Maternal and paternal strand - analyse both? (May/05/2006 )
Hi,
is it necessary to analyse both strands (maternal and paternal) for methylation status or is it sufficient to know which CpGs are methylated in one strand? Does hemimethylation occur in adult humans and if so, in which stages, on what scale and is it important?
Bram
Just a quick correction to the terminology relating to your question. Each chromosome contains two "strands" of DNA from the same parent. There will be one of each chromosome inherited from each parent.
When performing BSP the primers are designed to amplify one strand only (because bisulphite treatment makes the two strands non-complimentary). The PCR product will consist of amplified material from both the maternal and paternal chromosome. When sequencing the product it is usually not possible to tell which clone came from which parent unless a known informative polymorphism is present within the amplified region.
Perhaps your question is really asking is it necessary to amplify sequences from both strands? As far as I know there is no evidence for hemimethylation in adult humans and the forward and reverse strands show the the same methylation pattern for each chromosome.
There is plenty of evidence that maternal and paternal c'somes have different methylation patterns for specific genes however (these are known as imprinted genes). This forms the basis of MSPCR.
Thanks for the quick reply!
hi
best example of neccessity of methylaytion silencing, is regarding the sewual chromosomes.
A man holds one X only.
A woman holds 2 copies. So inactivation of one coy is necessary.
In contradiction, none all of the supplemental copies needs to be inactivated, as the turner syndrome shows. (sterile and also, in general, little wemen). This is an argument to the supposed theory in which, becaming a woman is the "pathway" contradicted only in man development.
Suppose I want to know the methylation state of an imprinted gene, just to know if there is a difference between the maternal and the paternal allele, will I see this on my sequence analysis? I mean, if one allele is methylated at some point and the other is not, I would see both a C and a T peak at that point in my sequence analysis, right?
When performing BSP the primers are designed to amplify one strand only (because bisulphite treatment makes the two strands non-complimentary). The PCR product will consist of amplified material from both the maternal and paternal chromosome.[i]
Are the primers for MSP designed to amplify one strand only because of the same reason as above? So It suggests that we can get hemimethylation PCR product if the chromosomes from paternal and maternal have different methylation status. Right?
Linksky
If you perform BSP on an imprinted region you will see a difference between the maternal and paternal strands following sequencing as you have described. Theoretically the ratio of methylated to unmethylated CpGs at any one site should be 1:1 and all/most of the methylated sites will be present in one sequence and the unmethylated sites in the other. eg.
O------O--O----O------O maternal strand (10 clones)
X-------X--X----X-------X paternal strand (10 clones)
Unless you have an informative polymorphism present within your sequence you cannot determine which is the maternal allele and which is the paternal one (if the imprinting status of the region is known then this is inferred).
If the methylation status of an imprinted gene is known then primers can be designed to preferentially amplify the methylated and unmethylated strands. This forms the basis of MS PCR investigation of imprinted genes. See this article
Are the primers for MSP designed to amplify one strand only because of the same reason as above? So It suggests that we can get hemimethylation PCR product if the chromosomes from paternal and maternal have different methylation status. Right?
Linksky
Yes MS PCR will amplify only one strand. The primers are designed to include CpGs so that the methylated sequence will include CG sites and the unmethylated sequence TG sites. In both cases the strands will be noncomplimentary following bisulphite treatment. If you apply MSPCR to an imprinted gene then one set of primers wil amplify sequence from the maternal c'some and the other set will amplify sequence from the paternal c'some (assuming the individual has a normal imprinting pattern - see article in my previous post).
The comment I made earlier in this string about hemimethylated DNA was referring to hemimethylation WITHIN the same DNA molecule. ie. one strand containing 5-mCG-3 and the other strand 5-GC-3. I don't believe there is any evidence for this in humans.
If you are taking about hemimethylation as it pertains to chromosomes (one chromosome being methylated for a region and the other chromosome being unmethylated for the same region) then yes, this occurs in humans for all imprinted genes and there is plenty of evidence for this. There are many well described diseases in humans where imprinting is defective.
I hope this helps.
Yes MS PCR will amplify only one strand. The primers are designed to include CpGs so that the methylated sequence will include CG sites and the unmethylated sequence TG sites. In both cases the strands will be noncomplimentary following bisulphite treatment. If you apply MSPCR to an imprinted gene then one set of primers wil amplify sequence from the maternal c'some and the other set will amplify sequence from the paternal c'some (assuming the individual has a normal imprinting pattern - see article in my previous post).
The comment I made earlier in this string about hemimethylated DNA was referring to hemimethylation WITHIN the same DNA molecule. ie. one strand containing 5-mCG-3 and the other strand 5-GC-3. I don't believe there is any evidence for this in humans.
If you are taking about hemimethylation as it pertains to chromosomes (one chromosome being methylated for a region and the other chromosome being unmethylated for the same region) then yes, this occurs in humans for all imprinted genes and there is plenty of evidence for this. There are many well described diseases in humans where imprinting is defective.
I hope this helps.
[/quote]
Thanks for your detailed explanation.
Then to all/most of the alleles, normally, only one of allele is expressed and the other is silent. Almost all of normal genes are imprinted genes.
If we find that the PCR products in MSP uing both M and U primers. it suggests this actually happens in normal situation. If we find the PCR product in MSP only M primer but not U primer group, it suggests that the gene has been methylated and is probably silent.
But If we find the PCR product only in U primer group, what does this mean? It means both the allele genes are expressed simultaneously?
hope the reply. Thanks.
Linksky
If we find that the PCR products in MSP uing both M and U primers. it suggests this actually happens in normal situation. If we find the PCR product in MSP only M primer but not U primer group, it suggests that the gene has been methylated and is probably silent.
But If we find the PCR product only in U primer group, what does this mean? It means both the allele genes are expressed simultaneously?
hope the reply. Thanks.
Linksky
Linsky
For imprinted genes, yes, one allele is expressed and the other is silent. However, you have said "almost all of normal genes are imprinted genes". This statement is not correct. Most genes are NOT imprinted genes and will have biallelic expression, not monoallelic expression. There have only been about 60 or so imprinted genes described in humans, usually organised into clusters or domains (eg. those at 11p15.5, 15q11.2).
If you are looking at the promoter region of a non-imprinted gene using MSPCR, then PCR product from only M primer indicates methylation (gene is probably silent). PCR product from only U primer indicates the promoter is not methylated (gene is likely to be expressed). If you get both M and U PCR product then you probably have a mixed cell population where some cells are expressing the gene and some are not.
Type "methylation AND epigenetics" into PubMed and you will get all the information you need (and more!).