrolyze carbapenems. Class B (metallo-beta-lactamases) include NDM (New Delhi metallo-beta-lactamase), VIM (Verona integron-encoded metallo-beta-lactamase), and IMP (imipenemase), which require zinc as a cofactor and hydrolyze virtually all beta-lactams including carbapenems but not aztreonam; these are not inhibited by serine-based beta-lactamase inhibitors. Class C (AmpC cephalosporinases) are chromosomally encoded in many gram-negative species and are typically inducible; they hydrolyze first-, second-, and third-generation cephalosporins and are not inhibited by clavulanate. Class D (OXA-type; OXA: oxacillin-hydrolyzing) enzymes include OXA-23 (predominant in Acinetobacter baumannii) and OXA-48 (predominant in Klebsiella pneumoniae), which are clinically important carbapenemases in Acinetobacter baumannii and Klebsiella pneumoniae, respectively.5
Extended-Spectrum Beta-Lactamases. Extended-spectrum beta-lactamases (ESBLs) are modified derivatives of TEM and SHV penicillinases, or the CTX-M family, that have evolved to hydrolyze third-generation cephalosporins and aztreonam (a monobactam) in addition to penicillins. ESBL (extended-spectrum beta-lactamase)-producing organisms are typically resistant to all penicillins and most cephalosporins, with susceptibility usually retained only to carbapenems and (variably) to beta-lactam inhibitor combinations such as piperacillin-tazobactam. ESBLs are inhibited in vitro by clavulanic acid, tazobactam, and sulbactam, but as demonstrated in the MERINO (multicenter randomized trial of piperacillin-tazobactam versus meropenem) trial results, this in vitro susceptibility does not reliably translate into clinical efficacy for bloodstream infections, likely due to inoculum effect (high bacterial burden in bacteremia overwhelms the inhibitor). CTX-M-15 (CTX-M: cefotaxime-Munich, a CTX-M family ESBL variant) is the globally dominant ESBL; it is carried on mobile genetic elements (plasmids and transposons) and spreads between species. Risk factors for ESBL infection include prior antibiotic exposure (particularly fluoroquinolones and third-generation cephalosporins), healthcare exposure, urinary catheterization, and travel to high-prevalence regions.6
Altered Penicillin-Binding Proteins: mecA and PBP2a. MRSA (methicillin-resistant Staphylococcus aureus) resistance is mediated by the mecA gene (or its homolog mecC), which encodes PBP2a, a transpeptidase with extremely low affinity for all beta-lactam antibiotics due to structural changes in the active site. PBP2a retains transpeptidase activity even in the presence of beta-lactam concentrations far exceeding clinically achievable levels, allowing the organism to continue cell wall synthesis and proliferate despite beta-lactam exposure. The mecA gene is carried on a mobile genetic element called the staphylococcal cassette chromosome mec (SCCmec), which has disseminated broadly among staphylococcal populations. Detection of MRSA (methicillin-resistant Staphylococcus aureus) is reliably performed by oxacillin MIC (minimum inhibitory concentration) testing, mecA gene PCR (polymerase chain reaction), or chromogenic agar culture. No beta-lactam antibiotic (with the limited exception of ceftaroline, a fifth-generation cephalosporin with affinity for PBP2a) is effective against MRSA.7
Efflux Pumps and Outer Membrane Permeability. Active efflux systems belonging to the resistance-nodulation-division (RND) family in gram-negative bacteria export a broad range of structurally unrelated antibiotics, including beta-lactams, from the periplasm to the exterior of the cell. In Pseudomonas aeruginosa, the MexAB-OprM and MexCD-OprJ efflux systems contribute to intrinsic and acquired resistance. Outer membrane permeability to beta-lactams is further reduced by downregulation or mutation of porin channels: in Pseudomonas aeruginosa, loss of the OprD porin (which provides the primary route of carbapenem entry) confers selective carbapenem resistance while preserving susceptibility to other classes. In Klebsiella pneumoniae producing KPC enzymes, the combination of carbapenemase production, porin loss (OmpK35 and OmpK36 downregulation), and efflux pump upregulation creates high-level, pan-resistant phenotypes.5
ESBL producers are resistant to most penicillins and cephalosporins but show in vitro susceptibility to pip-tazo and carbapenems; use carbapenems for serious infections regardless of pip-tazo susceptibility testing (MERINO trial). AmpC producers are resistant to first- through third-generation cephalosporins and cephamycins and are not inhibited by clavulanate; carbapenems are the treatment of choice. KPC-producing Klebsiella resist carbapenems; options include ceftazidime-avibactam, meropenem-vaborbactam, and imipenem-relebactam. Always correlate phenotypic resistance patterns with clinical decisions; susceptibility reports must be interpreted in the context of infection severity and inoculum burden.
Tipper DJ, Strominger JL. Mechanism of action of penicillins: a proposal based on their structural similarity to acyl-D-alanyl-D-alanine. Proc Natl Acad Sci USA. 1965;54(4):1133-1141.
doi:10.1073/pnas.54.4.1133Spratt BG. Penicillin-binding proteins and the future of beta-lactam antibiotics. J Gen Microbiol. 1983;129(5):1247-1260.
doi:10.1099/00221287-129-5-1247Harris PNA, Tambyah PA, Lye DC, et al. Effect of piperacillin-tazobactam vs meropenem on 30-day mortality for patients with E coli or Klebsiella pneumoniae bloodstream infection and ceftriaxone resistance: a randomized clinical trial. JAMA. 2018;320(10):984-994.
doi:10.1001/jama.2018.12163Wright AJ. The penicillins. Mayo Clin Proc. 1999;74(3):290-307.
doi:10.4065/74.3.290Bush K, Bradford PA. Beta-lactams and beta-lactamase inhibitors: an overview. Cold Spring Harb Perspect Med. 2016;6(8):a025247.
doi:10.1101/cshperspect.a025247Pitout JD, Laupland KB. Extended-spectrum beta-lactamase-producing Enterobacteriaceae: an emerging public-health concern. Lancet Infect Dis. 2008;8(3):159-166.
doi:10.1016/S1473-3099(08)70041-0Lakhundi S, Zhang K. Methicillin-resistant Staphylococcus aureus: molecular characterization, evolution, and epidemiology. Clin Microbiol Rev. 2018;31(4):e00020-18.
doi:10.1128/CMR.00020-18Craig WA. Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial dosing of mice and men. Clin Infect Dis. 1998;26(1):1-12.
doi:10.1086/516284