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
* * * * * 1 votes

MSP_Primer dimers?


Best Answer elenagardini, 08 June 2015 - 01:50 AM

hi again. Great, I just found the answer: 

https://eu.idtdna.co...not-a-diagnosis

 

Thank you very much again, Nick!!!!! I still have to optimize, but it has been really helpful discussing with you!!!!

 

Elena 

Go to the full post


  • Please log in to reply
17 replies to this topic

#1 elenagardini

elenagardini

    member

  • Active Members
  • Pip
  • 13 posts
1
Neutral

Posted 08 April 2015 - 01:12 AM

Dear All, 

 

I'm trying to establish a method to detect methylation in the NR3C1 gene but I detect an additional peak in the melt profile. I thought about primer dimers, but I'm not sure that is it the problem. Indeed, the "high" of the additional peak is proportional to the real peak (which depend on how much methylated the sequence is).

I attach the file with the melt curves. 

 

The difference plot also doesn't look very nice (attached). 

 

One of the primer is ACCTAATCTCTCTAAAACGACGTTAA. I think it could be prone to hairpins or self hybridization.? Could I slightly modify it? 

 

Thank you very much for any possible advise! 

 

Elena 

Attached Thumbnails

  • Difference Plot.jpg
  • Derivative Melt Curves.jpg


#2 5280

5280

    Veteran

  • Active Members
  • PipPipPipPipPipPipPipPipPipPip
  • 145 posts
14
Good

Posted 11 April 2015 - 02:55 PM

You should run the samples out on an agarose gel. Have you optimized your annealing temperatures or MgCl? Also, I am assuming your amplicons are larger than 100bp so you should be able to differentiate them from a primer dimer. If you are still unsure about the status of this amplicon and it is a single band on on a gel then you can gel purify and sequence it.

 

Secondly, have you fully characterized DNA methylation at the NR3C1 promoter to determine which CpG are differentially methylated? Or has someone characterized it in the literature? If you have not, MSP is not very informative. Furthermore, I would perform bisulfite clonal sequencing or pyrosequencing in conjunction with RT-qPCR for mRNA expression. 



#3 elenagardini

elenagardini

    member

  • Active Members
  • Pip
  • 13 posts
1
Neutral

Posted 30 April 2015 - 04:12 AM

Thank you very much for you helpful answer. I just saw your answer now. 

Yes, the first step is to run an agarose gel. Then I could purify it from the gel and run the HRM after it (and compare the methylation level with the one previously obtained). Something I didn't say before: I don't detect any peak in my blank control, which is strange since I should detect it in case of primer dimers.

 

The temperature doesn't make any difference, I always detect this second peak.

I didn't try to change the MgCl concentration because I'm using the Melt Doc mix (applied biosystem) that already contain MgCl. But I could try to add MgCl (?). 

 

My amplicon contains 8 CpG that have been described in the literature as differentially methylated and 8 for which I have no informations. 

 

Thanks again,

 

Best regards, 

 

Elena 



#4 methylnick

methylnick

    Veteran

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 266 posts
6
Neutral

Posted 10 May 2015 - 02:02 PM

The messy melt profiles suggest you have inefficient bisulfite conversion and primers that are not directed to fully converted DNA strands. You are amplifying a mixture of incompletely converted templates and therefore you have that messy melt profile.

 

Best way to resolve this is to revisit your primer design, we know bisulfite is not 100% efficient. But designing primers that specifically amplify fully converted DNA in your sample will help with the melt profile.

 

Nick


All comments and communication are my own personal ones, and are not tied to any of my affiliations. 


#5 elenagardini

elenagardini

    member

  • Active Members
  • Pip
  • 13 posts
1
Neutral

Posted 13 May 2015 - 05:10 AM

Thank you Nick.

In the gel electrophoresis I see no primer dimers but a shadow of a second band at 300 pb (my sequence is 200 pb), so I will try to slightly modify the primers. I will have a CG more, maybe that will improve the specificity. 

 

Elena 



#6 methylnick

methylnick

    Veteran

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 266 posts
6
Neutral

Posted 13 May 2015 - 12:55 PM

Hi Elena, by Gel Electrophoresis, you would only see one band of expected size if it is high enough resolution, you may see a thicker band at the expected size because if you have enough base differences within your amplicon of interest, because of inefficient bisulfite conversion, this could be seen as 1-5bp difference. 

 

good luck

 

Nick


All comments and communication are my own personal ones, and are not tied to any of my affiliations. 


#7 elenagardini

elenagardini

    member

  • Active Members
  • Pip
  • 13 posts
1
Neutral

Posted 21 May 2015 - 03:45 AM

ok, I see, that's a problem and I will try to design new primers. Concerning my second peak in the melt curves profile, could be the data collection a problem? I settled the data collection at the beginning of the HRM (60°C), at the end (95°C) and also in between. Could this first peak be the result of the data collection settlement? 

 

Thanks, 

 

Elena 



#8 methylnick

methylnick

    Veteran

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 266 posts
6
Neutral

Posted 21 May 2015 - 04:57 PM

no it is not a data collection problem and more likely to be a conversion artefact problem given that the primers and/or bisulfite conversion are not efficient to give rise to multiple species of amplicons.

 

Cheers

 

Nick


All comments and communication are my own personal ones, and are not tied to any of my affiliations. 


#9 elenagardini

elenagardini

    member

  • Active Members
  • Pip
  • 13 posts
1
Neutral

Posted 01 June 2015 - 07:05 AM

Hallo, 

I see, is not a data collection problem. 

 

I modified the primers (they are longer). With one of them modified I have less the problem of the second peak but the melt profile is messy, and with the other one, I have a clean melt profile, but still the two peaks (pictures attached). It is not a problem of primer dimers, since I don't see any residus when running the gel. 

I don't really understand why a product of 300 pb (which I saw on the gel) melt before my product of 200 pb, and why this two peaks are always proportional? 

 

 

Otherwise the primer works, in the sense that it detect until 0.1% of methylation and it don't amplify the unmethylated sequences. 

 

Thank you very much for any feed-back or suggestion!!! 

 

Elena 

Attached Thumbnails

  • Derivative Melt Curves.jpg
  • Difference Plot.jpg


#10 methylnick

methylnick

    Veteran

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 266 posts
6
Neutral

Posted 01 June 2015 - 01:22 PM

Hi Elena,

 

great to hear your experiments are working more efficiently now, with regard to the melt profiles, this is indicative of differential DNA methylation. So the two peaks may represent fully methylated and fully unmethylated templates, the messiness would be varying methylation levels in between. 

 

Good luck!

 

Nick


All comments and communication are my own personal ones, and are not tied to any of my affiliations. 


#11 elenagardini

elenagardini

    member

  • Active Members
  • Pip
  • 13 posts
1
Neutral

Posted 01 June 2015 - 10:50 PM

Dear Nick,

 

thanks:) I don't know, I used 0, 0.1, 1, 10, 100% methylated standard. So, at least in the upper profile for example, the 100%, I should have only one peak I think, or? 

 

Elena 



#12 elenagardini

elenagardini

    member

  • Active Members
  • Pip
  • 13 posts
1
Neutral

Posted 01 June 2015 - 11:07 PM

..or what you mean is that my primers amplify another sequence and that the two peaks are proportional since I'm using standard between 0 and 100 % methylated and so, also the additional sequence I amplify has the same rate of methylation and, therefore, it shows the same kind of profile. 

But how is it possible that 300 pb melt before? Or is it possible that my peak is the first one and it could be that it is smaller because the sequence is shorter and so it has less CpG (in number)?

 

thanks a lot,

 

Elena



#13 methylnick

methylnick

    Veteran

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 266 posts
6
Neutral

Posted 02 June 2015 - 12:57 PM

well how sure are you your standards are what they say they are? also, could it be that your 300bp template is unmethylated? if so, it would have a low GC-content and therefore low melt. 

 

N


All comments and communication are my own personal ones, and are not tied to any of my affiliations. 


#14 elenagardini

elenagardini

    member

  • Active Members
  • Pip
  • 13 posts
1
Neutral

Posted 02 June 2015 - 01:40 PM

I'm completely sure about my standards, because I converted them from the commercial fully methylated and fully unmethylated human DNA. The  300 pb also correspond to the same methylation rates because is also converted from these commercial standards, so it is not possible that the 300 pb is unmethylated. Furthermore, my MSP primer don't amplify unmethylated DNA fragments.. I really don't get what happen with my analysis.. also because the primers are theoretically specifics. What I could do is purify my 200 pb band from the gel and to HRM on it, to see at least to which peak correspond my gene... 

 

Elena



#15 methylnick

methylnick

    Veteran

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 266 posts
6
Neutral

Posted 02 June 2015 - 03:41 PM

Hi Elena,

 

maybe if you feed your sequences into a melt simulator such as uMelt, then you can see, you can then feed in your bisulfite converted sequence and then your methylated and unmethylated sequences for good measure.

 

cheers

 

Nick

 

https://dna.utah.edu/


All comments and communication are my own personal ones, and are not tied to any of my affiliations. 





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

©1999-2013 Protocol Online, All rights reserved.