SEQanswers

Go Back   SEQanswers > Applications Forums > RNA Sequencing



Reply
 
Thread Tools
Old 08-30-2017, 01:44 AM   #1
Radek
Junior Member
 
Location: Belgium

Join Date: Jun 2017
Posts: 3
Default Single-cell transcriptomic: rare transcripts

Hi!

I performed a smart-seq2 experiment on B-cells. After some minor changes in the protocol I got a satisfying amplification with no primer dimers and a 3 to 15 ng of cDNA (see attached bioanalyzer result).

- Biotinylated primers (TSO, oligo-dTNV, ISPCR) and reducing the amount of TSO by 5 and ISPCR by 2.
- SPRI beads 0.6x instead of ampurebeads 1x.

http://www.hostingpics.net/viewer.ph...6Sanstitre.png

Unfortunately I am only interested in a subpopulation of B-cells which are expressing a given transcript (poly-adenylated, ranging from 600 to 2000 bp) and my only option currently is to detect it after performing the amplification. Those cells can't be sorted before the smart-seq2 procedure.
I tried to detect it by qPCR and got systematic amplification in my 10 cells and 100 cells controls. Unfortunately in the 1-cells wells only a small fraction of my cells were positive while I was expecting almost all of them. Given the results I have the impression that this transcript is sometimes retrotranscribed sometimes not.

I have several questions related to this issue:
  1. Do you have any recommandation on how to improve the retrotranscription of relatively rare transcripts?
  2. Do you think that adding a transcript specific primer during the retrotranscription could help?
    I was thinking to use add an oligo:

    ISPCR sequence - linker - specific primer

    at the RNA denaturation step in addition of the oligo-dTNV.
  3. What is the minimal length of transcript which can be retrotranscribed with smart-seq2 (and why)? I see clearly that 500 bp seems like the limit but could not find a good explaination for it.

Thanks in advance for your answers!
Radek is offline   Reply With Quote
Old 08-30-2017, 10:04 AM   #2
cmbetts
Member
 
Location: Bay Area

Join Date: Jun 2012
Posts: 87
Default

1) These methods typically have 10-20% overall efficiency. If your transcript of interest has low expression levels, it will frequently drop out. This is purely binomial statistics and can't be avoided.
2) You can certainly try adding a gene specific primer. It would probably require a good bit of optimization though. Since there isn't a cleanup between RT and PCR, it would need to function as both an RT primer and a PCR primer pair with ISPCR. I'd anticipate some wonky PCR artifacts.
3) It only functions on polyadenylated RNAs, and not many of those are shorter. Also, the PCR is semi-suppressive which helps reduce artifact amplification by making amplification of shorter fragments less efficient.
cmbetts is offline   Reply With Quote
Old 08-30-2017, 01:25 PM   #3
alec
Member
 
Location: Cambridge, MA

Join Date: Apr 2011
Posts: 18
Default

I believe the low efficiency is due to a combination of low of processivity of the reverse transcriptase and low efficiency of the template switching reaction, rather than initiation of RT. A gene-specific primer would only help with initiation.

My guess is you would be better off trying to design a gene-specific primer for the complimentary strand that you would add during PCR. That way your gene could still be amplified if template switching failed or the reverse transcriptase fell off (depending on which part you are targeting). It would still probably require a good bit of optimization, and you'd lose some information about splicing/allelic expression if that is important to you.
alec is offline   Reply With Quote
Old 08-30-2017, 11:13 PM   #4
Radek
Junior Member
 
Location: Belgium

Join Date: Jun 2017
Posts: 3
Talking

Thank you both!

Quote:
Originally Posted by alec View Post
I believe the low efficiency is due to a combination of low of processivity of the reverse transcriptase and low efficiency of the template switching reaction, rather than initiation of RT. A gene-specific primer would only help with initiation.

My guess is you would be better off trying to design a gene-specific primer for the complimentary strand that you would add during PCR. That way your gene could still be amplified if template switching failed or the reverse transcriptase fell off (depending on which part you are targeting). It would still probably require a good bit of optimization, and you'd lose some information about splicing/allelic expression if that is important to you.
I am only interested in knowing which cells are positive or not so this might be a good way to go.
Radek is offline   Reply With Quote
Old 08-31-2017, 05:59 AM   #5
plyguy69
Junior Member
 
Location: america

Join Date: Apr 2015
Posts: 9
Default

Hi Radek. Thanks for the post. I am doing a very similar protocol currently and was wondering what RT enzyme are you using? Is it Superscript II/III?
plyguy69 is offline   Reply With Quote
Old 08-31-2017, 06:00 AM   #6
Radek
Junior Member
 
Location: Belgium

Join Date: Jun 2017
Posts: 3
Default

Superscript III lacks the strand switching activity so it could not work here.
Radek is offline   Reply With Quote
Old 08-31-2017, 11:00 AM   #7
plyguy69
Junior Member
 
Location: america

Join Date: Apr 2015
Posts: 9
Default

Hmmmm. I thought all MMLV enzymes had the template switching activity. Anyway, in our hands the Primescript RT from Clontech gave the best results. As to your questions it seems like we get longer cDNA products than 500bp, but I haven't tried any specific primers mixed in.
Attached Files
File Type: pdf 2017.8.31.cDNA.pdf (57.8 KB, 4 views)
plyguy69 is offline   Reply With Quote
Reply

Tags
single-cell, smart-seq2

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off




All times are GMT -8. The time now is 09:31 AM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2017, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO