LM-PCR (ChIP-chip) questions - (Sep/28/2005 )
I am trying to ampify ChIPed DNA for microarray hybridization, using the LM-PCR method. This method involves blunt-ending, linker-ligation, and PCR with linker-sequence primer. I have just started it, and I already have a myriad of questions. If you can provide answers point-to-point, it would be greatly appreciated. I'm particularly anxious to hear from Methylnick, who helped me in my previous question, but anybody who has experience with this method, please share your thoughts!
FYI, I am using the protocol used in the Ren Lab at UCSD (Science 290:2306; Nature 436:876) which was recommended by Nimblegen microarray company.
1. Do you phosphorylate the linker?
OK, so the linker is made of two oligo nucleotides; longer one and shorter one. I phosphorylated the 5'-end of the shorter oligo (which would form the blunt end after annealed with the longer oligo), because I figured sonicated (ChIPed) DNA ends don't necessarily carry 5'-phosphate. However, some published protocols (including this Nimblegen one) don't mention the need of linker-phosphorylation.
2. How many cycles of PCR do you do? How much DNA do you get?
This Nimblegen protocol suggests 22+22=44 cycles, and I'm supposed to be able to get ~10ug from ~20ng of starting material. However, my current yield is substancially lower than that. I'm not entirely sure why they divide the whole cycles into two parts (first 22 cylces and second 22 cycles). Moreover, I don't get so much amplification from the second 22 cylces (only several fold increase from the first 22 cycles) which is also puzzling me.
3. When you run the amplified products in agarose gel, what exactly do you see? Do you include a negative control i.e. H2O as a starting template instead of ChIPed DNA? In this negative control, do you get absolutely no amplification of anything? (I'm asking this because there could be artifacts involving the linker DNAs.)
...there are more questions, but I think I will stop here for now.
our method is also based on Ren's paper.
1. we don't phosphorylate our linkers, they seem to ligate quite well without phosphorylating.
2. we also see that we don't get as much yield from the second round of PCR when compared to the first so we are now routinely using one round of PCR of 22 cycles to yield ~5ug of DNA
3. run on a gel, the LM-PCR should give a nice smear on the gel from 100bp up to 5 kbp. negative controls yield no amplification or primer dimer.
I am also having similar problems as you with LM-PCR... (low yields!). Methylnick answered your questions but I just want to confirm that you should not phosphorylate the linkers (otherwise they could ligate to each other).
You might want to take a look a this paper : Development and validation of a T7 based linear amplification for genomic DNA", BMC Genomics 2003 May 9, 4:19. It is an alternative method to LM-PCR for amplifying small quantities of DNA. There is also this website: TLAD. It am testing this method now so I can't say how it compares to LM-PCR but it seems promising!
Hey, does anyone know why they add 0.01ul (!) of PFU Turbo polymerase in addition to Taq in LM-PCR?
I tried the T7-based amplification once according to that paper that you mentioned. It did not work in my hand. Tell me how it works for you, please! Now I'm still using LM-PCR. It gave me quite good result.
I have been having problems with LM-PCR. My total yield is only about 1-2 ug after 24 cycles. Also I loose enrichment after LMPCR. Has anybody else had similar problems? Any suggestions are appreciated.
Need help, struggling with LMPCR since months, i amplify chip DNA, with input samples i get upto 5-6 ug DNA, while with IP sample i end up with 1-2 ug of DNA, starting amount is 20 ng in both the cases. Can anybody help me out where i might be going wrong???? And i also lose enrichment.............................
Has anybody done it successfully using phosphorylated linkers, or everybody just uses normal linkers.............
Need ur replies..................