Best way to stick adaptors on degraded mRNA

[eab]

We need to make Illumina libraries from small quantities (<1ng) of human mRNA than has been degraded to a fragment size <150 nt.

We can do this by randomly-primed RT, second strand synth, end prep and adaptor ligation, a la standard RNASeq.

Or we can do it by sequential ssRNA adaptor ligation, a la miRNASeq.

Which would people advise using?

In favor of the standard approach is that we do this method all the time.

In favor of the miRNA approach is that it gives strand specific libraries.

I don't know how the two compare at small input quantities.

And I don't know the best reagents for the miRNA approach. Would probably buy a kit. Would people just use Illumina's?

Any feedback would be very much appreciated. Thanks!

[kerplunk412]

You can also get directionality from a first strand-second strand approach, using a dUTP-based kit. There are plenty out there, and you can probably save some money going with someone other than Illumina. You will probably need to modify some of the cleanup steps to make sure you retain smaller insert sizes, and you may need an extra cleanup post-PCR. If I remember correctly two 0.8x AmpureXP cleanups post-PCR worked well for for me in removing adapter-dimer products without much loss of products with >80 bp inserts.

The NEXTflex Rapid Directional qRNA-Seq kit from Bioo Scientific also offers the advantage of molecular indexing, which allows precise discrimination of PCR duplicates from distinct but identical cDNA molecules. You can PM me for more information. For disclosure, I work for Bioo Scientific.

[nucacidhunter]

1- Standard RNAseq with random priming will result in small insert sizes. This would require careful removal of primer and adapter dimers as noted above. Bead clean up step have to be adjusted to minimise library fragment loss. Other issue is low input and finding a kit that can give the best results with input amount available. A strand specific kit can be used to maintain directionality.

2- Adapter ligation approach will maintain input fragment sizes and using one of kits designed for small RNA can yield good results considering low input. The issue would be terminal structure of RNA fragments. For small RNA or miRNA profiling the assumption is presence of a 5’-phosphate and 3’-hydroxyl group. Degraded RNA may lack one or both. It may be possible to add these moieties to RNA by enzymatic reaction.

[eab]

Thanks to both! You raise good points. We are experienced with controlling the size cut using AMPure (or PAGE, when absolutely necessary), so I have confidence there. If we wanted strand specific, do you both think it would be best to try the mitomycin/dUTP method, rather than sequential RNA adaptor ligation?

We made a few miRNA libraries some years back, and felt it went well. But what I don't recall is how good the yield is, compared to making a library from the same material by standard RNAseq protocols. Any thoughts about that?

Thanks again!

[nucacidhunter]

Thanks to both! You raise good points. We are experienced with controlling the size cut using AMPure (or PAGE, when absolutely necessary), so I have confidence there. If we wanted strand specific, do you both think it would be best to try the mitomycin/dUTP method, rather than sequential RNA adaptor ligation?

We made a few miRNA libraries some years back, and felt it went well. But what I don't recall is how good the yield is, compared to making a library from the same material by standard RNAseq protocols. Any thoughts about that?

Thanks again!

dUTP method would be better as sequential adapter ligation would be successful only if RNA fragments have 3'OH and 5' phosphate group which may not be the case with degraded RNA. To my knowledge miRNA library prep is based on sequential 3' and 5' adapter ligation followed by RT and PCR amplification. miRNA library yield is depend on the kit and also presence of required chemical groups at RNA ends. I have had good miRNA libraries with 100 ng total RNA input which contains very small amount of small RNA.