Ampicillin vs Kanamycin: What Each Plasmid Marker Protects

Ampicillin versus kanamycin plasmid selectable markers are two common E. coli cassettes with different failure modes. Ampicillin fails when a secreted enzyme eats the drug around a true colony. Kanamycin fails less that way and asks for a longer recovery. Neither cassette is a quality ranking of the backbone. Labs pick them to keep transformants — and to predict what the plate will lie about.

What These Two Plasmid Markers Are

Each marker is a selectable ORF that lets a transformed cell grow in the matching antibiotic. Ampicillin resistance is usually bla, a beta-lactamase. Kanamycin resistance is usually NPTII (aminoglycoside phosphotransferase). The audience is anyone reading a resistance annotation before pouring plates. The purpose is selection, not a prettier map.

Addgene’s antibiotic 101 is the chooser table behind this pair. This page keeps the two everyday cassettes and the ways they fail. It does not rank every antibiotic on that table.

One Enzyme Leaves the Cell; the Other Does Not

Addgene’s amp problem is physical: beta-lactamase is excreted. A true AmpR colony clears ampicillin in a halo. Plasmid-free neighbors then appear as small satellite colonies. Overnight liquid cultures have the same leak — the drug is gone, and plasmid-free cells can accumulate. A plate read after a long weekend is more likely to lie than a plate read at 16 hours. That is a timing fact, not a reason to declare ampicillin useless.

NPTII inactivates kanamycin by phosphorylation inside the cell. Addgene does not describe a satellite-halo mechanism for kan. That is why kan plates are usually cleaner after a long incubation, not because kanamycin is morally better. The drug stays in the agar unless a resistant cell takes it up and inactivates it locally. Neighbors do not get a cleared halo to grow in.

Recovery Time and Dual-Use Are Operational, Not Moral

Addgene notes a longer recovery — about 60 minutes — before plating kanamycin transformants, and that NPTII can also support G418 selection in eukaryotic cells. Those are protocol facts. They are not reasons to retire ampicillin. Typical working concentrations are commonly cited in the 50–100 µg/ml (amp) and 30–50 µg/ml (kan) bands. Treat those as typical, not as a law for every strain. A short recovery that works for amp can starve a kanamycin plate of true colonies. That looks like a failed ligation when the cassette simply needed more time to express NPTII.

A Marker Is Not an Ori and Not an Expression Cassette

Copy number is the origin’s job. That split lives on the pBR322 versus pUC page. Promoters and whether the backbone is a cloning vector or an expression vector live on the feature page. Two plasmids in one cell still need two markers as well as two origin families; compatibility is the p15A page. A high-copy pUC plasmid with bla still leaks ampicillin. Copy number does not rewrite the marker’s failure mode.

Carbenicillin Is a Related Beta-Lactam, Not a Third Marker Class

Addgene treats carbenicillin as a more stable beta-lactam still aimed at bla. Switching to carb can reduce satellites. It does not turn the cassette into kanamycin. After the marker ORF is named, a map such as ZettaGene is one place to check the annotation. It does not pick the drug. The takeaway is the failure mode — leakage versus retention — not a better-antibiotic trophy.

Frequently Asked Questions

Why do ampicillin plates grow satellite colonies?

Excreted beta-lactamase lowers the drug around a true colony, so plasmid-free cells can appear nearby.

Is kanamycin always the better plasmid marker?

No. It is more stable on the plate and needs a longer recovery. Marker choice also has to match the second plasmid and any eukaryotic step.