Jump to content

  • Log in with Facebook Log in with Twitter Log in with Windows Live Log In with Google      Sign In   
  • Create Account

Submit your paper to J Biol Methods today!
Photo
- - - - -

HELP: Validation of a polyclonal knock out (KO) cell line when you dont have a g

knock outKO validation Crispr Cas9

  • Please log in to reply
6 replies to this topic

#1 Natalia KM

Natalia KM

    member

  • Active Members
  • Pip
  • 25 posts
0
Neutral

Posted 13 August 2020 - 04:43 PM

Hi there,

 

As some of you might be aware I am new to cloning and Crispr/Cas9 gene editing. As a result I've also not carried out many DNA related experiments.

 

I have recently attempted to make 4 different knock out cell lines using pLentiCrispV2 plasmids. The guides I designed worked for 2 of these cell lines and these were easy to validate as I had good antibodies.

 

I have terrible antibodies for the other 2 cell lines and have had confusing results with these so I cant tell whether there is a knock out there at all.

 

I wanted some advice on how to verify these cell lines for a knock out. If I extract DNA from WT and these polyclonal 'KO' cell lines, then PCR amplify a region, can I hand this in for sequencing directly. Considering this is a polyclonal cell line, I'm guessing the results won't be very clean?

 

Please advise how I can verify these cell lines and the simplest protocol to follow?

 

Any help would be greatly appreciated.

 

Thanks!!!



#2 bob1

bob1

    Thelymitra pulchella

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 6,706 posts
567
Excellent

Posted 13 August 2020 - 05:24 PM

If you know the region of the knock-out, and it is a substantial KO (anything more than say 20-50 bp), you can simply design an array of PCRs - use one primer anchored in the KO region, and one each on either side. The perform 2 PCRs - first with a primer outside, and the anchored inside primer, then second with primers on either side. You should get 5 potential results - no amplification with the anchored primer PCR (if there is amplification, you haven't knocked out, at least not a complete KO). The other reaction will have either a full-length product (for no KO) or a truncated/shorter product (if there is a KO), or a combination of both if it is a chimeric population. It pays to run both PCRs because if the first doesn't work, the second will tell you if you can amplify from that region.

 

This only works if you can separate the different bands on a gel - small KOs will need a greater resolving power gel than you will get with agarose, so look at acrylamide gels for these ones.

 

You can also look for expression of the RNA and see if it is full-length or not, or not expressed at all.



#3 Natalia KM

Natalia KM

    member

  • Active Members
  • Pip
  • 25 posts
0
Neutral

Posted 14 August 2020 - 07:37 PM

Thanks for the suggestions.

 

I do plan to run some PCRs as you suggest on an agarose gel but I have a few questions:

 

1. How can you tell how many bp the KO is? What % acrylamide gels would I need to run to see these small differences?

 

2. Can I take these PCR products purified on the gel and carry out Sanger sequencing? If so what would I expect to see from a polyclonal population of cells? Is the sequencing just going to be a mess if its not a monoclonal population?

 

How can you definitively identify a KO if theres no good antibody by the above methods?

 

Thanks so much for your help!



#4 bob1

bob1

    Thelymitra pulchella

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 6,706 posts
567
Excellent

Posted 15 August 2020 - 02:32 AM

1) it depends on how you did the KO.  The % acrylamide (or agarose) depends on how big your PCR product would be with and without the KO.

 

2) Yes you can sequence. If you have a mix of KO and no KO in the polyclonal cells, then you will get a mess from the sequencing if you sequence directly from the PCR. If you run the products on a gel and separate the two potential products, then gel extract the bands and sequence, you should be fine.

 

3) Whole genome sequencing (e.g. MiSeq) might work. The above methods work well and are widely established, they are the methods that were used before CRISPR/CAS was around. You can also use the same methods on RNA rather than DNA.



#5 Natalia KM

Natalia KM

    member

  • Active Members
  • Pip
  • 25 posts
0
Neutral

Posted 15 August 2020 - 04:15 AM

Thank you so much for your answers. 

 

1. Sorry I still do not understand how you can tell how many bp your KO will be prior to running an agarose gel without knowing whether you're getting an insertion or deletion? How can you work that out? I used the plentiCrispr V2 system.

 

2. If you have a polyclonal population, the cut could be at various sites within the guide sequence? So wouldn't the sequencing data still be messy?

 

Sorry for the dumb questions but its the only way to learn I guess!



#6 bob1

bob1

    Thelymitra pulchella

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 6,706 posts
567
Excellent

Posted 17 August 2020 - 01:46 PM

1) My very limited understanding or CRISPR/Cas9 is that it causes targeted cuts in the genome, and because you were making a KO, I assumed that this would be a deletion, because that's what they normally are. If you were using the CRISPR-cpf method then you might be doing insertions. I don't know how you determine the size of the KO, but it is presumably related to the size of your guide RNA. To be conservative I would design primers that give you a product of approx 100-150 bp with the KO, presumably longer without. You should be able to resolve this to single base-pair resolution on a 8 or 12% TBE-acrylamide gel.

 

2) possibly. I don't have experience with CRISPR, so I don't know the details of exactly how it works.



#7 Natalia KM

Natalia KM

    member

  • Active Members
  • Pip
  • 25 posts
0
Neutral

Posted 20 August 2020 - 05:24 PM

Thanks for your answer Bob1. I have spoken to various people and e everyone seems to have different opinions. 

 

Re running PCR amplified genomic DNA around the cut site, because Crispr generates indels leading to mismatches by just a few bps, these changes might be difficult to assess on an agarose gel. If a premature stop codon is introduced by these indels, this may be visible? I dont know.

 

mRNA expression is often unchanged/unreliable and Sanger sequencing of a mixed population will be messy. :(







Home - About - Terms of Service - Privacy - Contact Us

©1999-2013 Protocol Online, All rights reserved.