RIP & CLIP
Immunoprecipitation of the protein from a native lysate with RNase inhibitor, at 4 °C and in physiological salt, then sequencing or qPCR of the co-purified RNA. It recovers far more material than any crosslinking method and gives complex libraries, but because the complex is non-covalent the washes have to stay mild, and nothing in the protocol distinguishes binding that happened in the cell from binding that happened in the tube.
Our ruleSimulations prepare you for the bench but don't replace hands-on practice.
RIP
Immunoprecipitation of the protein from a native lysate with RNase inhibitor, at 4 °C and in physiological salt, then sequencing or qPCR of the co-purified RNA. It recovers far more material than any crosslinking method and gives complex libraries, but because the complex is non-covalent the washes have to stay mild, and nothing in the protocol distinguishes binding that happened in the cell from binding that happened in the tube.
How it differs from the standard method
- Mild washing is not a choice but a constraint: high salt, SDS or urea simply dissociate the complex, leaving the RNA that sticks to beads.
- Post-lysis reassociation is the central weakness, and the control for it is to mix two lysates after lysis (one with the protein, one with the RNA) and see what is recovered.
- Formaldehyde fixation before lysis is sometimes used as a middle course; it captures complexes but crosslinks protein to protein as well, so it reports neighbourhoods rather than contacts.
- The read-out is transcript-level: without a footprint there is no site, so RIP cannot tell you where on an RNA the protein sits.
- It remains the right first experiment for a protein with no crosslinkable contact, for a large ribonucleoprotein assembly, or when a candidate interaction needs checking in a day.
1Pull down a protein, ask what it was holding
The experiment is simple to state: immunoprecipitate an RNA-binding protein and sequence the RNA that comes down with it. The difficulty is a single fact about lysis.
RIP
No crosslink, gentle buffers, RNase inhibitor. Gives abundant, complex libraries and transcript-level answers. Cannot be washed stringently and cannot exclude reassociation, so it is weaker evidence, and still the right first experiment when no crosslinkable contact exists.
CLIP-seq / HITS-CLIP
UV crosslink, stringent wash, RNase footprint, radiolabel and excise the shifted complex from a membrane. Efficiency is only a few per cent, so the material is precious and the library is small.
iCLIP / eCLIP
Reverse transcription stops at the crosslinked amino acid remnant, so the read start marks the crosslink site itself. eCLIP adds a size-matched input, barcoded adapters and no radioactivity.
PAR-CLIP
A thiolated base absorbs 365 nm light, so crosslinking is far more efficient and much gentler. The crosslinked uridine is misread as cytidine during reverse transcription, and the T→C conversion both locates the site and marks a cluster as real.
2The three dials: crosslink, wash, digest
Crosslinking efficiency sets how much material you have. At 254 nm it is a few per cent at best, which is why CLIP protocols start from tens of millions of cells and why CLIP libraries are low-complexity: a library built from a million molecules and amplified through eighteen cycles contains thousands of copies of some fragments.
- Reassociation
- An RNA–protein interaction that forms in the lysate rather than in the cell. Indistinguishable from the real thing by stability, which is why crosslinking exists.
- Footprint
- The length of RNA that survives digestion while protected by the bound protein. Set by the RNase level, and the thing a peak width actually measures.
- Crosslink-induced truncation
- Reverse transcriptase stopping at the amino acid remnant left on the crosslinked nucleotide. The read start then marks the site: the basis of iCLIP and eCLIP resolution.
- T→C conversion
- The characteristic misincorporation opposite a crosslinked 4-thiouridine in PAR-CLIP. Locates the site and distinguishes a genuine cluster from background.
- Size-matched input
- A library made from the same lysate and the same gel region without the immunoprecipitation. The background model that turns read pile-ups into enrichment.
- UMI
- A random barcode in the adapter, read with the fragment, so identical PCR copies collapse to one molecule and peaks are called on molecules rather than reads.
- Library complexity
- The number of distinct molecules in the library. The quantity that limits a CLIP experiment, and the reason depth beyond it is wasted.
- Mock immunoprecipitation
- An isotype-matched IgG, a beads-only or, best of all, an untagged parental line or a knockout lysate, processed identically. Defines what the protocol recovers without the protein.
3Controls, and what each one rules out
Three controls do most of the work, and they are not interchangeable. A mock immunoprecipitation (isotype IgG, or better, the same pull-down from cells that do not express the tagged protein) shows what the beads and the antibody recover on their own.
4How it fails
- A hit list of the most abundant transcripts in the cell. No input library, or a native IP whose background is abundance-shaped. Call peaks against a size-matched input and crosslink if you can.
- Almost no library. Either the lamp was off, or the crosslinking efficiency was simply too low for the amount of starting material. More cells, not more PCR cycles.
- Peaks a hundred nucleotides wide. The RNA was never digested to a footprint, so no site can be located within them.
- Forty million reads collapsing to a million molecules. Depth far beyond the library's complexity. Sequence less, or build a bigger library from more cells.
- Inserts shorter than the read, so most of each read is adapter. Over-digestion; also a sign that the mappable fraction has collapsed.
- A beautiful native RIP that cannot be reproduced by CLIP. Often reassociation. The mixing control would have shown it before the paper did.
- PAR-CLIP clusters with no conversions. Too little 4-thiouridine or too little 365 nm dose, and at the other extreme, half a millimolar of 4-thiouridine inhibits ribosome biogenesis, so the cells being measured are stressed.
Key terms
Practise these as flashcards →Immunoprecipitation of an RNA-binding protein from a native lysate, followed by identification of the associated RNAs by qPCR, array or sequencing.
Crosslinking and immunoprecipitation: UV-crosslink the RNA to the protein in living cells first, so the complex survives stringent purification.
The high-throughput sequencing version of the original CLIP, giving peaks at the resolution of the footprint.
Enhanced CLIP: no radioactivity, barcoded adapters ligated on the beads, a size-matched input library, and single-nucleotide resolution from read starts.
Photoactivatable-ribonucleoside-enhanced CLIP: metabolic labelling with 4-thiouridine and 365 nm crosslinking, read out as T→C conversions.
Finding regions where immunoprecipitated reads are enriched over a background model. The choice of background (input, mock, or nothing) largely determines the answer.
The fraction of reads that are PCR copies of another read. High in CLIP by nature, and only interpretable if the library carries UMIs.
Where peaks fall across a normalised transcript: 5′ UTR, CDS, 3′ UTR, intron. A sanity check against what the protein is known to do.
When to use it
- Finding the transcriptome-wide target set of an RNA-binding protein, as a starting point for asking what it does to them.
- Locating the binding site within a transcript, to mutate it in a reporter and test whether the regulation is direct, which needs iCLIP, eCLIP or PAR-CLIP rather than RIP.
- Testing whether a regulatory effect on a transcript is direct binding or a downstream consequence.
- Mapping where a modification-reading or modification-writing protein sits on RNA, or which RNAs a ribonucleoprotein complex assembles on.
- Checking a single suspected interaction quickly: a native RIP followed by qPCR for a handful of candidates is a day's work, provided the result is treated as provisional.
Compared with alternatives
The same pull-down chemistry, read out for proteins rather than RNA. Prefer it when the question is which proteins are in the complex; the antibody validation needed is identical.
The DNA counterpart, where formaldehyde crosslinking and ChIP play the role UV and CLIP play here. The logic (crosslink, immunoprecipitate, fragment, sequence against an input) is the same, and so are the control arguments.
Measures abundance, not binding. Prefer it alongside a CLIP: targets that are bound and whose level or isoform changes when the protein is depleted are the ones worth pursuing.
Tests one purified protein against one RNA in vitro, with no cellular context. Prefer it to confirm a site found by CLIP, since it shows direct binding and gives an affinity.
Sub-techniques
Choosing one here switches the whole page, including the protocol and the lab, to it.
Immunoprecipitation of the protein from a native lysate with RNase inhibitor, at 4 °C and in physiological salt, then sequencing or qPCR of the co-purified RNA. It recovers far more material than any crosslinking method and gives complex libraries, but because the complex is non-covalent the washes have to stay mild, and nothing in the protocol distinguishes binding that happened in the cell from binding that happened in the tube.
- Mild washing is not a choice but a constraint: high salt, SDS or urea simply dissociate the complex, leaving the RNA that sticks to beads.
- Post-lysis reassociation is the central weakness, and the control for it is to mix two lysates after lysis (one with the protein, one with the RNA) and see what is recovered.
- Formaldehyde fixation before lysis is sometimes used as a middle course; it captures complexes but crosslinks protein to protein as well, so it reports neighbourhoods rather than contacts.
- The read-out is transcript-level: without a footprint there is no site, so RIP cannot tell you where on an RNA the protein sits.
- It remains the right first experiment for a protein with no crosslinkable contact, for a large ribonucleoprotein assembly, or when a candidate interaction needs checking in a day.