Do you have any controls?

Hi Karls

1) As suggested above the control should be included to see if the things are working properly

2) Do analysis with more number of primers

3) As the MS-AFLP involves PCR amplification subsequent to the Restriction digestion..ensure complete digestion .....

bye

HpaII and MspI are isoschizomers recognizing the same restriction site CCGC, So no surprise that you got the same profile although one is methylation sensitive. Better use other combinations of enzymes if possible. Good luck!

Hi guys and thanks for the input so far, I will try to comment on your suggestions:

1: What do you guys mean with control in this case? I have tested the restriction enzymes separately and together and run the product on agarose gel, and I got a nice smear

2: I have run the digestion step for 5 hrs, most protocols use 1-3 h so I do believe the digestion to be complete (250 ng DNA was used)

3: I know that the MspI and HpaII recognize the same sequenece. They are, however, differently sensivite to DNA methylation in the recognition sequence and this is what should result in different number of fragments (which I don't get)

This combination of enzymes is not something I came up with myself, but is well established in the litterature

The only explanation that seems obvious, is that there is absolutely no DNA methylation going on. This seems highly improbable. The more methylation, the bigger the (potential) difference between the two restriction reactions and in turn, in the number of fragments generated.

How about parameters such as primer concentration, and number of selective nucleotides? I have had the suggestion to only have 1/6 the amount of  EcoRI primer compared to MspI/HpaII in the selective PCR, something that I will try next. Any comments on this?

Again, thanks alot for your input so far! All suggestions are good suggestions at this point!

/Karl

I don't clearly understand when you're doing PCR, but after PCR, there is no methylation on DNA. Methylation is only present on genomic DNA. You can also have sometimes digestion with methylation sensitive enzyme if you are using too much enzyme for the DNA quantity you use, or incubating too much. You should perhaps try to optimize the conditions of restriction for HpaII and MspI, varying quantities of methylated DNA, quantities of enzymes and incubation time, to define a window in which the differential sensitivity of both enzymes is observable. You can also try other methylation sensitive enzyme (Xho I ...) to check the methylation status of your DNA.

Cheers;

Thanks Nadine

The PCR is done in two successive steps after the digestion of the DNA, to simply reduce the complexity of the samples as there are thousands of fragments present.

I see your point, but how to observe the differential methylation sensitivity of both enzymes if not by subsequent PCR, like I do now?

The protocol I use is not an own invention, but is from the litterature (so is the amounts of enzyme, DNA etc)....

But brainstorming is good! Keep it coming! ;)

How many bands/peaks are you getting.  These bands/peaks are only showing a piece of DNA that is the same length in BP from your two different enzymes.  If you have a lot of bands it could be that for each length of band you are getting multiple different sites across the genome but as they are the same length they are producing the same band.  Does that make sense?  Perhaps you could try increasing the specificity of you PCR primers to reduce the number of bands you get?

Hi Hazel!

As for the number of bands, I would say it is around 250 peaks (depends a bit on where I set the cut-off)

I understand that one peak can be composed of many different bands of the same BP length, but it seems odd that I would end up with the same number of bands from both reactions. I should see significantly less bands in the EcoRI-HpaII reaction, as HpaII is methylation sensitive and can therefore not cut as often as MspI.

Yes, I have been thinking about adding more selective nucleotides in the primes. The question is if this would simply reduce the sample complexity (just give fewer bands) but also reveal a difference between the two reactions.... just have to try and see.

I am now working part-time, so I don't really have the possibility to go all wild and crazy in the lab, testing everything back and forth, so now I really try to get as much input as possible before doing another attempt at the method.

Thanks!!

/Karl

as a fast solution . run your product on polyacrylamide gel this will increase the resolution and bands differentiation . led to high informatics result  

As a controls I would:

1. Include DNA digests - only EcoRI, only HpaI, only MspI

2. Include PCR reactions - only EcoRI primer, only HpaI/MspI primer, not the pair

Thank you for these suggestions so far! I will try to sunmarize them and see what I will try next.

Strange thing is that I had some sucess with the method on Xenopus tropicalis DNA, so one thing I will try is running it on Xenopus, toad and human DNA to see what gives....

Based on the images in the attached pdf, it looks like the MS-AFLP pattern differences can be fairly subtle--differences are in band intensity not total number.  As a possible control, I would suggest trying a "no ligase" reaction followed by the usual PCR to verify that the products you are seeing are coming from fragments with ligated adapters.  If the adapter ligation is not working as you expect, the PCR products that you are detecting would not be dependent on HpaII/MspI cut site differences.

Hi David!

Thank you for your excellent input, I will have a look at the paper you attached.

I was under the impression, largely because our french collaborators are doing the method (successfully) on Zebrafish, and they are enumerating the fragments in the two parallel reactions as the measurement of methylation level.

I am performing the ligation @ room temperature (appr 20 C) overnight, which is one supported possibility reading from the T4 ligase spec sheet from New England Biolabs. But it is a good idea to try the "no ligase" control, I will do that next. I am currently running the assay again, this time with primer quantities identical to that of the "mother" protocol from Xu et al 2000; this also features a 2h digestion instead of the 5h I have been using so far.

Thanks!