Written and maintained by CASRAI Editorial Board
Last updated
Short answer
A proximity ligation assay (PLA) is a method that detects two proteins sitting close together inside cells or tissue and turns each such event into a single fluorescent spot you can count. A spot means the two targets were within roughly 40 nm of each other. It does not by itself prove direct binding, and the absence of spots does not rule out a transient interaction.
Definition and scope
PLA is an in situ technique: it reports where in a cell or tissue section two proteins are close, rather than reading an average from a lysate. Its detection reagents are called PLA probes or proximity probes, which are antibodies that carry attached oligonucleotides. The targets are, in the words of one methods paper, “interactions or close proximity between pairs of proteins” (Sci Rep, 2018, PMC5876389).
That scope matters for interpretation. PLA reports proximity. A study of the hemochromatosis proteins Hfe and transferrin receptor 2 (Tfr2) describes the signal as forming only when the two proteins are “closer than 40nm” (PLoS One, 2013, PMC3796466). Two proteins can be within that distance without being in direct contact.
How it works in practice
The principle runs in four stages:
- Probes bind. Two PLA probes, each an antibody carrying an oligonucleotide, bind their respective targets.
- Ligation. If the probes are close enough, the conjugated oligonucleotides template the ligation of a secondarily added pair of oligonucleotides, closing a DNA circle.
- Amplification. Rolling circle amplification (RCA) replicates the circle, primed by one of the PLA probes.
- Detection. The amplified products are visualised with fluorophore-labelled detection oligonucleotides, and each product shows up as a discrete dot.
The Hfe and Tfr2 study treats each red dot as one molecular interaction event, so the output is a set of countable dots rather than a single intensity value. Exact incubation conditions and reagent volumes depend on the kit and the manufacturer’s instructions, so this page does not reproduce a bench protocol.
What a spot does and does not tell you
| Observation | What is supported | What it does not establish |
|---|---|---|
| Spots present | The two targets were closer than about 40 nm, and each spot represents one proximity event (PMC3796466). | Direct physical binding between the two proteins. |
| No spots, with a working positive control | The authors read the absence of signal as indicating no interaction between Hfe and Tfr2 (PMC3796466). | That a transient interaction never occurs. The same authors went on to test that possibility with BS3 crosslinking and co-immunoprecipitation. |
| Very dense spots | At higher probe concentrations the RCA products “become so abundant that they start to coalesce” (PMC5876389). | An accurate count: merged spots cannot be counted individually. |
Controls checklist for reading a PLA result
Two published studies document the following controls. Treat them as the minimum evidence that a signal is specific to the pair of interest.
- Single-probe or single-antibody reactions. In the Hfe and Tfr2 study, reactions using only one antibody (FLAG, Tfr1 or Tfr2) served as negative controls, and the authors state that “the absence of any spots indicates that the assay is specific.”
- Omit one probe. The UnFold probe paper (Sci Rep, 2018) reports that without both PLA probes present, minimal staining was detected, which shows that the dots depend on the pair of probes.
- A known interacting pair. The Hfe and Tfr2 study detected the previously characterised Tfr1-Tfr2 interaction, which shows the assay can return a positive signal in that system.
- Probe concentration within the countable range. Record the probe concentration used, because over-dense signals coalesce and compromise counts.
- An orthogonal method for negative results. A negative PLA is a statement about proximity. The authors of the Hfe and Tfr2 paper tested for a transient interaction by crosslinking and co-immunoprecipitation.
Related concepts
PLA sits alongside other ways of asking whether two proteins associate. Co-immunoprecipitation pulls a complex out of a lysate, so it can show association but loses the spatial context that PLA keeps; see our guide to co-immunoprecipitation controls and interpretation. Standard immunofluorescence shows two proteins in the same region, but overlapping fluorescence does not report closeness at the tens-of-nanometre scale that PLA is designed for. Because PLA relies on antibodies, the choice between polyclonal and monoclonal antibodies and the way antibodies are identified in methods sections (see citing antibodies with RRIDs) both affect how reproducible a PLA result is. Normalisation questions for blot-based comparisons are covered in western blot loading controls.
Where it matters for labs
For a lab deciding whether PLA can answer its question, the sourced points reduce to three. First, PLA needs two antibodies that each recognise their target in situ, because the probes are antibodies. Second, a claim of interaction should be backed by at least the single-probe negative control and a result in a system with a known interacting pair. Third, quantification should be planned around dot counts that stay below the coalescence range. Reagent selection and kit specifications are set out in the manufacturer’s own documentation.
Common misunderstandings
“A PLA spot proves the proteins bind each other.”
It shows they were within about 40 nm. Proximity alone does not distinguish direct binding from other close associations.
“No signal means no interaction.”
No signal under the conditions tested means no detectable proximity under the conditions tested. The Hfe and Tfr2 authors themselves considered that a transient interaction could be missed and tested it separately.
“More dots always means a stronger interaction.”
Dot number depends on probe concentration as well as biology. At high probe concentration, RCA products coalesce and the count stops tracking the number of events.
FAQ
What does a proximity ligation assay detect?
It detects close proximity between pairs of proteins in situ, reported as individual fluorescent spots, one per proximity event.
How close do two proteins have to be for PLA to give a signal?
The Hfe and Tfr2 study describes signal formation when the two proteins are closer than 40 nm.
What are PLA probes?
Antibodies with oligonucleotides attached. When two probes are close, those oligonucleotides template ligation of added oligonucleotides into a circle that is then amplified by rolling circle amplification.
Which controls should a PLA experiment include?
Single-antibody reactions as negative controls, a reaction lacking one PLA probe, and a known interacting pair as a positive reference, as documented in the studies listed under Sources.
Sources
- PLoS One (2013), “In situ proximity ligation assays indicate that hemochromatosis proteins Hfe and transferrin receptor 2 (Tfr2) do not interact”, https://pmc.ncbi.nlm.nih.gov/articles/PMC3796466/, checked 2026-10-29.
- Scientific Reports (2018), “Improved efficiency of in situ protein analysis by proximity ligation using UnFold probes”, https://pmc.ncbi.nlm.nih.gov/articles/PMC5876389/, checked 2026-10-29.
