Penicillins and β-Lactamase Inhibitors

Russell E. Lewis

2026-04-14

Penicillins and β-Lactamase Inhibitors

Russell E. Lewis
Associate Professor of Infectious Diseases
Department of Molecular Medicine
University of Padua

russelledward.lewis@unipd.it
https://github.com/Russlewisbo
Slides and course materials: www.padovaid.com

Learning Objectives

By the end of this lecture, you should be able to:

1. Describe the basic chemical structure of penicillins
2. Explain how the side chain determines spectrum and stability
3. Describe the mechanism of action involving PBPs
4. Explain peptidoglycan synthesis and cross-linking
5. List and explain the four mechanisms of β-lactam resistance
6. Classify β-lactamases by Ambler molecular class
7. Differentiate the five classes of penicillins by spectrum
8. Describe pharmacokinetic properties and dosing adjustments
9. Recognize adverse effects and hypersensitivity reactions
10. Select appropriate β-lactam/β-lactamase inhibitor combinations

PART 1: History and chemistry

We’ll begin with the historical discovery and chemical basis of penicillins. Understanding structure is essential for understanding function.

The Discovery of penicillin

• September 1928
• Alexander Fleming at St. Mary’s Hospital, London
• Noticed bacterial lysis around Penicillium mold contamination
• Published findings in 1929
• Initially unable to purify the active compound

Fleming was studying staphylococcal colonies when he noticed a contaminating mold had created a zone of bacterial inhibition. He named the antibacterial substance “penicillin” but couldn’t purify it for clinical use.

From discovery to clinical use

Year	Milestone
1928	Fleming discovers penicillin
1929	Fleming publishes findings
1940	Florey & Chain purify penicillin
1941	First human treated
1943	Mass production begins
1945	Nobel Prize awarded

The gap between discovery (1928) and clinical use (1941) was 13 years. Florey and Chain at Oxford developed the purification process. The first patient treated was a policeman with staphylococcal sepsis. Mass production was driven by WWII needs.

The Impact of penicillin

• Before penicillin: Minor infections could be fatal
• Bacterial endocarditis: 100% mortality
• Pneumococcal pneumonia: 30% mortality
• Staphylococcal sepsis: Often fatal
• Penicillin transformed infectious disease medicine
• Initiated the “antibiotic era”

It’s hard to overstate the impact. Diseases that were death sentences became curable. Surgery became safer. Emphasize that despite being nearly 100 years old, penicillins remain essential antibiotics.

Basic penicillin structure

Thiazolidine ring (5-membered with sulfur), β-lactam ring (4-membered, essential for activity), and the side chain (R group). The carboxylic acid makes penicillins weak acids.

The three key structural components

1. Thiazolidine ring
   • 5-membered ring containing sulfur
   • Provides structural stability
2. β-Lactam ring
   • 4-membered ring under strain
   • ESSENTIAL for antibacterial activity
   • Target of β-lactamases
3. Side chain (R group)
   • Variable component
   • Determines properties of each penicillin

The β-lactam ring is under considerable strain due to its geometry. This strain makes it reactive with PBPs but also susceptible to β-lactamase attack. The side chain is where all the modifications occur to create different penicillins.

The β-Lactam ring: Why it matters

• Strained 4-membered ring — chemically reactive
• Mimics D-Ala-D-Ala terminus of peptidoglycan
• Covalently binds to PBP active site serine
• Opening the ring = loss of activity
• This is why β-lactamases cause resistance

Important

No β-lactam ring = No antibacterial activity

The β-lactam ring mimics the natural substrate of PBPs, which is the D-alanyl-D-alanine terminus of the peptidoglycan pentapeptide. When the ring opens (either by PBP binding or β-lactamase hydrolysis), the compound changes structure.

Classification of β-lactamases

Ambler Class	Major Subtype	Preferred Substrates	Inhibitor	Genetic Localization	Representative Enzymes
A	Gram-positive β-lactamase 2a	Penicillins	Clavulanic acid	Chromosome or plasmid	PC1
A	Gram-negative β-lactamase 2b	Penicillins, 1st-gen cephalosporins	Clavulanic acid	Plasmid or chromosomal	TEM-1, SHV-1
A	Extended-spectrum β-lactamase 2be	Penicillins, extended-spectrum cephalosporins, aztreonam	Clavulanic acid	Plasmid	TEM-24, SHV-12, CTX-M-15
A	Inhibitor-resistant TEM β-lactamase 2br	Penicillins	Clavulanic acidc	Plasmid	TEM-30, SHV-10
A	Carbenicillin-hydrolyzing β-lactamase 2c	Carbenicillin	Clavulanic acidc	Plasmid	PSE-1, CARB-3
A	Cephalosporin-hydrolyzing β-lactamase 2e	Extended-spectrum cephalosporins	Clavulanic acid	Chromosome	CepA
A	Carbapenem-hydrolyzing β-lactamase 2f	Carbapenems	Avibactam, Relebactam, Vaborbactam	Chromosome or plasmid	KPC-2, SME-1
B	Metallo-β-lactamase 3a	All β-lactams except monobactam	EDTA, divalent cation chelators	Chromosome or plasmid	IMP-1, VIM-2, NDM-1
C	AmpC-type β-lactamase	Cephalosporins	Cloxacillin, avibactam, relebactam, vaborbactam	Chromosome or plasmid	AmpC, CMY-2

The side chain: Determining properties

The side chain modifications determine:

• Acid stability — Can it survive gastric acid? (oral absorption)
• β-Lactamase stability — Is it protected from enzymatic destruction?
• Spectrum of activity — Which bacteria are susceptible?
• Protein binding — How much free drug is available?
• Tissue penetration — Where does the drug distribute?
• Cross-reactivity - Risk of allergies?

All the different penicillin classes arise from different side chain modifications. A bulky side chain near the β-lactam ring provides steric protection against β-lactamases.

6-Aminopenicillanic Acid (6-APA)

• The penicillin nucleus (core structure without side chain)
• Isolated from Penicillium chrysogenum fermentation
• Allowed creation of semisynthetic penicillins
• Enabled systematic modification of the side chain
• Led to development of all modern penicillins

The isolation of 6-APA was a breakthrough. By attaching different side chains to this nucleus, chemists could create penicillins with different properties. This is how we got methicillin, ampicillin, and all the other semisynthetic penicillins.

PART 2: Mechanism of Action

Now we’ll explore how penicillins actually kill bacteria. Understanding the mechanism helps explain why they’re bactericidal and why certain bacteria are resistant.

Overview: How penicillins work

• Target: Final step of peptidoglycan synthesis
• Mechanism: Penicillins inhibit bacterial cell wall synthesis by binding to penicillin-binding proteins (PBPs)
• Result: Weakened cell wall
• Outcome: Osmotic lysis and cell death
• Effect: Bactericidal (kills bacteria)

Penicillins are bactericidal, not bacteriostatic. However, they only kill actively growing bacteria that are synthesizing new cell wall.
Dormant bacteria (persister cells) can survive penicillin treatment.

Why bacteria need cell walls

• Bacteria have high internal osmotic pressure
• Without cell wall → osmotic lysis
• Cell wall = peptidoglycan polymer
• Gram-positive: thick layer (50-100 molecules)
• Gram-negative: thin layer (1-2 molecules) + outer membrane
• Human cells have no cell wall → selective toxicity

This is why penicillins have such a good safety profile. They target a structure (peptidoglycan) that human cells don’t have. The selective toxicity ratio is excellent. Video source: https://pdb101.rcsb.org/

Peptidoglycan structure

• Backbone: Alternating NAG-NAM disaccharides
  • NAG = N-acetylglucosamine
  • NAM = N-acetylmuramic acid
• Pentapeptide stems: Attached to NAM
  • Terminate in D-Ala-D-Ala
• Cross-links: Connect adjacent chains
  • Provide strength and rigidity

The cross-links are what give peptidoglycan its strength. Without cross-linking, the cell wall is weak and the bacterium lyses. This is exactly what penicillins prevent.

Transpeptidation: The target reaction

1. Pentapeptide ends in D-Ala-D-Ala
2. Transpeptidase (PBP) binds to D-Ala-D-Ala
3. Forms covalent intermediate with penultimate D-Ala
4. Terminal D-Ala is released
5. Cross-link formed with adjacent chain
6. Penicillin mimics D-Ala-D-Ala and gets stuck

Penicillin structurally mimics the D-Ala-D-Ala substrate. When the PBP tries to bind penicillin, it forms a stable covalent complex instead of completing the reaction. The enzyme is “stuck.”

What are penicillin-binding proteins (PBPs)?

• Membrane-bound enzymes in all bacteria
• Catalyze final steps of cell wall synthesis
• Serine proteases — related to β-lactamases!
• Named for their ability to bind penicillin
• Multiple PBPs in each bacterial species
• Different PBPs have different functions

The evolutionary relationship between PBPs and β-lactamases is important. β-lactamases evolved from PBPs but developed the ability to rapidly hydrolyze (rather than be inhibited by) β-lactams.

Classes of PBPs

Class	Size	Function
High-MW Class A	>50 kDa	Bifunctional: transglycosylase + transpeptidase
High-MW Class B	>50 kDa	Transpeptidase only
Low-MW	<50 kDa	Carboxypeptidases

High-molecular-weight PBPs are the essential targets. Class A PBPs can do both glycan chain elongation and cross-linking. Class B only do cross-linking. Low-MW PBPs are carboxypeptidases that trim the peptidoglycan.

PBP functions in E. coli

PBP	Function	Inhibition result
PBP1a/1b	Transglycosylase + transpeptidase	Rapid cell lysis
PBP2	Cell elongation, rod shape	Round cells (cocci)
PBP3	Septum formation, cell division	Long filaments
PBP4-6	Carboxypeptidases	Minor effects

Different β-lactams have different affinities for different PBPs, which explains why some cause rapid lysis while others cause filamentation. The essential PBPs are 1, 2, and 3.

PBPs vs β-Lactamases: Key difference

PBPs (Cell Wall Synthesis)

• Bind penicillin tightly
• Slow deacylation rate
• Enzyme stays inhibited
• = Antibacterial effect

β-Lactamases (Resistance)

• Bind penicillin
• Fast deacylation rate
• Enzyme regenerates quickly
• = Drug destruction

This is the crucial biochemical distinction. Both enzymes bind penicillin via an active-site serine. The difference is how fast they release it. PBPs hold on (inhibition); β-lactamases let go quickly (resistance).

Tolerance vs Resistance

• Resistance: Bacteria grow in presence of antibiotic
  • MIC is high
• Tolerance: Bacteria survive but don’t grow
  • MIC is low (susceptible)
  • MBC is high (not killed)
  • Examples: stationary phase cells, persisters in biofilm

Tip

Tolerance explains why some infections relapse despite “susceptible” organisms

This distinction is important clinically. A “susceptible” organism may still cause treatment failure if many cells are tolerant (not actively growing). This is relevant for endocarditis and other difficult infections.

PART 3: Resistance Mechanisms

Four mechanisms of β-Lactam resistance

1. β-Lactamase production — enzymatic destruction
2. Decreased permeability — porin mutations
3. Efflux pumps — active drug removal
4. Altered PBPs — low-affinity binding

Warning

Multiple mechanisms often coexist, especially in MDR gram-negatives

Mechanism 1: β-Lactamases

• Most common resistance mechanism
• Enzymes that hydrolyze the β-lactam ring
• Open the ring → inactive compound
• Gram-positive: Secreted extracellularly
• Gram-negative: Located in periplasmic space
• Can be chromosomal or plasmid-encoded

The location matters. In gram-negatives, β-lactamases are concentrated in the periplasm where they can intercept drugs before they reach PBPs. In gram-positives, they’re secreted and protect nearby bacteria too.

β-Lactamase mechanism

1. β-Lactamase binds penicillin (like a PBP)
2. Forms acyl-enzyme intermediate
3. Rapid hydrolysis — water attacks the bond
4. Ring opens, releasing penicilloic acid
5. Enzyme regenerates immediately
6. Cycle repeats (catalytic)

One β-lactamase molecule can destroy thousands of penicillin molecules. This is why even low levels of β-lactamase can cause resistance. The key is the fast deacylation rate compared to PBPs. Video source: https://pdb101.rcsb.org/

Ambler classification of β-Lactamases

Class	Active Site	Mechanism	Examples
A	Serine	Acyl intermediate	TEM, SHV, CTX-M, KPC
B	Zinc (metallo)	Direct hydrolysis	NDM, VIM, IMP
C	Serine	Acyl intermediate	AmpC, CMY
D	Serine	Acyl intermediate	OXA enzymes

The Ambler classification is based on amino acid sequence. Classes A, C, D are serine enzymes (evolved from PBPs). Class B uses zinc ions and has a completely different mechanism. This matters for inhibitor selection.

Class A β-Lactamases: The most common

• TEM-1: Most common plasmid enzyme worldwide
• SHV-1: Common in Klebsiella
• CTX-M: Dominant ESBL globally
• KPC: Carbapenemase (major threat)
• Generally inhibited by clavulanic acid
• Exception: KPC (not well inhibited)

TEM-1 was one of the first plasmid-encoded β-lactamases discovered. CTX-M-15 has become the dominant ESBL worldwide over the past 20 years. KPC is concerning because it hydrolyzes carbapenems.

Extended-Spectrum β-Lactamases (ESBLs)

• Definition: Class A enzymes that hydrolyze extended-spectrum cephalosporins and aztreonam
• Evolution: Point mutations in TEM/SHV expanded spectrum
• CTX-M family: Now most common ESBL
• Inhibited by clavulanic acid (in vitro)
• Plasmid-encoded — spread easily

Warning

Clinical outcomes with BLI combinations may be unpredictable for serious ESBL infections

ESBLs arose from mutations in TEM and SHV that expanded their substrate range. CTX-M enzymes are a separate family. Despite in vitro inhibition by clavulanic acid, treatment failures occur with serious infections.

Class B: Metallo-β-Lactamases (MBLs)

• Use zinc ions instead of serine
• Hydrolyze all β-lactams EXCEPT aztreonam
• NOT inhibited by current approved inhibitors
• Examples: NDM-1, VIM, IMP
• Major global health threat
• Limited treatment options

Important

The “aztreonam loophole” — MBLs cannot hydrolyze monobactams

NDM (New Delhi metallo-β-lactamase) is spreading globally. The lack of effective inhibitors is a major problem. The aztreonam exception is being exploited clinically (aztreonam + avibactam combinations).

Class C: AmpC β-Lactamases

• Cephalosporinases — preferentially hydrolyze cephalosporins

• Often chromosomally encoded

• Can be inducible (expressed when exposed to β-lactams)

• Classic “SPACE” organisms: Serratia, Pseudomonas, Acinetobacter, Citrobacter, Enterobacter

• Revised “HECK-YES” organisms: Hafnia, Enterobacter, Citrobacter (freundii complex), Klebsiella (aerogenes), Yersinia, Enterobacter, Serratia

• NOT inhibited by clavulanic acid

• Inhibited by avibactam, relebactam, vaborbactam

The SPACE organisms have inducible chromosomal AmpC. However, the SPACE mnemonic is overly broad and partly incorrect as not all listed organisms have clinically relevant inducible AmpC. It lumps together organisms with very different resistance risks and it leads to overtreatment (e.g., unnecessary carbapenems). The “HECK-YES” organisms have clinically meaningful inducible chromosomal AmpC β-lactamases. Exposure to certain β-lactams (e.g., ceftazidime) can induce high-level expression. Cefepime is more stable to AmpC hydrolysis than ceftriaxone.

Class D: OXA enzymes

• Named for ability to hydrolyze oxacillin
• Heterogeneous group with variable spectra
• Some are carbapenemases (OXA-48, OXA-23)
• OXA-48: Poorly inhibited by most inhibitors
  • Exception: Avibactam inhibits OXA-48
• Increasingly important in Acinetobacter (OXA-23)

OXA-48 is a challenging enzyme because it doesn’t dramatically elevate MICs to carbapenems but does cause clinical failures. It’s important in K. pneumoniae in certain geographic areas.

Carbapenemases: The greatest threat

Enzyme	Class	Distribution	Inhibitors
KPC	A	Americas, worldwide	Avibactam, vaborbactam, relebactam
NDM	B	South Asia, global	None currently approved
VIM	B	Europe, global	None currently pproved
OXA-48	D	Middle East, Europe	Avibactam

Carbapenemases compromise our “last-resort” antibiotics. Geographic distribution varies. KPC is most common in the US. NDM is dominant in South Asia. Treatment options are limited.

Mechanism 2: Decreased permeability

• Relevant only for gram-negative bacteria
• Outer membrane = barrier to drug entry
• Porins = channels for drug entry
• Porin mutations/loss → reduced drug entry
• Common in Pseudomonas aeruginosa
  • OprD loss → carbapenem resistance

The gram-negative outer membrane is an additional barrier. Drugs must pass through porin channels. Loss or modification of porins reduces drug accumulation. This alone usually causes low-level resistance.Image source: https://pdb101.rcsb.org/

Mechanism 3: Efflux Pumps

• Active transport of drugs out of the cell
• Use energy (proton motive force or ATP)
• Can be constitutive or inducible
• Often have broad substrate specificity
• Example: MexAB-OprM in P. aeruginosa
• Contribute to multidrug resistance

Efflux pumps often work with other mechanisms. In Pseudomonas, efflux + β-lactamases + porin loss creates high-level resistance. Efflux pumps can affect multiple drug classes simultaneously. Image source: https://pdb101.rcsb.org/

Mechanism 4: Altered PBPs

• PBPs with low affinity for β-lactams
• Drug binds poorly → ineffective inhibition
• Examples:
  • MRSA: Acquires PBP2a (mecA gene)
  • Penicillin-resistant pneumococci: Mosaic PBP genes
  • Enterococcus faecium: Low-affinity PBP5

Important

MRSA resistance is NOT due to β-lactamases — it’s due to PBP2a

This is a critical concept. MRSA produces a novel PBP (PBP2a) with very low affinity for most β-lactams. No amount of β-lactamase inhibitor will help. That’s why MRSA requires different antibiotics.Video source: https://pdb101.rcsb.org/

MRSA: Mechanism of resistance

1. MRSA carries mecA gene (on SCCmec element)
2. mecA encodes PBP2a (also called PBP2’)
3. PBP2a has very low affinity for β-lactams
4. Native PBPs are inhibited, but PBP2a continues working
5. Cell wall synthesis continues
6. All β-lactams ineffective (except ceftaroline, ceftobiprole)

This mechanism explains why we need different drugs for MRSA. The newer cephalosporins ceftaroline and ceftobiprole have enhanced binding to PBP2a, which is why they’re active against MRSA.

PART 4: Classification of Penicillins

Now we’ll classify penicillins by their antibacterial spectrum. Understanding the classes helps with drug selection.

Five classes of penicillins

1. Natural penicillins — Penicillin G, Penicillin V
2. Penicillinase-resistant — Nafcillin, Oxacillin, Dicloxacillin
3. Aminopenicillins — Ampicillin, Amoxicillin
4. Carboxypenicillins — Ticarcillin (obsolete)
5. Ureidopenicillins — Piperacillin

Each class has different side chain modifications. The modifications determine spectrum, stability, and pharmacokinetics. We’ll discuss each class in detail.

Class 1: Natural penicillins

Penicillin G (parenteral) and Penicillin V (oral)

• The original penicillins
• Narrowest spectrum but highest potency against susceptible organisms
• Acid-labile (Pen G) vs Acid-stable (Pen V)
• Susceptible to β-lactamases
• Still drugs of choice for many infections

Natural Penicillins: Spectrum

Excellent activity:

• Streptococcus pyogenes (Group A strep)
• Streptococcus agalactiae (Group B strep)
• Streptococcus pneumoniae (susceptible strains)
• Neisseria meningitidis
• Treponema pallidum (syphilis)
• Most oral anaerobes
• Listeria monocytogenes

Group A strep has essentially 100% susceptibility to penicillin. Penicillin G remains first-line for syphilis. These are important “penicillin G diseases” to remember.

Natural Penicillins: Clinical uses

Infection	Drug of Choice
Group A strep pharyngitis	Penicillin V
Syphilis (all stages)	Penicillin G
Neurosyphilis	IV Penicillin G
Meningococcal meningitis	Penicillin G
Actinomycosis	Penicillin G
Gas gangrene (C. perfringens)	Penicillin G

These are infections where natural penicillins remain preferred.
For syphilis, penicillin G is still the drug of choice alhough outcomes may be similar with ceftriazone.
Penicillin-allergic patients with syphilis should undergo desensitization.

Class 2: Penicillinase-resistant penicillins

Nafcillin, Oxacillin, Dicloxacillin, Flucloxacillin

• Bulky side chains create steric hindrance
• Resist hydrolysis by staphylococcal β-lactamases
• Spectrum: MSSA and streptococci
• No gram-negative activity
• NOT effective against MRSA

The bulky side chain physically blocks β-lactamase access to the β-lactam ring. Remember: they don’t work against MRSA because MRSA resistance is PBP-mediated, not β-lactamase-mediated.

Antistaphylococcal penicillins: Details

Drug	Route	Protein Binding	Elimination	Special Considerations
Nafcillin	IV	90%	Hepatic	Hypokalemia, phlebitis
Oxacillin	IV	90%	Mixed	Hepatotoxicity
Dicloxacillin	PO	96%	Renal/hepatic	Highest protein binding
Flucloxacillin	PO/IV	96%	Renal	Not available in US

Nafcillin is primarily hepatically eliminated, so no renal dose adjustment needed.
Oxacillin can cause hepatotoxicity. Dicloxacillin has the highest protein binding of any penicillin.

Class 3: Aminopenicillins

The amino group on the side chain improves penetration through gram-negative outer membranes.
Amoxicillin’s hydroxyl group improves oral absorption. Both are susceptible to β-lactamases.

Aminopenicillins: Expanded spectrum

• Same as natural penicillins PLUS:
  • Enterococcus faecalis
  • Haemophilus influenzae (non-β-lactamase producing)
  • Escherichia coli (non-β-lactamase producing)
  • Proteus mirabilis
  • Salmonella and Shigella spp.
  • Listeria monocytogenes

The expanded gram-negative coverage is useful but limited by β-lactamase production.
Many E. coli are now resistant due to β-lactamases. Enterococcal coverage makes ampicillin important for certain infections.

Ampicillin vs amoxicillin

Property	Ampicillin	Amoxicillin
Oral absorption	30-55%	74-92%
Effect of food	Decreased	None
Preferred route	IV	Oral
Bioequivalence	—	Better than ampicillin

Tip

For oral therapy, amoxicillin is almost always preferred

Oral amoxicillin achieves much better levels than oral ampicillin.
Unless IV is needed, amoxicillin is the better choice.

Classes 4 & 5: Antipseudomonal penicillins

Carboxypenicillins (ticarcillin) — largely obsolete Ureidopenicillins (piperacillin) — in wide use

• Extended gram-negative spectrum
• Activity against Pseudomonas aeruginosa
• Excellent anaerobic coverage
• Susceptible to β-lactamases
• Now used with β-lactamase inhibitors

Spectrum summary table

Class	Gram+	Gram-	Pseudomonas	Anaerobes	MRSA
Natural	+++	-	-	++	-
Antistaphylococcal	++ (Staph)	-	-	-	-
Aminopenicillins	++	+	-	++	-
Antipseudomonal	+	++	++	+++	-

This summary helps with quick selection. Note that NONE of the penicillins work against MRSA.
Antistaphylococcal penicillins are narrow but highly effective against MSSA.

PART 5: Pharmacokinetics

Oral absorption

Penicillin	Absorption (%)	Food Effect
Penicillin V	60	None
Ampicillin	30-55	Decreased
Amoxicillin	74-92	None
Dicloxacillin	37	Decreased
Flucloxacillin	44	Decreased

:::{style=“width: fit-content; margin: auto;”} ::: {.callout-tip} Amoxicillin has the best oral bioavailability ::: :::

Absorption varies dramatically. This determines which penicillins can be used orally.
Amoxicillin is exceptional with 74-92% absorption unaffected by food. Lower doses (e.g., 250–500 mg)→ High bioavailability (~85–95%), fairly linear but with higher doses (e.g., ≥1000 mg)→ Reduced fractional absorption (bioavailability drops to ~60–70%)→ but total exposure still increases, but less than dose-proportional.

Most others oral peneicllins should be taken on an empty stomach.

Protein binding

• Ranges from 17% (aminopenicillins) to 97% (dicloxacillin)
• Only free drug is active
• High protein binding:
  • Reduces active drug concentration
  • Prolongs half-life
  • May reduce tissue penetration
• Clinical significance debated

Protein binding affects interpretation of serum levels. A drug with 96% protein binding has only 4% free drug.
Whether this is clinically significant is debated, but it may matter for deep-seated infections.

Distribution

• Generally good tissue penetration
• Achieve therapeutic levels in:
  • Lung, liver, kidney, muscle
  • Pleural, peritoneal, synovial fluid
  • Bone (variable)
• Poor penetration without inflammation:
  • CNS, eye, prostate

Penicillins are polar molecules with limited lipid solubility. They distribute well to most tissues but penetrate the CNS poorly. Inflammation increases CNS penetration (important for meningitis treatment).

CNS Penetration

Condition	Penicillin G	Ampicillin
Normal meninges	<1%	<1%
Inflamed meninges	5-10%	13-14%

Note

Inflammation is required for adequate CNS levels. As meningitis resolves, drug penetration decreases.

This has clinical implications. Meningitis treatment requires high doses because penetration is still limited even with inflammation. As treatment succeeds and inflammation decreases, drug levels in CSF decrease too.

Elimination

• Most penicillins: Primarily renal excretion
  • Glomerular filtration + tubular secretion
  • Short half-lives (0.5-1.5 hours)
  • Probenecid can block tubular secretion
• Exceptions:
  • Nafcillin: Primarily hepatic
  • Oxacillin: Mixed hepatic/renal

The short half-lives necessitate frequent dosing (every 4-6 hours).
Renal impairment requires dose adjustment for most penicillins. Nafcillin is the exception—no renal adjustment needed.

Renal dosing: When to adjust

CrCl (mL/min)	Adjustment Needed?
>50	Usually no
30-50	Consider for some agents
10-30	Yes for most agents
<10	Yes, significant reduction
Hemodialysis	Dose after dialysis

The threshold for dose adjustment varies by agent. Natural penicillins and aminopenicillins need adjustment below 30 mL/min. Antistaphylococcal penicillins (except oxacillin) generally don’t need adjustment.

Dosing in renal failure

Agent	CrCl 10-29	Hemodialysis
Penicillin G	75% dose	Dose post-HD
Ampicillin	0.5-2g q12h	0.5-1g q12-24h
Amoxicillin	500mg q12h	500mg q12-24h
Piperacillin	3g q8-12h	3g q12h
Nafcillin	No change	No change

Nafcillin doesn’t require renal adjustment. For dialysis patients, give the dose after dialysis to avoid drug removal.
Extended-spectrum agents like piperacillin require adjustment.

Optimizing β-Lactam dosing

• β-Lactams are time-dependent killers
• Efficacy correlates with %T>MIC (time above MIC)
• Target: 40-70% of dosing interval above MIC
• Strategies to optimize:
  • More frequent dosing
  • Extended infusions (3-4 hours)
  • Continuous infusions

This PK/PD concept is fundamental. Unlike aminoglycosides (concentration-dependent), β-lactams need to maintain levels above MIC for as long as possible. Extended infusions are now standard for piperacillin-tazobactam.

Extended infusion dosing

Example: Piperacillin-tazobactam

Traditional	Extended Infusion
4.5g over 30 min q6h	4.5g over 4 hours q8h
Higher peak, lower trough	Lower peak, higher trough
Less time above MIC	More time above MIC

Tip

Extended infusion may improve outcomes, especially for organisms with higher MICs

Extended infusion dosing has become standard of care in many institutions. The total daily dose may be similar or even lower, but the pharmacodynamics are superior.

PART 6: Adverse effects

Overview of adverse effects

• Hypersensitivity reactions — most important
• Gastrointestinal effects
• Hematologic effects
• Neurologic effects
• Nephrotoxicity
• Hepatotoxicity
• Electrolyte disturbances

Hypersensitivity is the most clinically significant concern.
Most other effects are dose-related or specific to certain agents.

Hypersensitivity: Types

Type	Timing	Mechanism	Manifestations
Type I	Minutes-hours	IgE-mediated	Anaphylaxis, urticaria, angioedema
Type II	Days	Antibody-mediated	Hemolytic anemia, cytopenia
Type III	1-3 weeks	Immune complex	Serum sickness, drug fever
Type IV	Days-weeks	T-cell mediated	Maculopapular rash, contact dermatitis

Type I reactions are the most dangerous and require avoidance. Type IV reactions (delayed maculopapular rashes) are most common and generally don’t preclude future use.

Penicillin allergy: By the numbers

• ~10% of patients report penicillin allergy
• <1% have true IgE-mediated allergy when tested
• ~2% will react if challenged
• True anaphylaxis: <0.01%
• Allergy often wanes over time
  • 50% lose sensitivity within 5 years
  • 80% lose sensitivity within 10 years

These numbers are important. Most “penicillin allergy” labels are inaccurate. Over-labeling leads to use of broader-spectrum, more toxic, and more expensive antibiotics. De-labeling is an important intervention.

Penicillin allergy de-labeling

1. History assessment — Was it really an allergic reaction?
2. Risk stratification — High risk vs low risk features
3. Skin testing — Detects IgE-mediated allergy
4. Graded oral challenge — Confirms tolerance
5. Update medical record — Remove incorrect allergy label

Tip

De-labeling programs are safe and improve patient care

Penicillin allergy de-labeling is an important quality improvement initiative. Many patients can safely receive penicillins after proper evaluation. This improves outcomes and reduces costs.

Cross-reactivity with cephalosporins

• Historical estimates: 10% cross-reactivity (overestimate)
• Current data: ~1-2% cross-reactivity
• Cross-reactivity relates to side chain similarity
• Highest risk: Similar R1 side chains
  • Ampicillin → Cephalexin, Cefadroxil
• Lower risk: Dissimilar side chains
  • Ceftriaxone, cefepime

The 10% cross-reactivity rate was based on flawed data. Current evidence suggests much lower rates. Side chain similarity predicts cross-reactivity better than the β-lactam ring alone.

Other adverse effects

Effect	Most Common With	Notes
Diarrhea	Ampicillin, amoxicillin-clav	Disruption of gut flora
C. difficile	All	Risk with any antibiotic
Neutropenia	Prolonged high-dose therapy	Reversible
Seizures	High-dose penicillin G	Especially in renal failure
Interstitial nephritis	Methicillin, nafcillin	Allergic mechanism

GI effects are common with oral penicillins. Neutropenia is associated with prolonged courses (>2 weeks). Seizures occur with very high doses, especially in patients with renal impairment.

Agent-specific toxicities

• Nafcillin: Hypokalemia, phlebitis
• Oxacillin: Hepatotoxicity, interstitial nephritis
• Ampicillin: Maculopapular rash (especially with EBV)
• Amoxicillin-clavulanate: Diarrhea, hepatotoxicity
• Piperacillin: Platelet dysfunction, hypokalemia
• High-dose Penicillin G (K+ salt): Hyperkalemia

The ampicillin rash with infectious mononucleosis is not a true allergy and doesn’t predict future reactions.
Piperacillin can cause platelet dysfunction with prolonged use.

PART 7: β-Lactamase inhibitors

β-Lactamase inhibitor classes

Traditional (β-lactam)

• Clavulanic acid
• Sulbactam
• Tazobactam

Novel (non-β-lactam)

• Avibactam
• Relebactam
• Vaborbactam

The traditional inhibitors are β-lactam compounds themselves that act as “suicide substrates.”
The novel inhibitors have different mechanisms and broader coverage, including KPC and AmpC.

Mechanism: Traditional inhibitors

1. Inhibitor binds to β-lactamase active site
2. Forms stable acyl-enzyme complex
3. Complex undergoes irreversible fragmentation
4. Enzyme is permanently inactivated
5. “Suicide inhibitor” mechanism
6. One inhibitor molecule = one enzyme molecule

The traditional inhibitors sacrifice themselves to inactivate the β-lactamase. This is different from competitive inhibition—the enzyme is destroyed, not just temporarily blocked.

Traditional inhibitor combinations

Inhibitor	Partner	Formulations
Clavulanic acid	Amoxicillin	Oral, IV
Sulbactam	Ampicillin	IV
Tazobactam	Piperacillin	IV

Sulbactam has intrinsic activity against Acinetobacter.
Tazobactam is always used with piperacillin.

Spectrum of traditional inhibitors

Inhibited:

• Class A β-lactamases (TEM, SHV, many ESBLs)
• S. aureus β-lactamase
• Bacteroides β-lactamase

NOT inhibited:

• Class B (metallo-β-lactamases)
• Class C (AmpC)
• KPC (weak inhibition)

This is a critical limitation. Traditional inhibitors don’t work against AmpC or carbapenemases. That’s why piperacillin-tazobactam doesn’t work reliably against AmpC-producing organisms or KPC producers.

Novel β-Lactamase inhibitors

Inhibitor	Partner	Unique Feature
Avibactam	Ceftazidime	Inhibits KPC, OXA-48, AmpC
Relebactam	Imipenem	Inhibits KPC, AmpC
Vaborbactam	Meropenem	Inhibits KPC, AmpC

Important

None inhibit metallo-β-lactamases (NDM, VIM, IMP)

These newer combinations have expanded activity against carbapenemases and AmpC.
They’re important for treating CRE. However, the gap remains for metallo-β-lactamases.

Inhibitor spectrum summary

Inhibitor	Class A	ESBLs	KPC	AmpC	MBLs	OXA-48
Clavulanate	✓	±	✗	✗	✗	✗
Tazobactam	✓	±	✗	✗	✗	✗
Avibactam	✓	✓	✓	✓	✗	✓
Vaborbactam	✓	✓	✓	✓	✗	✗

Note that metallo-β-lactamases (MBLs) are not inhibited by any approved inhibitor. Avibactam has the broadest spectrum, including OXA-48.

MIC values: BLI combinations

Organism	Amp/Amox	Amox-Clav	Pip-Tazo
S. aureus (MSSA)	16	1	1
H. influenzae (BL+)	>16	0.5	0.06
E. coli	>16	4	2
K. pneumoniae	>16	2	4
B. fragilis	>16	0.5	2
P. aeruginosa	>16	>16	4

These MICs show how inhibitors restore activity. Note that P. aeruginosa remains relatively resistant even to pip-tazo—the MIC is at the breakpoint.
This explains why pip-tazo may fail for serious Pseudomonas infections.

Clinical limitations of BLI combinations

Warning

In vitro activity does not guarantee clinical success

• ESBL infections: Treatment failures reported with BLI combinations
• Inoculum effect: High bacterial loads may overwhelm inhibitor
• Serious infections: Carbapenems often preferred for ESBL bacteremia
• AmpC producers: Traditional inhibitors ineffective

This is an important nuance. Just because an organism tests “susceptible” to pip-tazo doesn’t mean it’s the best choice for serious infections.
The inoculum effect means that with high bacterial loads, β-lactamases can overwhelm the inhibitor.

PART 8: Clinical Application

Clinical Case 1: Skin Infection

55-year-old man with cellulitis

• Erythema, warmth, swelling of left lower leg
• No drainage, no crepitus
• No systemic symptoms
• No diabetes, no recent hospitalization
• No history of MRSA

What is the most appropriate oral therapy?

Case 1: Answer

• Likely pathogens: S. aureus, Group A strep
• MRSA risk: Low (no risk factors)
• Best choice: Dicloxacillin 500mg QID or Cephalexin 500mg QID
• Rationale: Narrow spectrum, covers MSSA + strep
• Avoid: Amoxicillin-clavulanate (too broad), Fluoroquinolones (no added benefit)

This case illustrates appropriate narrow-spectrum selection. Without MRSA risk factors, we should target MSSA and streptococci. Broader agents aren’t better and contribute to resistance.

Clinical Case 2: Pneumonia

68-year-old woman with community-acquired pneumonia

• Cough, fever, dyspnea for 3 days
• CXR: Right lower lobe infiltrate
• O2 sat 94% on room air
• COPD (mild), no recent antibiotics
• Outpatient treatment appropriate

What is the most appropriate oral therapy?

Consider the likely pathogens, resistance patterns, and guideline recommendations. Is she low or high risk for resistant organisms?

Case 2: Answer

• Most likely pathogen: S. pneumoniae
• Other possibilities: H. influenzae, atypicals
• COPD: Increases H. influenzae risk
• Best choice: Amoxicillin-clavulanate 875mg BID + Azithromycin (for atypicals)
• Alternative: Respiratory fluoroquinolone (moxifloxacin)
• Rationale: Covers pneumococcus (including most resistant strains), H. influenzae

For CAP with COPD, coverage for H. influenzae is important. Amoxicillin-clavulanate provides this. Adding a macrolide covers atypical pathogens. This follows IDSA/ATS guidelines.

Clinical Case 3: UTI

32-year-old woman with uncomplicated cystitis

• Dysuria, frequency for 2 days
• No fever, no flank pain
• No recent antibiotics
• No structural urinary abnormalities

What is the most appropriate therapy?

Is this a case where penicillins are first-line? Consider resistance patterns in uropathogens.

Case 3: Answer

• Most likely pathogen: E. coli (75-95% of uncomplicated UTIs)
• Resistance concern: >40% of E. coli resistant to ampicillin
• First-line options (per guidelines):
  • Nitrofurantoin 100mg BID x 5 days
  • TMP-SMX (if local resistance <20%)
  • Fosfomycin single dose
• Amoxicillin/ampicillin: NOT first-line due to resistance

This illustrates that penicillins aren’t always the right choice. For uncomplicated UTI, resistance to ampicillin is too high to use empirically. Nitrofurantoin has maintained good activity.

Clinical Case 4: Intra-Abdominal Infection

52-year-old man with perforated appendicitis

• Post-operative day 1, now febrile
• WBC 18,000
• CT shows pelvic abscess
• No recent antibiotics, no recent hospitalization

What is the most appropriate empiric therapy?

Consider the polymicrobial nature of intra-abdominal infections. What organisms need to be covered? What about resistance?

Case 4: Answer

• Pathogens: Gram-negatives (Enterobacterales), Anaerobes (B. fragilis), possibly Enterococcus
• Best choice: Piperacillin-tazobactam 4.5g IV q6h (extended infusion)
• Alternative: Ceftriaxone + metronidazole
• Rationale: Broad gram-negative coverage + anaerobes
• Duration: Until source controlled, then course completion

Pip-tazo is an excellent choice for community-acquired intra-abdominal infections. It covers all the expected pathogens. Extended infusion dosing optimizes pharmacodynamics.

Clinical Case 5: Endocarditis

45-year-old man with S. aureus bacteremia

• IV drug user, new murmur
• TEE: Tricuspid vegetation
• Blood culture: MSSA (oxacillin MIC 0.25)
• No contraindications to β-lactams

What is the most appropriate therapy?

This is a serious infection requiring prolonged therapy. What’s the optimal agent for MSSA endocarditis?

Case 5: Answer

• Diagnosis: MSSA tricuspid valve endocarditis
• Duration: 4-6 weeks IV therapy
• Best choice:* Cefazolin 2 g IVq8h less nephrotoxicity (Burdet et al. 2025)and possibly better outcomes (Prosty et al. 2025; McDanel et al. 2017)than Nafcillin 2g IV q4h or Cloxacillin 2g IV q4h
• NOT vancomycin: Inferior outcomes for MSSA
• Monitoring: Renal function, signs of drug fever, eosinophilia

Important

For MSSA, nafcillin/oxacillin are superior to vancomycin

This is important: vancomycin is inferior to nafcillin for MSSA infections. The antistaphylococcal penicillins penetrate better and have faster bactericidal activity. Don’t use vancomycin for MSSA just because it’s “easier.”

Clinical Case 6: Meningitis

22-year-old college student with meningitis

• Headache, fever, stiff neck, rash
• CSF: 2000 WBC (95% PMN), protein 250, glucose 20
• Gram stain: Gram-negative diplococci
• Suspected Neisseria meningitidis

What is the most appropriate therapy?

Case 6: Answer

• Diagnosis: Meningococcal meningitis
• Best choice: CCeftriazone 2 grams IV q12h
• Alternative: Penicillin G 4 million units IV q4h
• Duration: 7 days
• Chemoprophylaxis: Close contacts need rifampin, ciprofloxacin, or ceftriaxone
• Dexamethasone: Controversial for meningococcus (consider for pneumococcus)

Penicillin G remains an effective option meningococcal meningitis even through resistance rates vary (between 3-74%).Ceftriaxone is often used empirically before organism identification.

Clinical case 7: Syphilis

28-year-old man with primary syphilis

• Painless chancre on penis
• RPR reactive, TP-PA positive
• Reports penicillin allergy: “rash as a child”

What is the most appropriate management?

This case combines two important topics: syphilis treatment and penicillin allergy management.

Case 7: Answer

• Treatment of choice: Benzathine penicillin G 2.4 million units IM x 1
• Alternative Ceftriaxone 2 grams IV daily 10-14 days
• Penicillin allergy: Likely not true allergy (childhood rash)
• Management options:
  1. Penicillin skin testing → if negative, treat with penicillin
  2. Penicillin desensitization → then treat
• NOT acceptable: Doxycycline or azithromycin (inferior efficacy)

Penicillin is the only proven treatment for syphilis, especially neurosyphilis. If a patient has documented penicillin allergy, desensitization is recommended. This is one of the few infections where there’s no acceptable alternative.

Quick selection guide

Infection	First-Line Penicillin
Strep pharyngitis	Penicillin V
MSSA skin/soft tissue	Dicloxacillin
MSSA bacteremia/endocarditis	Nafcillin
CAP (outpatient, no comorbidity)	Amoxicillin
CAP (outpatient, with COPD)	Amoxicillin-clavulanate
Intra-abdominal infection	Piperacillin-tazobactam
Syphilis	Benzathine penicillin G
Listeria meningitis	Ampicillin

This quick reference summarizes the preferred penicillin for common infections. Emphasize matching the narrowest effective spectrum to the infection.

Summary and Key Takeaways

Key point 1: Structure determines function

• β-Lactam ring is essential for activity
• Side chain determines spectrum, stability, and pharmacokinetics
• Modifications create different penicillin classes
• Understanding structure explains clinical properties

Key point 2: Know the resistance mechanisms

• β-Lactamases: Most common; hydrolize β-lactam ring
• Permeability: Porin changes in gram-negatives
• Efflux: Active drug removal
• Altered PBPs: Low-affinity binding (MRSA, PRP)
• Multiple mechanisms often coexist

Key point 3: Match spectrum to pathogen

• Natural penicillins: Streptococci, syphilis, meningococcus
• Antistaphylococcal: MSSA only
• Aminopenicillins: Add enterococci, some gram-negatives
• Pip-tazo: Broad including Pseudomonas, anaerobes
• None work against MRSA

Key point 4: β-Lactamase inhibitors have limits

• Traditional inhibitors: Class A only
• NOT effective against:
  • AmpC (Class C)
  • Metallo-β-lactamases (Class B)
  • KPC (limited)
• Newer inhibitors: Broader coverage but still gaps
• In vitro activity ≠ clinical success

Key point 5: De-Label penicillin allergies

• Most reported allergies are NOT true allergies
• True IgE-mediated allergy is rare (<1%)
• Skin testing can identify true allergy
• De-labeling improves patient care
• Don’t avoid penicillins unnecessarily

References

Burdet, Charles, Nadia Saïdani, Céline Dupieux, Adrien Lemaignen, Etienne Canouï, Laure Surgers, Marc Olivier Vareil, et al. 2025. “Cloxacillin Versus Cefazolin for Meticillin-Susceptible Staphylococcus Aureus Bacteraemia (CloCeBa): A Prospective, Open-Label, Multicentre, Non-Inferiority, Randomised Clinical Trial.” The Lancet 406 (10517): 2349–59. https://doi.org/10.1016/s0140-6736(25)01624-1.

McDanel, Jennifer S, Mary-Claire Roghmann, Eli N Perencevich, Michael E Ohl, Michihiko Goto, Daniel J Livorsi, Makoto Jones, et al. 2017. “Comparative Effectiveness of Cefazolin Versus Nafcillin or Oxacillin for Treatment of Methicillin-Susceptible Staphylococcus Aureus Infections Complicated by Bacteremia: A Nationwide Cohort Study.” Clinical Infectious Diseases 65 (1): 100–106. https://doi.org/10.1093/cid/cix287.

Prosty, Connor, Dean Noutsios, Todd C. Lee, Nick Daneman, Joshua S. Davis, Nynke G. L. Jager, Nesrin Ghanem-Zoubi, et al. 2025. “Cefazolin Vs. Antistaphylococcal Penicillins for the Treatment of Methicillin-Susceptible Staphylococcus Aureus Bacteraemia: A Systematic Review and Meta-Analysis.” Clinical Microbiology and Infection 31 (8): 1272–82. https://doi.org/10.1016/j.cmi.2025.04.045.