Written and maintained by CASRAI Editorial Board
Last updated
Short answer
A lox-stop-lox (LSL, or loxP-STOP-loxP) cassette is a transcriptional termination sequence flanked by two loxP sites and placed in front of a gene or reporter. The STOP blocks expression until Cre recombinase removes it, so Cre switches the allele on. It is the standard design for conditional gain-of-function alleles.
Definition and scope: lox-stop-lox, Cre-lox and loxP
Three search terms describe layers of one system. loxP is the short DNA recognition site. Cre recombinase is the enzyme that recombines DNA between two loxP sites; “cre lox”, “cre-loxP” and “Cre-lox recombination” all name this pairing. A lox-stop-lox cassette is one specific way of using it: two loxP sites placed around a STOP element so that recombination removes the block. The loxP site itself and the way site orientation decides between excision and inversion are covered in our guide to Cre-lox driver lines and Cre toxicity, which also deals with choosing a Cre driver; this page covers the cassette.
The Ingenious Targeting Laboratory glossary (a gene-targeting service provider) describes the cassette as blocking gene expression until Cre excises it, and states that the STOP and polyA termination sequences, often accompanied by selectable markers, sit between the loxP sites. It lists conditional gene activation and lineage tracing as uses.
The logic is the reverse of a conditional knockout. In a knockout, loxP sites flank an essential exon and Cre deletes it, so the gene is lost. With an LSL cassette the gene is intact but silent, and Cre removal of the STOP makes it active. Gain-of-function and reporter alleles use the second design.
How Cre switches the allele on: a worked example, LSL-K-ras G12D
The most-cited demonstration is Jackson et al. (2001, Genes Dev). The authors built a mouse strain, LSL-K-ras G12D, in which expression of oncogenic K-ras is controlled by a removable transcriptional termination Stop element flanked by loxP sites. The endogenous K-ras locus is targeted, so after excision the mutant protein is made at endogenous levels rather than from an over-expressing transgene. Missense mutations at codons 12, 13 and 61 reduce GTPase activity and cause constitutive signalling; G12D is a codon 12 change.
Excision was triggered with AdenoCre, an adenovirus expressing Cre, delivered intranasally. The authors report that this gave control over the timing and multiplicity of tumour initiation: the allele stays silent in every cell until a Cre-expressing virus reaches it, and the amount of virus sets how many cells are switched on.
How excision was confirmed: the 265 bp and 305 bp products
The same paper offers a concrete genotyping logic. PCR was run with primers flanking the Lox-Stop-Lox cassette. The wild-type K-ras allele gave a 265 bp product. The activated allele, left with a single loxP site after the cassette was removed, gave a 305 bp product, 40 bases larger. That larger band appeared only in DNA from the lungs of infected Lox-K-ras G12D mice and was not detected in the other tissues examined, which told the authors that infection and activation were confined to the lung.
Two points follow for anyone designing a check of their own. First, flanking-primer PCR tells you that recombination happened in a tissue, and in which tissue. Second, it does not tell you the protein is made. In the same study, protein expression was checked separately, by immunoprecipitation followed by Western blotting of lung extracts. Gel-based readouts of this kind are explained in our gel electrophoresis guide, and primer design checks in our NCBI primer-BLAST guide.
Three ways labs use the same cassette
| Use | What the STOP controls | Source |
|---|---|---|
| Knock-in oncogene at its own locus | Oncogenic K-ras G12D at the endogenous K-ras locus; Cre removal gives endogenous-level expression | Jackson et al. 2001 |
| Cre reporter mouse | A loxP-flanked EGFP and STOP sequence sits after the CAG promoter. Cre removes both, allowing bicistronic firefly luciferase and beta-galactosidase (lacZ) expression. Cre-negative littermates gave no luciferase signal. | Ishikawa and Herschman 2011 |
| Plasmid for Cre-dependent expression | Addgene plasmid 11584, “Lox-Stop-Lox TOPO”: a 5,700 bp loxP-STOP-loxP insert in a pCR-Blunt II-TOPO backbone, for Cre-dependent activation of a gene of interest, with puromycin selection in mammalian cells; deposited by Tyler Jacks and citing Jackson et al. 2001 | Addgene |
The reporter in the second row is a loxP-EGFP-STOP-loxP design and the authors do not use the abbreviation LSL; it is the same floxed-STOP principle, with a fluorescent protein that is lost on recombination. Reporter assays of the downstream luciferase kind are described in our luciferase reporter assay guide.
Where it matters for labs and animal facilities
Because the cassette makes a silent allele, two things decide what you actually observe: where Cre is delivered or expressed, and when. In the K-ras example that control came from a virus; in other models it comes from a transgenic driver line, which is a separate design decision. The cassette also changes what a genotyping protocol has to show. Written genotyping and handling steps belong in a laboratory standard operating procedure, and any in-vivo work with a viral Cre vector sits under the institution’s animal-use oversight; see our IACUC protocol guide.
Common misunderstandings
Excised DNA is not the same as expressed protein
A recombined band on a gel shows the STOP is gone. The K-ras study confirmed protein by a separate immunoprecipitation and Western analysis, and the two readouts answer different questions.
A reporter allele is not your allele
A Cre reporter such as the CAG-driven EGFP-STOP line shows where Cre recombined its own reporter. It does not prove that a different floxed allele in the same animal recombined in the same cells, because each locus is a separate recombination target.
“Cre-lox” does not always mean deletion
Cre acting on two loxP sites in the same orientation excises the DNA between them. In an LSL design the excised piece is the STOP, and the result is activation, not loss of function.
Cre delivery is a variable, not a constant
In the K-ras work the virus was the timing control. How much Cre reaches how many cells changes how many cells switch on, so it belongs in the methods and in the controls.
FAQ
What does lox-stop-lox mean?
It names a DNA cassette in which a transcriptional STOP sequence is flanked by two loxP sites. Cre recombinase removes the STOP, which turns on the gene sitting downstream of it.
What is the difference between Cre-lox and lox-stop-lox?
Cre-lox is the general recombination system (Cre enzyme plus loxP sites). Lox-stop-lox is one design built on it, where the thing flanked by loxP is a STOP element, so the outcome is activation.
What does loxP do?
loxP is the site Cre recognises. Two loxP sites define the DNA segment Cre acts on; their arrangement determines whether the segment is excised or inverted.
Is “cre lox recombination” the same as “cre-loxP recombination”?
Yes. The terms Cre-lox, Cre-loxP, Cre/lox and “Cre recombinase loxP” all refer to the same system.
Sources
- Jackson EL et al. Analysis of lung tumor initiation and progression using conditional expression of oncogenic K-ras. Genes Dev 2001;15(24):3243-3248. https://pmc.ncbi.nlm.nih.gov/articles/PMC312845 (date checked: 2026-10-29).
- Ishikawa T, Herschman HR. Conditional bicistronic Cre reporter line expressing both firefly luciferase and beta-galactosidase. Mol Imaging Biol 2011;13(2):284-292. https://pmc.ncbi.nlm.nih.gov/articles/PMC3051074 (date checked: 2026-10-29).
- Addgene plasmid 11584, Lox-Stop-Lox TOPO. https://www.addgene.org/11584/ (date checked: 2026-10-29).
- Ingenious Targeting Laboratory, mouse genetics glossary, Lox-Stop-Lox cassette (vendor definition). https://www.genetargeting.com/mouse-genetics-glossary/lox-stop-lox-cassette (date checked: 2026-10-29).
