RNA-seq from single cells?

[Simone78]

Originally Posted by kobeho24

Hi Simone,
Glad to know that you have already tried the TGIRT on your hand. I wonder if 0.2% SDS can strip the sticky enzyme off the DNA efficiently. And have you tried SDS along with high temperature incubation and also RNase H/ RNase A treatment? From the bioanalyzer, things seem much better than the retroviral RT. Hope to see its application in single-cell RNA-seq.

Gary

Hi Gary,
I just tried with 0.2% SDS, in the same way as I am doing after tagmention to strip the Tn5 off the DNA before PCR and it didn´t work. I also tried SDS + NaOH/HCl (skipping columns or precipitation, as they recommend in the protocol) but, as I said, there was probably too much salt at that point and the PCR didn´t work.
/Simone

[JJMS]

Originally Posted by Simone78

I am still convinced that SMARTScribe is repackaged Superscript II since in my hands they perform exactly the same
good that it works for you as well!
Btw, I also talked to Life Tech and they also said to me that they had no complaints from other customers (that was back in July or August, I think), but since our SSII is fine I didn´t pursue it further. They probably tell the same thing to everybody, clever guys

Hi Simone,
Yes it sure looks like that. Our profiles with high input and SMARTscribe indeed match ssII nearly to a 100%. But with single cells SMARTscribe deviates a little. I have to leave out Mg because otherwise a smear around 500 bp's becomes visible. But for now, I'm very happy!

[jwfoley]

Originally Posted by JJMS

Our profiles with high input and SMARTscribe indeed match ssII nearly to a 100%. But with single cells SMARTscribe deviates a little. I have to leave out Mg because otherwise a smear around 500 bp's becomes visible.

SuperScript II's reaction buffer is 50 mM Tris-HCl pH 8.3, 75 mM KCl, 3 mM MgCl2. ClonTech doesn't say what's in the SMARTScribe mix but according to the documentation from their "legacy" single-cell RNA-seq kit, the first-strand reaction buffer is (at 1X) 50 mM Tris-HCl pH 8.3, 75 mM KCl, 6 mM MgCl2. If that's the same thing they provide when you buy SMARTScribe separately (though it might not be, since they also give you 20 mM DTT instead of 100 mM, as I hope you noticed), then of course you have to adjust the magnesium to compensate. Note that the SMARTScribe enzyme is also provided at 100 U/µL vs. SuperScript II's 200 U/µL.

[Simone78]

Originally Posted by jwfoley

SuperScript II's reaction buffer is 50 mM Tris-HCl pH 8.3, 75 mM KCl, 3 mM MgCl2. ClonTech doesn't say what's in the SMARTScribe mix but according to the documentation from their "legacy" single-cell RNA-seq kit, the first-strand reaction buffer is (at 1X) 50 mM Tris-HCl pH 8.3, 75 mM KCl, 6 mM MgCl2. If that's the same thing they provide when you buy SMARTScribe separately (though it might not be, since they also give you 20 mM DTT instead of 100 mM, as I hope you noticed), then of course you have to adjust the magnesium to compensate. Note that the SMARTScribe enzyme is also provided at 100 U/µL vs. SuperScript II's 200 U/µL.

that´s true, thanks for the explanation. I should have said "repackaged Superscript...with a twist" (changed few details to justify the new packaging)
However, nowadays most or all of the RT protocols around are done with some kind of MMLV-based enzyme and won´t differ so much from each other. My opinion, at least.

[jwfoley]

Originally Posted by Simone78

that´s true, thanks for the explanation. I should have said "repackaged Superscript...with a twist" (changed few details to justify the new packaging)

Well, SuperScript II is wild-type MMLV RTase plus point mutations to kill the RNase H domain. In principle different companies could actually be selling different enzymes with different point mutations...

Originally Posted by Simone78

However, nowadays most or all of the RT protocols around are done with some kind of MMLV-based enzyme and won´t differ so much from each other. My opinion, at least.

Even the fancier enzymes like SuperScript III and Maxima are, as far as I can tell, just engineered to be heat-resistant. That's what makes them more processive: you could run SuperScript II at 50 °C and it would probably be just as fast, except the enzyme would degrade too quickly to be very useful. Unfortunately this engineering tends to kill the C-tailing activity for some reason. But if you aren't using template-switching, mere heat resistance is still a legitimate improvement. (For sequencing I might still worry about increased error rates at the higher temperatures.)

[kobeho24]

Originally Posted by Simone78

Hi Gary,
I just tried with 0.2% SDS, in the same way as I am doing after tagmention to strip the Tn5 off the DNA before PCR and it didn´t work. I also tried SDS + NaOH/HCl (skipping columns or precipitation, as they recommend in the protocol) but, as I said, there was probably too much salt at that point and the PCR didn´t work.
/Simone

Hi Simone,
I don't know if you noticed that the newly published TGIRT paper in the journal RNA. They use the TGIRT to do the RT of human plasma RNA. And I realized that there is a fragmentation step prior to RT when dealing with whole cell total RNA as they suggested. Thus, on my perspective, defeat its better processivity compared to retroviral RT. You mentioned that the average size of cDNA generated by TGIRT is 4-5k. I just wonder if you have an in-house protocol to do so. It would be much appreciated if you let me know.
BTW, I noticed that the TGIRT can template switch not only RNA but also DNA, which means it might not be compatible with single-cell RNA-seq, unless we extract the total RNA from a single cell and do rRNA-depletion before RT, am I right?

Best!
Gary

[amcg]

Hi all,
I'm hoping to use the Smart-seq2 protocol to make rna-seq libraries from small populations (200-500) of cells collected by FACS. In this case would you advise sorting them directly into the lysis buffer (triton-x and RNase inhibitor) described in the Nature Protocols paper, or would it be better to extract the RNA first and then proceed with the reverse transcription on purified RNA?

Many thanks!

[chaichontat]

Hi Simone,

I used the Smart-seq2 protocol using the Superscript IV. For some reason, the reaction failed completely (no cDNA products were observed). When I did not add 1M betaine (Sigma) and 6mM MgCl2, the RT reaction went beautifully, but there's no template switching. The template was 10 ng of purified PBMCs RNA and the TSO was synthesized by Eurogentec. Could you provide me with the reaction conditions you used with SSRT IV?

Thanks!
Chaichontat

[Simone78]

Originally Posted by amcg

Hi all,
I'm hoping to use the Smart-seq2 protocol to make rna-seq libraries from small populations (200-500) of cells collected by FACS. In this case would you advise sorting them directly into the lysis buffer (triton-x and RNase inhibitor) described in the Nature Protocols paper, or would it be better to extract the RNA first and then proceed with the reverse transcription on purified RNA?

Many thanks!

Hi,
even if your cells have a lot of RNA I wouldn´t do it. This is only my opinion, others might think differently. Every time you transfer RNA before any amplification step you lose molecules. I would sort directly into the lysis buffer, but be aware that 0.2% Triton might not be sufficient for lysing so many cells.
Best,
Simone

[Simone78]

Originally Posted by chaichontat

Hi Simone,

I used the Smart-seq2 protocol using the Superscript IV. For some reason, the reaction failed completely (no cDNA products were observed). When I did not add 1M betaine (Sigma) and 6mM MgCl2, the RT reaction went beautifully, but there's no template switching. The template was 10 ng of purified PBMCs RNA and the TSO was synthesized by Eurogentec. Could you provide me with the reaction conditions you used with SSRT IV?

Thanks!
Chaichontat

Hi,
I did exactly as described in the Nat Prot paper, just replacing SSRTII with SSRTIV and doing the RT for 15 min @ 50 degrees and inactivated the enzyme for 10 min @ 80 degrees.. I tested it only on tot RNA (10 pg) and I actually noticed that when I didn´t add betaine and MgCl2 I got worse results than when using them...the opposite of what you found. in my case the template switching went ok, at least based on the cDNA yield and size.
Best,
Simone

[Simone78]

Originally Posted by kobeho24

Hi Simone,
I don't know if you noticed that the newly published TGIRT paper in the journal RNA. They use the TGIRT to do the RT of human plasma RNA. And I realized that there is a fragmentation step prior to RT when dealing with whole cell total RNA as they suggested. Thus, on my perspective, defeat its better processivity compared to retroviral RT. You mentioned that the average size of cDNA generated by TGIRT is 4-5k. I just wonder if you have an in-house protocol to do so. It would be much appreciated if you let me know.
BTW, I noticed that the TGIRT can template switch not only RNA but also DNA, which means it might not be compatible with single-cell RNA-seq, unless we extract the total RNA from a single cell and do rRNA-depletion before RT, am I right?

Best!
Gary

Hi Gary,
yes, I saw the paper, thanks! I didn´t know that the TGIRT can do strand switch on DNA as well (this was not mentioned when I contacted the Lambowitz lab). When I first did my tests some time ago I used tot RNA. Unfortunately, it didn´t work at single cell level (pg) and I just dropped it. therefore, I never sequenced anything and never tried on a real cell. I do have a protocol that worked with 100 ng (!) RNA but after the publication of this new article, I want to try again using some modifications and see if I can get it to work with less RNA. So I can´t really share it right now, sorry!
Best,
Simone

[chaichontat]

Originally Posted by Simone78

Hi,
I did exactly as described in the Nat Prot paper, just replacing SSRTII with SSRTIV and doing the RT for 15 min @ 50 degrees and inactivated the enzyme for 10 min @ 80 degrees.. I tested it only on tot RNA (10 pg) and I actually noticed that when I didn´t add betaine and MgCl2 I got worse results than when using them...the opposite of what you found. in my case the template switching went ok, at least based on the cDNA yield and size.
Best,
Simone

Hi Simone,

It turned out that my PCR primers were the problem. I was too fixated on the template switching process that I completely ignored the preamplification. Upon replacing the primers, everything works beautifully. I really thank you for this brilliant protocol!

Chaichontat

[KroSeq]

Originally Posted by Simone78

Hi,
I did exactly as described in the Nat Prot paper, just replacing SSRTII with SSRTIV and doing the RT for 15 min @ 50 degrees and inactivated the enzyme for 10 min @ 80 degrees.. I tested it only on tot RNA (10 pg) and I actually noticed that when I didn´t add betaine and MgCl2 I got worse results than when using them...the opposite of what you found. in my case the template switching went ok, at least based on the cDNA yield and size.
Best,
Simone

Hej Simone,
what is the benefit of using SSRT IV instead of SSRT II?

[Simone78]

Originally Posted by KroSeq

Hej Simone,
what is the benefit of using SSRT IV instead of SSRT II?

the RT reaction is way faster than with the SSRTII but, apart from that, I don´t see any improvement in the reaction (at least with my settings!). The prices are almost the same, the SSRTIV is a bit more expensive.

[longwood]

E. Coli contamination of superscript II has been confirmed by the manufacturer. I know a bunch of people reported this on this forum a few months back, so thought I'd share this letter (attached) I received from Thermo Fisher. The letter basically says they have started testing for E. Coli DNA as part of their routine test now and those who received the tainted lots will be sent free replacement upon request.

[kobeho24]

Originally Posted by Simone78

Hi Gary,
yes, I saw the paper, thanks! I didn´t know that the TGIRT can do strand switch on DNA as well (this was not mentioned when I contacted the Lambowitz lab). When I first did my tests some time ago I used tot RNA. Unfortunately, it didn´t work at single cell level (pg) and I just dropped it. therefore, I never sequenced anything and never tried on a real cell. I do have a protocol that worked with 100 ng (!) RNA but after the publication of this new article, I want to try again using some modifications and see if I can get it to work with less RNA. So I can´t really share it right now, sorry!
Best,
Simone

Hi Simone,
Sounds really great that you do have a protocol. What I wonder is RT condition for non-fragmented RNA. Just incubate the RNA with TGIRT at proper temperature for 15 min? Looking forward to your successful modified protocol.

Gary

[chaichontat]

Hi Simone,

I'm trying to amplify one gene of interest; however, I couldn't get a specific band when I added the whole 10 ul of the RT reaction (final PCR volume was 25 ul). The band only appeared when I added 1 ul of the RT mix, suggesting that there's some inhibition of PCR by the RT mix.

However, there was one deviation from your protocol: I used the component version (not ReadyMix) of the KAPA HiFi Hot Start. Do you think that this could have been a factor?

Thanks!
Chaichontat

[kobeho24]

Hi all,
Recently I’ve done 3 runs of single-cell RNA-seq (Smart seq2) trial. I just followed the original protocol. A highly reproducible background curve can be observed even in negative controls (only nuclease free water) when using our new SuperScript II RTase, which cannot be seen when using SSIII, or without RTase. Our bad SSII lot no. is 1688365.
QQ??20160121185909.png

QQ??20160121190101.png

QQ??20160121190137.png

Does any of you guys have any recommendation of substitutive MMLV RTase? I saw ProtoScripts II & SS IV so far, any candidate else?

Best wishes!
Gary

[Simone78]

Originally Posted by kobeho24

Hi all,
Recently I’ve done 3 runs of single-cell RNA-seq (Smart seq2) trial. I just followed the original protocol. A highly reproducible background curve can be observed even in negative controls (only nuclease free water) when using our new SuperScript II RTase, which cannot be seen when using SSIII, or without RTase. Our bad SSII lot no. is 1688365.
Attachment 4169

Attachment 4170

Attachment 4171

Does any of you guys have any recommendation of substitutive MMLV RTase? I saw ProtoScripts II & SS IV so far, any candidate else?

Best wishes!
Gary

I would use either SMARTscribe (Clontech) or Maxima H- (Thermo Scientific).

[kobeho24]

Originally Posted by Simone78

I would use either SMARTscribe (Clontech) or Maxima H- (Thermo Scientific).

Hi Simone,
Between these two, which one's better? It seems that SMARTscribe is similar to SSII and Maxima H- is similar to SSIV. Besides, as SMARTscribe is 100U/ul while Maxima H- is 200U/ul, do we still make it 100U in the 10ul RT reaction as it is in the Smart-seq2 protocol? Am I right?

Best!
Gary