Dear Osama Sweef,

it cannot be excluded that your treatment conditions have an effect on the expression of the reference gene. Therefore, you should start with a run that only includes different potential reference genes in order to find the one (or more) that shows the lowest differences in Ct values between control and treatments. Afterwards you can choose one (or more) suitable reference gene that was not affected due to your treatment. In addition to 18S, you could check GAPDH, HPRT1, GUSB, ACTB or TUBA1A, for instance.

Good luck. - CL

an other possibility is a difference of quality in treated samples, leading to degraded RNA and therefore a shift in the expression.

did you check for the RIN?

fred

Using the same amount of RNA is no guarantee that you started with the same amount of the housekeeping gene. If your treatment turns on (or off) expression of a LOT of genes then you would see a different percent of the RNA from the housekeeping gene compared to the total RNA.

That's why you normalize expression to the housekeeping gene and don't worry about the amount of starting RNA as long as you're in the proper range for your protocol.

Thanks for of you

I applied the concept of. Carsten Lange and it works very well

thanks a lot

Excellent