Superbug is a journalistic label rather than a microbiological category, but the organisms it usually points at are specific and the list is short. What they share is not one clever trick; it is several mechanisms stacked in the same isolate, so that the options remaining after the first failure are already gone too.
Which organisms earn the name
In practice the term covers the ESKAPE group, an acronym coined because these organisms escape the effects of common antibacterials:
| Organism | Gram | What makes it hard |
|---|---|---|
| Enterococcus faecium | Positive | vanA operon remodels the peptidoglycan target away from glycopeptides |
| Staphylococcus aureus | Positive | mecA gives a low affinity PBP2a, removing the beta-lactam class |
| Klebsiella pneumoniae | Negative | Carbapenemases plus porin loss, frequently on the same plasmid background |
| Acinetobacter baumannii | Negative | Low permeability, AdeABC efflux, OXA type carbapenemases, environmental persistence |
| Pseudomonas aeruginosa | Negative | Intrinsic low permeability, multiple RND pumps, inducible AmpC, biofilm formation |
| Enterobacter species | Negative | Inducible AmpC that selects rapidly under beta-lactam pressure |
Others appear in the same conversation for good reasons: drug resistant Neisseria gonorrhoeae, where the remaining options are genuinely few; Mycobacterium tuberculosis, where multidrug and extensively drug resistant strains are a distinct field; and Clostridioides difficile, which is a consequence of antibiotic use rather than a resistance problem in the usual sense.
Why stacking is the real problem
A single mechanism is usually survivable. A carbapenem resistant K. pneumoniae isolate often carries a carbapenemase, a porin change, an extended spectrum beta-lactamase and a plasmid borne quinolone resistance determinant at the same time. Each one alone would leave a route through. Together they close most of them, and the remaining agents are older, more toxic or less well supported by outcome data.
This is also why screening against a single organism is misleading. The useful question is which of a panel a compound holds, and by which mechanism it loses the rest. The antibiotic resistance page sets out the four mechanism classes that do the stacking.
What is left when the usual agents fail
- Colistin and the polymyxins, reintroduced out of necessity, with real toxicity and now with plasmid borne mcr resistance in circulation.
- Newer beta-lactam and inhibitor combinations such as ceftazidime-avibactam and meropenem-vaborbactam, covered on the carbapenem resistance page.
- Cefiderocol, a siderophore cephalosporin that uses iron transport to get itself across the outer membrane, which is a neat answer to the uptake problem.
- Gram-positive agents including linezolid, daptomycin and the newer glycopeptides, for MRSA and VRE.
What this means for discovery
Three practical consequences for a program:
- Activity against a laboratory reference strain says little. Circulating isolates carry the stack.
- The mechanism a compound loses to matters more than the MIC it loses at, because it determines whether chemistry can recover the series. Efflux and permeability are workable. Target mutation usually is not.
- Uptake beats affinity as a design constraint on Gram-negatives, for the reasons set out under Gram-negative bacteria.
Screen against the panel
The screen below runs the full ESKAPE set. It returns a predicted MIC band per organism, the expected resistance mechanism, and a confidence level that tells you how much published data sits behind each row. It is a computational prediction and it is research use only, which is exactly the right status for a triage step.