Dilute your cDNA ? - for real time PCR (Jul/10/2008 )

Sorry if this has been asked here already. I am in the process of trying to "perfect" my real time assay. As a first step I have prepared 10 fold dilution of cDNA and performed real time to get the amp efficiency (want to use PFAFFL method). The cDNA was prepared using 1ug of RNA in a 20ul RT reaction. I then pooled equal volumes of 10 individual cDNA samples to prepare a stock from which the dilutions were prepared. The dilutions are stock, 1:10, 1:100, 1:1000 and 1:100,000. A 80ul master mix containing 3ul of cDNA from each dilution was prepared and 25ul per well was amplified. Looking at the data from my efficiency reactions I get the following

Actin - Ct Range from 17 to 31 , slope = -3.4
Target gene 1 - Ct range from 22 to 34, slope = -3.1
Target gene 2 - Ct range from 19 to 33, slope = -3.3

I have run a subset of my samples (controls and various treatments) to see what the Ct values are.
Actin - Ct Range from 16 to 20
Target gene 1 - Ct range from 22 to 24
Target gene 2 - Ct range from 20 to 23

As youn see there is very little change in Ct values. This is a little concerning as I expect that I should see significant changes in gene expression of the target genes. From the preliminary data the biggest change I get is a ratio of 3.5. My question is, should I be diluting my cDNA (say 10 or 100 fold) prior to real time PCR? Are my Ct values for my samples too close to the high end of my amplification efficiency curves? Will diluting the cDNA give me a greater ability to detect changes in the Ct values?

Any suggestion would be greatly appreciated.

Hi,

I don't know if I got it right, you think you'll detect differences better if you dilute your samples, right? There is one problem, if you have a look at your replicates in your dilution series you should see, that the lower the amount of cDNA is the higher is the variance within your replicates. This is just logical, because if you pipette one molecule more or less just by chance has an greater effect upon small amounts of starting material than on big ones. I just heard talks of Michael Pfaffel and Jan Hellmans last week and one of them (don't remember who) said your Ct values should be around 15 to 25 and everything from 30 on you should not analyse due to effects like this. So I would suggest not to dilute because you will increase your effect of pipetting errors and might measure differences you really did not want to measure ;-)

good luck