RIP-Seq: Does difference in RNA pulldown conc. between Samples and Negative Control matter?

Question

[samantha_jeschonek]

I am optimizing conditions for an RNA-IP Sequencing (RIP-seq) experiment. The goal is to identify associated RNAs that co-IP with a particular transcript.

For my experimental samples and positive control, I am getting about 10-fold more RNA pulled down than for the negative control (as measured by Qubit). An initial screen with RT-qPCR shows that my transcript of interest is much more enriched in the samples than the negative control (great news).

My concern is the difference in total RNA pulled down in the IP. Will the discrepancy between RNA concentrations affect downstream steps in library preparation for RNA-seq? Is it ok to use a negative control for RIP-seq that pulls down significantly less RNA than the experimental samples? Do I need to find a negative that pulls down the same total concentration of RNA but only fails to pull down my transcript of interest?

Any advice would be appreciated. Thanks!

Answer 1

[Ryan Dale]

I agree with your concern. If you use something like IgG as the control, you'd be pulling down a small amount of nonspecific noise, and as a result you'll have a low-complexity library likely to have lots of PCR duplicates. I've always thought it made more sense to use input RNA as the control so you can normalize to the total amount of RNA for each transcript. That's based on the assumption that transcription levels vary over several orders of magnitude (e.g., from low expression to high expression) while this is unlikely to be true for nonspecific binding levels.

Ideally you'd do something like enrichment = IP_RNA / (total_RNA - nonspecific_RNA) but I'm not aware of any tools that support that kind of normalization.

[samantha_jeschonek]

Exactly -- this is my concern. One of the negative controls I'm using is IgG and another is a protein that should not bind a positive control transcript.

In both cases, the negatives pull down very little RNA (undetectable even by Qubit RNA High-sensitivity assay).
Even though the qRTPCR looks good, I'm worried the library will have a high representation of "sticky" RNAs. That because there is so little RNA, there will be spurious amplification and perhaps make it seem that RNAs are highly abundant when it is just an artifact of there being so little RNA to start with.

[harold.smith.tarheel]

This is the same problem for ChIP-Seq using IgG as the control Ab for IP, which is why most people use input chromatin as the control.

Using an equivalent amount of mock-treated RNA (X-linked, incubated, etc) for control library prep is probably best. Alternatively, if your RNAs of interest are mRNAs, you could use poly(A) binding protein Ab (available commercially) as your IP control.

[samantha_jeschonek]

Thanks for the reply, Harold. This is very useful. I've been doing a little reading into ChIP-seq, but still have a concern.

Is it necessary to still do an additional negative control (like IgG) as well as input? And then the comparison is between enrichment of the two IPs? As a rough example, if transcript-Y shows a 10% enrichment over input with Protein-X pulldown, but the IgG (negative) pulldown shows only a 0.1% enrichment, can it be concluded that Protein-X interacts with transcript-Y? Or is a threshold of enrichment set and only those hits are examined? Ie: Deciding to only consider the those hits that represent the top 5% of transcripts enriched?

Thanks for you advice and time!