RE: DNA-guided genome editing using the Natronobacterium gregoryi Argonaute

[SKim]

Hello,

I guess most of you are aware of a recent paper published in Nature Biotechnologies.

http://www.nature.com/nbt/journal/vaop/ncurrent/full/nbt.3547.html

It's been a week since the publication and I have been expecting that there would be flaming discussion about this new genome editing tool but this forum is relatively quiet.

We understand that there will be more work to be done for further validation.

However, based on clear advantages over CRISPR/Cas9, I am wondering how many of you are thinking that NgAgo is going to be the one in the next round.

[YIng Dang]

I can't wait to see a follow-up report.

[Shouguang Huang]

Hi, 

I have read the literature and found this technique really attracts our eyes. There are more advantages than CRISPR/Cas9, such as the description in the literature:

1. It has a low tolerance to guide-target mismatch.

2. 5' phosporylated short ssDNA are rare in mammalian cells, which minimizes the possibility of cellular oligonucleotides misguiding NgAgo.

3. NgAgo follows a "one-guide-faithful" rule.

4. it is easy to design and synthesize ssDNA and to adjust their concentration.

In sum, I have to say NgAgo-guide genome editing is a super nice technique. However, I guess most of labs are still focusing on CRISPR/Cas9, they might hard to switch their direction to this new stuff. But I strongly believe that there will be more and more scientists harbor this new emerging tool in next few months. 

LET'S SEE!

[Shouguang Huang]

Me too.

[tmcmor...@gmail.com]

I read this paper, and attracted by this method. if the detailed mechanism uncovered, probably, more applications will be published.

[Jiabiao Hu]

Combination of Cas9 and NgAgo would greatly expand the tools for study biology. Cas9 still has its own advantage, as it is easy to repurpose for Imaging (Cas9-GFP), transcriptional regulation (CRISPRi & CRISPRa). Does anyone want know how to express a ~24 ssDNA in mammalian cells?

[SKim]

Jiabiao,

I agree with you in that repurposing could be easier with CRIPSR. Lots of Cas9 engineering have been reported. In addition, gRNA has rooms to accommodate different functional domains.
As you pointed out, if short ssDNA can not be stably expressed in the cells, this could be limitation with NgAgo for further application.
However, people always find a way to overcome limitations. Hope to see further validation and how this field is moving with this new genome editing tool!

[pei...@gmail.com]

If this technology can be easily repeated in other labs, it will be very useful for editing genome.  I couldn't access and read the full text of the paper.  Do you know if a synthetic oligo can be used as guide-DNA?  How did the authors perform transfection, use plasmid or lentivirus? 

On Monday, May 9, 2016 at 11:02:30 AM UTC-4, SKim wrote:

[Fatwa Adikusuma]

I am wondering whther the plasmid will be available on Addgene? Would love to give this a try.

Cheers,

Fatwa

[Harrys Kishore Charles Jacob]

Hi Fatwa,

We checked with the authors and they said that it has been submitted to Addgene and should soon be available.

CJ

[Aaron Zhang]

Please let people know if one finds it available on Addgene. 

[Nam]

If this works efficiently, at least similar level to the CRISPR/Cas9 system, I am willing to change my method from the CRISPR/Cas9 to the Argo method.

Some company already stop providing TALEN service because there is no new order. The CRISPR/Cas9 might have same destiny.

[Manuel Kaulich]

Just wondering if anybody is concerned about the fact that ssDNA can easily be incorporated into the host genome, generating unpredictable off-target effects.

On Monday, May 9, 2016 at 5:02:30 PM UTC+2, SKim wrote:

[Joe Miano]

This has my full attention as they targeted multiple genes, including one (HBA2) that was inaccessible to Cas9.  We need the protein/mRNA so we can generate some mice and compare with Cas9  That there is no PAM requirement opens up the sequence landscape considerably though their testing of some 40-odd ssDNAs revealed a low efficiency (20-40 percent versus ~80% for CRISPR gRNAs). Also need to think about secondary editing after HDR integration though it sounds like this may not be an issue with the NgAgo-ssDNA since the editing is transient and once NgAgo is loaded with a ssDNA, it does not exchange.  The big question is how transient?  Will we see secondary editing?  

On Monday, May 9, 2016 at 11:02:30 AM UTC-4, SKim wrote:

[YIng Dang]

HI Joe,

Would you mind point out which 40-odd ssDNA you are talking about in the paper. I don't see they comparing NgAgo with Cas9 side by side except in figure 3c and figure 4e 4f. and supplemental figure 10. None of these showed a lower efficiency. Thanks for help.

Ying

[Joe Miano]

Hi Ying,

I was referring to Suppl FIgure 5 and the T7E1 assay.  They tested 47 ssDNA guides across multiple genes.  The efficiency of cutting was 20-40%.  This is quite good and yes, they did not compare with Cas9 in that specific case.  WHat I meant to indicate, which we all recognize, is that there looks to be superior ssDNA accessibility to the genome for DSB versus Cas9 where the latter did not appear to occur at the HBA2 locus   SOrry for the confusion.

My question, which we won't know definitively until we all try it out, is the extent of secondary editing and mosaicism in the zygote.  One thing revealed to me, not surprisingly, is need for 5' utr in NgAgo for mouse zygote injections.

I want badly to compare head to head with Cas9 in mouse zygote!

Joe

On Monday, May 9, 2016 at 11:02:30 AM UTC-4, SKim wrote:

[Yuting]

Me too! Has anybody asked the correspondence author for the availability of the Ng-Ago plasmid?

在 2016年5月9日星期一 UTC-7下午3:46:32，YIng Dang写道：

[Martin]

NgAgo plasmid is currently undergoing quality control at Addgene according to their twitter

On Monday, May 9, 2016 at 5:02:30 PM UTC+2, SKim wrote:

[Joe Miano]

Cool.  SInce my kids tell me I am too old for Twitter Ha! I have no info on progress of QC.  Can you find out about the presence of a 5'utr which will be needed for efficient translation in context of mouse zygote injections?

I hear though from my postdocs that through Social Media in CHina, the data are reproducible and so it will be just a matter of time before this new innovation spreads quickly.'

The Sr author has not responded to my email requests for protein etc.  He probably is getting 1000s of emails each day!

On Monday, May 9, 2016 at 11:02:30 AM UTC-4, SKim wrote:

[Martin]

https://www.addgene.org/78253/

On Monday, May 9, 2016 at 5:02:30 PM UTC+2, SKim wrote:

[Aaron Zhang]

Thanks for posting, Martin!

Cannot wait for people's result!

[Genome-editing]

This is very exciting and promising. Could be a game changer if it turns out to be better than CRISPR-Cas9 method.

https://dnascissors.wordpress.com/2016/06/05/genome-editing-by-a-dna-guided-enzyme/

[Joe Miano]

By now, many of you have obtained the NgAgo plasmid from addgene.  We are gearing up for some zygote injections and would like to hear from others who plan on testing this in mice as well

It is tempting to try injecting the plasmid from addgene with 5'phospho-ssDNA oligo and HDR template though we did not have any luck with this approach using the pX330 plasmid containing 

Further, the amino acid sequence is not codon optimized for mice.

So......we plan on making a new version with optimal codons for the 887 aa protein with 5' utr and 3' polyA signal 

Does anyone have other ideas etc?

Thanks

Joe

On Monday, May 9, 2016 at 11:02:30 AM UTC-4, SKim wrote:

[dirk....@monash.edu]

Hi Joe

 I also tried injecting pX330 early on with no luck.  Are you planning to make protein from the NgAgo plasmid for your zygote injections?  if so any hints would be appreciated.

Many thanks

Dirk

[AT]

Hi all,

Just curious what's known about the 5'phosphorylated short ssDNA and if this is only rare in mammalian system.

Wonder if the DNA guided editing can be adapted for use in other organisms such as plant, fungus etc...

[Joe Miano]

Dirk,

We are codon optimizing for mouse and having the entire RNA (with 5' and 3' utr plus polyA for stability) generated for injection with the ssDNA plus HDR template (same strand as the ssDNA).

Looking forward to hearing from the forum on the use of NgAgo for mouse studies.

JOe

[Aidan]

Hi all,

I am trying out the NgAgo system in hard to transfect cells in which the CRISPR-Cas9 system never produced any edits. I'm hoping for a non-coding large deletion mediated by 2 cuts. As I am relatively inexperienced in transfection I wanted to seek some advice on the amounts of ssDNA to transfect etc.

I used electroporation (Nucleofector IIB) and pMaxGFP looks very good, >75% transfected. However, I wasn't sure how much 5' P-ssDNA to transfect. Since this nucleofector kit indicated 2 microg plasmid DNA, I tried various amounts and ratios:
     2 microg total, 1:1 ratio of plasmid to ssDNA --> 1 microg NgAgo plasmid + 1 microg 5' P-ssDNA (500 ng each 24 mer)
     2 microg total, 2:1 ratio of plasmid to ssDNA --> 1.333 migrog NgAgo plasmid + 0.667 migrog 5' P-ssDNA (0.333 microg each 24 mer)
     2.6 microg total, 1:1 ratio of plasmid to ssDNA --> 1.333 migrog NgAgo plasmid + 1.333 migrog 5' P-ssDNA (0.667 microg each 24 mer)

Attached is the gel where I expected 579 bp Wt band and ~172 bp deletion band.  As you can see its all Wt.
Lane 1 - 100 bp ladder
Lane 2, 3 - PCR +ve controls
Lane 4, 5, 7 - NgAgo transfected with above DNA
Lane 6 - NgAgo transfected with just 5' P-ssDNA
Lane 8 - PCR -ve control

I will sequence the wt bands I got to see if there are small indels at each 24mer site - it could be that NgAgo won't create large deletions via 2 cut sites? Also to double check that there are no SNPs overlapping the 24mers that could affect targeting/cutting.

Any advice on how much 5' P-ssDNA to transfect would be great,

Thanks,
Aidan

[Isaac]

Hi all,

Here's a NgAgo specific group to allow topical discussion. 

https://groups.google.com/forum/#!forum/ngago 

Mine is in the mail, can't wait.

Isaac

[1872n...@gmail.com]

Hi all,

Does anyone have some exciting result using NgAgo. I have transfected the 293 cells with NgAgo and phosphorylated ssDNA targeted two genes, however, there is no Indels at all. Even I did multiple transfections of ssDNA, no positive result. 

[Aidan]

Me again,

Tried a different region again in my cell type that I had edited successfully with CRISPR-Cas9 in HEK293 (note that these are not the cells I'm trying now). The guides are essentially the same except for 2 extra bp on either side of the sgRNA sequence. Tried a number of different combinations for nucleofection but no deletions. The first lane is PCR from edited HEK293 gDNA that I had from before, the rest of the lanes are various combinations of NgAgo plasmid with two 24mer 5'P guides.

Will do a side by side comparison with the sgRNAS in HEK293 cells but from the other posts around here I'm seeing it doesn't look like it is going to work.

So what is it? did they send the wrong plasmid? is there no nls even though addgene has one annotated on their plasmid map? I've also sequenced the whole Ago coding region in the plasmid and it matched the sequence from addgene. double digest diagnostic gave the appropriate sized bands.

so what's the deal, are we missing something in the methods?? has anyone had success with NgAgo? if so please let us know....

Aidan

[1872n...@gmail.com]

I do not think it is because of the plasmid or nls signals. I think it might be very difficult to assemble the NgAgo-ssODN complex in cells after transfection.

[SKim]

Thanks Aidan for sharing your results with us.

It seems that the NgAgo system is not cooperating well for now. It would be a great idea to share successful results whoever gets there first.
I believe we are still in a validation stage. Whether it is a reliable method compatible to CRISPR or not would be answered in the near future.

[Francisco J Sanchez Rivera]

Hmmm... This is definitely not looking promising at all...

I guess for reproducibility purposes, one would need to carry out the experiments pretty much identically to the way the authors performed them in the paper. That said, a robust system should be very straightforward to validate, so the fact that it hasn't worked properly for 2/2 testers here is a bit worrisome.

I remember seeing a previous post from someone saying that the system had been used efficiently in China?

It would be great if this system works.

F

[Zhiqi Sun]

Has anybody reproduce the NgAgo-ssODN assembly experiment?

在 2016年6月27日星期一 UTC+2下午10:26:33，1872n...@gmail.com写道：

[Debojyoti Chakraborty]

Hello everyone,

we have successfully used NgAgo in HeLa cells following a protocol where we used 700ng NgAgo plasmid and 500 ng gDNA (non purified, heat inactivated PNK). We confirmed knockout of GFP from a plasmid and could see the effect even after 96h. About 40% reduction in Western blot could be seen although there is scope for more optimization.

cheers
debo

[Chikai Zhou]

Hi Debo

do you have some sequencing results? are you sure it is not RNAi?

thanks.

[Debojyoti Chakraborty]

we are currently sequencing it.. will be able to let you know once the results come

--
You received this message because you are subscribed to a topic in the Google Groups "Genome Engineering using CRISPR/Cas Systems" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/crispr/V2B0YTlZlFo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to crispr+un...@googlegroups.com.
To post to this group, send email to cri...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Dr. Debojyoti Chakraborty
Room 129, CSIR-Institute of Genomics and Integrative Biology
Mathura Road,
New Delhi - 110020
+919433790132

[crisp...@gmail.com]

Dear Debo,

Very exciting news.

However, I'm a little worrying about the PNK treating. How do you exclude the effects of RNAi activated by non-phosphorylated ssODN? Do you have the controls with the ssOND not treated by PNK?

Best Regards

On Wednesday, June 29, 2016 at 7:17:02 AM UTC+2, Debojyoti Chakraborty wrote:

[Debojyoti Chakraborty]

Hi,

No way to know yet actually since we have done this once just to check if there is a reduction in gfp which we observed both by facs and western. We are currently sequencing the products to rule out Rnai if any. Will let you know if that's confirmed

Best
Debo

Dr. Debojyoti Chakraborty

Room 129, CSIR-Institute of Genomics and Integrative Biology

Mathura Road, 

New Delhi - 110025

www.rnabiologylab.com

   

--

[ilasadar]

That's very good news Debo.

Need some clarification tho. Did you use the Addgene plasmid or you design the construct w/ gblocks. Did you use the same protocol as it was mentioned on the paper? Can you share some pcr/blot pictures? Thank you for sharing. 

[kyle...@gmail.com]

Hi, Debo,

Good to hear that, the author, Dr. Chunyu Han, also suggested to do the GFP targeting first, but RNAi may be an issue that involved in. I am looking forward to hearing from you. Thanks!

Kyle

在 2016年6月29日星期三 UTC-7上午1:38:35，Debojyoti Chakraborty写道：

[Jonathan Geisinger]

Looking at the posts here and (much more importantly) at the second and third sections of the results in the paper, I am unsure as to why one would want to pursue using this enzyme over any of the Cas9/Cas analogs that already exist.  

If NgAgo has an extremely limited window for loading the ssDNA into itself (in HEK239T cells of all things!), it's probably going to be very hard to get any significant editing.  This, to me, outweighs any targeting benefit claimed to be associated with NgAgo.

Also, the reduction in GFP expression may just be the complex sitting there, similar to the whole CRISPRi silencing idea.  

[Debojyoti Chakraborty]

Hi Ilasadar,

We did not use the Addgene plasmid but a plasmid sold by miaolingbio (miaolingbio.com). It's called NLS-NgAgo-pCDNA3.1. It has a CMV promoter which can be shutdown in certain types of cells (like ES cells) so putting in a CAGGS might help in expressing the protein more.

Our protocol used lipofectamine 3000, 500ng gDNA and 700 ng plasmid. The gDNA was phosphorylated with PNK but was not purified. We have done it once to check for GFP knockdown (co-transfected GFP plasmid) so we need to verify with more controls/sequencing which we are currently doing.

hope that helps

debo

On Wed, Jun 29, 2016 at 5:43 PM, ilasadar <o.ku...@gmail.com> wrote:

That's very good news Debo.

Need some clarification tho. Did you use the Addgene plasmid or you design the construct w/ gblocks. Did you use the same protocol as it was mentioned on the paper? Can you share some pcr/blot pictures? Thank you for sharing. 

--

You received this message because you are subscribed to a topic in the Google Groups "Genome Engineering using CRISPR/Cas Systems" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/crispr/V2B0YTlZlFo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to crispr+un...@googlegroups.com.
To post to this group, send email to cri...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Dr. Debojyoti Chakraborty
Room 129, CSIR-Institute of Genomics and Integrative Biology
Mathura Road,

New Delhi - 110020
+919433790132

[S Lin]

Plasmid GFP expression reduction by NgAgo co transfection is reproducible but this cannot be confirmed by sequencing. Many factors affect GFP expression. So far many labs have failed to repeat the genomic DNA targeting as shown by the paper，which successfully targeted over 50 loci with 100% success rate and 20% more efficiency. If anyone has obtained NgAho induced genomic DNA indels confirmed by sequencing please share your data and tricks.

[GK]

No indels in zebrafish!!! Any body got it to work in Zebrafish? Curious to Know!!!

[1872n...@gmail.com]

I do not think we can repeat the experiment just following the published paper. Either there is some tips that are not described in the paper or the result are false positive in the paper.

[Jonathan Geisinger]

If those either of those are true, NgAgo is a non-starter for deletion experiments.  It may still be useful as a blocker of transcription, but that raises a a different question of efficiency versus dead-Cas9.

[James Gagnon]

Thanks for the updates everyone, interesting to hear about people's experience. Here is an update on my implementation of NgAgo in zebrafish embryos. TL;DR - it doesn't work for me (yet).

Longer story below,

Jamie

------------------

I cloned NgAgo into a vector for in vitro transcription of mRNA (I can share this vector if anyone wants it). I targeted 5 genomic regions within a pigmentation gene (tyr) that are well targeted by Cas9 to give a strong phenotype. I used IDT 5'phos oligos, and co-injected with NgAgo mRNA into 1-cell. 

I saw no pigmentation loss at 48 hrs (with Cas9, I see near complete pigmentation loss). I did MiSeq amplicon sequencing of the targeted sites and found no indels.

So looks like it's not working for me. However, I haven't completely given up yet for several reasons:

1. Potential issues with 5'phos oligos vs. PNK treated oligos (although i'm skeptical this is an issue)

2. NgAgo loading issues. Zebrafish grow at 28C, all experiments in paper were done at 37C or 55C. It's possible NgAgo won't load gDNA at 28C. 

If I have more time to commit to NgAgo (and that's a big if) I'd like to express and purify NgAgo protein, load it with gDNAs in vitro, and inject the complex. This would avoid issues with oligo degradation and loading.

--
You received this message because you are subscribed to the Google Groups "Genome Engineering using CRISPR/Cas Systems" group.
To unsubscribe from this group and stop receiving emails from it, send an email to crispr+un...@googlegroups.com.

[GK]

Hi Jamie,
  I also tried to grow zebrafish embryos at 37c after 1hpf, desperate to see any indels...but i found nothing.  About 50% of the embryos managed to grow with some abnormalities at 37c but the severely abnormal or the normal ones did not had any indels!!! In total i had 8 targets (the pigmentation and early essential genes), but found nothing.

[Haijun Tong]

Hi debo:

    I guess your GFP kncokdown, is probably due to the low transfection efficiency when you cotransfect with gDNA. Because when you complex the lipo with DNA, the gDNA is huge amount compared with your plasmid, so they will compete with your GPF package. So you transfect less GFP, that is why by facs and western blot, you see a decrease of the signal. So far, in my hand, it not work at all. 

Haijun

[Debojyoti Chakraborty]

Hi Haijun,

Our transfection rates are pretty high, reaching 70-80% and NgAgo reducing plasmid expressed gfp is pretty reproducible. We have controls including non Phosphorylated gDNAs which do not give a reduction in GFP so what we see cannot be explained by transfection related issues. We are doing more tests to look at all possibilities and would be able to comment once we are done.

Best
Debo

[Xiaobing Qing]

Dear Debo,

Very good news. I'm curious about how you set up the controls.

Did you have a control group with 5'-P-ssDNA but without NgAgo? At least from what you said I couldn't see whether the 5'-P-ssDNA will affect the GFP expression by its own. If I remember it correctly, phosphorylated oligos are toxic to cells.

Waiting for your sequencing results.

Best

Xiaobing

在 2016年7月12日星期二 UTC+2下午5:55:50，Debojyoti Chakraborty写道：

[Debojyoti Chakraborty]

Dear Xiaobing,

I would guess that too. So far controls have been scrambled and non Phosphorylated gDNAs. Haven't finished looking for indels or sequencing yet, guess that would be the most telling experiment..

[Joe Miano]

All in vitro studies we have conducted using either in house PNK mediated phosph of ssDNA or 5' phosph ssDNA from IDT against a target we have published on with CRISPR are negative.  We are awaiting first litters of mice and a modified NgAgo we are building that is codon optimized etc.  This is not looking good and I am hearing more of the same from colleagues using other animal model systems.

There obviously is a disconnect with what is in the paper and what we are all doing.  I would like to hear directly from those who have had clear success (especially in vivo).  Thanks,  Joe

On Monday, May 9, 2016 at 11:02:30 AM UTC-4, SKim wrote:

[Jingjin He]

Hi Debo,

Was the GFP plasmid cotransfected with NgAgo plasmid and gDNA? Or the HELA cell is an endogenous GFP reported cell line?

If GFP plasmid was cotransfected with NgAgo/gDNA. Just curious, is there a phosphorylated gDNA control which is not targeting GFP gene? There is a possibility that phosphorylated gDNA might affect transfection efficiency. I am wondering if a non-GFP targeted control will still decrease the GFP expression.

Best,

Jingjin

在 2016年7月12日星期二 UTC-7上午8:55:50，Debojyoti Chakraborty写道：

[Chris Koceja]

There are some strange discrepancies between the NgAgo coding sequence provided by http://www.miaolingbio.com/ of pCDNA3.1-NLS-NgAgo and Addgene's nls-NgAgo-GK. Maybe someone else can take a look...

http://www.miaolingbio.com/cn/product/plasmid/1826.html

https://www.addgene.org/78253/


Chris

[Aidan]

Well hopefully Chris you have figured it out. Here is a screenshot of the alignment I did comparing the addgene deposited sequence with the MiaoLingBio sequence. on the top is the addgene sequence which the ATG of the Ago coding region should start at 634. as you can see they don't match remotely, its not just a matter of some SNPs. so maybe this is the reason why everyone is struggling to see it work.

but it still begs the question, if addgene did verification and provide the sequence they must have had a sequence to compare it to provided by the authors. and one would assume that they wouldn't have observed a discrepancy and proceeded to ship it out as if it was fine. therefore they must have been giving the matching addgene plasmid sequence to compare it to.

strange indeed. maybe they wanted MiaoLingBio to have a head start on other companies trying to make commercial NgAgo products??

Aidan

[Debojyoti Chakraborty]

Hi guys, did you consider that this could be codon optimization for mammalian systems?

best

debo

--

You received this message because you are subscribed to a topic in the Google Groups "Genome Engineering using CRISPR/Cas Systems" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/crispr/V2B0YTlZlFo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to crispr+un...@googlegroups.com.
To post to this group, send email to cri...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Dr. Debojyoti Chakraborty
Room 129, CSIR-Institute of Genomics and Integrative Biology
Mathura Road,
New Delhi - 110020

www.rnabiologylab.com

[Aidan]

Yes, but which one is optimized for mammals? as everyone who got it from addgene has issues and you reported success and got it from miaolingbio it seems the one sent to addgene is not optimized for mammals.

so again I ask why no documentation supporting that? why not say for the addgene plasmid this is not codon optimized for mammals. but why even deposit that when most people interested are working with mammals, and they already have worked with human cells from the paper so they presumably could have sent addgene the mammalian optimized plasmid.

[Debojyoti Chakraborty]

Hi Aidan,

I agree that if it is codon optimized it should have been mentioned. I don't find such a thing on their webpage.

We do manage to reproducibly knockdown GFP from plasmid but so far haven't seen anything at the DNA level (indels in T7 assay). So we are still clueless about what could be going on there.

debo

On Fri, Jul 15, 2016 at 12:04 AM, Aidan <aidan.e...@gmail.com> wrote:

Yes, but which one is optimized for mammals? as everyone who got it from addgene has issues and you reported success and got it from miaolingbio it seems the one sent to addgene is not optimized for mammals.

so again I ask why no documentation supporting that? why not say for the addgene plasmid this is not codon optimized for mammals. but why even deposit that when most people interested are working with mammals, and they already have worked with human cells from the paper so they presumably could have sent addgene the mammalian optimized plasmid.

--

You received this message because you are subscribed to a topic in the Google Groups "Genome Engineering using CRISPR/Cas Systems" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/crispr/V2B0YTlZlFo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to crispr+un...@googlegroups.com.
To post to this group, send email to cri...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

-

[zhu luchang]

Hello Aidan,

This is what I found: If you take the addgene sequence and blast against the "Expression vector NLS-NgAgo-pcDNA3.1" sequence in NCBI, you may find that the NgAgo gene sequences of these two are identical. 

I attached a screenshot of the result.

Thanks,

Luchang

在 2016年7月14日星期四 UTC-5下午1:25:10，Aidan写道：

[Xia]

Hi, Luchang and Debo and all others, 

I have also done some sequence alignment.

Protein sequence alignment shows NLS-NgAgo-pcDNA3.1 (commercial)  and nls-NgAgo-GK (addgene)

is almost identical to the NgAgo in GeneBank (ANC90309.1), except the following differences:

1. Commercial one is codon opertimized for mammalian expression.
2. Commercial one has a SV40 NLS at the N-terminal and a nucleoplasmim NLS at the C-terminal while the addgene one only has N-terminal SV40 NLS. 
3. Addgene one has an additional short peptide tail “RDPYVSK” which does not belong to NgAgo or EGFP-N1 backbone. 

I don’t think codon optimization is that important. Although the commercial plasmid might express better than addgene one, still no Indel has been reported by Debo’s lab. 

So why do we see a GFP knockdown?  One possibility is that the NgAgo in cells might inhibitor transcription through binding to the target gene but it doesn't make any DSB.

It is quite disappointing, but I do believe that NgAgo system can be further improved. We certainly need more information from the author. 

Best,

Xia

[Guangshuai Jia]

Please have a look at these pictures from Han's paper. 

[Genome-editing]

hello guys, 

we tried NgAgo plasmid cotransfection with 5'phospho-ssDNA guides and I must say cell mortality rates are very high (toxic). We got 5'phospo-ssDNA guides synthesised from IDT, desalt purification. Does anybody have tested desalted verus PAGE purified ssOligos and looked for cell viability.

cheers!

 

[Genome-editing]

[Michael Wilson]

Hello, 

We have twice attempted to test the NgAgO addgene plasmid in HEK 293 cells using lipofectamine transfection and the EGXXFP split reporter assay. As many are aware, this assay requires DNA cleavage in the EGXXFP plasmid to restore GFP expression. 

We ordered oligos with 5' phosphorylation and transfected directly or PNK-treated an aliquot prior to transfection, just to ensure 5' phosphorylation; neither of these methods restored GFP expression. In contrast, the positive controls (Cas9 and gRNA expressed from pX330) gave robust GFP expression.

I will echo others here that there is a discrepancy between what the authors have shown and what others have been able to repeat. I don't believe anyone has reported indels yet, correct?  

[Aaron Zhang]

Gaetan Burgio claims in his twitter that they got efficient editing in mice, although it is not confirmed by sequencing (yet!). 

link to his post

https://twitter.com/GaetanBurgio/status/753942181442879488

On Friday, July 15, 2016 at 8:57:32 AM UTC-5, Aaron Zhang wrote:

The band shift, distortion, and intensity raise some concern. But the images are probably heavily compressed during publication and quite poor in quality. High quality raw files are needed to address the concerns. 

[mike]

These are serious allegations. 

在 2016年7月15日星期五 UTC+8下午8:27:45，Jia Xiaofang写道：

[mike]

I took a closer look to your questions. The intensity problems can be easily explained if pre-dyed gels were used. Smaller bands could have dragged some of the dyes with them, creating a miss match in molar weights and intensities. (However I didn't know that you could pre-dye a polyacrylamide gel in any way.)

And yes I agree with Aaron that raw file should be made available to address these issues you found.

在 2016年7月15日星期五 UTC+8下午8:27:45，Jia Xiaofang写道：

Please have a look at these pictures from Han's paper. 

[Genome-editing]

hello guys,

A small survey to assess the usage and success of NgAgo system. Once you complete the survey, you can see a summary of the results :-)

Please do take part.

http://goo.gl/forms/MLuSv0mUDgxDvJRB2

[Yan Cui]

Today I re-visited the two papers about DNA-guided Ago from Dr John van der Oost's Lab.

(http://www.nature.com/nature/journal/v507/n7491/full/nature12971.html)

(http://nar.oxfordjournals.org/content/early/2015/04/28/nar.gkv415.full)

In their papers, it was clearly demonstrated that  Ago  need gDNA for both strands to linearize plasmid DNA. If only one gDNA is present, Ago will only nick the target strand, which pairs with the gDNA, but will not cut the other strand.

Will this be the explanation that NgAgo failed to cleave DNA because only gDNA for one strand was supplied?

People who are interested to validate the NgAgo's ability to make double strand break should consider to supply gDNA for both strands. 

[Xiang Li]

I tried paired gDNAs, still didn't detect any indel.

[Xiang Li]

Hi Xiaofang,

I also have some questions about the sizes in Fig 4. For others, the NBT editor should already checked whether the Figs came from combination of different gels.

[SL]

We also tried paired oligos. No luck.

[SL]

edit the survey to include both company and lab P oligos. Some of us used both.

[Xiang Li]

Oligo phosphorylation percentage from IDT usually more than 80%

[Yan Cui]

paired gDNAs are prone to form a double-stranded complex. This may dramatically reduce the amount of single-stranded gDNA.

I would consider to shift the paired gDNA position to make them non-pairing, like the case of CRISPR-Cas9 nickase (http://www.nature.com/nmeth/journal/v11/n4/full/nmeth.2857.html).

If CRISPR-Cas9 nickase  can generate DSB (with two adjacent sgRNAs targeting the two DNA strands respectively), in principle NgAgo should also be able to generate DSB with two gDNA.

Hope somebody consider to test this strategy.

[Xiang Li]

Actually we tested paired gDNAs like that.

[Yan Cui]

Then this NgAgo system is really suspicious.
look at the Fig.1b in the paper, NgAgo needs FW guide and RV guide to cleave plasmid in vitro. In Fig. 1C, there is no statement whether both FW guide and RV guide were supplied.
In Supplementary Figure 1, it was showed that if NgAgo and one guide were co-transfected into 293T cells, and subsequently purified from 293T cells, then NgAgo can cleave plasmid in vitro.
I cannot understand this at all. If this is true, does it mean, if one guide is transfected, the cell will synthesise the complementary guide and also will be loaded into NgAgo?

[Andrea Rossi]

We did give a try but did not have any luck.
We have also a question for Professor Han:

Concerning your "one-guide-faithful" claim, what indications do you have that a NgAgo protein doesn't swap guides in vivo? Based on your experimental setup (fig2 and SupFig1), I understand that when NgAgo is incubated at 37C it doesn't bind the guide (using lack of activity as a proxy). So, why the incubation (at 37C) of NC pre-loaded NgAgo, extracted from cells, with the FW guide tells you that there is no guide swapping? I would like to see if swapping happens at 55C. Even though this is not a "physiological" temperature and this incubation would reduce the activity of NgAgo (to nickase at best), this experiment would reveal if there is an intrinsic ability of guide swapping.

[Joe Miano]

"ssDNA" is the first part of the title of the paper!  I hardly think anyone interpreted it any differently. 

Anyone going to CRISPR meeting at CSHL 17-20 August?  I hope to hear some updates on NgAgo there.

[Petr Tomek]

Hi everyone, this all sounds very dubious. Has anyone noticed how long did it take for this paper from first submission to the publication? Nearly 1 year!!! Heck, you can redo the whole project in 1-year. I wonder what was going wrong. Only Nature editors and reviewers would know... 

Perhaps authors were having some issues but what editor would decline to publish a work like that. Nature publications are very frequently full of holes. We have to be extra careful these days, since these high profile journals take anything that sounds cool but the science may not be bulletproof. There are always some issues reproducing high profile papers. That's the way it is.  Any thoughts on this??

Cheers, Petr

Dne neděle 17. července 2016 9:42:35 UTC+12 Joe Miano napsal(a):

[NoMore Drama]

Well the corresponding author Han Chunyu explained this to Chinese press on mutiple occasions: This work was first submitted to Science but eventually turned down after half a year. After that collaborator Xiao Shen from Zhejiang University joined in (late 2014) and made suggestions on things, the work was latter submitted to NBT(June 2015). Then one of the NBT reviewers made a fuss and replicated in his/her own lab the results in Fig3 (GFP reduction in cells) before giving the green light. The paper was accepted nine months after submitting.

[Yolanda Cheng]

I have never heard of such a case that reviewers would replicate the work he/she is reviewing. So weird!

[NoMore Drama]

Sounds weird to me too. This part of the story was told by Han multiple times as a prove of the reproducibility of his work. You can check out some of the materials I've posted in this threads here.  https://groups.google.com/forum/#!topic/crispr/It8gtql5X8s

There were mentioning of this in both the online posts and phone recordings.

[Petr Tomek]

Hahaaa, that's ridiculous! An utter waste of time if this successful replication of this particular aspect of the paper was not reported to the community. It would be a very useful information for many readers. I wonder who could waste public grant money and not publishing the results openly. Weird!

Thanks for the links NoMoreDrama, I will check those out.

--
You received this message because you are subscribed to a topic in the Google Groups "Genome Engineering using CRISPR/Cas Systems" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/crispr/V2B0YTlZlFo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to crispr+un...@googlegroups.com.
To post to this group, send email to cri...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

RNDr. Petr Tomek, PhD
Auckland Cancer Society Research Centre
University of Auckland, Faculty of Medical and Health Sciences
85 Park Road, Auckland 1023
New Zealand

email: sara...@gmail.com
phone: 022 100 6994

www.TrioPohadka.com

www.facebook.com/TrioPohadka

[Jonathan Geisinger]

More than likely (assuming all good actors acting in good faith) it was rejected from Science for not being good enough.  Most of their assays report efficiencies below what is usually obtained with HEK293T.  Similarly, their CRISPR/Cas9 efficiencies are low for the cell type.