guide
Antibiotic selection and blue-white screening
How antibiotic plates keep a plasmid in a cloning strain, and why a white colony on X-gal is a candidate rather than proof of the insert.
- Author
- EVRINTH Editorial Team
- Published
- 8 October 2026
- Updated
- 8 October 2026
- Reading time
- 9 min

Antibiotic selection and blue-white screening are two different questions that happen to share a Petri dish. Selection asks whether a cell is keeping a plasmid that confers resistance. The colour screen asks whether a short peptide encoded on that plasmid is still intact enough to complement a host enzyme. Neither question is "is this the DNA I designed?". This page is for people plating non-pathogenic E. coli K-12 cloning strains after a transformation. The culture and the competence step itself are in bacterial culture and transformation. Reagent classes sit in the reagents and chemicals catalogue, the surrounding methods on the molecular biology pathway, and any design question can be framed with the custom gene synthesis reference as an enquiry, not as a service this site claims to perform. Specifications go through the quote request.
The decision the plate is allowed to make
Use antibiotic agar to discard cells that did not take up a resistant plasmid, or that cannot express the marker. Use a blue-white overlay only when the vector and the host were built for alpha-complementation. Then stop. A clear no-DNA plate and a white, isolated colony are permission to test that colony. They are not a map of the insert. Laboratories that skip the test discover the error after a week of expression from the wrong plasmid.
The organisms in view are the same laboratory cloning strains as on the parent page: K-12 derivatives chosen for plasmid work. This is not guidance for plating clinical specimens or any organism your institution has not already accepted for the bench.
How antibiotic selection works
A resistance gene on the plasmid encodes a protein that modifies or destroys the drug, or that replaces a sensitive target. Without that protein, a sensitive host stops dividing at the dose written for that marker. With it, the transformant forms a colony. The host must actually be sensitive. A strain that already carries a chromosomal kanamycin cassette will grow on kanamycin whether or not your plasmid arrived. Read the genotype before you pour.
The drug has to survive preparation. Many antibiotics are heat-labile. They are mixed into agar or broth only after sterilisation and cooling. A plate poured with the drug already boiled is a rich medium with a label. Working concentration is specific to the marker, the copy number and the strain. Take it from the plasmid map and the antibiotic note you trust. Do not borrow a single number for every drug and every vector. Public cloning notes at Addgene show how laboratories record that pairing. Enzyme and reagent classes used to build the plasmid are surveyed in catalogues such as NEB's product list; those pages are orientation, not a transcribed protocol.
Ampicillin deserves a separate sentence because it fails in a distinctive way. The enzyme that confers resistance can leak into the agar and clear a halo. Neighbouring cells that never received the plasmid then form small satellite colonies. They look like a successful dense transformation. They are unselected growth. Pick only a well-isolated large colony, restreak onto fresh antibiotic agar if you are unsure, and do not interpret a halo of pinpricks as extra clones. Markers that inactivate the drug inside the cell, rather than destroying a cloud of it outside, give cleaner plates. They still need the right dose and plates that have not been stored until the drug is gone.
Plates age. Warm storage, light and time reduce effective dose. A plate that "used to work" can become a lawn. Date them, store them as the antibiotic requires, and keep a no-DNA control on the same batch. That control is the only honest test of the plate in front of you.
How blue and white colour is made
Many cloning vectors carry a short lacZ alpha fragment in the multiple-cloning site. The alpha peptide does not hydrolyse substrate by itself. The host supplies the rest of beta-galactosidase from a lacZ delta M15 allele. The two pieces associate, enzyme activity returns, and 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside, usually called X-gal, is cleaved to a blue insoluble pigment. An insert that interrupts the alpha reading frame usually abolishes that activity, so the colony stays the colour of the cells, which readers call white.
An inducer such as IPTG is often added because the lac promoter is repressed without it. Some vectors express the alpha fragment from a promoter that does not need IPTG. Adding inducer "because the paper did" can be unnecessary or, on a repressed system, essential. Know which promoter you have. The sequence of the cassette, if you need to check a named accession, can be looked up in GenBank. The accession does not tell you that your particular miniprep matches it.
X-gal is poorly soluble in water and is usually dissolved in a compatible organic solvent before it is spread or poured. The solvent volume should be small enough not to poison the lawn. The chromogenic substrate is light-sensitive and is not a dye you add after colonies have already been picked. Put it in or on the agar before you interpret colour.
A workflow with branch points
Pour or spread antibiotic agar and, if you are screening colour, X-gal and inducer as the vector requires. Plate a no-DNA control and, when you can, an empty-vector control beside the ligation or assembly. Incubate inverted. Read antibiotic selection first. If the no-DNA plate has colonies, stop. Colour on a non-selective plate is not selection.
If the no-DNA plate is clear and the empty vector is blue, then white colonies on the insert plates are candidates. If the empty vector is also white, the host may lack delta M15, the X-gal may be missing or degraded, or the plates were read too soon. Do not pick "whites" from a screen that cannot turn blue. If everything is blue, including wells you expected to be disrupted, consider an in-frame insert, a ligation that never accepted the insert, or an incubation so short that you are looking at background colour. Restreak a few colonies onto fresh screening agar before you trust a pale result.
| Observation | What it supports | What it does not prove |
|---|---|---|
| No-DNA plate clear | This antibiotic batch is selective against these cells | That sample colonies contain the right insert |
| Empty vector blue | Host, X-gal and alpha peptide can produce colour | That every white colony is recombinant |
| Isolated white colony | Alpha-complementation failed in that colony | Sequence, orientation or an intact open reading frame |
| Blue colony | Alpha peptide is working well enough to cleave X-gal | That the colony is empty, if a short in-frame insert is possible |
| Satellites around a large colony | Local drug depletion, especially with ampicillin | Extra legitimate transformants |
Failure modes
White colonies that miniprep as empty vector usually mean the backbone was damaged in lacZ alpha, or you picked a satellite, or the colour had not developed. Blue colonies that nevertheless contain an insert are the in-frame case; design the screen knowing your insert length, or do not use colour as the decision. A lawn means selection failed: drug omitted, cooked, expired, or the strain was resistant already. Mixed blue-white colonies that were picked with a thick loop are two clones in one tube. Restreak. Toxic inserts can disappear entirely, so a plate of only blues can mean the recombinant never formed a colony. That is a cloning result, not a cue to incubate for three days and pick tiny satellites.
Confirm with a method that sees the DNA. A diagnostic restriction digest, a PCR that spans the junction, or a Sanger read will contradict a colour call often enough that the contradiction is the point of the assay. The colour got you to a shorter list.
Safety and research-use limits
Plates of K-12 cloning strains are still biological waste. Inactivate them by the method your institution specifies before disposal. Antibiotics and X-gal solvent are chemical hazards; the solvent in particular does not belong in a biosafety cabinet as if the cabinet were a fume hood. ATCC culture guides are a public reminder that culture material comes with handling conditions. This article is not medical advice, not a diagnostic identification of an organism, and not permission to plate an isolate that has not been assessed.
Climate and the life of a plate
In a hot, dry incubator, thin agar cracks and colonies stay small. In humid weather, plates weep and X-gal can crystallise on the lid. Store antibiotic stocks and poured plates as their chemistry requires, protected from a warm bench. A power cut that warms a cold room full of ampicillin plates is a reason to include a no-DNA control with the next session, not a reason to trust last month's pour. Develop colour at the temperature the strain grows, then move plates somewhere cooler if you want the pigment to finish without the colonies overgrowing. Write the hour you scored them. "White" without a time is an incomplete observation.
What to send with an enquiry
Name the antibiotic, the need for X-gal or inducer, the host genotype, and the orthogonal check you will run after the screen. Ask for grade and specification. Use the quote request and the reagents and chemicals catalogue. The custom gene synthesis reference is only a prompt if the insert itself must be designed. It does not mean a synthesis run is being offered as a performed service on the strength of this paragraph.
Read an antibiotic plate and a blue-white screen without over-calling a colony
- 01Match marker, drug and hostConfirm the plasmid resistance gene, the antibiotic in the agar, and that the host cannot grow on that drug without the plasmid. Heat-labile drugs go in only after the medium has cooled.
- 02Confirm the strain can actually turn blueBlue-white screening needs a lacZ alpha donor on the vector and a lacZ delta M15 host. Plate an empty vector on the same X-gal agar. If that control stays white, colour cannot be read on the sample plates.
- 03Score colour only after it has had time to developIncubate as the method states, then hold plates cold if your laboratory uses that step so the blue pigment finishes. A colony that is white at hour twelve can be pale blue at hour twenty.
- 04Treat a white colony as a candidate and test the insertPick from the centre of a well-isolated white colony, away from satellites. Check size, orientation or sequence before you call the construct correct. Colour is a screen, not an identity assay.
Questions from the bench
Does a white colony prove the insert is present and correct?
No. Blue-white screening is not proof of a correct insert. A white colony means alpha-complementation failed often enough that the colony did not turn blue. That can be a disrupted lacZ alpha gene, and it can also be a damaged empty vector, a colony that never received the right plasmid, or a colony scored before the pigment developed. Sequence, a diagnostic digest or a PCR across the junction is what identifies the insert.
Why are there small colonies crowded around one large colony?
On ampicillin plates, a resistant colony can secrete beta-lactamase that destroys the drug in a local halo. Sensitive cells then grow as satellites in that halo. They are not extra transformants. Pick the centre of the large colony, or switch to a marker whose resistance does not leak in the same way, and do not store ampicillin plates until the drug is exhausted.
Can a colony be blue and still contain an insert?
Yes. A small in-frame insert can leave a working alpha peptide, so the colony turns blue even though foreign DNA is present. The reverse error is treating every blue colony as empty. If the insert is short or the junctions were designed in frame, colour is a weak screen and a restriction digest or a sequence read has to carry the claim.
What should an enquiry about screening reagents include?
The antibiotic class, whether you need X-gal and an inducer, the host genotype, and how you will confirm candidates after the colour screen. Ask for the written specification of each chemical. A family name on a catalogue page is not a concentration table and not evidence about any particular lot.
References
Manufacturer names identify published method classes. Trademarks remain with their owners. Catalogue records on this site are independent references for enquiry. They are not a statement of inventory, distribution rights or a supply commitment. This page is educational. It is not medical advice, a diagnostic protocol or a biosafety approval.
Catalogue
Related products and categories
These links follow the subject of the article into published manufacturer references. A listing is a reference for an enquiry, not a statement of stock or distribution rights.
Continue in this cluster
Related reading
Bacterial culture and transformationHow non-pathogenic E. coli cloning strains are grown and how plasmid DNA is introduced, with the controls that make a colony meaningful.
A glossary of bench microbiology termsPlain definitions of colony, lawn, streak, pellet, transformant and the other bench words a cloning afternoon depends on.
Antibiotic stocks and outdated platesHow to retire antibiotic stocks and agar plates for cloning strains when heat, age or drying would fake a lawn or an empty plate.